Datasets:

mimiklee

/

masterthesis-longt5-sum-8k

like 0

Introduction In the UK, as in other industrialised nations, there is considerable government interest and concern over an ageing workforce and a strong policy thrust toward extending working lives. In the future it is hoped that more people will work until state pension age and beyond, and indeed in many countries state pension ages are rising (OECD, 2011). It is routinely asserted that greater opportunities to work flexibly will encourage older workers to continue in work and delay retirement (e.g. OECD, 2006: 98-101). In addition, survey evidence often indicates a demand amongst older workers (those aged 50 plus) to work flexibly. Much of this policy and practice discussion of the utility of increasing flexible work options (FWOs) for older workers is uncritical about the availability and quality of flexible work. In addition, it is typically gender blind with remarkably little scrutiny of gender differences, notwithstanding the high proportion of older women who already work flexibly in the sense of part-time work. This gap reflects two related shortfalls in theorisation: the 'taken-forgranted' associations between gender, age and flexible working -that it is primarily an issue for people, and particularly women, with school age or younger children; and a lack of understanding of gender and retirement. Drawing upon a qualitative study of over-50s in GB, this article interrogates the intersections of gender and age in flexible working to begin to fill these gaps. The article commences with a brief overview of the proposed attractions and utility of FWOs for older workers, i.e. those aged 50 and above, before drawing upon existing research into gender and flexible working and gender and retirement to hypothesise ways in which gender and age might interact to shape attitudes, opportunities and experiences of flexible working in later life. It then explores these assumptions through a qualitative study of 96 people aged 50-64, drawing upon their revealed discourses of flexible working. Finally, it discusses the implications of the findings for theorising relationships between work and life in later life and policy around extended working lives. --- Flexible working and the older worker For the purposes of this article, we define 'workplace flexibility as the ability of workers to make choices influencing when, where, and for how long they engage in work-related tasks' (Hill et al, 2008: 152). The main emphasis of policy, practitioner and academic interest in FWOs has centred round combining paid work with caring, most often familial responsibilities, typically carrying an implicit age and/or life-stage bias towards men and women with younger children. However, a body of work seeking to explore the utility of FWOs in sustaining older workers in employment for longer (Alden, 2012;Barnes et al, 2009;Gardiner et al, 2007;Hirsch, 2007:108-113;Loretto et al, 2005;Moen and Sweet, 2004;Yeandle, 2005) has broadened this conceptualisation. As outlined in the introduction, prompted by policy attention, it has been espoused that flexible working offers an attractive way for older workers to stay in work for longer than they could or would if they were working full-time, to extend their working lives beyond expected or normal retirement ages and/or to exercise choice in the nature and timing of their retirement. FWOs, especially part-time working, have been held up as a means of improving choices for older workers (Loretto et al, 2005;2007), for example by downshifting at the end of their working careers with their current employer and hence delaying full retirement, or by providing 'bridge jobs' between the career occupation and full retirement (Loretto et al, 2005;Vickerstaff, 2010). UK Government policy has also supported the need to create greater flexibility for people to 'choose a phased approach to retirement' (Department for Work and Pensions, 2006:139). Survey data show a strong expressed preference for working flexibly at the end of working life (Maitland, 2010;Smeaton et al, 2009); or for re-engaging in flexible work after retirement, mainly for financial reasons (Smeaton et al, 2009: 58-59). Beyond these attractions, FWOs offer opportunities to the third of the 50+ workforce with a long-term health condition (Barnes et al, 2009: 2) to remain in work, and to the long-term unemployed or inactive (often because of health reasons) in order to ease them back into the labour market (Humphrey et al, 2003;Hudson et al, 2009: 69). One in five people in the age group 50-59 are providing informal or unpaid care (Howard, 2004). A minority of those between 45-65 find themselves as the'sandwich or pivot generation', combining care for their own children with informal care for elderly parents (Evandrou and Glaser, 2004: 771;Mooney et al, 2002: 1). Research suggests that those who combine employment and caring would welcome a range of flexible working options. In particular they would appreciate: better access to flexible working hours; the opportunity to reduce working hours if necessary; the right to timeoff for caring responsibilities; and the ability to work from home where feasible (Howard, 2004;Mooney et al, 2002). Given the myriad of attractions of FWOs, it is not surprising that there is some evidence from the US that older workers may value access to FWOs more highly than younger workers (Pitt-Catsouphes and Matz-Costa, 2008). However, what is more remarkable is the lack of attention paid to the role of gender in the policy and practice debates. We contend that this gap has arisen because of two, inter-related, oversights: the limited focus on the relationships between gender and flexible working which has focused mainly on one life stage; and the prevailing tendency to think of end of working life and retirement as gender-neutral or following a typical male trajectory. We now explore each of these in turn. --- Gender, age and flexible working The predominant policy and research focus on flexible working has centred round parents, especially mothers, who are trying to combine paid work with childcare (see for example, Craig and Powell, 2011;Lyonette et al, 2011). In the UK, this was signalled in the genesis of legislative approaches to flexible working and work-life balance as 'family-friendly' policies. As such, Emslie and Hunt (2009) argue that most studies [of work-life overlap] 'either ignore gender or take it for granted' (p152). This assertion is reflected in the findings that women are more likely to take up FWOs to accommodate childcare and family responsibilities, whereas men tend to use FWOs to extend their working hours (Hofacker andKonig, 2013, Martinengo et al, 2010), thereby reinforcing the traditional male-breadwinner family model and gender roles. However, thinking about flexible working in later life when childcare responsibilities in theory are less dominating, allows us to go beyond gender roles associated with parenthood and to more fully interrogate the relationships between gender and flexible working. This opens up the opportunity to use life stage as 'a focus issue in explorations of gender differences' (Martinengo et al, 2010(Martinengo et al, : 1364) ) rather than simply controlling for age as a 'noise' variable. Only a few studies to date have examined the relationships between gender and flexible working in later life stages. Emslie and Hunt's (2009) qualitative examination of gender and work-life balance in mid-life (age 50-52) found very strong evidence of traditional gender identities among women, even where their roles had been non-traditional, for example women's identity remained family-bound even in cases where a women took up full-time working because her partner had been made redundant. Amongst the men, the study found some who changed their mentality over time, moving from a live to work orientation to a work to live focus. In another qualitative study, this time of Australian women in two age groups (53-58 and 65-70), Everingham et al (2007) discerned three patterns of engagement with flexible working in the transitions to retirement. The first was non-engagement, i.e. women who simply stopped work, adhering to the more traditional 'cliff-edge' retirement. The second was the 'transitional model' where the women used flexible working to gradually retire. Although the authors acknowledge that this model appears to adhere most closely to the government's model for using flexible working as a transition to retirement, for women as opposed to men, this was only one of many such transitions they had made over their working lives. A third model arose from a group of women who wanted to use flexible working to open up new working opportunities -to transform the nature of their work. Cebulla et al (2007) conducted secondary analysis of major panel data sets to examine the repercussions of working longer on the private lives of men and women beyond the UK's state pension age (SPA -60 for women and 65 for men at the time of the study). They set out to investigate the effects of flexible working, but their main finding was that working longer in any capacity (full-time or part-time) affected the gender divisions of household tasks, with traditional gender divisions being much less pronounced in non-working SPA couple households, but retained among those who stayed in work (whether or not that work was flexible). While these studies are welcome, their insights are inevitably partial. Emslie and Hunt's work, although extending the parameters of work-life balance and flexible working beyond parents of small children, still retained a fairly narrow focus on family life. The Australian study looked only at women; whereas Cebulla et al's only examined behaviour of men and women beyond SPA and did not provide illumination of the relationships between gender and flexible working per se. We also maintain that any opportunity for, or disposition towards, FWOs in later life builds on previous experience of FWOs and therefore we need to understand the relationship between gender, age and FWOs across the lifecourse. For many women FWOs have characterised their labour market activity across the lifecourse and do not suddenly emerge in later life as a possible response to pending retirement. --- Gender and retirement The traditional notion of retirement as 'an event experienced by men at the end of their careers' (Ho and Raymo, 2009: 153) has been disrupted by the move to extend working lives. In their comprehensive review of retirement, Wang and Shultz (2010) challenge the concept of retirement as a discrete event, by examining the main ways in which retirement is now conceptualised: as decision-making; as an adjustment process; as a career development stage; and as a human resource management concern. Indeed, retirement is increasingly seen as a reversible state rather than an end destination (see Maestas (2010) on 'unretirement'). However, in these discussions, as in research and policy attention more generally, gender is largely absent or loosely implied into notions of'societal norms' of family roles and/or marital status. In particular, understanding of women's retirement processes is limited because of 'a tendency to view women's retirement and retirement planning relative to normative understandings already established for men.' (Wong and Hardy, 2009: 77). A small body of work (most of it from the USA and quantitative in nature) has challenged the prevailing tendency, asserting that because of differences in work trajectories, financial security and family responsibilities (see Price, 2003), we need to consider men's and women's retirement separately. However, Wang and Shultz (2010) remind us that little empirical research exists to confirm the theorising, and that gender is often reduced to the status of a moderator variable in retirement transitions (see also Pleau, 2010). An additional element to the lack of attention to gender has been the tendency to view decisions around retirement as not only male or de-gendered, but also as individualised (see Lewis, 2007's concept of the 'adult worker'), whereas in reality, many decisions are embedded in domestic contexts which combine established family and gender roles with more temporally proximal factors (O'Rand et al, 1992: 83). Existing research has been instructive in showing that women's retirement is typically more influenced by their husband's retirement and by their family-roles and patterns of labour market participation across the life-course than is the case for men (e.g. Price andNesteruk, 2008, Pleau, 2010). Nevertheless, research to date 'disproportionately represents the experiences of married women' (Price, 2003: 351), and of professional women. It has also been focused mainly on the timing of retirement, with relatively little investigation into how the nature, meaning and processes of retirement may differ between men and women (Flynn, 2010;Loretto and Vickerstaff, 2013). In summary therefore, limited attention to women's retirement and to the ways in which gender and flexible working inter-relate in later life have left us with some key knowledge gaps surrounding the attitudes and behaviours of older men and women towards FWOs. We argue that a fuller understanding is necessary in order to provide a currently absent evidence-base (see Gielan, 2008) for the efficacy of FWOs in policies aimed at extending working lives. --- The qualitative study The project was commissioned by the UK government's Department for Work and Pensions (DWP) to investigate the factors encouraging labour market activity among people aged 50-64 (Vickerstaff et al, 2008). Ninety-six people in three contrasting British locations were interviewed. Edinburgh was chosen as representative of an area with relatively low unemployment and generally good standards of living; in contrast Thanet in Kent is an area of high unemployment and socio-economic disadvantage. Nottingham is a site of average unemployment and a younger and more ethnically-diverse population. Overall, as can be seen in Table 1, the study focus was on people/households with low-medium income levels i, with the spread of household income in each of the three locations reflecting the socio-economic characteristics of that region. Within each location, the sample was purposively selected to capture a range of personal (age, gender, marital status), health, financial, caring and employment statuses. The economic activity status categories in Table 1 are based on respondents' self-definitions. All but two in the retired group had retired 'early', i.e. before current SPAs. Three men in the retired group who described themselves as'semi-retired' were self-employed, but in practice had a very loose attachment to the labour market. Table 1 here Respondents were interviewed in couples or as individuals using a biographical time-line and a semi-structured schedule. The interviews typically lasted for about an hour and a half and were digitally recorded. The transcribed interviews were analysed using NVivo 7. First, each interview was analysed for a set of respondent descriptive attributes, i.e. age, gender, employment status, location, household income etc. Interviews with couples were coded twice, once for the female and once for the male respondent. Next, the interviews were coded for references or answers to direct questions that respondents were asked, for example direct experience of working flexibly. Third the interviews were coded interpretively for references to factors affecting continued working, retirement and flexible working. In the account which follows, quotations from the interviews are used as examples representative of views and recurring themes or discourses expressed by a number of different respondents (on drawing conclusions from qualitative data see Miles and Huberman, 1984: 215-221). The reference number at the end of quotations indicates the location of the interviewee (E= Edinburgh, N=Nottingham and T= Thanet) and whether they were interviewed as an individual or as part of a couple (S/C), their gender (M/F) and their economic activity status at the time of the interview. --- Results --- Overview Although the sample exhibited heterogeneity of behaviours and attitudes towards flexible working in later life, overall there was a relatively low incidence of flexible working across all groups (Table 1; those who were retired or not working were classified on the basis of their most recent job). This low incidence reflects national patterns. Even though there have been increases in employment rates of the over-50s and over 65s in recent years (ONS, 2013a; 2013b), analysis of official figures (see Loretto et al (2005) and updated personal analysis -details from authors on request) suggest that the majority of older men are working on full-time permanent contracts and that despite a much higher proportion of women work part-time (as is the case throughout the working life) the proportion of men and women over-50 in employment who work part-time has fallen in recent years, i.e. the increase in older worker employment has been in full-time, not part-time jobs. In addition, most over-50s are working in permanent positions. The extent of other forms of flexible employment (e.g. locational flexibility, flexible contracts) is relatively insignificant. Most men work full time and then'retire'; there is little evidence for gradual retirement. These findings for the British case are echoed in comparative review, which found that across a range of European and North American countries there were only limited "signs of flexibilisation of work patterns" for older workers, especially in comparison to changes in the youth labour markets (Hofacker, 2010:184-199). Returning to our sample, also mirroring national patterns, the most common form of flexibility was part-time working, with women being markedly more likely to have worked part-time across their lifecourse. Typically they had stopped work when their first child was born and then had returned to work, often different work, part time as the children got older. Many then further modified their hours in later life to undertake caring for elderly parents, an ailing partner or grandchildren. In order to explore more fully the relationships between age, gender and flexible working in later life, we have drawn on the main discourses presented by our respondents with regard to the role of flexible working throughout their lives and as they approached or experienced retirement. In some cases, men and women presented quite different discourses; in others we show how the same terminology had separate meanings for men and for women. Although separated for analytical purposes, it should be noted that these discourses do overlap. --- Freedom This was a dominant discourse amongst the interviewees, whether retired or still working -many respondents were strongly attracted to the freedom that release from work had given or was expected to give them. As such, they would not or had not considered flexible working in any guise. However, the way in which this freedom was articulated was highly gendered. For several women who were retired, this freedom manifested itself as escape from a low quality and unsatisfying job, most often involving 'flexible' working. The following quote is from a respondent who retired aged 56. Her work experience reflected that of many of the women in the sample, who had often give up skilled jobs to work flexibly when raising their families and who had ended up in a lifetime of underemployment. '...I couldn't get out of there quick enough, basically. I only worked 3 days which was fine. The hours suited me, the days suited me and had the job been different then it might have encouraged me to stay on longer, but it was just so, so boring and I just hated it' (EC50F_retired) Men, on the other hand, were more likely to comment that they had 'done their bit' and were now 'off the treadmill' (EC43M_retired) and had settled into their new phase of life and new routines. The majority had left school at 15 or 16 and had been working continuously since then, so by their mid-50s they had completed 40 years of working life. 'I thought well it [retirement] gives me an opportunity to finish in fulltime employment because basically you get to a stage where you've just had enough. You think well I've done the stint. I've done my time in the trenches and I think that's it.' (NS79M_retired). --- Time and balance The importance of time sovereignty (Atkinson and Hall, 2009;Fagan, 2001) was one of the strongest recurring theme across the interviews, as illustrated by the case of this Edinburgh couple, both in their mid-50s. They planned to retire at same time (when he reaches 63). They said that they would not consider flexible working as retirement means: '...time to do things, just time to do things, more time to yourself. Not having to get up at 6.30 or 7 in the morning. Go for more walks, time with the dogs.' (EC37M+F_working) This feeling was echoed by many respondents (working and retired, men and women), and it altered their attitudes towards flexible working, and to work-life balance, in later life. Where, previously, flexible working allowed them to accommodate life around working needs, they were now expressing change in their orientation to work; work took a back seat, and if done at all would need to fit around the rest of their lives. 'I want to be in the position where I want the job to suit me more than me suiting the job....You know, three days a week would be fine, so This redressing of the balance between life and work was felt strongly by several men who wanted to focus on their families as they had missed out on this earlier in their lives. A typical case was ES48M_working, aged 63, who wanted to retire from a career in the building industry to spend time with his grandchildren. He viewed this as compensation for missing out on his own children's early years. --- Choice and control In the main, this was a male discourse, articulated by men who had chosen to work flexibly as well as those who had rejected flexible working in their transition to retirement or as a way of extending their working lives. Only a minority of respondents had worked flexibly to accommodate health problems and these were all men. Most were in Edinburgh or Nottingham (i.e. the more affluent areas of our study), and had taken ill-health retirement from their principal job, e.g. police force, and were now working in quite different jobs. Flexible working for these people was characterised by a high degree of control and autonomy, achieved by being self-employed, e.g. as a taxi driver. A mere four men reported working flexibly in their current jobs to allow them to deal with health issues. A man in Nottingham who had cancer and was going back to work after chemotherapy commented on how good his employer had been: 'You don't think about it when you take a job but like over the last five years I've probably had I think 10 months off work on full pay and you never think about that when you go for a job. I mean a lot of jobs, after a month, they'll probably say you're on your own. It's been very important.' (NS72M_ working) Choice also manifested itself in relation to choosing not to work flexibly: for example, a retired man in Edinburgh who took voluntary severance when he was 60. He had considered other employment (e.g. taxi driver, delivery man) but 'didn't want to be at beck and call...if I was going to do something I wanted to be able to do it when I wanted to, when it suited me.' He was financially worse off, but subsequently his wife has been diagnosed with breast cancer and this reinforced his sense that he needed to manage his own time (EC32M_retired). Where women spoke of choice, this was almost exclusively in terms of rejecting flexible working as they wished to continue building their careers, to upgrade rather than downgrade. For some women who had put their careers Discourses of choice and control from both men and women were often predicated on perceptions and previous experiences of flexible work as being poor quality work. While retailers such as B&Q have raised the profile of employing older workers, it is quite clear that such publicity has also served to reinforce the idea that the jobs available are low-skilled, low-paid and with few prospects, a perception with empirical basis (see Lain, 2012;McGovern et al, 2004). This view, or'suspicious perspective' (Simpson et al, 2012) acted as a very real deterrent for people across all employment statuses in the study, and had especial resonance for women, many of whom had a history of parttime and fragmented working in jobs for which they were over-skilled. --- Fitting in This was an almost exclusively female discourse. Several women, especially those in less affluent situations, spoke of working flexibly throughout their lives in order to fit in paid work around childcare and other domestic commitments. For many, as in the case of TS9F_retired talking about work in her 30s, it was also clear that this 'fitting in' often involved trying to juggle several paid jobs 'I did a little bit of child minding at that time, you know, other than my own children....I worked as a school dinner lady. I did that for 10 years..... I did a little bit of seasonal cashier work, because everything fitted around the children at that stage. Interviewer: OK. So then we go on to your 40s unless there's anything else. --- Oh I... Sorry... Interviewer: Yeah? My only hesitation is because of my memory. I think I did a little bit of... a little bit of almost holiday fostering. I was a registered foster parent but I really only did like respite care for families, so only short term.' Later in life, even with the imperative of childcare removed, women still spoke about fitting in. Instead of working flexibly to fit in with their domestic roles, they opted to fit in by retiring completely from paid work, either to complement their husband's plans for retirement, or with their perceptions of what their employer expected of them. The desire to please their employer manifested itself in the finding that women were less aware of opportunities to work flexibly or retire gradually, less likely to want to ask in case it upset their employer, and more likely to retire in order not to be a nuisance. For example, TS9F_retired above who had fitted in so many jobs in her earlier life had retired partly because of job-related stress. However, when asked if she had considered working flexibly she replied, '...there's so much unemployment you've got to make space somewhere for young blood hadn't you?' A key implication of this point is that it problematizes the concept of gradual retirement. While previous research has observed that older women may leave labour market earlier than men because of lack of gradual retirement opportunities (e.g. Gielan, 2008), these findings show that for women already working part-time the scope for further reduction of hours may be limited. Reflecting national patterns, many of our female respondents (and some of the men) had replaced childcare responsibilities with obligations to care for elderly dependents or partners who were ill. While in theory flexible working offers a way to combine employment and caring needs, the reality showed a mismatch between employee and employer needs. The overarching issue here was that existing FWOs did little to accommodate the often unpredictable nature of caring needs as in the case of NS79M_unemployed/retired who had to give up working completely to care for his mother who had dementia. Her needs meant that he could not fit in any work, even on a part-time basis, around his caring responsibilities. --- Helping out Again, this was a predominantly female discourse and shows how the boundaries for women between paid and unpaid work become even more blurred in later working life and in retirement. Firstly, women who had combined flexible working with childcare earlier in their lives typically spoke in terms of their entire time in work in terms of helping out (e.g. NC62F_working who helped with sustaining family work-life balance by working night shifts to complement her husband's working schedule, and who also helped through voluntary activity most often around the children's social activities). As with the discourse of fitting in, this discourse was extended into their approach to working later and/or into retirement. A woman in Nottingham was working for 8 hours a week as a personal carer for a doctor she had worked for when he was still in practice: 'I sort of help look after an old retired doctor and the job, now I've been with him for 15 years and so now I tend to go round about 10, 10.30 and come home around 2.30. Whichever days, but mainly, like I'm going again tomorrow, very often it's Monday and Thursday, but it can change depending on... I mean at the moment it's changing a bit because he's ill.' (NS64F_working). A variant of this was those women who were working flexibly (or had given up work altogether) so as their daughters could work. These women 'helped out' by caring for grandchildren, as in the case of ES40F_working, who reduced her job at a GP practice to 8 hours per week in order to care for her grandson. Their flexibility often ensured that their daughters could work full-time. As one, a widow, said: 'I didn't want to leave the girls [her daughter and granddaughters] in the lurch.' (NS71F_retired) Secondly, many women in our sample, who in addition to caring for dependents, 'helped out' by caring for neighbours and friends, all on an unpaid basis. For these women, flexibility meant something altogether wider than standard labour market definitions. One woman, who described herself as retired, was looking after a couple of elderly neighbours, doing their shopping, cleaning and personal care: 'it's unpaid as we've been friends for a long time' (TS21F_retired ). Secondly, for many women, flexibility was achieved outside the formal, paid work arena. Thirdly, our findings also showed that some older women may eschew flexible work in later life in order to take advantage of the opportunities offered by a 'proper' job, while some older men may wish to work flexibly (or not to work at all) in order to spend time with their families that they missed out on earlier in their lives. In line with Emslie and Hunt (2009;167), such attention highlights 'the way gender identities are continually reassessed and reconfigured', and contribute to the current calls for adopting a 'lifecourse' perspective to the understanding and managing work and careers (see e.g. Gardiner and Tomlinson 2009). It is important to consider to what extent these gendered discourses of flexible work are limited to the current cohort of 50+ workers. It is posited that attitudes and orientations to work are highly influenced by people's expectations at the time they entered the labour market (Yeandle, 2005: 14) and that changing societal norms around maternal employment have influenced and will continue to influence labour market participation, especially for women (Martinengo et al, 2010). However, we note that despite the radical change in social norms and attitudes over the lifetime of our respondents, gender-roles still extend across the generations and may extend into future cohorts of older workers. Several of the women in our study had been forced to leave work under the'marriage bar' or when they had their children, but their daughters now worked full time. However this apparent 'cohort difference' was only achieved by the older women'sacrificing' their working lives to care for grandchildren. This finding underlines the fact that structural supports for working parents (mothers) lag behind societal attitudes, and calls into question what will happen when these daughters' daughters enter the workforce? Thus, this more detailed consideration of older women's working trajectories problematizes once again rigid distinctions between paid and unpaid work in women's experience. Current notions of 'productive' or 'active' ageing through extending working life by flexible working are simply too narrow to portray women's circumstances or to assess adequately their contribution to society. In particular we found that many of our older women were working flexibly and productively, but invisibly, on an informal basis. Moreover, much of this unpaid work is arguably even less visible than that undertaken by younger women because caring responsibilities for the elderly are viewed as less legitimate than obligations to small children, with the result that this type of caring is often hidden from employers (Gautin andHagan, 2010, Alden, 2012). In addition, this lack of recognition of older women's 'work' renders it less visible in debates over extending working lives and of work-life balance in later lives. We concur with the observations made by Cebulla et al (2007: 865) that 'working longer in life may well lead to an extension or perpetuation of unequal domestic divisions of labour.' As such, the consequences for women of ignoring gender in later working life may be to intensify cultural and structural discrimination. One way to overcome this lack of attention and to better address the relationships between gender and later working life is by considering more inclusive concepts of work, such as Glucksmann's (1995Glucksmann's (, 2005) ) Total Social Organisation of Labour (TSOL). TSOL highlights connections, overlaps and interactions across diverse socio-economic relations, and has been particularly instructive in understanding the complex nature of women's work in a changing society. Its two overarching principles are firstly to emphasise the interconnections between work activities, and secondly calling for a higher level of analysis than work itself. We feel there would be merit in extending this framework to better theorise women's relationship to work in later life and to retirement. For example, a key dimension of TSOL is how women articulate work and non-work. We have shown how the blurring of boundaries extends into later working life and can render'retirement' meaningless for many women. Furthermore, in relation to TSOL's principle of raising the level of analysis, our study contributes to critiques of WLB which draw attention to the fact that WLB is still perceived as a personal, not a societal issue. As Emslie and Hunt (2009) note, this has led to the focus on individual strategies to better accommodate conflict or to achieve balance, as opposed to structural consideration of FWOs and attention to more affordable care. The consequences of this individual focus have also been examined in other critical accounts of flexible working. For example, Dick (2010) and Rigby and O'Brien-Smith (2010) both highlight the ways in which part-time working in practice may actually inhibit temporal flexibility on the part of the employee who feels they have to be available 'on-call' all the time. Moreover, Ford and Collinson (2011: 268) raise concern at the ways in which 'WLB discourses have intensified the pressure...to strive to be the perfectly balanced human being.' Being released from this pressure is attractive to many reaching the end of their working lives. Our analysis has provided insight into the gendered nature of these pressures, with a particular focus on later working life. In conclusion, a critical view of flexible working and the benefits it provides to older workers, their employers, and to society in general remains largely absent in policy debates on extending working lives. Our article has presented a far more nuanced view by turning attention to the relationships between gender, flexible working and working in later life. We have argued that, not only are these relationships underexplored in empirical research, but that they are also under-theorised. Further attention to the ways in which gender roles are maintained, intensified or even challenged over the course of the working life is needed to more fully understand the complexity of constructions of later-life working and WLB, and to appreciate how retirement is changing and will continue to change into the future. --- Biographies Wendy Loretto is Professor of Organisational Behaviour at the University of Edinburgh Business School. Her main research field

In many countries economic and social concerns associated with ageing populations have focused attention onto flexible forms of working as key to encouraging people to work longer and delay retirement. This article argues that there has been a remarkable lack of attention paid to the role of gender in extending working lives and contends that this gap has arisen because of two, inter-related, oversights: little consideration of relationships between gender and flexible working beyond the child-caring phase of life; and the prevailing tendency to think of end of working life and retirement as gender-neutral or following a typical male trajectory. The findings of a qualitative study of people aged 50+ in the UK challenge some of the key assumptions underpinning the utility of flexible work in extending working lives, and provide insight into the ways in which working in later life is constructed and enacted differently for men and women.

'WLB discourses have intensified the pressure...to strive to be the perfectly balanced human being.' Being released from this pressure is attractive to many reaching the end of their working lives. Our analysis has provided insight into the gendered nature of these pressures, with a particular focus on later working life. In conclusion, a critical view of flexible working and the benefits it provides to older workers, their employers, and to society in general remains largely absent in policy debates on extending working lives. Our article has presented a far more nuanced view by turning attention to the relationships between gender, flexible working and working in later life. We have argued that, not only are these relationships underexplored in empirical research, but that they are also under-theorised. Further attention to the ways in which gender roles are maintained, intensified or even challenged over the course of the working life is needed to more fully understand the complexity of constructions of later-life working and WLB, and to appreciate how retirement is changing and will continue to change into the future. --- Biographies Wendy Loretto is Professor of Organisational Behaviour at the University of Edinburgh Business School. Her main research field is age and employment, with a particular focus on changes in employees' and employers' attitudes and practices in extending working lives. She is especially interested in the ways in which gender and age interact to affect work and retirement experiences amongst older men and women. Her research has been funded by research councils, Scottish and UK Governments, charities, trade unions and employers. She is currently working with Sarah Vickerstaff on the Uncertain Futures project. i The DWP were less interested in higher income earners as they typically have maximum choice over their retirement afforded by good pensions and privileged work histories. Instead they were more interested on the majority of the population who often face a variety of pressures at the end of their working lives. --- Sarah

In many countries economic and social concerns associated with ageing populations have focused attention onto flexible forms of working as key to encouraging people to work longer and delay retirement. This article argues that there has been a remarkable lack of attention paid to the role of gender in extending working lives and contends that this gap has arisen because of two, inter-related, oversights: little consideration of relationships between gender and flexible working beyond the child-caring phase of life; and the prevailing tendency to think of end of working life and retirement as gender-neutral or following a typical male trajectory. The findings of a qualitative study of people aged 50+ in the UK challenge some of the key assumptions underpinning the utility of flexible work in extending working lives, and provide insight into the ways in which working in later life is constructed and enacted differently for men and women.

Introduction Loneliness has become a significant public health concern for older people, especially within high-income countries [1]. Over 20% of community-dwelling older adults in Taiwan report feeling lonely sometimes or usually [2]. Perceived loneliness can have detrimental effects on mental health and well-being, including an increased risk of depression [3] and reduced life satisfaction [4]. High levels of loneliness in later life can accelerate the degeneration of the nervous system, hippocampus and other brain regions that are related to emotional regulation and cognition. Therefore, loneliness has been identified as a predictor of accelerated cognitive decline and the emergence of dementia [1,5]. To minimize the spread of COVID-19, Taiwan enforced restrictions on physical and social contact, ranging from recommendations to maintain physical distance and stay at home orders to lockdowns of businesses and communities from May to August 2021 [6]. Notably, the suspension of community-based services and programs have caused significant and enduring impacts on the daily life and social interaction of older adults residing in communities, making them particularly vulnerable to increased loneliness and mental health problems [7][8][9][10]. However, the loneliness of older individuals has been affected differently by the COVID-19 pandemic. The resource perspective assumes that the degree to which an individual can access resources directly determines his or her feelings of loneliness and indirectly influences mental health by changing his or her social relations or levels of participation in activities during the pandemic [11]. Specifically, high levels of accessibility to resources across multiple dimensions, ranging from individual and materials to non-material, have enabled older persons to better manage restrictions and to maintain a desired lifestyle during the pandemic [11]. Thus, older adults with more resources would be less likely to experience increased loneliness, as well as decreased mental health and wellbeing because of COVID-19. As loneliness refers to a subjective emotional state reflecting a lack of satisfaction with social relationships or discrepancy between desired and achieved levels of social relations [3,12], loneliness literacy represents one's behavioral, cognitive and social skills to access loneliness-related information [13]. The Loneliness Literacy Scale (LLS) [13] was derived from behavioral change theory and consists of four constructs, including: (1) awareness of and willingness to use loneliness-related services (motivation); (2) self-perception of one's ability to find information and manage daily living (self-efficacy); (3) the availability of support from family, friends, and neighbors and the importance of their opinions (perceived social support); and (4) beliefs of importance to stay active within one's social context (subjective norm). The validity and reliability of the LLS, as well as its association with loneliness, have been confirmed in a sample of older adults in the Netherlands. To the best of our knowledge, the LLS has not been tested in older adult populations of different cultures. Additionally, how loneliness and loneliness literacy influence changes in loneliness during the COVID-19 pandemic and the mental well-being of community-dwelling older people remains unknown. Guided by the resource perspective [14], several resources that may be supplementary to loneliness or change in loneliness during COVID-19 and have the potential to influence the mental well-being of older people were considered, including (1) personal factors, such as age, sex, education, financial satisfaction, health condition, physical function, physical activity, coping strategies, COVID-19 worry, and COVID-19 quarantine experience [1,3,8,[15][16][17]; (2) interpersonal factors, such as marital status, living arrangement, employment status, social support and social participation, internet use, pet ownership, and changes in social interaction due to COVID-19 [15,[18][19][20][21][22]; and (3) societal factors, such as the area of residence [18]. Due to the gap in existing knowledge of this topic, the study aimed to, first, evaluate the extent to which loneliness literacy is related to loneliness and increased loneliness during the COVID-19 pandemic, and second, to determine how loneliness, loneliness literacy, and increased loneliness during the COVID-19 pandemic are associated with the mental well-being of older adults. Mental well-being that has been recognized as a consequence of loneliness includes depressive symptoms [1,3,5] and life satisfaction [4]. This research advances current knowledge by comprehensively assessing the resources at different levels and various manifestations of loneliness, the emotional response of loneliness, cognitive and behavioral abilities to access loneliness-related information (loneliness literacy), changes in loneliness during COVID, and how these all relate to depressive symptoms and life satisfaction. --- Methods --- Data and sample The survey data were collected by telephone interviews from November to December 2021. The participants experienced the COVID-19 lockdown from May to August 2021. The sample was drawn from six areas in Taiwan based on the region. The sample size for each region was based on the age and sex distribution of the population, and the sample was stratified into four layers (age 65-74 and age 75; males and females). The community-based sample of older participants was recruited from the computer-assisted survey telephone interview system. The inclusion criteria included community-based older adults who were 65 years or older, able to communicate and with no severe cognitive impairment. Those who did not have autonomy due to physical or psychological impairments were excluded. If there was more than one person in the household, only one person was interviewed. No proxy was allowed. There were a total of 804 complete cases. Interviewers were trained in conducting interviews and following research ethics. Participants were informed of the research purpose and their permission was obtained before the interview was conducted. Approval for the study was granted by the Institutional Review Board before it was conducted. --- Measures The questionnaire draft was designed based on the literature review. Two experts in gerontology reviewed the questionnaire draft to assure content and face validity. Twenty-four older adults of different ages, sexes, and living areas (rural or urban) were recruited for the pretest to test the construct validity and reliability and to modify the wording of the questions. The final version of the questionnaire was then modified. --- Loneliness, change in loneliness, and loneliness literacy Loneliness was measured by the 6-item revised UCLA loneliness scale-6 [23] by asking about the frequency of the following feelings, for example: how often do you feel that you lack companionship, feel alone, or are no longer close to anyone? Each item was scored from 1 to 4, ranging from never to always. The total score of the 6 items was defined as loneliness. The Cronbach's alpha of the loneliness literacy scale was 0.781. The participants were also asked if they agreed that loneliness has positive effects, scored from 1 to 5 from strongly disagree to strongly agree. In addition, an open-ended question about when the participants would feel lonely was asked to understand the possible reasons or context for loneliness. The change in loneliness due to COVID-19 was assessed by asking how the feeling of loneliness changed during COVID-19. Responses included: loneliness increased a great deal, increased a little, did not change, reduced a little, or reduced a great deal. The response was then categorized as "reduced or no change" vs. "increased". Loneliness literacy was modified from the Loneliness Literacy Scale (LLS) [13]. Based on the results of the pretest and the evaluation of the validity by two experts, only the 15 items from the scale were used (please see Supplement Table S1). The sample questions were as follows: able to attend activities alone, able to find information in daily life, family would help if you ask, friends would help you if you ask, you are aware of information regarding exercise programs, entertainment or courses, etc. The items were scored from 1 to 5, from strongly disagree to strongly agree. The Cronbach's alpha of the loneliness literacy scale was 0.769, indicating high internal consistency. The factor analysis result of loneliness literacy is shown in Supplement Table S2. Four dimensions of loneliness literacy were extracted by factor analysis: self-efficacy, social support, in-home support, and socialization; the explained variance was 55.94%. Then, the four levels of literacy were calculated as the average score of the sum of the items for each dimension. --- Mental well-being Mental well-being included depressive symptoms and life satisfaction. Depressive symptoms were measured by the revised 10-item version of the Center for Epidemiological Studies Depression scale [24]. The sample items were as follows: I felt depressed; I felt that everything I did was an effort; I could not get going, etc. Each item was scored from 0 to 3; the total score ranged from 0 to 30. The CESD-10 scale was validated and tested in previous research using T transformation on nation-representative data in Taiwan [25]; a score of 8 or more was defined as depressive. The Cronbach's alpha of the CESD-19 in this study was 0.59, indicating a slightly lower internal consistency. Life satisfaction was measured by one question asking if participants felt satisfied about their life, and the score was from 1 to 5 and then categorized into two groups (unsatisfied/fair and satisfied). --- Personal factors Individual factors included health conditions, health behaviors, coping strategies, and demographics. Physical health conditions included self-rated health and physical function. Self-rated health was scored from 1 to 5, indicating poor to excellent health. Physical function was measured by three kinds of activities: activities of daily living (including dressing, transferring, walking indoors, using the toilet, taking a bath, and eating), indoor instrumental activities of daily living (cooking or preparing meals, laundry, doing housework), and outdoor instrumental activities of daily living (walking outdoors, going out by transportation alone, and shopping for groceries). Each item asked if there was any difficulty in any of the activities of that kind (based on the most difficult one). Difficulty was scored from 0 to 3 to represent not difficult, slightly difficult, very difficult, and unable to do. The total score of the three kinds of physical function difficulty ranged from 0 to 9. Regarding vision and hearing, participants were asked if there was any impairment (yes/no). Health behaviors included smoking (never, quit, current smoking), physical activity or exercise (frequency in a week and minutes every time), and drinking frequency (scored 0 to 4, indicating never to every day). Physical activity was measured in minutes per week, and regular physical activity was defined as 150 min or more (yes or no). Coping strategies were measured according to the coping strategies of older Taiwanese individuals [26] by asking the frequency with which the participants engaged in coping behaviors when they were in distress. Each item was scored from 1 to 4, indicating never, seldom, sometimes, and always. According to the results of factor analysis, the coping strategies were categorized into two types: problem-focused coping (taking action to reduce worry/distress, doing something else to distract from distress, leaning to live with worry/distress, and praying or meditation to get comfort from religious beliefs) and emotion-focused coping (seeking out care from others, not believing that worry/distress has happened, continuously venting, and blaming yourself for letting it happen). The two kinds of coping strategy scores were the summed scores of the items of each kind. Other personal demographic variables included age, sex, education (illiterate, elementary school, junior high school, senior high school, college or university, graduate and above), religious belief (yes/no), and financial satisfaction (scored 1-5). --- Interpersonal factors Interpersonal factors included marital status (having a spouse or not), living arrangement (living with others or alone), social support, social participation, leisure activities, and social interaction changes during COVID-19. Social support was measured by social contact frequency and satisfaction with social support. The contact frequency of family and of friends/neighbors at least once per week was defined as high; otherwise, it was defined as low. Satisfaction of social support from family or from friends/neighbors was scored from 0 to 5. Social participation was defined as working (yes/no), taking care of family or children (yes/no), and participating in volunteer or social groups (yes/no). Leisure activities included physically active leisure (exercise, dancing, traveling) or sedentary leisure activity (such as reading, watching movies, painting, and calligraphy) (yes/no). Using the internet at least once per week was defined as being an active user. The participants were also asked if they had pets (yes/no). The influences of COVID-19 were added to this survey, including whether social participation, leisure activity and interaction with relatives and friends changed during the COVID-19 pandemic (reduced a great deal, reduced a little, no change, increased a little, and increased a great deal); worry or fear because of COVID-19 (not at all, occasionally, usually, and always); and experience of COVID-19 quarantine (yes/no). --- Societal factors Societal factors were assessed by the place where the participants lived, including regions and rural/urban areas. Four regions of Taiwan were coded: North, Central, South, and East. The living area was also defined as rural or urban, which was defined according to the city development definition [27]. --- Analysis Descriptive analysis, bivariate analysis, factor analysis, linear regression analysis, and logistic regression analysis were conducted. Factor analysis was used to extract the components of loneliness literacy. Linear regression analysis was used to examine factors related to loneliness. The factors related to increased loneliness during COVID-19 and mental well-being were first analyzed by logistic regression. --- Results Table 1 shows a description of the sample. The average age was 73.74 years old; males and females accounted for 45.6% and 54.4% of the participants, respectively. The average mean loneliness score was 8.01 (score 6 to 24). During COVID-19, 61.1% of the participants had reduced interactions with family or friends, and 26.4% reported change in loneliness increased. Most people did not worry or feel afraid due to COVID-19 (mean = 1.50, SE = 0.83). Only 0.9% had experienced COVID-19 quarantine. Bivariate analysis was conducted to examine all the related factors to loneliness and changes in loneliness; the significant factors were included in the following multivariate analysis. Table 2 shows the factors related to loneliness using a linear regression model. The older participants who had lower financial satisfaction (B = -0.402, p <unk> 0.01), had physical function difficulties (B = 0.315, p <unk> 0.01), had lower self-efficacy (B = -0.638, p <unk> 0.05), had lower in-home support (B = -0.184, p <unk> 0.001), performed more problem-focused coping (B = 0.187, p <unk> 0.001) and more emotion-focused coping (B = 0.171, p <unk> 0.01), did not participate in social groups (B = -0.561, p <unk> 0.05), and reported lower satisfaction of family support (B = -1.169, p <unk> 0.001) were more likely to have a higher level of loneliness. Table 3 shows the factors related to increased loneliness during COVID-19 by a binary logistic regression model. The older participants who were male (OR = 1.878, p <unk> 0.05), felt more loneliness (OR = 1.095, p <unk> 0.05), had lower self-efficacy (OR = 0.458, p <unk> 0.01), had more social support (OR = 1.595, p <unk> 0.01), had more socialization (OR = 1.606, p <unk> 0.01), performed more emotionfocused coping strategies (OR = 1.188, p <unk> 0.001), had lower satisfaction with family support (OR = 0.651, p <unk> 0.05), performed more physically active leisure activities (OR = 1.737), and were more worried about COVID-19 (OR = 2.095, p <unk> 0.001) were more likely to experience increased loneliness during COVID-19. Table 4 shows the associations of loneliness and other factors related to mental well-being by logistic regression analysis. Having depressive symptoms was significantly related to lower financial satisfaction (OR = 0.718, p <unk> 0.05), poor self-rated health (OR = 0.682, p <unk> 0.01), more loneliness (OR = 1.240, p <unk> 0.001), --- Discussion This cross-sectional study examined the associations of loneliness literacy with loneliness and changes in loneliness during the COVID-19 pandemic among older --- Loneliness literacy and loneliness The concept of loneliness literacy used in this study was modified from the Loneliness Literacy Scale [13]. The original scale consisted of dimensions of self-efficacy, perceived social support, motivation, subjective norm, and two unused dimensions (i.e., knowledge and attitudinal belief ). In this study, four dimensions of loneliness literacy based on factor analysis were found, namely, selfefficacy, social support, in-home support, and socialization. Some items in the original scale were not included because the pretest participants scored them very low, such as formal service knowledge, motivation and subjective norm items. Additionally, support from family and help at home were separated from the original perceived social support factor. The results of factor analysis suggest that as a family-centered culture, a family bond may exert a stronger influence on older people than bonds from other sources, such as friends or neighbors. Older people in Taiwan also expect more from their family than from friends or neighbors. Thus, support from family and help at home were separate from social support from friends or neighbors in this Taiwanese sample. Higher self-efficacy and in-home support were related to a lower level of loneliness among community-based older adults in this study. This finding supports the hypothesis of the resource perspective. Self-efficacy represents information and resources known and available in the neighborhoods of older people and they have the confidence to use these resources in daily life. In-home support means support at home, either from their family or from in-home services. In fact, from the qualitative descriptions of the context of loneliness in the open-ended question, many participants felt lonely when they were sick, disabled, or needed help but were without assistance. Self-efficacy and in-home support are modifiable dimensions that can be addressed to reduce the manifestation of loneliness. To facilitate the provision of neighborhood resources, the accessibility of community networks and information should be developed. In-home support can be encouraged and strengthened by providing homebased or alternative community-based services. --- Changes in loneliness during COVID-19 During a pandemic event, such as COVID-19, disease control policies, social distancing, and worry about the risk of COVID-19 can increase the risk of feeling lonely [12]. Among the four dimensions of loneliness literacy, self-efficacy, social support, and socialization were related to changes in loneliness during COVID-19, but in opposite directions. Having more self-efficacy reduced the possibility of a change in loneliness (increase) due to COVID-19. It is possible that when older people have more information and knowledge about COVID-19 or have a greater ability to acquire resources from the community, they are less likely to experience loneliness during the pandemic. In contrast, older adults who had high levels of social support (i.e., higher social support from friends or neighbors) and higher socialization reported an increase in loneliness during the COVID-19 pandemic. One explanation is that these people may be affiliated with people outside their families, but usual social interaction was not allowed during the COVID-19 pandemic, thus contributing to an increase in loneliness. In addition to loneliness literacy, it was found that being male, engaging in emotion-focused coping strategies, and having lower satisfaction with family support were related to increased loneliness during COVID-19, which is in line with the literature. Females tend to have better psychological resilience and are less likely to feel lonely [28]. Engaging in problem-focused coping, rather than emotion-focused coping strategies, would be more realistic under the uncertainty of a new pandemic disease and, thus, may make older persons more robust against increased loneliness [29]. A satisfactory family relationship could lower loneliness, especially among older males. Interestingly, performing physically active exercise was related to higher odds of reporting feeling lonelier because of COVID-19. It is possible that physical activity can expedite the ability to regulate emotion during the COVID-19 outbreak, and physical and leisure activities may have been prohibited or restricted. Therefore, those who used to perform physical and leisure activities may feel more disruptions in their daily routine and perceive their loneliness level as high. --- Loneliness and mental well-being Previous studies indicated that loneliness was related to more depressive symptoms [3,[30][31][32] and poor life satisfaction [4,17,[33][34][35]. In accordance with previous studies, high levels of loneliness were significantly associated with having depressive symptoms and poor life satisfaction when other covariates were controlled for. In the four dimensions of loneliness literacy, high self-efficacy was related to high life satisfaction, and high in-home support was related to few depressive symptoms. This investigation advances current knowledge by confirming that increased self-efficacy and in-home support may be helpful to increase mental well-being. Providing more information to increase loneliness-related knowledge and efficacy may help older people access home and community-based care and manage their lives in the community, leading to higher life satisfaction. Additionally, social support and assistance from family members are particularly important for lowering the depressive symptoms of community-dwelling older individuals in a family-centered society, such as Taiwan. Engaging in more emotion-focused coping strategies was related to more depressive symptoms, which is consistent with previous research [25,36]. Worry about COVID-19 was also related to higher depressive symptoms [12], implying that educating people on how to adopt active coping strategies in response to stress caused by the pandemic is important in protecting mental health and has the potential to prevent further mental conditions. Additionally, consistent with previous studies, older adults performing physically active and leisure activities are less likely to develop cognitive impairment and depressive symptoms [37,38], suggesting that physical activity or leisure may increase physical function and protect cognitive function and emotional health. --- Limitations There were some limitations in this study. First, due to the COVID-19 pandemic, the survey data was not collected through face-to-face interviews; rather telephone interviews were conducted. The interview time had to be shortened, and some of the questions could not be included in this survey, such as the resilience scale and details of health conditions. In addition, some questions about loneliness were not easy to explore, such as the open-ended questions regarding the context of loneliness and perceived consequences. Second, as a cross-sectional study, the causal relationship of the variables cannot be confirmed. Third, the sample was drawn by the Computer-Assisted Survey Telephone Interview System, and the sampling was based on the population distribution of the region, not by cities. Thus the sample may not be generalizable to cities, and the sample of older people may not be generalizable to the entire population. --- Conclusion In conclusion, higher self-efficacy of loneliness literacy, engaging in problem-focused coping rather than emotion-focused coping strategies, and having higher levels of satisfaction with family support were related to a lower chance of increased loneliness during the pandemic and were also protective factors for mental well-being in older adults. Building up an age-friendly community to provide home-and community-based services is suggested. Providing accessible information and enabling social connectedness in the community for older adults should be actions taken by the government. Learning positive (problem-focused) coping strategies and building up resilience under stress are highly suggested in health education to promote mental health. A longitudinal survey should be conducted for following up on the mental health of older adults during the pandemic and to explore the effects of geographical resources on loneliness in future research. An intervention study is also suggested to examine whether the provision of community-based services and an age-friendly social network may reduce loneliness. --- Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. --- Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s12877-022-03396-7. Additional file 1: Table S1. Descriptive analysis of the 6-item Revised UCLA Loneliness Scale. Table S2. Factor analysis of loneliness literacy scale. --- Authors' contributions HCH conceptualized the study, obtained the funding, and analyzed the data; HCH & SFC designed the study and wrote the manuscript. Both authors have read and approved the manuscript. --- Declarations Ethics approval and consent to participate The study was conducted according to the guidelines of the Declaration of Helsinki. This study used a secondary de-identified data, and informed consent was not applicable. The study was approved by Taipei Medical University Joint Institutional Review Board (N202101047). --- Consent for publication Not applicable. --- Competing interests The authors declare that they have no competing interests. --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background: Loneliness has become a significant public health concern for older people. However, little is known about the association of loneliness, loneliness literacy, and changes in loneliness during the COVID-19 pandemic with mental well-being. The purpose of this study was to explore whether loneliness literacy is related to a lower risk of loneliness, increased loneliness during the COVID-19 pandemic, and improved mental well-being for communitybased older adults. Methods: A telephone survey was conducted to collect data from older adults aged 65 years or older in Taiwan (n = 804). Loneliness, change in loneliness during COVID-19, and loneliness literacy were the main variables. Mental well-being was assessed by depressive symptoms and life satisfaction. Related factors included personal level (demographics, health conditions, health behaviors, and problem-focused/ emotion-focused coping strategies), interpersonal level (marital status, living arrangements, social support, social participation, leisure activities, and social interactions during COVID-19), and societal level (areas and regions) factors. Results: Four dimensions of loneliness literacy were identified by factor analysis: self-efficacy, social support, socialization, and in-home support. Self-efficacy and in-home support were related to lower loneliness. Lower self-efficacy, higher social support, and higher socialization were related to changes (increases) in loneliness during COVID-19. In-home support may prevent depressive symptoms, while self-efficacy was beneficial for better life satisfaction. In addition, emotion-focused coping may increase loneliness during COVID-19, while satisfaction with family support would be a protective factor against loneliness.Loneliness literacy is related to loneliness and increased loneliness during the COVID-19 pandemic. Building up an age-friendly community with embedded services/information and learning positive coping and mental resilience strategies are suggested.

INTRODUCTION In Brazil, the Primary Health Care (PHC) network adopts PHC guiding principles, both essential and derivative attributes under the terms established by Bárbara Starfield (1), in addition to the bond and responsibility of teams for a given territory and the population that there lives. Similarly, the indigenous health care subsystem, established by Law 9,836/99, has a health care network for these minorities managed by the Special Office of Indigenous Health (SESAI -Secretaria Especial de Sa<unk>de Ind<unk>gena) of the Ministry of Health, which, through Special Indigenous Health Districts (DSEI -Distritos Sanitários Especiais Ind<unk>genas), offering PHC shares in indigenous lands (2). The existence of an indigenous health subsystem expresses the recognition, by the legislator, of the need to offer an alternative capable of promoting extension of coverage and providing specific attention to health needs shaped by linguistic and cultural singularities, in addition to geographical and access barriers faced by such culturally differentiated groups (3). However, eventual singularities in DSEI' performance do not nullify the fact that they offer PHC actions, allowing us to assume that the guiding principles and attributes of the PHC assistance model remain in force there. On the other hand, the challenges to the provision of PHC in indigenous lands have similarities with those faced by municipal health systems in remote and difficult to access areas, showing a potentially enriching dialogue between both realities. However, when starting the research, in the literature review on the subject, no studies were found that compared the two realities, making the approach even more relevant. In social research, the comparison between differences and similarities faces the challenge of properly selecting the elements to be compared, because, unlike what happens in experimental methods, there is no way to control the variables involved in the comparison (4). The research that gave rise to this article carried out an analysis of essential attributes (first contact access, longitudinality, integrality, and coordination of care) and derivatives (family centered care, community orientation and cultural competence) (1). PHC attributes lend themselves to dimensioning the resoluteness and quality in service organization, based on the assumption that the greater proportion of their presence, in the daily routine of the basic network, is equivalent to an increase in the positive impact on the health of the population assisted. Such attributes have also been used in the country through PCATool-Brasil, as performance indicators in the assessment of the PHC network of SUS (Sistema <unk>nico de Sa<unk>de -Brazilian Unified Health System) (5), which increases the reliability in its use. For this article, analysis of an essential attribute (longitudinality) and a derivative (community orientation) was prioritized. The first is seen by Starfield (1) as a central and exclusive characteristic of PHC, as it provides, over time, the monitoring of episodes of illness and the provision of preventive care to families accompanied by professionals from PHC teams. The good performance of the logitudinality attribute is also linked to the guarantee of continuity of care, with positive repercussions on the teams' performance, which justifies its use for PHC assessment purposes. Community orientation is also considered in the qualification of health actions in PHC, as it has relevance based on the professional's knowledge about the users' need in their social context through epidemiological, social, cultural data, in addition to the experience in the community to favor identifying existing healthcare problems in the community, assessment of services and readjustment of these to the population's health and social needs, being also an attribute considered in the qualification of the actions of PHC services (1). Understanding the characteristics of health in the community and the resources available favors a broader way of assessing needs than just in interactions with patients or their families (1). The two attributes were selected because they were more sensitive to the comparative process, since the previous scrutiny of the data showed that, of a wide set of attributes, longitudinality and community orientation were those implemented with greater constancy and regularity by healthcare professional teams, both from DSEI-ARN and from the investigated municipal health systems, lending themselves better to the exercise of comparability. Despite the local character of the analysis developed here, it has the potential to contribute to the improvement of health care provision for rural, border and other remote and difficult-to-reach locations, including non-indigenous populations. --- OBJECTIVES To analyze, from healthcare professionals' perspectives, longitudinality and community orientation in Primary Health Care, offered both in the Special Indigenous Health District and in the primary network, which assists non-indigenous populations in municipal health systems in the Upper Rio Negro region. --- METHODS --- Ethical aspects In accordance with Resolution 466 of 2012 of the Brazilian National Health Council (Conselho Nacional de Sa<unk>de), which regulates the ethical aspects of research involving human beings, the study was submitted to and approved by the Research Ethics Committee of Universidade de S<unk>o Paulo School of Nursing. --- Study design, period and location This is a cross-sectional quantitative study. The instrument selected to guide the methodology was the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE). The study was carried out in the Upper Rio Negro Indigenous Special Health District (DSEI-ARN) and in PHC services of the municipalities of S<unk>o Gabriel da Cachoeira (SGC), Santa Isabel do Rio Negro (SIRN) and Barcelos in Amazonas State, from June to August 2015. The municipal health services studied were selected due to the interfaces they maintain with DSEI-ARN. DSEI-ARN is distributed among these three municipal territories, maintaining important operational relations with such offices. They receive referral patients from indigenous lands for health care that are not offered by DSEI-ARN. Altogether, the three municipalities, which are in the district territory, form an extensive region (295,917.10 km 2 ), known as Upper/Middle Rio Negro, representing 35% of the total area of Amazonas (6). In this area, there is a population of approximately 96,616 people, 75% of them living in rural areas calling themselves indigenous, linked to one of the 25 (twenty-five) ethnic groups demarcated by one of the four linguistic groups (Eastern Tukano, Aruak, Maku, and Yanomami) (7) distributed there. Municipalities have low social indices, with AHDI (Mean Human Development Index) ranging from 0.47 in SIRN, 0.50 in Barcelos and 0.62 in SGC (6). --- Population; inclusion and exclusion criteria The eligible population corresponded to 131 professionals (physicians and nurses), who were selected based on the registration of the Brazilian National Health Establishment Register (CNES -Cadastro Nacional de Estabelecimento de Sa<unk>de) and confirmed with the health offices' local managers. After confirming data, prior contact was made to schedule the start of data collection. Of the total of 131 professionals, 91 were linked to DSEI-ARN, which assists the indigenous population in the village. The rest of the study population (40 professionals) was linked to the Health Offices of the three municipalities, assisting the population that lives in municipal headquarters and in rural areas not demarcated, regardless of whether they declare themselves as indigenous. During the period of application of the research, 5 professionals were away for health treatment, 5 on vacation, and 5 refused to participate, resulting 116 professionals who participated in the study. Professionals were chosen because they are considered primary care providers of higher education who make up a minimum team of Family Health Strategy (FHS). This study included nurses and physicians who had worked for more than 6 months at DSEI-ARN and/or the PHC of the three municipalities, having an employment relationship with the municipal health offices and/or with SESAI/DSEI-ARN. Professionals who were on sick leave and/or vacation were excluded. --- Study protocol The Primary Care Assessment questionnaire (PCATool) -Brazil, healthcare professional version, was used, then validated (5). The longitudinality and community orientation components (dependent variables) and the degree of orientation for PHC, considered structural, were used for this study, used to assess the attributes' reach. The responses followed a Likert-type scale, with a score from 1 to 4, "Certainly yes" (value = 4); "Probably yes" (value = 3); "Probably not" (value = 2); "Certainly not" (value = 1). When the sum of the answer "I don't know/don't remember" (value = 9) reached 50% or more of the total items of the attributes, the questionnaire was excluded; otherwise, these responses were considered "Probably not" (5). In the case of this study, there was no loss of questionnaire. All scores were calculated by arithmetic mean of items' response values that make up each attribute or its component (5). The mean score for the attributes longitudinality (composed of 13 variables) and community orientation (composed of 6 variables) were calculated by adding the value of the variables, divided by the number of variables for each component, as established in the PCATool-Brazil manual (5). So, we have: mean score A = (A1 + A2 + A3 + A4 + A5 + A6 + A7 + A8 + A9 + A10 + A11 + A12 + A13)/13 and mean score B = (B1 + B2 + B3 + B4 + B5 + B6)/6. Then, these scores obtained for each component were transformed into a scale from 0 to 10, called adjusted score, following the formula: Adjusted score = (Score obtained -1) x 10/3 To assess the degree of PHC orientation, score values <unk> 6.6 were considered satisfactory, and, as unsatisfactory, <unk>6.6. --- Analysis of results and statistics The collected data were inserted into a single spreadsheet using Microsoft Excel software ®. Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS ® ), version 21.0 for Windows and in statistical package R. For statistical analysis, the alternatives "Certainly yes" and "Probably yes" were aggregated, represented in the tables by 1*, the same occurring for the alternatives "Certainly not" and "Probably no", which, in the tables, they are represented by 2*. Assessment of items that make up the two attributes of the PHC object of this article was made by crossing variables of PCATool and professionals' institutions of origin (DSEI-ARN and SEMSA). The association between these variables was assessed using chi-square test or maximum likelihood test, when appropriate. To compare the scores between professionals' institutions of origin, Kruskal-Wallis test was used, since data did not have a normal distribution. For all statistical analyzes, a significance level of 5% was considered. --- RESULTS Of the universe of 131 eligible professionals, 116 (88.5%) answered the questionnaire, 87 from DSEIARN and 29 from SEMSA. Of these, 84 (72%) were nurses and 32 (28%) were physicians, with a mean age of 35 years, prevalence of females (59.7%), mixed-race (42.2%), single (44%). Concerning contracting agent, 60.3% were hired by Non-Governmental Organizations (NGOs) and 39.7% by Direct Administration. Training time was 5 years, with 3 years of experience in PHC in the Upper Rio Negro region. Table 1 shows the variables that make up the longitudinality attribute. There was a statistically significant difference between variables A1 (p = 0.001), A4 (p = 0.001), A7 (p = 0.007) and A11 (p = 0.001). Still in relation to the data presented in Table 1, it is relevant to highlight that, in variable B6, a statistically significant difference was observed (p<unk>0.001). The degree of orientation for PHC had a low score (6.1; 5.6) and a statistically significant difference (p<unk>0.004). For the group of institutions studied, the consolidated of the answers given by professionals shows an unsatisfactory score for the degree of orientation towards longitudinality (6.4 and 6.5 for DSEI-ARN and SEMSA, respectively). Values below the cutoff point (<unk> 6.6) recommended in PCATool were totalized for an adequate orientation of this attribute in PHC activities, even though the municipal system has achieved a performance closer to that recommended. Table 2 systematizes the data related to the Community Orientation component and the degree of its orientation to PHC. In Table 2, only variable B6 showed a statistically significant difference (p<unk>0.001) in the comparison between the two types of service providers. The degree of orientation for PHC also had a low score (6.1; 5.6); for both, it is a statistically significant difference (p<unk>0.004). It should be noted, however, that absence of a statistically significant difference does not prevent an appreciation of the variables in which such difference was absent, as it is a small universe, therefore subject to fluctuation in analyzes. --- DISCUSSION Using a validated instrument to assess health services, based on the attributes of PHC, as is the case of PCATool, is important, as it allows assessing how healthcare professionals perceive the services they provide to the populations they assist, also contributing to improve accessibility, quality and effectiveness of the characteristic attributes of PHC (8)(9). When comparing the responses of the respondents of DSEI-ARN and the Municipal Health Offices to the longitudinality attribute and with regard to the stability of care by the same professional, the 71.3% of negative responses from those working in DSEI-ARN, in counterpoint to the 34.5% of negative responses from those linked to municipal health systems, suggest a high workforce turnover in DSEI. Although the data from the research do not allow us to conclude conclusively that turnover is the cause of the problem, it is a known difficulty and pointed out in publications (2)(3)(4)(5)(6)(7)(8)(9)(10) that dealt with service organization in the indigenous health subsystem, pointing out DSEI's difficulty in ensuring the maintenance of qualified professionals to provide health care in the villages. For other variables of the same attribute, there is limited knowledge about the living and working conditions of patients assisted, both for those working at DSEI-ARN (58.6% know their patients' co-residents and 59.8% the work performed by them) and by non-indigenous primary health care professionals, whose percentages of favorable responses are even lower (27.6% know their patients' relatives and 17.2% the work performed by them). In other words, although the turnover in caring for indigenous people may be high, the work in the villages seems to favor a more personalized service, in contrast to the anonymity of patients assisted at the municipal headquarters, suggesting a lack of bonds and humanization in PHC teams' work that there act. The limitation of the bond established with patients is reaffirmed by the answers given to variable A4, in which it was asked whether, when facing problems, patients can call healthcare professionals. The 17.2% of affirmative responses given by non-indigenous primary care professionals shows the fragile link provided by services in meeting unforeseen situations faced by users. The 56.3% of affirmative answers given by DSEI-ARN professionals demand additional clarification on working conditions at DSEI-ARN. Although it is low, this percentage of affirmative responses shows an effort to establish a professional-patient relationship capable of instituting some welcoming to the healthcare problems faced by indigenous users, even if geographically distant. Publications describing conditions of access, organization and provision of health services in indigenous lands are limited. However, those available (2)(3)(4)(5)(6)(7)(8)(9)(10) show that geographical characteristics, such as great distances and natural obstacles, hinder the regular offer of care in indigenous villages, either in the manner recommended by FHS or by the Brazilian National Indigenous Health Policy (PNASPI -Pol<unk>tica Nacional de Atenç<unk>o à Sa<unk>de dos Povos Ind<unk>genas). Such publications (2)(3)(4)(5)(6)(7)(8)(9)(10) also show infrastructure limitations, such as lack of electricity and of means of communication such as telephone network, making it impossible to provide health actions and services on a daily basis. In this context, PHC care is done through periodic trips by the teams, who need to travel in painful conditions, from the municipal headquarters to the villages, in order to provide care. Communication takes place mainly through radio, installed in a restricted number of locations. In this context, it must be inferred that the answer given by the interviewees who work at DSEI-ARN to the question contained in variable A4 (if patients can call healthcare professionals when facing problems) refers, in fact, to the search for contact through radio, connoting an important effort to meet such needs through a precarious means of communication as radio is. The working conditions described for DSEI-ARN must be understood as structural barriers to access services both with regard to the physical network's structuring and health action management, organization and provision. Such limitations increase, among other consequences, the difficulty of establishing workforce and obtaining and exercising the cultural competence (11) necessary to provide culturally appropriate care. Moreover, they limit the longitudinality of care and compliance with PNASPI provisions (7). The description of the circumstances that shape the care provided by DSEI-ARN professionals makes it understandable -and to some extent justifiable -the reach of low percentages of affirmative responses to the variables that make up the longitudinality attribute of the care offered to the indigenous, highlighting elements of difficulty for DSEI-ARN to become a regular and daily source of care. On the other hand, the professionals of Municipal Health Offices work in the urban network of the small municipal headquarters in the usual way of FHS, through a regime of 8 hours of work and daily operation of the units. Although they enjoy more favorable working conditions, the scores obtained by this group are low due to the high number of negative responses to the questions set out in variables A4, A8, A10, A11 and A13 (Table 2), which assess professionals' knowledge on health history, lifestyle and work of patients as part of the longitudinality attribute. For this group of respondents, professional turnover should not be overlooked, because, although they operate in more favorable conditions than their counterpart in DSEI-ARN, it is an event known in the national scenario for the primary care network (12) and, in particular, for small municipalities, where the PHC network also faces difficulties to act as a regular source of care for the population. The unsatisfactory score of the consolidated of the answers given by professionals of both institutions for the longitudinality attribute (6.4 and 6.5 for DSEI-ARN and SEMSA, respectively) does not differ essentially from the results obtained in other areas of the country, including in regions richer and with better HDI, such as southern and southeastern Brazil (9,13). These are results that express all barriers faced by SUS to promote equity, quality, effectiveness of the primary care network (14). For community orientation, the second attribute evaluated, the results obtained are somewhat paradoxical. For both services, more than 70% of respondents believe they have adequate knowledge of the community's health needs (variable B2, Table 2). However, this belief is not supported by consistent assessments, such as user satisfaction surveys (variable B4 in Table 2, with 14.9% of favorable responses for DSEI-ARN and 13.8% for professionals serving non-indigenous populations) or investigations into the main healthcare problems in the community (variable B5, Table 2 = 23% and 21.1%, respectively). The results obtained for Upper Rio Negro in this attribute are difficult to compare with the scientific literature, due to the selection by some authors, of variables other than those recommended in PCATool-Brasil and the adoption of different cutoff points to classify the scores (15). The most feasible results of comparison were found in a survey conducted in Maranh<unk>o (16), which used variables compatible with those adopted in our research in Upper Rio Negro. It should be noted, however, that the research conducted by Alencar et al (16) in Maranh<unk>o was not oriented towards the indigenous population, as it covered a wider range of interviewees (users, professionals and managers), using another scale to group the answers obtained. Even so, isolating only the results obtained from professionals working in the basic network, it was possible to observe that, in Maranh<unk>o, only 15% of interviewees believed they knew the community's main healthcare problems. Among the same respondents, only 7.5% (against 14.9% of DSEI-ARN and 13.8% of workers in health offices in Upper Rio Negro) conducted surveys to assess user satisfaction. For the question, we sought to identify the population's main problems through surveys. Only 3.7% of professionals from Maranh<unk>o's primary network declared doing so, against 23% of respondents from DSEI-ARN and 21.1% of those working with non-indigenous populations in Upper Rio Negro. The expressive recognition by DSEI-ARN professionals regarding the participation of indigenous users in health councils (97.7%) contrasts with the 37.9% of positive responses given by the interviewees who work in the municipal health systems covered by the research, as well as the 59.3% recognition for the same item, obtained in the research by Alencar et al (16) carried out with PHC professionals in Maranh<unk>o. In the case of Rome, it can be inferred that such results are mainly due to the strong role of indigenous associations in health policy (17). This fact makes their participation in health councils (18) and in other spaces for indigenous policy management, whether at the local or national level, a daily event impossible to be ignored by professionals working at DSEI-ARN. This is a unique scenario that is not repeated for non-indigenous people from adjacent municipalities, contributing to explain the high percentage of negative responses from interviewees who work with nonindigenous populations. For the set of the community orientation attribute, the consolidated score is low, but that achieved by professionals working in DSEI-ARN (6.1) is closer to the cutoff point of 6.6 stipulated in the research as a satisfactory score, in counterpoint to the result of 5.6, achieved by respondents from the municipal health offices. An analysis undertaken by Gonçalves and Bógus (19) supports the findings of this study, as it reinforces the idea that the participation of social action and the performance of health councils are still only incipient in the routines of family health teams. --- Study limitations The main limitation of the study is due to the scarcity of research that favors the comparison of the findings presented here with those found in other realities. --- Contributions to nursing, health, and public policies This study contributes to the critical reflection on the need to improve the work process of professionals who work in remote and difficult to access regions, also pointing out findings that not only express the unique Amazonian sociodiversity, but also identify innovative situations capable of revitalizing the PHC profile in the region. It also has the potential to contribute to the necessary reorganization of PHC health services and actions, in order to make them more permeable to local populations' needs. Thus, it can be a data source for future studies that seek evidence that the guarantee of equitable and comprehensive access to PHC network care in the Upper Rio Negro region continues to be processed and improved. --- CONCLUSIONS This study demonstrates the assessment method's contribution to the social reality knowledge and contextual information provision that shape healthcare professionals' experiences in their daily work, enabling them to outline strategies for the improvement of their practices in a given health territory. It reaffirms the relevance of using validated instruments, such as PCATool, in to assess the central characteristics of PHC in SUS. As a whole, the data reveal that longitudinality and community orientation appropriation by healthcare teams that offer care to indigenous and non-indigenous populations of Upper Rio Negro still needs improvement. Even so, these findings are not contradictory with those found in other regions of the country, demonstrating limitations faced by the Primary Care National Policy (Pol<unk>tica Nacional de Atenç<unk>o Básica) in providing coverage extension to remote places in the national territory. However, analysis of the scores obtained by the interviewees who work in indigenous health, in comparison to those who work in the health systems of the municipalities adjacent to indigenous lands, suggests that, despite facing adverse work conditions and problems common to PHC teams allocated in poor regions and with low HDI, DSEI-ARN professionals express, comparatively, greater adherence to the attributes longitudinality and community orientation, assessed in the research. The results found cannot be considered satisfactory; however, they demonstrate the importance of mapping the societal and contextual elements that surround PHC teams' work and that can contribute to improving their performance even in adverse conditions. The research findings also highlight the need to understand the ways of life of populations that inhabit the territories assisted by healthcare teams, which have a marked influence on the healthcare profiles found there. This premise was demonstrated, for instance, in the finding that high recognition of social control activities by teams working at DSEI-ARN is related to the leading role of ethnopolitical struggles that contribute to the politicization of teams' daily activities. On the other hand, the low recognition of community participation in social control among professionals working with non-indigenous people shows the limited political role played by non-indigenous regional populations in the municipalities studied. Despite this, the research findings are still worrisome, since the studied attributes shape the temporal, longitudinal monitoring of health situations and injury occurrence in the population assisted. The limited appropriation by professionals of essential characteristics and derivatives of PHC indicates the need to improve work management in health units, seeking to guarantee the quality and effectiveness of their professionals.

Objectives: to analyze, from healthcare professionals' perspectives, the longitudinality and community orientation in Primary Health Care, offered both in the Special Indigenous Health District and in the primary network that assists non-indigenous population in municipal health services in Upper Rio Negro region. Methods: a cross-sectional study with 116 professionals, 87 (75%) of indigenous health, 29 (25%) of municipal services. Primary Care Assessment Tool, professional version, used by Upper Rio Negro for Social Sciences region. For association of variables, chi-square test and Kruskal-Wallis were used. Results: longitudinality obtained an unsatisfactory score (6.4 and 6.5), as well as community orientation (6.1 and 5.6) for both services. Weaknesses refer to professional turnover, little knowledge about users' living conditions, precarious employment relationship. In indigenous health, satisfaction was higher when compared to municipal services. Conclusions: it is necessary to improve work management in health services, seeking to guarantee the quality of performance of professionals.

Introduction Language shift from Spanish to English has been studied in different communities in the United States (e.g., D<unk>az, 2011;Potowski, 2004). However, few studies have focused on the language shift of native and foreign-born (FB) aged (here defined as those whose age is 65 and beyond) Mexican origin Latinos/as (hereafter only refer to as Mexicans). Weinreich (1968) defines language shift as a situation in which the habitual use of one language is being replaced by the habitual use of another. In general, when a minority group comes into contact with a dominant host group, language shift frequently results (Portes & Hao, 1998;Wong-Fillmore, 1991)-a phenomenon that has been documented in a number of countries where minority language groups come in contact with more dominant cultural groups (Fishman, 2001). Mexicans who abandon the use of Spanish as they increase their English language usage are said to have experienced a language shift-a transition that may at times be the product of socially coercive processes. We introduce the idea of language resistance to highlight how language shift can at times be borne out of socially oppressive systems that coerce minority-language speaking members to abandon their mother tongue. We interpret the act of retaining Spanish (either in the presence or absence of learning the majority-group host language) as language resistance. Although Mexicans in the United States have historically been relegated to a minority and disenfranchised social status (Siordia, 2011), their population will play a key role as non-Latino/a-Whites become a minority population in the next 40 years (Siordia, Panas, & Delgado, 2012). This study fills a gap in the literature by investigating how contact with the majority group influences the degree to which minority group members use English over Spanish. Our substantive contribution to the literature comes from showing how an aged Mexican's co-ethnic context can provide the resources for language resistance. Our study investigates individual-level (i.e., microlevel) language shift in a sample of aged Mexicans in the Southwest (see more details in the Method section). We make use of a hierarchical linear model to include a U.S. Census tract-level (i.e., macrolevel) ethnic context measure to examine how the microlevel relationship between the degree of "contact with Anglos" and English language use is affected by the percentage of Mexicans in their tract of residence. Our first research question is as follows: Does the degree of contact with Anglos influence Mexicans' language shift? Given our review of the literature (below), we hypothesized a direct (i.e., positive) microlevel relationship between the degree of language shift and the level of contact with Anglos. In simpler terms, we expect a language shift from Spanish to English to be more evident in aged Mexicans who report having a high degree of contact with Anglos. This hypothesis follows classical research on language shift. Our substantive contribution to the literature comes from our second research question: How does the macrolevel percentage of co-ethnic concentration moderate the microlevel relationship between English language use and contact with Anglos? After framing language use as an available instrument for hegemonic control by the dominant group (see Discussion section), we hypothesize that a tract's level of "Mexican concentration" will significantly moderate the microlevel relationship between English language use and contact with Anglos. In particular, we expect that as the tract's Mexican concentration level increases, the influence of "contact with Anglos" on English language use will be weakened. We posit such a hypothesis because we believe that tract-level co-ethnic concentration exerts an effect on language shift, above and beyond the effects of other microlevel attained (e.g., education) and ascribed (e.g., gender) characteristics, where geographical proximity with co-ethnic members provides the resources for choosing to abstain from English language use. In sum, we expect that although contact with Anglos will be associated with language shift, its effect will be mediated by the area's level of co-ethnic concentration. If this proves to be the case (e.g., we find no evidence to refute the tentative hypothesis), then we will interpret that racial-ethnic contexts have the potential for offering social resources (e.g., availability of others who speak Spanish; employment opportunities not requiring the use of English) that support language resistance. From this point of view, we argue that highly concentrated co-ethnic areas offer Mexicans an option to resist the abandonment of Spanish language use by providing them with access to economic and social rewards were English language use is not required. --- Method Our analytic sample was obtained from the Hispanic Established Populations for the Epidemiologic Study of the Elderly (HEPESE) baseline data. The HEPESE is a longitudinal study initiated in 1992 (Markides et al., 1999). The project collected data on 3,050 community-dwelling Mexican origin Latinos aged 65 years and above who resided in the five southwestern states of Arizona, California, Colorado, New Mexico, and Texas (Black, Markides, & Ray, 2003). HEPESE respondents were interviewed in-person by trained field operatives. To investigate the cross-level interaction (CLI) between percentage of Mexicans and contact with Anglos, we used two different levels of measurement: the individual level and tract level. We begin by outlining the measures at the individual level. At "Level-1" we have our continuous dependent variable: a scale on English language use. The scale was created from seven questions in the survey. In the first three questions, individuals were asked how well they: (1) understood spoken English, (2) spoke English, and (3) read English. Their answers were recorded as 1 = not at all, 2 = not too well, 4 = pretty well, 5 = very well, or 8 = don't know. Individuals with an "8" on any of these answers received a missing on the English Language use Scale. In the latter four questions, participants were asked what language they usually used to speak: (4) with their spouse, (5) with friends, (6) with neighbors, and (7) at family gatherings. Their answers were recorded as: 1 = only English, 2 = mostly English, 3 = Spanish and English equally, 4 = mostly Spanish, 5 = only Spanish, 6 = not applicable, 8 = don't know, and 9 = refused. All those answers with a 6, 8, or 9 received a missing on the English Language use Scale. In order to have all responses running on a scale where a high number indicates more English language use, we reverse coded the last four items, so that 1 = 8, 2 = 7, 3 = 6, 4 = 5, and 5 = 4. These strict procedures produced our English Language use Scale that, as shown in Table 1, ranges from a 19 (low English language use) to 44 (language shift from Spanish to English is most present). There were other language use questions available but we decided to exclude them because of the characteristics of the sample. For example, one of the question was what language was used "at work." Since the majority of the respondents do not work, the question was not applicable. These type of questions were omitted from entering the scale formation; as a result, we reduced the number of individuals from the analysis (total missing = 187). Our Level-1 continuous independent variable of interest is the "degree of contact with Anglos" an individual reported to have. In HEPESE, aged Mexicans were asked: "Throughout your adult life, have your neighbors been mostly Mexican-Americans, mostly Anglo, or about equal numbers of each?" They were also asked: "(Are/Were) the people with whom you work closely on [the job/your last job] mostly Mexican-Americans, mostly Anglo, or about equal numbers of each? And, they were lastly asked: "Throughout your adult life, have your close personal friends been mostly Mexican-Americans, mostly Anglo, or about equal numbers of each?" For these three questions, their answers were recorded as: 0= Mexican, 1 = mostly Mexican-American, 2 = mostly Anglo, 3 = about equal numbers of each, 8 = don't know, and 9 = refused. Individuals with a "2" on these questions received a value of "1" on one of three binary variables to indicate they had mostly Anglo contact. The binary variables were then added to create the Contact with Anglo Scale, which as can be seen in Table 1, ranges from 0 (mostly no Anglo contact) to 3 (mostly had Anglo contact). After exploring several other possible related variables (e.g., depression symptomatology) at Level-1, we settled on controlling for age, gender, marital status (married vs. not married), if they have a ninth-grade education and beyond, and whether they are FB or native born. We also include other exploratory covariates. We included their cognitive status using the Mini-Mental State Examination (MMSE; Matallana et al. 2011), in which, those with a score of 25 or greater are said to have high cognitive capacity. As a socioeconomic indicator, in addition to whether they have a high school diploma or not, we considered if they were receiving a pension. In reference to the social network variables we assigned a "1" to those who reported most of the time when answering the following two questions: "In times of trouble, can you count on at least some of your family or friends most of the time, some of the time, or hardly ever?" and "can you talk about your deepest problems with at least some of your family or friends most of the time, some of the time, or hardly ever?" These were included as a proxy measure of the person's social network health status. Our article is unique because it investigates language shift in aged Mexicans while accounting for their generational status, and this HEPESE variable is being used in publication for the first time. The HEPESE generation type algorithm produces six potential categories: At "Level-2" (i.e., the tract level) we focused on the moderating macro variable of the percentage of Mexican. The percentage of Mexicans in the tract was computed using 1990 U.S. Census Bureau Summary File 1 tract-level data. Besides accounting for the in-poverty concentration in the person's tract of residence, we control for population density-this was calculated by dividing the total population in the tract by the tract's geographical area. To specify our hierarchical linear model we used the HLM 6.04 software (Raudenbush, Bryk, Cheong, & Congdon, 2004). We opted to use a multilevel model to investigate the CLI because it has been convincingly argued that multilevel statistical tools provide advantage over traditional methods (Flaherty & Brown, 2010) when investigating macro on micro relationships. This software estimates equations that help explain cross-level statistical associations by accounting for the fact that tract measures are between-people dependent, a violation of classical regressions where independence is assumed. Our multilevel linear model is preferred over the classical ordinary least square regression because it does not assume that both microlevel and macrolevel factors come from simple random samples (Arnold, 1992). Our model runs a single regression for each tract (i.e., level-2 unit) by nesting our HEPESE respondents "within-tract" equations, so that CLIs are the result of "across-tract" equations. The variance around each parameter at level-1 is taken into account in the regression at level-2 (Arnold, 1992). In short, both intra-and intertract coefficients have their own error measurements where maximum likelihood and generalized least squares estimation procedures generate HLM coefficients and variances (Poston, 2002). The use of the HLM software empowers our research by giving us the ability to understand how the microlevel contact with the Anglo factor is on average associated with English language use across and within tracts, as we control for the various microlevel and macrolevel factors included in the equation. We have 2,573 individual-level units being nested in 94 tract-level units and we interpret "population-average" coefficients with "robust standard errors" from our HLM outputs (Mass & Hox, 2004a, 2004b). Please note that existing research making use of hierarchical models has not made it customary to reveal the minimum amount of level-2 units allowed per each level-2 unit. Since the modeling, in effect, executes a regression for each level-2 unit, having a regression slope based on one or two people may be of concern. Researchers making use of such models should be explicit with the minimum amount of units allowed per cluster. In our project, we seek to reduce the chances of making a Type-I error (i.e., finding false statistical significance) with our model and parameters, by only using observations where there are at least 10 HEPESE respondents per-tract. --- Results We begin this section by giving a brief description of the analytic sample from the descriptive statistics (Table 1). At Level-1, on the language scale, the average score is 26. The majority of the subjects have very little contact with Anglos and an average age of 74. About 58% are females, 56% are married, 43% are FB, and only a few (15%) have a ninthgrade education and beyond. Most people in our sample report having somebody they can "count on" and "talk with" within their social networks. Only 14% have a pension and 51% have a high MMSE score, which signals a high level of cognitive functioning in the sample. At Level-2, we observed that the average Census tract in our analytic sample has a high level of co-ethnic concentration (71%). On average, our tracts have a 34% in-poverty concentration and a 1.73 km 2 population density. We now turn our attention to our empirical multilevel model. To justify the need for a multilevel model, we first execute the following simple two-level random intercept model with the English Language use Scale as the dependent variable. At Level-1 of the equation we have: Language ij = <unk> 0j + r ij, and at Level-2: <unk> 0j = <unk> 00 + u 0. From the intercept-only model outputs, we calculate the intra-class correlation (ICC), by using the following equation:. The results were: a <unk> 00 = 10.61 (p =.000) and a <unk> 2 = 35.68, so that: 10.61/(10.61 + 35.68) = 0.229. This means that 23% of the variance in English language use can be explained from between-tract factors (i.e., social-structure phenomena). As per Hox (1995), the ICC is "the proportion of the variance explained by the grouping structure in the population" (p. 15). Since ICC measures the proportions of variance in language use between-tracts, we could infer that almost one fourth in the variance of language use can be explained at the macrolevel. Since <unk> 00 is statistically significant and the ICC is greater than 0, we conclude that a multilevel model is not only useful, but it is necessary in modeling English language use aged Mexicans. After justifying the use of the multilevel linear model, we specified the following "full equation" at Level 1: In the Level-1 equation, Language ij represents the predicted score on the English Language use Scale; i and j refer to the ith HEPESE person in jth tract; <unk> 0j is the intercept in jth tract; <unk> 1j through <unk> 11j are the 11 average slopes for the individual-level variables, in jth tract; and r ij is the error term for the ith HEPESE person in jth tract. At Level-2, our abbreviated equation is:... and so forth until we reach <unk> 11 : In the Level-2 equation, <unk> 00 is the "king" intercept of the full model, <unk> 01 is the intercept of percentage of Mexican, <unk> 02 is the intercept of population density, <unk> k0 (kth ranging from 1 to 11) are the direct effects of the individual-level factor on the English Language use Scale, <unk> k1 are the indirect effects of percentage of Mexican on the microlevel slope, <unk> k2 are the indirect effects of population density on the microlevel association, and u 0 is the error measurement for all intercepts. No other Taus are included, thus, CLIs are assumed to be similar across all tracts. The "mixed" equation between Level-1 and Level-2 would look something like: After specifying the above equation in HLM, we obtained the results displayed in Table 2. The order of variables is prioritized as per our research agenda. Please note that although much could be infer from the results, we focus our interpretations on the contact variable and the moderating effects of co-ethnic concentration. As expected at the microlevel, we find that as the level of contact with Anglos increases, the use of the English language is greater (<unk> 100 =4.86; <unk> = 0.00). That is, Mexican Americans are more likely to experience a language shift when there is high contact with Anglos. Thus, our findings demonstrate that there is a positive microlevel relationship between the degree of language shift and the level of contact with Anglos. And, in answer to our first research question, we find that the degree of contact with Anglos does influence a Mexican's language shift from Spanish to English. Although the confirmation that contact with Anglos matters is important, our main focus is on understanding the effect the tract's level of Mexican population has on the microlevel slope. We observed that, as the level of Mexican concentration increases, the effect of the contact with Anglo factor on English language use weakens (<unk> 110 = -3.88; <unk> = 0.00). The increasing co-ethnic concentration reduces the speed by which the microlevel positive slope between contact and language rises. This finding shows that we are unable to falsify our hypothesis that a tract's Mexican population concentration significantly diminishes the effect of contact with Anglos on English language use. Of notable importance is our findings that being FB (<unk> 40 = -5.81; <unk> = 0.00) reduces English language usage, while having a ninth-grade education and above (<unk> 60 = 3.90; <unk> = 0.00) is related to higher levels of English language usage. The mean scores for English language usage by generation type is as follows: G1 = 23.3, G2 = 28.9, G3 = 30.8, and G4= 32.1. Although the pattern makes it clear that those who are of a latter generation on average use more English, we note generational status is not predictive of language shift (<unk> 111 = 0.00; <unk> = 0.31) in our model. In sum, our findings demonstrated that the ethnic context measure has a moderating effect on the microlevel relationship between English language use and contact with Anglos. In answer to our most substantive question, we find that the macrolevel percentage of co-ethnic concentration moderates the microlevel relationship between English language use and contact with Anglos. We interpret these findings as evidence that co-ethnic contexts are capable of offering the social and economic resources for language resistance. We interpret this CLI as a signal that aged Mexican's Spanish language capital in highly co-ethnic concentrated areas provides them with the ability of abstaining from experiencing a language shift. --- Discussion In this study, we have demonstrated that the direct relationship between the amount of contact with Anglos and aged Mexicans' English language use is moderated by the co-ethnic concentration. As the presence of co-ethnics increases, the direct microlevel relationship weakens. The study of Spanish language use is complex, since the maintenance or shift of a language depends on different factors (e.g., Bills, Hernández-Chávez, & Hudson, 1995). In our study, we interpret the tract-level co-ethnic effect as signaling that when the social resources for retaining the mother tongue are present, many aged Mexicans, despite their level of contact with Anglos, will have the ability to resist shifting to only English language use. Several limitations came to light during this study. In particular, we are making several assumptions on how the macrolevel percentage of Mexican is causally related to the microlevel relationship between contact with Anglos and English language use. For example, it is possible that as the numbers of Mexicans increases, the availability to learn, practice, and thus use more English is diminished. Unfortunately, the questions from our database do not directly ask why individuals use English over Spanish; thus, we were unable to dissect whether individuals are choosing to either abandon Spanish, or shift to English language use in relation to English language availability. Another limitation is that our dependent variable only measures the participants' self-report in English language use, and does not ask the motives or perceived causes of doing so. Therefore, future research should seek out to ascertain how minorities assigned volition factors in their language use decisions. Since the scientific accuracy of indirect and nontheory-driven tract measures remains uncertain, future research should focus on minorities' subjective evaluation of co-ethnic cohesion in their "neighborhood." Notwithstanding the limitations, we believe our study is unique, insightful, and substantively contributes to the literature by providing support for the contact hypothesis and adding that co-ethnic concentration may play a role on how microlevel processes evolve in language shift. This project compliments previous work on language shift among Mexicans in the United States (e.g., Mart<unk>nez, 2009) by enhancing knowledge produced from studies on young participants (Lutz, 2006;Potowski, 2004) and older adults (e.g., Burr & Mutchler, 2003). Our study helps expand scientific literature dedicated to language loss and/or maintenance among minority populations. The investigation makes a substantive contribution to social science research by frame the research project from the point of view of those who abandon their mother tongue out of necessity and not from an inborn volition. We give voice to this underrepresented group by pointing out that socially coercive process are born when dominant groups instrumentalize the labeling of languages as either superior or inferior to retain power (Kroskrity, 2006)making the dominant group's hegemonic determination possible as they subjugate minority groups' native tongue through legal and cultural mean. By using this "conflict" theoretical position we add to academic discourse on how and why aged Mexicans resist the adoption of a new language. By using this theoretical frame, our study explores how social environments affect language resistance and advances the idea that Mexicans are not equally equipped, either at the individual and/or context level, to retain Spanish when/if they desire it. This study is important because it gives voice to the views of those who may have felt theirs, or their ancestors' abandonment of their mother tongue, traditions, and thus, their identity was the product of coercion (see Portes & Schauffler, 1994). Language resistance needs not be seen as a revolutionary behavior disrupting the equilibrium of a homogeneous society. Language resistance could instead be understood as one of many factors contributing toward the "heterorization" of the U.S. society-where, perhaps in the future, the desire for monolinguism will be an archaic social value. --- Biographies Carlos Siordia obtained his PhD in sociology at Texas A&M University, TX, USA. He is the current project director for the Hispanic Established Populations for the Epidemiologic Study of the Elderly (HEPESE) and is an assistant research professor at the University of Texas Medical Branch. His research interests range from using quantitative methods (spatial and hierarchical) for modeling human behavior (like language use) to data science (i.e., production of quantitative knowledge) in the study of inequality and health in minority populations. Mar<unk>a E. D<unk>az received her EdD in curriculum and instruction, with a specialization in bilingual studies from the University of Texas at Brownsville, Brownsville, TX, USA. Dr. D<unk>az is a lecturer at the College of Education, University of Texas at Brownsville. Her research interests include biliteracy, family literacy practices, and digital literacies.

In this study, we investigate individual-level language shift in a population of Mexican origin Latinos/as aged 65 and up. By using data from the Hispanic Established Populations for the Epidemiologic Study of the Elderly, we investigate their English language use as the dependent variable in a hierarchical linear model. The microlevel independent continuous variable is their level of contact with "Anglos"; the macrolevel continuous independent variable is the percentage of Mexicans in tract of residence. After accounting for their generational status, other microlevel social and health covariates, and tract-level attributes, we found a direct relationship between contact with Anglos and a "shift" toward more English language use, where as co-ethnic concentration increases, the influence of contact with Anglos decreases. We frame this article with a discussion on language shifting, and explain how co-ethnic concentration may provide the resources for engaging in a language resistance.

Introduction Depending on the cause, the literature assumes a basic division into two types of childlessness. One is the childlessness conditioned by medical factors (involuntary childlessness) and the second is the lack of children as an eff ect of a conscious choice of lifestyle -the so-called voluntary childlessness (Kalus, 2002). Data collected by the World Health Organization suggest that around 10-15% of women at --- Wojciech <unk>wi<unk>tkiewicz, Sylwia Wac<unk>awik Poland the reproductive age who desire to have a child, experience primary or secondary infertility lasting longer than 5 years (Mascarenhas et al., 2012). On the other hand, researchers studying contemporary family changes also point to a growing trend of childlessness by choice (Slany, 2002). Th eories concerning human development in late adulthood and old age are usually based on the assumption of a fundamental role of typical life events, related to family formation and raising children -and, later on, grandchildren. In his concept of psychosocial development, Erikson (2002) assumes that the motive power of the human maturing is a psycho-sexual pursuit of procreation. In this context, the lack of children, especially unintentional, constitutes a threatening factor, because it limits the creative possibilities of the individual and deprives them of the experience of procreation. Childless persons can meet with negative social reactions: stigmatization or exclusion. Lampman and Dowling-Guyer (1995) proved that childlessness is generally evaluated as negative, but the disposition of the respondents is much more pejorative towards childlessness by choice, whereas unintentional childlessness meets also with sympathy and understanding. Th e negative valuation of childlessness in marriage is an expression of the strength of pro-family and pronatalistic values. From the point of view of the sustainability of the social system, it is important for the social situations deviating from traditional norms of family life -if their elimination is not possible -to be included in the symbolic universe in force, so that they will not pose a threat to the current standards (Berger, Luckmann, 2010). Normalizing cultural models are produced (Szacka, 2003, p. 78), which indicate how to think, feel and act towards the troublesome social phenomenon. Th ese standards are transmitted in the process of secondary socialization (Szacka 2003, p. 138), which is remarkably intensive towards persons who take up non-typical social roles. Th e process of the neutralization of non-standard institutions may take on two forms: therapy and nihilation (Berger, Luckmann, 2010). Th e aim of the therapy mechanism is to defi ne non-standard situations in socially approved categories. It leads to defi ning childlessness as a lack, failure in a social gender role and contributes to the medicalization of this phenomenon. In the process of therapy it is assumed that not having children is always unintentional. Th e proper way of playing the role of a childless person is standardized by psychological conceptions of a crisis situation and the course of the process of adaptation to childlessness, (Kalus, 2002, p. 24). Th e nihilation, in turn, relies on negating the reality of phenomena or their interpretation and conferring on them a cognitive status that cannot be treated seriously (Berger, Luckmann 2010, p. 168). Nihilation is applied towards people who are childless by choice, because they reject the fundamental assumptions that each married couple should have children and each adult human being, especially each woman, desires to have children. --- Research method Th e aim of the presented research was an in-depth study of childlessness as a life experience as well as the detection of varying cultural models of experiencing childlessness in marriage. A qualitative research model in methodological terms of the humanistic coeffi cient was deemed to be the most adequate for the chosen research purposes (Znaniecki, 1988). A tentative hypothesis was proposed: among persons who experienced marital childlessness for diff erent causes, there will be diff erences in the defi nition of their own life situation. In accordance with the ideas of the grounded theory methodology, it was deemed appropriate to limit the initial research conceptualization to broad research issues in order to remain open to emerging empirical data (Konecki, 2000). In order to obtain adequate empirical data, problem-centered interview was utilized (Witzel, 2000, p. 1). Purposive sampling method was selected. Th e respondents were childless persons over the age of 50, who live or have lived in a formal marriage or informal partnership with a person of the opposite sex for at least 5 years. Th e empirical data was analyzed using the constant comparative method and the negative case analysis (Konecki, 2000). Th e sample encompassed 21 persons: 9 men and 12 women. Th e median age was 65 (range 52 to 85). Th e median length of relationship was 25 years (range 6 to 55). In most cases, physiological factors were the objective cause of childlessness. Th is concerned 12 respondents. For the remaining 9 persons the lack of children resulted from their lifestyles. --- Findings Th e analysis of the research material allowed for the isolation of two basic orders of defi ning the situation of childlessness in marriage. Th e order of motivation refers to the subjectively conceived reason for childlessness and it includes six categories (Table 1), whereas the order of evaluation stands for the emotional and moral valuation of having children -and it encompasses four categories (Table 2). In their individual reasons, the respondents usually employed multiple categories of motives, and sometimes also multiple categories of childlessness evaluation. --- Personal defi nitions of the causes of childlessness Table 1. Ways of defi ning childlessness in marriage according to motivation --- Causes of childlessness Number of indications --- Physiological childlessness 12 Postponed childlessness 7 Childlessness by God's will 6 Childlessness by choice 5 Childlessness caused by illness 3 Childlessness for economic reasons 3 Th e most frequent causes of childlessness were various physiological reasons. Th ese respondents were mostly persons who desired to have children, were aware of the non-normativity of their situation and received treatment for their reproductive diffi culties. Among these respondents, however, there were also persons who underestimated the objective causes of childlessness, drawing attention to their choice in the form of discontinuation of treatment. Th e statements of the respondents also indicate a defi nitely negative evaluation of infertility. It is viewed as a trait that lowers the value of a woman or man and requires compensation. When speaking of their infertility, the respondents used depreciating terms: A bull without balls (m, 55, se, ph1 ); Unfi t for children (f, 77, pe, ph). Th is interpretation is consistent with the traditional understanding of the function of family and gender roles. Th e second most frequent interpretation of the cause of childlessness was the postponement of the decision to start a family. Th is defi nition of the situation was cited mainly by men (5 persons). Th e sex of the respondents turned out to diff erentiate the subjective motives for postponing the decision to have children. For the men, postponing the decision to get married was a result of reluctance to start a family and take up the responsibilities connected with that -or a result of diffi culties with becoming independent from their parents, e.g.: Because I was like, as they say, they called me Casanova. I was more in the parks, to have fun there, and stuff (...) and this bachelor's life of mine, I liked it, having fun, those parks (m, 66, ve, po). In the case of two women the postponement of the decision was associated with achieving educational ambitions and involvement in professional work. Another relatively frequent interpretation of the cause of childlessness was seeing it as a result of fate, the will of a force majeure, e.g.: God didn't give me children, because there was a reason (...) I always say that someone decides for us. Everything in life is for a reason, and we have no infl uence on that (f, 54, se, ph). Th is category was oft en cited by persons who were childless for causes evaluated as independent of their will. In the studied sample, those were exclusively women. Another version of this explanation was seeing childlessness as an eff ect of the malice of fortune, an incomprehensible and unfair judgment of God: I always say that I wanted to have children (...), but the Lord God refused me this, this pleasure (f, 80, se, ph). Th e foundation of the non-normative lifestyle upon the socially-propagated faith in God's providence gives it a very strong legitimization. Th e female respondents subscribing to that interpretation believed in a traditional system of values -for each of them family was a dominant value, and having children was very meaningful. Th e next category of having no children is personal choice. Th is cause was most frequently cited by the persons who consciously planned for childlessness, but also those who discontinued the treatment of infertility. Th e reasons for choosing childlessness in the studied sample are consistent with the fi ndings of a number of previous studies (e.g.: Agrillo, Nelini, 2008). Th ese were: a sense of incompetence, a fear of doing harm to the child, a fear of responsibility, no felt need, the infl uence of family experiences, economic reasons, professional ambitions or focusing on a high quality of relationship. Th e people who were childless by choice had a strong sense of the uniqueness of their lifestyle, at the same time, however, they usually did not see it as deviating from the norm. Th ey opposed the stiff ness of the rule of having children, claiming that a complete freedom of choice should operate in that matter: Apparently, for me this imperative of motherhood wasn't dominant. For somebody, this is unimportant. Never in my life have I said to anyone: Why do you have a child? (...) I think that every human being should have this freedom of decision (f, 69, he, ch). In the studied group there were three persons whose lives were subordinated to an illness or disability and in their interpretation this was the fundamental cause of their childlessness. Not having children was understood by these persons as a consequence of a broader health issue, as an element of adaptation to life with an illness. Th e ill persons are of the opinion that in their life condition the lack of children was necessary, because their health would make it impossible to raise a child. Th e respondents pointing to this category of explanation have a high sense of uniqueness, otherness of their lives. In the case of a situation of illness, childlessness is, on the one hand, associated with a sense of regret and loss, but it is also rationally considered a desired norm in the given situation. Th ese persons feel that they are victims of external circumstances. Th e last of the identifi ed categories of subjective causes of childlessness is a lack of children for economic reasons. In the statements of the respondents there were two kinds of material factors interpreted as causes for childlessness. Th e fi rst one is the fear of lack of fi nancial measures to provide for the children due to a low economic status, and the second is the reluctance to limit one's fi nancial possibilities as a consequence of expenses for the child, when one's consumption aspirations and status are very high. Th e respondents citing economic causes were strongly concentrating on the insuffi ciency of fi nancial measures and perceived the material needs of the child as very substantial, lowering the desired standard of living. In their evaluation, this cause was seen as resulting from fate. --- Emotional and moral evaluation of childlessness Th e second basic order allowing for interpretation of the narrations of the respondents was the emotional attitude to their childlessness and the value of having children in their individual hierarchy of values. Based on this order, four ways of evaluation can be identifi ed (Table 2). --- Table 2. Ways of evaluating childlessness in marriage --- Order of evaluation Number of indications --- Reconciled childlessness 9 Preferred childlessness 7 --- Unreconciled childlessness 5 Burdensome childlessness 4 Th e most frequently cited evaluation of childlessness is becoming reconciled with it. It is the standpoint of the persons who perceived their lack of children as a result of external factors. In their lives, these persons were faced with the necessity to modify their reproductive plans. At that time they mostly felt regret because of not having children, but they worked through their situation and became reconciled with it. Th ey do not exhibit a sense of shame or wrong regarding childlessness. In that group, childlessness in marriage is evaluated in concordance with the social norm as a negative phenomenon. Th e interpretation of the personal situation in the categories of norm, however, varies according to the importance attached to having children. Th e persons for whom parenthood was very meaningful evaluated their family situation as abnormal. Th is resulted in a strong need to compensate for the lack of norm: Th e main reason is, well, unfortunately, a stroke of fate (...) whether you're sorry or not, you have to make a decision and not brood over it forever (...) well, one simply needs to accept this state, you can't pity yourself (f, 76, he, ph). Th ese persons engaged in helping their family of origin and sought alternative forms of meeting their need for caring, such as taking care of the children of their siblings or working with children; whereas the respondents for whom procreation was not a fundamental value did not have the feeling that their family life deviated from cultural norms. Th ese persons attached more importance to the quality of life in a marriage union than to having children from it: I know also such who are already over 50 and they haven't been married even once, so with me it's not yet as bad, because I've had a wife (m, 66, ve, po). Also, the respondents who perceived their childlessness as a result of living with an illness, presented an attitude of reconciliation with not having children: I became reconciled with it, I don't deplore that, because I told myself that I don't have to despair over this and that's it (...) with my illness that's for the better that I didn't have children (f, 65, se, il). Th e second most frequently mentioned evaluation category of lack of children is the preference of childlessness. Th is evaluation is applied by the respondents who consciously chose childlessness or postponed this decision. Th e persons preferring childlessness evaluate it in a very positive way. Th ey think that the lack of children brought to their lives a pronounced majority of advantages. Th ey do not experience regret because of childlessness, and without exceptions they have a positive attitude towards their decision. Th ey perceive not having children in the category of freedom from limitations and an opportunity for self-realization. Th ey are satisfi ed with their way of life and their decision. Th ey describe their life as interesting and active and they are convinced that they owe its shape to childlessness, because parenthood would have deprived them of the means to pursue all their passions. In the experience of these persons, there is no space for a sense of guilt or inadequacy resulting from the non-typical form of their family life. Th ey present a non-standard, postmodern way of thinking about family, in which children are a relative value and do not constitute a necessary fulfi llment of a mature marriage: I was a quality manager in a large enterprise in a construction corporation in Gliwice (...) all our free time were trips (...) theater, meetings (...) I think that with a child I can't imagine something like that (f, 69, he, ch). Th e persons who, in turn, postponed the decision to have children, think that thanks to childlessness they gained a pleasant, carefree young age and quiet and freedom in the old age: I was like a free bird, I liked that (...) And if I had children now, I would have to strive for these children, for grandchildren, and it wouldn't be enough because of the paltry pension (m, 66, ve, po). Th e persons who have never become reconciled with the lack of children present an opposite evaluation of the situation. Th ose are the respondents who wanted to have children very much and who regarded the impossibility of procreation as a result of their own actions or unfavorable fate. Th ey had a decidedly negative attitude to childlessness as a phenomenon. Th ose respondents experienced strong sadness, anger at themselves and a sense of guilt. Th e following statement is an accurate illustration of such attitude: I have never concealed the fact that I miss this child, I have never been like this to speak somewhere with the family: I don't need this, what is it for, it's only a nuisance, in order to build up my person as so happy, because I wanted that. I never say that, that it was my choice. I don't believe that somebody can choose like this, not to have children, it's deceiving oneself, everybody needs a child (f, 54, he, po). For them, the deviation from the norm of starting a family is combined with a sense of poverty of life and regret: As they say, a woman is unfulfi lled. Likewise about a man they say that he should build a house, beget a son and plant a tree. So I am also unfulfi lled as a woman (m, 55, se, ph). Th ese respondents had a strong need to compensate for the lack of children. Th ey demonstrated a strong commitment to infertility treatment and also to helping their family of origin, work with children or other caring behaviors. It seemed that these persons had not worked through their situation psychologically, and that blocked the possibility to become reconciled with their childlessness for them. Th e last of the identifi ed categories of evaluating childlessness is a sense of a painful burdensomeness of this situation. Th e persons presenting this attitude concentrate on the negative consequences of not having children such as the lack of company, help and care. Th ey represent the type of attitudes characterized by a certain instrumentality towards the child as a value: I always used to think that I would be healthy, that I am healthy, why a child, I will make it on my own. And now, aft er a brain stroke, paresis, a child would help me, or do the shopping. Now I have to go by myself, now one regrets it (m, 55, pe, po). Th is defi nition of family situation is applied by the persons who decided not to have children in their young age, but aft er a time they started regretting it -two men whose heath situation became much worse and a woman overburdened by the need to care for her parents. Th e respondents did not want to have children because they considered them an unnecessary nuisance. For the most part of their lives they were very content with childlessness, and did not perceive the lack of children as negative or burdensome. In the case of the men, only the change of their health situation turned out to be of importance, and thus the intensity of the felt regret is very high. Th e lack of full self-reliance and the lack of care caused an overestimation of the signifi cance of having children. Th e memory of a carefree youth has faded and at present the sense of the burdensomeness of life without children dominates. As for the woman, this experience is much less severe and she still experiences life satisfaction because of lack of children. --- Conclusions Th e way of experiencing childlessness in each individual case results from a confi guration of the employed order of motivation and evaluation. It seems, however, that the order of evaluation has a fundamental meaning for the entire functioning of the respondents. It is relatively independent from the objective cause of childlessness. When defi ning and explaining their childlessness, the majority of the respondents referred to the traditional symbolic universe, based on Christian principles and collectivist values concerning the child as a desired and expected goal of family life. Th e dominant view of family is an institution based on a formal relationship of a woman and a man, called to realize the reproductive function. Th is norm was referred to even by those respondents who did not attach much importance to having children and allowed the deviation from this norm due to signifi cant external factors. Th e vast majority of the respondents successfully acquired in the process of socialization models of experiencing childlessness based on the mechanism of therapy. What dominated was the acceptance of childlessness, in accordance with the psychological model of the process of handling this crisis, as well as a negative evaluation of childlessness as a phenomenon. Th e sense of regret or burdensomeness of not having children also conforms to cultural messages concerning the consequences of childlessness ("In old age there won't be anybody to hand you a glass of water"). When it comes to postmodern categories of recognizing childlessness, interpreted as a choice, method of self-realization or development of individuality, they were only employed by the persons who deliberately chose childlessness. It is worth mentioning, however, that the respondents in this study were persons who made their reproductive decisions in the sixties and seventies of the twentieth century.

Childlessness in marriage is still socially defi ned as otherness. It is a deviation from the current cultural standard and -as a consequence -it requires explanation and legitimization, especially when the otherness is deliberate. Th e aim of the conducted qualitative research was to establish personal defi nitions of the situation of childlessness in marriage applied by persons aged 50 and older. Th e study focused on cultural patterns of experiencing and validating this non-standard lifestyle. 21 interviews centered on this issue were conducted. Th e analysis of the obtained data showed that the most important variables determining the model of experiencing childlessness in marriage by the subjects were the causes of childlessness and the signifi cance of having children in their individual hierarchy of values. In each individual case, the way of experiencing childlessness results from the confi guration of the recognized order of motivation and evaluation. It also seems that the order of evaluation is signifi cant to the entire functioning of these persons. It is relatively independent from the objective cause of childlessness.

Introduction Health information technologies, like electronic patient records, have for several years been a crucial part of clinical practice. Currently, we also witness to the may benefit from using technologies (typically video conferences) in their professional development and learning. Technologies may be effective for learning purposes and represent a potential for developing new professional communities (for example, Newman et al, 2009). However, technologies can only be successful when they make sense within the existing social relations which they are to function. This means that translation or re-invention of technologies into everyday context of use is needed (Webster, 2002;Allen, 2012). This study is situated within the research field concerning the introduction and use of ICT in health care and the aim is to contribute to an enhanced understanding of mediating technologies in health care. Several authors have addressed how the introduction of ICT in health care has a profound effect on the organisation of health care (Webster, 2002;Heath et al, 2003;Timmermans and Berg, 2003;H<unk>land, 2012;Lyngstad et al, 2013;Melby and Helles<unk>, 2014), professional practice and patients' experience of illness (Oudshoorn, 2011), and how implementation processes are challenging because of the complexities of health care (Robert et al, 2009(Robert et al,, 2010)). Few studies that address introduction of ICT in health care have looked at e-messages, video conferences and net-based discussion forums. As far as we know, none of these have investigated in detail the work involved when users negotiate how to communicate and collaborate via these new technologies. In this article, we explore how people relate to and experience different forms of mediating technology within health care. When different types of technology are introduced in interaction situationsin addition to, or replacing, the physical face-to-face meetingthe technology both enables and impedes different forms of interaction. People interpret and negotiate the new possibilities and limitations offered by the technology and by this create new practices. Drawing on Goffman's (1972Goffman's (, 1974Goffman's (, 2005) ) perspectives on social interaction, we argue that people work to maintain a more or less stable order for interaction in social situations. When technology is introduced to mediate interaction this creates a new frame (Goffman, 1974) for understanding and making sense of the specific situation. This new framework implies new ways of organising and making sense of experience, and requires work by the participants in the interaction. Health personnel negotiate framehow to understand and participate in mediated interaction. In this article, we investigate health personnel's work to make sense of technology-mediated interaction in health care. We analyse the impact of these new frames on interactions and the efforts by participants to change and sustain them (Goffman, 1974). In order to understand technology as a mediator, we follow Wyatt et al (2008) who draw on Latour's (2005) distinction between an intermediary and a mediator. An intermediary is someone (something) who transports meaning or force without transformation. Mediators, on the other hand, transform, translate, distort and modify the meaning of the elements they are supposed to carry (Latour, 2005, p. 39). We acknowledge that the technologies in our study have the role of mediators; they have the potential to change the meaning of the information being transmitted, and alter the relationship between the communicating partners. Our article builds on material collected in two different research projects in Norway. Both projects have addressed the introduction of ICT in municipal health care. Project A has studied public health workers' use of video conferencing and net-based discussion forums for competence development, while project B has looked at the introduction of e-messages in home care and at general practitioners (GPs) as a means of communication. Study designs, settings and methods are described in more detail in the Methods section. Even though the technologies are different, they have in common that they make users switch from face-to-face interactions, or telephone conversations, to different types of technology-mediated interactions. Hence, users encounter many of the same challenges and must learn to collaborate by the use of technology. --- The Interaction Order, Frame and Mediated Interaction Goffman's concern was to study the elementary processes that are fundamental to all social life, and by this establish a new and specific field of study within sociology, namely the interaction order (Jacobsen and Kristiansen, 2004). He was interested in focused gatherings/encounters, and it is the myriad of these encounters, he claims, that constitutes the interaction order. Encounters are examples of what Goffman (1972) identifies as focused interaction, that is, what happens when people agree to maintain a single cognitive and visual focus within a limited time frame, for example, in a conversation. The interaction order refers to the order and structure that are to be found in social situations where people meet face-to-face. In these meetings, an order that follows its own rules develops, as well as structures that ensure exchange and communication between people in these situations. Even if Goffman was mainly concerned with interaction that takes place face-to-face we will argue that his perspectives and concepts are relevant for understanding technology-mediated interaction. We follow Pinch (2010) in arguing that technologies are present in the interaction order and play a crucial part in how interaction is mediated, as the interaction order is 'embedded within, mediated by, and staged by material circumstances and mundane technologies' (p. 419). The self is, according to Goffman, first and foremost motivated by a desire to be well-regardedand encounters offer possibilities for eliciting that regard (Best, 2005). Face-to-face interaction happens in and because of the presence of others, where visible signs of orientation and involvement in the situation are glances, gestures, positionings and verbal statements that people feed into the situation, either intended or unintended (Goffman, 2005). In On face-work, Goffman (2005) studies the face-work people engage in to maintain their own face and the face of others in the interaction. Face is understood as the positive and symbolic value a person attaches to himself and others attach to him. Further in this essay, Goffman underlines how important it is for a person to save one's own face and describes that the safest way of preventing threats to his face is to avoid contacts where it is likely that this could occur. Goffman is mostly concerned with immediate encounters, but also mentions mediated encounters (for example written expressions). However, he emphasises that it is through direct personal contact that the unique informational conditions prevail and where the importance of face is particularly visible. When the possibility of spoken interaction arises, a system of practices, conventions and rules come into play that guides and organises the exchange of messages in the situation. For example, when a person sends out a message, the other participants in the encounter are obliged to show that the message is received and that the content is either acceptable for all involved or that it can be contradicted in an acceptable manner. In Alienation from interaction, Goffman (2005) emphasises again the unique qualities of a conversationa conversation creates a world and a reality for the participant where there are also other participants present. Further, he points to the importance of the many small signals in the interaction: 'When individuals are in one another's immediate presence, a multitude of words, gestures, acts, and minor events become available, whether desired or not, through which one who is present can intentionally or unintentionally symbolize his character and his attitudes'(p. 114). Goffman (1974) did not claim that interaction happens in a vacuum, in fact he devoted a 576 page essay to explore what surrounds an interactionto frame. Frame refers to the relational dimensions of meaning, how participants in interactions organise experience and make sense of events. It guides the response to the question 'what is it that's going on here'? Given the participants' understanding of what it is that is going on, they adjust their actions in relation to this understanding (Goffman, 1974). Goffman's concept is related to concepts like context, setting and background and concerns frameworks of interpretation, the frame participants interpret events within. Frame implies 'guided doings'participants are subjected to rules and expectations, to social appraisal of his or hers actions (Goffman, 1974, p. 22). Of particular interest to our study, is the importance Goffman makes of the struggle to make seemingly effortless actions work. He claims that competent performances are achieved through a process of negotiations 'in a cold sweat' (Goffman, 1974, p. xii). This implies that participants have to work to make sense of a situation and to negotiate its meaning and its implications for action. When new technology is introduced in interaction situations, this implies new framesnew schema for the interpretation of what it is that is going on and new efforts by participants to make sense of the situation and how to act in an appropriate manner. Several researchers have used the work of Goffman to investigate mediated communication, and in this article we will draw on a study of mobile telephone use (Rettie, 2009) in particular. One of the main arguments in Rettie's study is that we need to understand the difference between face-to-face interaction and mediated interaction, and to understand the difference between various forms of mediated interaction. There are differences between synchronous and asynchronous mediated communication; the study explores the boundaries of synchrony. Synchronous mediated communication can have characteristics that make it similar to a face-to-face encounter, as Goffman understands it, that is, an encounter where the participants cooperate to maintain focused interaction (Rettie, 2009). Even though the participants are not physically present in the same place in a telephone conversation or a video conference, they share the same time frame and a mediated co-presencea mediated encounter (Rettie, 2009). In a video conference the participants can see, hear and talk in real-time, and this gives the participants the possibility to coordinate their actions and cooperate on the interaction. In asynchronous communication, like e-messages, on the other hand, this common, shared time frame does not exist, implying that it is more difficult for the participants to coordinate their actions (Rettie, 2009). A net-based discussion forum can have characteristics in line with what we find in synchronous mediated communication, but that requires that the participants agree to'meet' in the forum on specific points of time. Otherwise, the use of netbased discussion forums can be an example of asynchronous mediated communication, where the participants are not present in the same time frame. Synchronous communication offers the participants a possibility to share an experience in real-time -'their shared practice of coordinated interaction creates the inter-subjective experience of a mediated encounter' (Rettie, 2009, p. 426). Synchronous media afford this possibility of interaction. Asynchronous media 'does not afford continuous cooperative practice and the experience of a moment-by-moment intersubjectivity in a shared social reality' (Rettie, 2009, p. 126). The participants in this type of communication have a possibility to respond to each other's expressions, but there is no cooperation in real-time. Rettie's ( 2009) study of mobile telephone use illustrates that the border between synchronous and asynchronous communication can be blurred and that it is not the technology itself that defines this border. SMS, e-mail and chat (for example via Facebook) all have the same transmission time, but expected time used for responses are very different and formed by normative expectations and user practices. E-mail was, in this study, perceived as rather asynchronous, while SMS was perceived as 'near-synchronous.' It was acceptable to wait several days for a response to an e-mail, while a response to a SMS was expected quickly. A delayed answer to a SMS was perceived as sending a meaningful message, that is, an intended delay/no answer. The study shows that the experience of mediated interaction is'shaped both by the temporal characteristics of the medium and by normative framing expectations' (Rettie, 2009, p. 436). As Hutchby (2003) says: 'there is a complex interplay between the normative structures of conversational interaction and the communicative affordances offered by different forms of technology '(p. 585). In this article we use Goffman's (and Rettie's) concepts of interaction and frame to understand the work involved when health personnel experience and make sense of the various forms of mediating technologies (e-messaging, video conferences and a net-based discussion forum). We explore the negotiations taking place in their efforts to make sense of the new framework. --- Methods This article draws on material collected in two different Norwegian research projects. Both projects used an explorative approach with open-ended interviews. This approach is well-suited for investigating the participant's own experiences of a phenomenon (Silverman, 2001). The Ombudsman for Privacy in Research (Norwegian Social Science Data Services) approved both studies. In project A, we investigated the introduction of video conferences and a netbased discussion forum for competence development of health personnel in home care (nurses and assistants in home care and nursing homes) in a Norwegian region with large geographical distances. We conducted group interviews and individual interviews with 14 of the total 19 participants in the course 'Ageing on the Internet' in 2011. The course aimed at strengthening the health-care workers' competence on elderly people and on how to use ICT for cooperation and learning at a distance. The interviews were taped and transcribed verbatim by the first author. In project B, we investigated the introduction of e-messaging between home-care services and GPs. The e-message system was developed as a module that can be integrated with the different electronic patient record systems in use in Norway, and the aim was to better facilitate (electronic) communication between healthcare workers across organisations. In 2008-2009, an e-messaging pilot was implemented in six municipalities, of which two were selected for this study. These were strategically chosen because the involved GPs and home-care nurses had the most experience with the use of e-messaging. We conducted interviews with 43 persons in total: 23 nurses, 11 GPs, 5 secretaries and 4 project managers. The interviews were a combination of individual interviews and group interviews and were conducted during 2011. The interviews were taped and transcribed verbatim by a research assistant. (Table 1) Analysis Projects A and B were developed independent of each other, but were both connected to research activity at The Norwegian Centre for Electronic Patient Records. We discovered that even if the projects had different points of departure, they bore many similarities related to how health-care personnel work to make sense of technology-mediated interaction. Within both projects a rough thematic coding of the data had already been conducted before starting on the common analysis for this article. The current analysis has been conducted in close collaboration between First author and Second author. The data were analysed using an interpretative and eclectic approach, described by Kvale (2007) as 'bricolage,' in which the aim was to generate meaning and see connections across the material. In an iterative process, we started out with discussing similarities and differences in how our interviewees had described their efforts to comprehend the technology as a tool for communication. Following the discussion, and introducing Goffman's perspectives to guide the analysis, we returned to our material to consider the suggested categories. This process was performed several times. We aimed at developing categories that were common for both projects and which, in our opinion, said something more general about technology-mediated interaction. Based on the interviewees' own expressions of their efforts to make sense of technology-mediated interaction, four categories which represented themes prevailing to a large extent in both projects, were developed. They are presented in the following section and exemplified with quotes from both projects. --- Making Sense of Technology-Mediated Interaction in Health care When new technologies are introduced in interaction situations in health care, health personnel have to figure out how to relate to these technologies and how to make sense of them in their particular context. This involves continuous work and negotiations on a detailed level. In this article, we explore and discuss this work in relation to four categories that were particularly visible in our empirical material. We bring forward health personnel's own expressions of how they relate to new technologies in these situations. First, we investigate health personnel's efforts to make sense of how to perform in a competent and professional manner in technology-mediated interaction. Second, we explore how health personnel reflect upon how and when to communicate in these situations, how they negotiate immediacy. Third, we discuss how health personnel relate to a different and limited repertoire of social cues in technology-mediated interaction. Last, we look into what kind of work needs to be performed in order to maintain a mutual commitment between actors in technology-mediated interaction. --- Performing Competence The introduction of technology-mediated interaction in addition to, or instead of, face-to-face interaction establishes new frames for addressing competence. To perform in a competent and professional manner then includes using new technology adequately. This represents work for all health personnel, and for some groups, it represents a major challenge. Some groups of health personnel in our study had never sent or received an e-mail, never opened an attachment, never used Power Point to present something and barely been on the Internet. Then being 'forced' to use video conferences and to participate in a net-based discussion forum implied a great effort. These groups of health personnel can provide good care to patients and have a strong professional identity, but are now faced with a reality where they no longer can avoid using technology to perform in a competent manner. This might represent a threat to their 'face,' as described by Goffman (2005), a threat to their positive and symbolic value. Goffman (2005) argues that the safest way to prevent threats to one's face is to avoid situations where that could happen, and some interviewees (in both projects) reacted by withdrawing from the situation. Others embraced the new situation and, through a lot of hard work, experienced to cope with the new demands in a larger extent: [For us that] has not even turned on a computer before, it's fun. To participate in the discussion forum and send e-mails and send in assignments and download and cut and paste and everything. It really is. Because that we have to be able to do, that I've been told. (Project A, Participant 1, Health-care worker) The health-care worker articulates how she has overcome a major barrier using technology and how this is experienced as something positive. She also acknowledges that using technology in these situations represents new demands that they, as health personnel, simply have to cope with. Using computers on a basic level is often regarded as something everybody copes with in our modern society, but our study shows that there are still groups where this represents a major challenge. Non-users of technology must still be reckoned with (Wyatt et al, 2002). When the first barrier of simply learning the required technical skills is overcome, work still remains on a very detailed level, when health personnel try to make sense of the new technology. In a group interview, reflecting on a discussion from a physical seminar the day before, two interviewees explain: The health-care workers articulate how they feel insecure expressing themselves in a net-based discussion forum and that they need training to do this, even down to the very detail of when to use smileys, and so on. A net-based discussion forum is often presented as an arena for informal, oral communication. However, since communication happens through writing and remains on the screen, it is difficult to withdraw unfortunate formulations, as one can do in oral communication. The health-care workers' wish for a 'delete-button,' even if it is said jokingly, shows that they are insecure regarding what should or should not be written in this forum and how this arena for communication should be interpreted. Following Goffman's (1974) concept of frame, it is unclear within what frame the communication should be understood. Is it a situation with 'oral' informal communication in a private setting, where one can'say' things that are not always completely thought through, or is it a more formal and public situation where one has to be precise in what is'said'? How do you perform in a competent manner in this setting? Some of the participants in the net-based Negotiating technology-mediated interaction in health care discussion forum also used the text from the discussions, using cut and paste, as a point of departure for the written assignment they had to hand in, adding even more uncertainty regarding type of communication. Frame implies 'guided doings,' meaning that participants in the interaction are subjected to rules and expectations (Goffman, 1974). In this setting, the rules and expectations are still unclearthe participants have to work to make sense of this situation and to negotiate its implications for action. The introduction of new technology can also imply that particular parts of what health personnel have learnt in their formal training is no longer valid regarding how to perform in a competent and professional manner. Health personnel using e-messages experienced that what they had learnt in their formal training regarding how to relate to people did not always apply to these new situations where they were communicating via technology: In training we learnt that we should be nice and write 'with best regards' and our name below. But I can see that when I send a message, my name appears at the top [of the text] as sender. So after a while I thought it was so stupid to write 'with best regards' when I saw my name anyway. So I stopped. (Project B, Participant 1, Home-care nurse) The quote exemplifies how 'traditional' ways of addressing people in formal communication is challenged with the introduction of e-messages. Moreover, here there is uncertainty regarding frame (Goffman, 1974). It is unclear if the e-messages should be understood as a 'letter' between participants in a communication, then following expected rules regarding formal communication, or if they should be understood as pieces of information intended for the patient record. Shortly after the introduction, health personnel expressed that they also used to write, for example, 'have a nice weekend' and add smileys to the messages, thus performing, as they would do in telephone conversations, before. However, after they had used e-messages for a longer period and after they had realised that the messages became part of the patients' record, they stopped adding this type of information. So, what we see is that the rules and expectations for how to perform in a competent manner in this setting are negotiated through a process that involves work on a very detailed level. --- Negotiating Immediacy Health personnel experience that technology-mediated interaction differs from face-to-face and telephone interaction in several ways. A common theme, identified among the interviewees in both Projects A and B, was how health personnel struggle to define how to relate to time in technology-mediated interaction: How and what should be communicated at what time? How long is it acceptable to wait for an answer? Rettie (2009) shows that the borders between synchronous and asynchronous communication can be blurry in mediated interaction and that expected time used for response are influenced by normative expectations and user practices. In project A, even though the net-based discussion forum affords 'nearsynchronous' communication (Rettie, 2009), it is not used in this way by the participants. They did not agree upon a time to meet and discuss in the forum, which means that the communication there is experienced as more asynchronous, less immediate and more split up: In the discussion forum I thought... it wasn't that easy because... if you post something there, it could take days before you got any response, so then you didn't get any further... [contrary to] when you talk to someone, then you get feedback at once, so it's... maybe... if you had known them a little better, then you could have sent an SMS in additioncan you check, I have sent you something, so... (Project A, Participant 3, Health-care worker) The health-care worker articulates how communicating in a net-based discussion forum is very different from face-to-face communication and that she does not receive the immediate response that you do face-to-face. Goffman (1972Goffman (, 2005) ) highlights the unique qualities of a face-to-face encounter, where people agree to maintain a single cognitive and visual focus within a limited time frame. When various forms of mediating technology replace the physical encounter, the participants need to negotiate what is an appropriate time for response. The frame (Goffman, 1974) is changing, meaning that the participants have to negotiate if this is more or less synchronous communication, implying an immediate response, or if it is asynchronous communication, where it is acceptable to wait days for a response. The health-care worker in the quotation above reflects upon her own use of this particular technology, that they could have texted each other to get a more immediate response, thus signifying the need for negotiations to take place to make sense of and use this technology. Health personnel using e-messages experienced that sometimes they had to wait for a long time for answers to the messages, even if the guidelines stated that replies should be given within three days: I experience that it may take a long time before you get an answer. I just checked, and it was 14 days back [that we sent a request]. Some answers immediately, whilst with others you have to send a new request or call them on telephone and ask if they have seen the e-message. (Project B, Participant 2, Home-care nurse) Negotiating technology-mediated interaction in health care The nurse describes how she often has to wait for a long time for a response to the messages she sends and that the response time varies between the different persons she communicates with. Also, several nurses described that with persons who normally would respond quickly, they would expect a quick response. This shows how the expected response time is formed by normative expectations and user practices (Hutchby, 2003;Rettie, 2009), implying that there are negotiations on a very detailed level between the different groups and persons who are supposed to use the e-messages to collaborate. They have to negotiate when and how to use e-messages and what is an appropriate response time. In home care, because of the often experienced long response time from GPs, they would never use e-messages for urgent matters. --- Enabling Social Cues A third category identified in our empirical material concerns health personnel's experience of reduced access to social cues in technology-mediated interaction, compared with face-to-face and telephone interaction, and how this challenges communication. Social cues refer to all the small signals people giveconscious or notduring interaction, signals that convey meaning and help people to interpret a conversation (Goffman, 2005). When introducing technologymediated communication, many of the social cues implied in seeing and hearing each other are lost (cf. Walther et al, 2005;Mentis et al, 2013), or they have to be'reinvented,' given a new shape, adapted to the technology. In both projects A and B our interviewees talk about difficulties in making communication and interaction 'flow,' because of the lack of social cues. On a general level, for example, some health-care workers talked about how they experienced it to be much easier to talk than to write. They considered the net forum to be an obstacle rather than an enabler for smooth communication in their group work discussions. More specifically, interviewees expressed that technology-mediated interaction made it more difficult to detect the communication partner's mood and to make a joke, and it was easier to be misunderstood when you had to write instead of talking. In an interview with health-care workers in project A, this is exemplified in the following: Health care worker: The disadvantages [in the net-based discussion forum] must be... maybe, that you don't see... the communication... you don't see if a person is mad or happy or... Interviewer: That you don't see the face? Health care worker: Yes. There can quickly be misunderstandings. The health-care worker describes how the fact that you cannot see each other creates possibilities for misunderstandings in the communication. The same concern is expressed by a GP in Project B, who had used e-messaging for a period of time: It's like with e-mail; if you have an unfortunate formulation, it might be comprehended completely wrong by the recipient, and it might escalate... So, sometimes it's much simpler, just picking up the telephone. (Project B, Participant 3, GP) The GP explains how e-mail and e-messages imply the possibility for misunderstandings, and that he sometimes actually prefers to use the telephone instead. The telephone is a well-established mediating technology where the fact that you can hear each other gives the communicating partners access to a richer repertoire of social cues. The introduction of technology-mediated interaction means the introduction of new frames (Goffman, 1974) for health personnel. A reduced repertoire of social cuesor a need to reinvent relevant social cuesis part of the new frames. Health personnel negotiate how to adjust their communication and actions within the new frames. Some of the health personnel in Project A, using the net forum, even asked for more training in how to behave and express themselves in the forum to avoid misunderstandingsthus implying that the lack of social cues were considered challenging. However, we do not argue that the introduction of technologies that imply less access to social cues is one-sidedly negative for interaction. In Project B, some nurses explained that actually being able to sit down in peace and quiet instead of meeting GPs or having telephone conversations with them meant that they wrote, according to themselves, more wellconsidered and precise messages. This indicates that sometimes it is beneficial for communication to switch from an interaction form rich with social cues to one with less social cues. In particular, it can be efficient when exchanging information where precision and details are important and there exists a language for describing the matter of case (for example, reporting a patient's blood pressure values) and the sender's frame of understanding is less relevant. What we see here, is that health personnel's efforts to make sense of social cues in technology-mediated interaction imply that they negotiate the frame of the interaction: does the interaction represent a formal exchange of information, Negotiating technology-mediated interaction in health care meaning access to social cues might be less relevant? Or does it represent a conversation close to what is expected in informal face-to-face interaction, meaning a reduced access to social cues might be challenging? These negotiations take place on a daily basis and involve continuous work to establish a common frame of reference. --- Establishing and Maintaining Commitment How to develop and maintain commitment to the technology and to your communication partners in technology-mediated interaction is the last category we will discuss in this article. In our empirical material we find examples of work that needs to be done and work that is done in order to establish and maintain commitment in technology-mediated interaction. The work necessary to facilitate participation and commitment also requires work by those other than the end-users, for example by management. In Project A, interviewees had ideas about how to facilitate participation and commitment in a net-based discussion forum. When asked about possibilities for improvements, two interviewees suggested: Interviewee 1: There should be some more information. Interviewee 2: Before the course, yes. Interviewee 1: I think maybe that the employer should take that part, then, because it was clear that there were some demands that people maybe didn't know about. So there should maybe be more information about how to work in groups. Interviewee 2: Online. Interviewer: How that works? Interviewee 1: Yes. Or, that you actually commit yourself even if you don't sit next to each other, that you have to commit, to go in[to the forum] and participate. (Project A, Participant 1 and 2, Health-care workers) The two health-care workers point to the need for more information, better arrangements from the employer to facilitate participation, and a consciousness raising that participating in a net forum means committing to take part. In Project A, participants also raised the issue of 'free riders' in the net forum. They argued that all participants should be active participants, and not only observe what the active actors did. This illustrates that the introduction of technology-mediated interaction creates a need for work to be doneto facilitate commitmentbut that the participants have not yet established how this should be done or who is responsible for doing it. As Pinch (2010) points out, new technologies create new interactional problems that the participants in the interaction need to solve. From Goffman's (1974) perspective, we can say that a more coherent view of the technology and the interaction, a common frame, needs to be negotiated among the participants in order to establish how it is appropriate to commit to each other within this new frame. The question of commitment was also an issue in Project B. Here we found examples of work when actors tried to make their communication partner commit to the interaction. In this project many nurses in home care were eager users of the e-messaging system, while some GPs used it infrequently. This imbalance caused problems in communication. For nurses the solutions could be to make a telephone call to the GP office, but also formal systems were developed to deal with discrepancies from the communication guidelines: Some [doctors] are really quick, while others use a lot of time. And then you have to send a message, a so-called 'exception message.' And if you don't receive a reply, you have to call the secretary, and you might be told that the doctor isn't present, or is somewhere else. And then you get your explanation on why you didn't receive a reply. (Project B, Participant 2, Home-care nurse) Nurses in Project B would occasionally send discrepancy reports to the municipal administration. In this respect their work represents an educating effort vis-à-vis the less committed users, thus being part of the negotiations taking place in order to establish rules and expectations within a new frame (Goffman, 1974). There is much work needed to make actors commit to participate in technology-mediated interaction, including promoting the use of technology in an appropriate and timely way. In this respect we see links back to the categories on performing competence and negotiating immediacy. Being committed to the interactionand being a 'good' communication partnermeans first of all starting to use the technology, and thereafter negotiating what should encompass 'correct' use. --- Conclusions The health-care sector is increasingly faced with different forms of technology that are supposed to mediate interaction, thus fully or partially replacing face-toface encounters (like for example many telemedicine initiatives). The rationale behind the introduction of these technologies is often to reduce costs and to enhance efficiency, but potential gains cannot be expected to be automatically Negotiating technology-mediated interaction in health care realised. As we have shown in this article, there remains work to be carried out. Technologies that can be understood as mediators influence the meaning of the elements they are supposed to carry (Latour, 2005), implying that face-to-face interaction cannot simply be replaced by technology-mediated interaction without involving active sense-making work from the participants. In this article, we have introduced and discussed four categories of work health personnel perform to make sense of technology-mediated interaction. Health personnel work to make sense of how to perform competently in technology-mediated interaction. They struggle to learn the basic technical skills, to define what type of communication this represents and to apply and adjust what they have learnt in their training within the new framework. It is unclear what frame (Goffman, 1974) the communication should be understood within, and health personnel negotiate the rules and expectations for how to perform in a competent manner in this setting. Further, health personnel negotiate how to relate to immediacy in technology-mediated

The health-care sector is increasingly faced with different forms of technology that are introduced to mediate interaction, thus fully or partially replacing face-to-face meetings. In this article we address health personnel's experiences with three such technologies, namely: electronic messages, video conferences and net-based discussion forums. Drawing on Goffman's perspectives on interaction and frame, we argue that when technologies are introduced to mediate interaction, new frames for understanding and making sense of situations are created. These new frames imply new ways of organising and making sense of experience, and require work by the participants in the interaction. In this article, based on interviews from two Norwegian research projects, we investigate health personnel's work to make sense of technology-mediated interaction in health care. We discuss this work represented in four categories: how to perform in a competent manner, how to negotiate immediacy, how to enable social cues and how to establish and maintain commitment. Concluding, we argue that the introduction of mediating technologies redefines what is considered up-to-date, 'good' health-care work and challenges health personnel to change (some of) their work practices and moves, as a result, far beyond simple interventions aimed at making work more efficient.

like for example many telemedicine initiatives). The rationale behind the introduction of these technologies is often to reduce costs and to enhance efficiency, but potential gains cannot be expected to be automatically Negotiating technology-mediated interaction in health care realised. As we have shown in this article, there remains work to be carried out. Technologies that can be understood as mediators influence the meaning of the elements they are supposed to carry (Latour, 2005), implying that face-to-face interaction cannot simply be replaced by technology-mediated interaction without involving active sense-making work from the participants. In this article, we have introduced and discussed four categories of work health personnel perform to make sense of technology-mediated interaction. Health personnel work to make sense of how to perform competently in technology-mediated interaction. They struggle to learn the basic technical skills, to define what type of communication this represents and to apply and adjust what they have learnt in their training within the new framework. It is unclear what frame (Goffman, 1974) the communication should be understood within, and health personnel negotiate the rules and expectations for how to perform in a competent manner in this setting. Further, health personnel negotiate how to relate to immediacy in technology-mediated interaction. Immediacy is experienced as something very different in technology-mediated interaction compared with face-to-face interaction, where you get immediate response, and health personnel negotiate how and what should be communicated at what time. The expected response time varies, even between different persons they communicate with, indicating that expected response time is influenced by normative expectations and user practices (Hutchby, 2003;Rettie, 2009). The third category in our material concerns how health personnel deal with less access to social cuesor the need to reinvent social cuesin technologymediated interaction. The reduced access to social cues can be perceived as a challenge, or an advantage, depending on what type of interaction the technology is supposed to facilitate. When the technology-mediated interaction is supposed to flow almost like informal face-to-face interaction, the reduced access to social cues represents a challenge for the participants. On the other hand, when the technology-mediated interaction represents a formal exchange of information, the reduced access to social cues is sometimes seen as an advantage, keeping the interaction focused and precise. Lastly, we have looked into the work involved to establish and maintain commitment in technologymediated interaction. There is work needed from both participants in the interaction and from managers and others to facilitate participation and sustain commitment in technology-mediated interaction. This work includes how to deal with the problem of 'free-riders,' and how to handle cooperating partners, who do not always commit to using the technology as expected. We outline three main findings from our study. First, we argue that the introduction of technology-mediated interaction implies new frames for the interaction to be understood within (Goffman, 1974). This means that health personnel work to make sense of the new situations and participate in negotiations on meaning and implications for action down to the very smallest detailsto put it in Goffman's words -'in a cold sweat' (Goffman, 1974, p. xii). Implications for technology introduction initiatives in health care include a need to acknowledge this work and give time for health personnel to understand the new situation and develop the new practices. Second, we argue that the introduction of technology-mediated interaction does not imply one new, defined form of interaction. Even within the frames of one technology, the outcome of communication and interaction can vary depending on the technology itself, user practices and expectations. Following this, our findings point towards the necessity of a complex hybrid of information and communication mediators in health care, meaning that technology introduction initiatives need to consider different types of technology for different types of communication they wish to facilitate. Third, we argue that the physical face-to-face meeting has some unique qualities (Goffman, 2005) that the participants in our study find it hard to replace, even if they commit to using the new technologies. This implies that when technology-mediated interaction is introduced, it is still necessary to include the possibility to meet face-to-face in some instances. The introduction of mediating technologies redefines what is considered up-todate, 'good' health-care work. It challenges health personnel to change (some of) their work practices and moves, as a result, far beyond simple interventions aimed at making work more efficient. Do we see the contours of a new type of health personnel rising from these demands? A thorough understanding of how health personnel make sense of technology-mediated interaction is required in order to integrate these technologies in health-care practices in a proper way. This article makes a contribution in this respect, and further research should look into how health and care are being performed in increasingly new ways. Pieter Toussaint, Trond Kongsvik and the editors for valuable comments on earlier drafts of this article. This work is licensed under a Creative Commons Attribution 3.0 Unported License. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/ --- About the Authors

The health-care sector is increasingly faced with different forms of technology that are introduced to mediate interaction, thus fully or partially replacing face-to-face meetings. In this article we address health personnel's experiences with three such technologies, namely: electronic messages, video conferences and net-based discussion forums. Drawing on Goffman's perspectives on interaction and frame, we argue that when technologies are introduced to mediate interaction, new frames for understanding and making sense of situations are created. These new frames imply new ways of organising and making sense of experience, and require work by the participants in the interaction. In this article, based on interviews from two Norwegian research projects, we investigate health personnel's work to make sense of technology-mediated interaction in health care. We discuss this work represented in four categories: how to perform in a competent manner, how to negotiate immediacy, how to enable social cues and how to establish and maintain commitment. Concluding, we argue that the introduction of mediating technologies redefines what is considered up-to-date, 'good' health-care work and challenges health personnel to change (some of) their work practices and moves, as a result, far beyond simple interventions aimed at making work more efficient.

Introduction Older adults' alcohol consumption has come under increasing popular and academic scrutiny in recent years. In the United Kingdom (UK), for example, recent media coverage has drawn attention to the 'hidden' reality of alcohol misuse among older adults (see e.g. Bingham <unk>; McVeigh <unk>). While reports of alcohol misuse among the elderly point out that the majority of older adults are moderate drinkers, popular interest in drinking practices in later life demonstrates the need for better understanding of this behaviour. Moreover, the unveiling of older adults' drinking behaviours reveals normative assumptions about alcohol and age, that drinkingparticularly when associated with loss of controlis a behaviour characteristic of youth and older adults are expected to internalise responsibility (see e.g. Griffin et al. <unk>; Johnson <unk>). Yet drinking is an important dimension of older adults' lives, for many it is a way of connecting with others, as well as with their past lives (Burruss, Sacco and Smith <unk>; Ward, Barnes and Gahagan <unk>). This paper responds to this public concern about drinking in later life through exploring how drinking behaviours in later life vary according to key socio-economic and lifecourse characteristics and how these behaviours change over a ten-year period. The analysis uses a lifecourse perspective, which considers how drinking behaviours are structured by trajectories and transitions within social contexts (Schulenberg, Maggs and O'Malley <unk>). The analysis uses data from the English Longitudinal Study of Ageing (ELSA) to investigate how socio-economic characteristics, partnership, employment and health statuses influence drinking behaviours among ELSA participants at the beginning of the survey, and how lifecourse transitions relating to health, employment and partnership are associated with changes in drinking behaviours over the ten-year period. The analyses are stratified by gender as gender identity shapes both lifecourse trajectories and transitions (Rossi <unk>) and drinking behaviours (Wilsnack et al. <unk>). By comparing men's and women's drinking over the lifecourse, we can contextualise the relationships between lifecourse events and alcohol consumption. The paper begins with an overview of existing research on drinking in later life to identify the main trends in drinking over the lifecourse and characteristics that have informed our analysis. We then present the data-set and the modelling strategy used. The empirical analysis investigates both the frequency and quantity of older adults' drinking to provide a more detailed understanding of how these two attributes of drinking behaviours change in later life. In particular, the analysis considers the importance of lifecourse transitions in accelerating or decelerating changes in alcohol consumption. --- <unk> --- Lifecourse transitions, gender and drinking in later life --- Trends in alcohol consumption and health in later life One important observation about the relationship between age and alcohol consumption is that there is no universal pattern of consumption trends with age that is consistent across time and place (Wilsnack et al. <unk>). However, the prevailing pattern in Anglo-European contexts is that alcohol consumption declines in later life (Molander, Yonker and Krahn <unk>; Platt, Sloan and Costanzo <unk>; Shaw et al. <unk>) after peaking in young adulthood (Meng et al. <unk>); and more older adults are non-drinkers compared to younger age groups (Meng et al. <unk>; Ng Fat and Fulller <unk>). This overall decline in consumption with age is found for all consumption levels, including heavy drinking (Karlamangla et al. <unk>). However, while older adults are more likely to be non-drinkers and consume less if they do drink, they are also more likely to drink more frequently; in contrast to younger people who are more inclined to concentrate consumption in binge-drinking episodes (Ng Fat and Fuller <unk>). English data on the frequency of drinking show that the regularity of drinking increases with age (Meier <unk>). Men in the oldest age group (<unk> and over) were more likely to drink five or more times a week than any other age group, and women aged <unk>-<unk> were the most frequent drinkers (Meier <unk>). The overall decline in alcohol consumption with age has meant that national alcohol misuse prevention strategies do not identify older adults' drinking as a priority for intervention (see e.g. the UK's <unk> Alcohol Strategy; Secretary of State for the Home Office <unk>). Yet this policy and research lacuna has been addressed in recent years. This has been driven by a number of factors. First, in a global ageing society the number of older drinkers is increasing (International Center for Alcohol Policies <unk>). Second, and a more fundamental concern, is that the age profile of drinking is changing along with trends in per capita consumption. This change is brought about by age, period and cohort effects, and disentangling these is not necessarily straightforward (Meng et al. <unk>). However, what is emerging from analysis of cross-sectional data is that there have been greater discernible changes in younger people's drinking compared to older age groups in recent years. For example, in the UK there was a substantial increase in per capita consumption during the latter half of the twentieth century, particularly among women (Smith and Foxcroft <unk>), which has been reversed slightly in the first decade of the twenty-first century (Meng et al. <unk>). This recent decline in drinking has been concentrated in younger age groups (Meier <unk>) as British young people born since <unk> are not reproducing the equivalent high consumption/ <unk> Clare Holdsworth et al. low abstinence behaviours of earlier cohorts (Meng et al. <unk>). However, their predecessors who had higher consumption and lower abstinence during youth and mid-adulthood are now entering later life. While these cohorts might experience the age effect of declining consumption/increasing abstinence, the overall impact of these age and cohort trends is that differentials in drinking across the lifecourse have declined in recent years (Office for National Statistics <unk>). In the UK, men aged <unk>-<unk> are more likely to drink above recommended limits than all other ages, though there is no age effect of heavy drinking for women (Meier <unk>). Turning to the health outcomes of drinking, data on alcohol-related mortality and morbidity also highlight a potential upward trend associated with drinking in later life. Alcohol-related mortality is higher among older age groups and is increasing among the elderly while stabilising and declining at younger ages (Knott, Scholes and Shelton <unk>; Office for National Statistics <unk>). Alcohol-related morbidity for older adults is also increasing. Between <unk> and <unk> in England and Wales, the number of alcoholrelated admissions into hospital for men aged <unk> and over increased by <unk> per cent and for women by <unk> per cent (Institute for Alcohol Studies <unk>). Increases in alcohol-related mortality and morbidity in later life will not just reflect changes in drinking among the elderly, but will also be due to longer-term health effects of drinking across the lifecourse (Bergmann et al. <unk>). However, relative increases in the number of older adults drinking in excess of recommended levels and incidences of old-age alcohol mortality and morbidity have focused policy attention on the need to respond to older person's drinking (Crome et al. <unk>). --- Characteristics of drinking in later life Research on drinking in later life has sought to identify factors associated with the age profile of drinking to understand better changing consumption over the lifecourse, particularly the overall decline in drinking among older adults (Brennan et al. <unk>; Platt, Sloan and Costanzo <unk>). Understanding the causality of why older adults consume less, if not less often, and the characteristics and lifecourse events that may accelerate or decelerate this decline, can inform estimates of the drinking behaviours of current generations entering later life (Shaw et al. <unk>). Factors that are associated with drinking in later life include health status, socio-demographic characteristics and lifecourse transitions, which all point to the significance of the social context of drinking (Platt, Sloan and Costanzo <unk>). Health is one of the most important factors associated with drinking at all ages, yet the causality between health and alcohol consumption is complex. There is considerable policy and research interest in how different levels of alcohol consumption impact on health and wellbeing and in discerning both the harms and possible benefits of drinking (Balsa et (Bergmann et al. <unk>). However, the association between poor health and abstinence is more debated. It may be explained by the'sick quitter' effect; that is, individuals with poor health stop drinking or reduce their alcohol consumption (Brennan, Schutte and Moos <unk>a; Shaper <unk>; Shaper, Wannamethee and Walker <unk>). Alternatively, it could occur because of health benefits associated with moderate drinking, as moderate drinkers report better health outcomes particularly for cardiovascular disease (Ronksley et al. <unk>) One of the challenges in unravelling this relationship is that it is difficult to control fully for the'sick quitter' effect in survey analysis and the causal explanation for this behaviour is unclear. Disentangling the causality between health and drinking is particular problematic in later life, as cross-sectional survey analysis does not find an association between heavy drinking and poor health at older ages (Frisher et al. <unk>). This could be due to a selection effect associated with the increased risk of dying for heavy consumers, resulting in an under-representation of older heavy drinkers with poor health. In older ages the relationship between poor health and less consumption/abstinence is intensified as health declines with age (Dawson, Goldstein and Grant <unk>). Cessation of drinking associated with poor health could be due to abstinence associated with previous as well as existing poor health; medical advice about reducing alcohol consumption in response to a health condition; possible interactions with medication; and, it also might reflect limited social connections for people living with poor health who as a consequence have fewer opportunities to drink (Ng Fat et al. <unk>). In summary, the relationship between health status and drinking is not straightforward and this appears to be intensified in later life. It is necessary to consider multiple causalities between alcohol consumption and health, as not only does drinking impact on health, but health status can also influence drinking behaviours. Other factors associated with declining alcohol consumption with old age point to the importance of individual resources and identities in mediating the relationship between drinking and age. Gender is an important determinant of drinking behaviours at all ages across the lifecourse. Women drink less, less often and are more likely to abstain compared to men (French et al. <unk>). In recent years this differential has declined, particularly for younger age groups (Alati et al. <unk>), and increases in younger women's consumption have been an important drive in increases in overall per capita consumption (Smith and Foxcroft <unk>). At older ages the differential by gender remains and in quantitative analysis is often one of the most important variables in predicting drinking behaviours, including problem drinking (Moos et al. <unk>). Studies of women's drinking behaviours have sought to situate these in the context of identity development and performance, and as such emphasise the conditional rather than essentialist nature of these behaviours (Day, Gough and McFadden <unk>). This also raises the possibility that gendered practices will vary with age, though the intersection between gender and age over the lifecourse has received less consideration. Gender also intersects with socio-economic characteristics in framing drinking behaviours across the lifecourse and in later life (Platt, Sloan and Costanzo <unk>). Individual resources are an important moderating factor for consumption levels, with an overall positive relationship between income and consumption occurring in different national contexts (Brennan, Schutte and Moos <unk>b). Recent analysis of ELSA data demonstrates that characteristics associated with'successful ageing' are also associated with increased risk in harmful drinking (Iparraguirre <unk>). However, the precise relationship between social disadvantage and alcohol consumption is not necessarily linear. Studies that focus on deprivation have found a U-shape relationship as more disadvantaged individuals are more likely to abstain or misuse alcohol drinking (Cerda, Johnson-Lawrence and Galea <unk>). At older ages this U-shape is less marked, and the evidence points to a closer association between disadvantage and abstinence. It is not just individual resources and identities that influence drinking behaviours over the lifecourse. Drinking is, in most contexts, a social behaviour and one that is conditional on relationships with others over time and place, though the social context of drinking has been explored mostly with reference to younger people's drinking (Kuntsche et al. <unk>). Yet it is reasonable to assume that changes in older adults' drinking behaviours may be brought about by lifecourse transitions that impact on social connectivity. --- <unk> --- Lifecourse transitions, gender and drinking in later life Dealing with isolation and loneliness is an important challenge in promoting active ageing and may also be associated with increased reliance on alcohol at older ages (Wadd et al. <unk>). Qualitative research with older adults has identified social isolation as a causal factor of drinking; in particular, widowhood is a recognised risk factor (Wilson et al. <unk>). This is partially confirmed in quantitative analysis where not being married is associated with greater alcohol consumption in older age, including binge and heavy drinking (Blazer and Wu <unk>; Platt, Sloan and Costanzo <unk>). Moreover, across the lifecourse the impact of marriage on men and women's drinking patterns has also been demonstrated (Demers, Bisson and Palluy <unk>; Windle and Windle <unk>). However, the dynamics of partnership change on drinking behaviours have received less consideration in existing studies. Another aspect of the social context of drinking that undergoes change in later life relates to working practices. Organisational cultures have been shown to be relevant in shaping employee's drinking behaviours (Ames, Grube and Moore <unk>). Moreover, employment can provide social connections that facilitate consumption, people to go drinking with as well as opportunities to drink. Therefore, it is reasonable to assume that retirement will impact on drinking behaviours. However, this association does not seem to be straightforward as there is no consistent pattern in existing research (Brennan, Schutte and Moos <unk>b; Kuerbis and Sacco <unk>; Wang, Steier and Gallo <unk>). While retirement might reduce drinking opportunities for some workers, for employees working in industries with alcohol controls retirement might increase opportunities to drink, and retirees might also have more time to drink. Furthermore, individual experiences of retirement transitions will not be independent of other characteristics, particularly income, as well as changes in health (Brennan, Schutte and Moos <unk>b). Retirement has very different meanings for older people, and these differences moderate the association with drinking behaviours (Kuerbis and Sacco <unk>). To summarise, existing research on drinking in later life has established the general overall decline in consumption in later life and has identified a number of factors associated with this behavioural change, particularly health. The review of existing research on drinking in later life demonstrates the importance of health, partnership and employment statuses for drinking, though the specific impact of transitions in these statuses on drinking behaviours remains unclear, as research has not to date considered the dynamics between lifecourse transitions in later life, such as changes in partnership status, health and retirement. We propose that a lifecourse perspective which considers the relevance of trajectories and transitions in key life events may be applied to drinking in later life. This approach considers <unk> Clare Holdsworth et al. https://doi.org/10.1017/S0144686X15001178 Published online by Cambridge University Press the influence of socio-economic characteristics and partnership and health statuses on drinking behaviours, the trajectories of drinking behaviours over time and how these might be influenced by lifecourse transitions. Our analysis examines how trajectories of drinking in later life are changed by transitions in these three domains (health, employment and partnership), and how this differs by gender. --- Methodology --- Data The longitudinal analysis of drinking in late life was carried out using ELSA (Steptoe et al. <unk>). The original ELSA sample was recruited from participants in the Health Survey for England (HSE) for the years <unk>, <unk> and <unk>, and the responses to the HSE are included in the ELSA study as wave <unk>. The first wave of ELSA was carried out in <unk>/<unk> with a sample size of <unk>,<unk> participants. The sample has been interviewed every two years and the last wave was surveyed in <unk>/<unk>. While ELSA has included questions on alcohol consumption in all waves, these have not been consistent in each wave and for this reason we have had to restrict our analysis of drinking behaviours to waves <unk> (<unk>/<unk>/<unk>), <unk> (<unk>/<unk>) and <unk> (<unk>/<unk>). Wave <unk> data do not contain the detailed socio-demographic indicators collected in the main ELSA surveys and for the baseline measures of wealth we have used a wave <unk> variable. The analysis was restricted to respondents in all three waves (<unk>,<unk> cases) and we excluded respondents who have either left the study over time or have recruited to refresh it over the ten years. Respondents from all three waves were used as the multi-level models require at least three repeated measures for most cases (Curran, Obeidat and Losardo <unk>). Analysis of attrition of key variables has shown that attrition was biased towards non-drinkers rather than drinkers and there was no discernible relationship between being present in all three waves and drinking behaviours for drinkers in wave <unk> (Nazroo, Zaninotto and Gjonça <unk>). --- Variables We use two variables to capture drinking behaviours: frequency of drinking in the last <unk> months and the number of weekly units consumed. <unk> We model frequency and quantity separately as they are both important determinants of drinking behaviour that may diverge in later life (Ng Fat and Fuller <unk>). In particular, analysis of cross-sectional data has found that older adults consume less alcohol but drink more frequently compared to <unk> Lifecourse transitions, gender and drinking in later life younger adults. Our analysis has considered baseline determinants and trajectories of both frequency and quantity of consumption to provide a more detailed picture of changes in drinking behaviours in later life. The analysis included various measures of socio-economic characteristics, lifecourse transition and other health behaviours, and the distribution of these variables is given in the Appendix. These include three lifecourse transition variables for partnership, employment status and health. There is considerable complexity in these transitions and in order to include these variables in the longitudinal model with sufficient cases for each transition status we have simplified these as follows. For partnership, we distinguish respondents who remained in a partnership over the study period, those who were not in a partnership in all waves and those who either formed or left a partnership between waves <unk> and <unk>. One of the limitations of this variable is that we have not distinguished between the reasons for ending a partnership due to the small numbers involved (this could be widowhood or divorce/separation). Employment transitions focus on transitions into retirement and distinguish those who remained economically active from those who were always retired or who retired during the study period. Self-rated health was used as an indicator of health as it is a strong predictor of mortality (Jylha <unk>; Lima-Costa et al. <unk>). Transitions in health distinguish between individuals who had constant good or not-good selfrated health (i.e. fair or poor health) and transitions between these two. For all the transition variables, more complex transitions that were not captured by the main categories were coded as 'other'. These residual transitions accounted for approximately <unk> per cent of partnership and <unk> per cent of health transitions and, due to the small numbers, are not reported here. However, for employment, other transitions accounted for approximately <unk> and <unk> per cent of male and female respondents, respectively, and this category is reported. All other socio-economic variables were captured at wave <unk> (wave <unk> for wealth) and include wealth quintiles, level of education and smoking behaviours. In the analysis we also considered variables for social support and social capital (as measured by membership of organisations) but these were not significant and we have not included these in the final models. Age at wave <unk> was also included to control for age affects across the sample. We also checked if the relationship with age was linear by including age squared in the analysis, this was not significant and we have not included this variable in the final model. --- Modelling strategies In our analysis we modelled how frequency and consumption of alcohol varied over the ten-year period for men and women. We have used two <unk> Clare Holdsworth et al. modelling strategies to analyse frequency and quantity of alcohol consumption because of the different nature of these outcome variables. Frequency of drinking was captured using a variable on drinking in last <unk> months which was a categorical ordered variable, and for this analysis we performed multi-level ordered logit analysis. The analysis of consumption used a measure of the number of units consumed in the previous week which is a continuous variable, and this analysis was carried out using a growth curve model of log of weekly units. We transformed the units of alcohol consumed in the last week into a logarithm because the distribution of units of alcohol was not normally distributed. The longitudinal ELSA data can be viewed as having a two-level hierarchical structure (Singer and Willet <unk>). Level <unk> described within-person change; that is, how respondents' frequency and quantity of alcohol consumption changed over time. Level <unk> described between-person differences in drinking behaviours over time. Data on frequency and consumption were available at three time-points (waves <unk>, <unk> and <unk>) and the measure of time in the analysis was wave which was fitted as a continuous variable. In this analysis, we investigated how the rate of change over time in frequency and consumption of alcohol varied according to lifecourse transitions. Time trend analysis was included by fitting an interaction term between transitions variables and wave. The analysis therefore provided information about the individuals' initial drinking behaviour (intercept) and trajectory during the study period (slope). Both types of multi-level analyses were estimated with Stata <unk>. Analyses were carried out separately by gender as existing research on drinking at all ages identified the significance of gender differences in drinking. The separate analysis by gender considered whether the impacts of both socio-economic characteristics and lifecourse events on drinking behaviours were similar for men and women. The models for frequency of drinking were carried out with all respondents (drinkers and non-drinkers) who were present in waves <unk>, <unk> and <unk> (<unk>,<unk> men and <unk>,<unk> women). Non-drinkers were not included in the analysis of weekly consumption as it was not possible to include a value of zero in the dependent variable. Thus, analysis of quantity of consumption was restricted to respondents who were drinkers in at least one wave (<unk>,<unk> men and <unk>,<unk> women). The models were built in a step-wise fashion to compare the impact of lifecourse events in more detail. The stepwise models were built up as follows: Model <unk> included age and wave variables only; Model <unk> included age, wave and partnership (including an interaction between partnership and wave); Model <unk> included age, wave and employment (including an interaction between employment and wave); Model <unk> included age, wave and health (including an interaction between health and wave); and Model <unk> (final <unk> Lifecourse transitions, gender and drinking in later life model) with all lifecourse transition variables, interactions between transition variables, and wave and other socio-economic characteristics included as covariates. --- Results The distribution of the frequency of drinking in last <unk> months and mean weekly units consumed in waves <unk> and <unk> by gender are given in Tables <unk> and <unk>. Men's greater and more frequent consumption is confirmed, as is the overall decline in quantity and frequency over time for both genders. There were though some notable trends beyond this overall pattern. For frequency, there was a marked increase in abstinence in wave <unk> compared to wave <unk> for both men and women. For women in wave <unk>, not drinking was the modal consumption frequency. There were small increases in the proportion of men and women reporting occasional drinking between the two timepoints. This increase in abstinence was concomitant with a decline in more frequent drinking, and most of this was accounted for by a reduction in daily drinking, particularly for women. In Table <unk>, the mean reduction in mean weekly units consumed was roughly equivalent for both men and women (around <unk> per cent), though it was slightly greater for men. --- Frequency of drinking The analysis of men's and women's frequency of drinking at baseline (wave <unk>) and over time (between waves <unk>, <unk> and <unk>) is summarised in Tables <unk> and<unk>. Taking the baseline coefficients first (i.e. variables fitted without an interaction with wave), negative coefficients for a category indicate that individuals with this characteristic drank less frequently than the reference category at wave <unk>, positive coefficients indicate more frequent drinking. The coefficients for the interactions between the transition variables and wave can be interpreted to indicate whether the rate of change in frequency of drinking over time varies in comparison to the reference category (in a partnership at all waves, in continual employment or in continual good health). A negative coefficient for the interaction terms indicates a steeper decline in frequency over time for a particular sub-category in comparison to the reference group. In other words, as wave is negative in all models, a negative interaction term indicates that respondents in the subcategory in question experienced a steeper decline in frequency of drinking over the ten-year period, compared to the reference category. A positive coefficient suggests that individuals in the particular sub-category experienced a smaller decline in drinking frequency over time. <unk> Clare Holdsworth et al. The coefficients for wave and age in Model <unk> demonstrate that over time and at older ages the frequency of drinking declined. The coefficient for wave was greater for women, indicating that over time women's frequency of alcohol consumption declined faster than men's. There were also important differences by gender and between the different lifecourse transition variables. Taking partnership status first, this had a limited impact on the frequency of drinking for men and women. Partnership status was not significant for the frequency of drinking at baseline for men, though there was a marginally significant finding for men not in a partnership (p <unk> <unk>.<unk>). This group of men, after controlling for wealth, education and smoking (i.e. Model <unk> only), reported drinking more frequently at wave <unk> in comparison to men who remained in a partnership. The interaction with wave shows that men who were not in a partnership at each wave experienced a steeper decline in frequency of drinking over time compared to men who remained in a partnership. For women, any baseline associations between frequency and partnership status found in Model <unk> were non-significant when other socio-economic variables were included in the full model. However, as for men, in comparison to those who remained in a partnership, women not in a partnership at all waves experienced a more pronounced decline in drinking frequency over time, and the same was also found for women whose partnership ended during the observation period. Retirement also had a limited impact on the frequency of drinking and there were no consistent relationships between employment transitions and drinking frequency for men and women. In the full models for both genders there were no significant coefficients (p <unk> <unk>.<unk>), though for women being retired all the time or those who experienced an alternative employment transition marginally increased the frequency of drinking at wave <unk> (p <unk> <unk>.<unk>). In contrast to partnership and employment, self-rated health was more strongly associated with the frequency of drinking at the baseline and over time. The coefficients for both baseline and slope (i.e. the interaction terms between health and wave) were larger in comparison to the other transitions variables and were, mostly, significant. These coefficients were also consistent for men and women, though they were greater for the latter. At wave <unk> respondents with poor self-rated health drank less frequently; as measured by the coefficients for those in continual poor and improving self-rated health as both groups had poor self-rated health at wave <unk>. The interaction with wave shows that those in continual poor self-rated health experienced a sharper decline in drinking frequency compared to those who remained in good health, and the same effect is observed for those whose health deteriorated between waves. For women, this latter group had a more pronounced decline in the frequency of drinking than those in permanent poor health. The final group of variables included in the full model adjusted for socioeconomic characteristics. These were significant for both men and women and illustrate that frequency of drinking in later life increased with higher wealth, levels of education, and for current and former smokers. For both men and women, and particularly for the latter, the largest coefficients were found for these variables, indicating the importance of resources and other health behaviours for patterns of alcohol consumption in later life. --- Weekly units consumed Turning to the results for the analysis of the number of units of alcohol consumed in the last week (Tables <unk> and<unk>), the findings were broadly similar to frequency though there were some important differences. In these tables, a negative coefficient for the baseline variables indicates that the category of interest drank less at wave <unk> compared to the reference category; positive coefficients illustrate the opposite relationship. For the interactions between the lifecourse variables and wave, negative coefficients indicate that the category of interest was associated with a greater rate of change over time when compared with the reference category, i.e. the decline in weekly units consumed was accelerated. In Model <unk> both age and wave variables were negative and significant, illustrating that the amount of alcohol that older adults consumed declined with age and over time. The coefficients for wave were similar for men and women, indicating a similar rate of decline over time by gender, though the coefficient for age at wave <unk> was greater for men than women. Turning to the lifecourse variables, for partnership we found differences by gender and a contrast to the results for the models of frequency. Taking men first, men not in a partnership in wave <unk> (including those who subsequently re-entered a partnership by wave <unk>) were drinking more than men in a partnership at the baseline. These baseline coefficients for men not in a partnership were greater than coefficients for all other variables except for smoking, which demonstrates the importance of partnership status for men's baseline drinking. For women, partnership status had no significant association with baseline weekly alcohol consumption. However, women whose partnership ended during the study period experienced a steeper decline of weekly units over time compared to women who remained in a partnership and this trend replicates the same finding for frequency. It would appear that partnership was protective for men regarding how much they drank at the baseline, but not how often. While for women the end of a partnership led to a decline in both the frequency and quantity of alcohol consumption over time. Retirement was not important for men's baseline drinking, which was similar to the results for the frequency model, though men who retired over the period of observation had a steeper decline in drinking than men who remained in employment. Retired women, those who retired between the waves and women who experienced another employment transition all drank less at wave <unk> than employed women. For health, at wave <unk> men with poor self-rated health whose health subsequently improved drank less, while women whose health deteriorated drank more, though these coefficients were only marginally significant (p <unk> <unk>.<unk>). Both men and women with deteriorating self-rated health reported a steeper decline in weekly consumption of alcohol over time compared to older adults who remained in good health. This suggests that worsening self-rated health was associated with a reduction in the amount of alcohol consumed over time. The results for the socio-economic variables were similar to the model of frequency. However, unlike the frequency models in which there were <unk> Lifecourse transitions, gender and drinking in later life differences between all categories, the differences by wealth quintile and education status for quantity were only found between older adults at the top and bottom of the wealth distribution and those with no or higherlevel qualifications. For women, it was only the top two income groups that drank more than those in the bottom group. For men, there was a Ushape relationship; those in the second and top quintile reported more drinking. For education, both men and women with degrees drank more than those with no qualifications, and the same association was found for women with post-compulsory education. All female former and current smokers drank more in wave <unk> than non-smokers, while for men this was restricted to current and former regular smokers. --- Discussion The analysis of frequency and quantity of drinking in later life confirmed an overall decline in drinking behaviours in later life (Brennan et al. <unk>; Breslow and Smothers <unk>; Platt, Sloan and Costanzo <unk>). We have found that this sample of people aged <unk> and over experienced a decline in drinking over the ten years, and they also drank less often, which contrasts with cross-sectional data that show more frequent drinking in later life (Ng Fat and Fuller <unk>). This suggests that while the current generation of older adults report drinking more often than younger drinkers, this behaviour is not necessarily associated with an increase in the frequency of drinking with age. Instead these longitudinal findings, when compared with recent cross-sectional data, indicate that there are cohort differences in the frequency of drinking and current younger cohorts have developed more time-focused drinking behaviours. The cohort of older adults considered in this analysis has experienced a particular constellation of economic and social changes which will shape their drinking behaviours. In particular, younger members (aged roughly between <unk> and <unk> in the late <unk>s when data for wave <unk> were collected), entered adulthood at the same time as per capita alcohol consumption increased in post-war UK (from the early <unk>s onwards, peaking in the early <unk>s). Moreover, they have also entered later life at the same time as economic prosperity for older people has improved relative to younger age groups (Belfield et al. <unk>). Recent survey data that demonstrate continuation in older people's drinking practices while younger ages are drinking less may, therefore, reflect the relative prosperity of older and younger cohorts. Older adults have potentially more opportunities and more resources to facilitate the continuation of younger and mid-adult drinking practices into later life. Yet, as this analysis demonstrates, older people's drinking behaviours are mediated by the ageing <unk> Clare Holdsworth et al. process associated with overall declines in alcohol consumption and lifecourse transitions. The drinking behaviours observed in the current generation of older adults reflect a hybrid of established behaviours from younger and mid-adulthood; responses to an ageing body and associated lifecourse transitions; and, the prevailing social-economic context, which for the current generation of older people may facilitate the continuation of regular alcohol consumption. Drinking is, for most people, a deeply social activity (Emslie, Hunt and Lyons <unk>) and one that resonates with groups and individuals (Backett and Davison <unk>); it is therefore reasonable to assume that lifecourse events in later life which bring about reorientations of older people's social interactions will impact on their drinking behaviours. However, the direction and magnitude of this impact has not been explored, which has been addressed in this analysis. The findings for health confirm that poor self-rated health was associated with drinking less often at the baseline and that those with poor or deteriorating health were more likely to reduce the number of times that they drank over time compared to those with good health (Balsa et al. <unk>; Platt, Sloan and Costanzo <unk>). Older adults whose self-rated health improved over the observation period initially drank less often when their health was poor, but did not reduce the frequency of drinking over time

Older people consume less alcohol than any other adult age group. However, in recent years survey data on alcohol consumption in the United Kingdom have shown that while younger age groups have experienced a decline in alcohol consumption, drinking behaviours among the elderly have not reduced in the same way. This paper uses data from the English Longitudinal Study of Ageing to analyse both the frequency and quantity of older adult's alcohol consumption using a lifecourse approach over a ten-year period. Overall drinking declined over time and the analysis examined how socio-economic characteristics, partnership, employment and health statuses were associated with differences in drinking behaviours and how these changed over time. Higher wealth and level of education were associated with drinking more and drinking more frequently for men and women. Poorer self-rated health was associated with less frequent consumption and older people with poor and deteriorating health reported a steeper decline in the frequency of alcohol consumption over time. Men who were not in a partnership drank more than other men. For women, loss of a partner was associated with a steeper decline in drinking behaviours. These findings have implications for programmes to promote responsible drinking among older adults as they suggest that, for the most part, characteristics associated with sustaining wellbeing in later life are also linked to consuming more alcohol.

to an ageing body and associated lifecourse transitions; and, the prevailing social-economic context, which for the current generation of older people may facilitate the continuation of regular alcohol consumption. Drinking is, for most people, a deeply social activity (Emslie, Hunt and Lyons <unk>) and one that resonates with groups and individuals (Backett and Davison <unk>); it is therefore reasonable to assume that lifecourse events in later life which bring about reorientations of older people's social interactions will impact on their drinking behaviours. However, the direction and magnitude of this impact has not been explored, which has been addressed in this analysis. The findings for health confirm that poor self-rated health was associated with drinking less often at the baseline and that those with poor or deteriorating health were more likely to reduce the number of times that they drank over time compared to those with good health (Balsa et al. <unk>; Platt, Sloan and Costanzo <unk>). Older adults whose self-rated health improved over the observation period initially drank less often when their health was poor, but did not reduce the frequency of drinking over time compared to those in good health. The findings for the impact of self-rated health on the number of units consumed were similar, yet overall the changes in self-rated health were more strongly associated with the frequency of drinking rather than with the quantity of alcohol that drinkers consumed. This could reflect fewer opportunities to drink among older adults with poor self-rated health. These findings suggest that changes in drinking behaviour associated with health occur because individuals are unwell and receive medical advice to reduce alcohol consumption (Shaper et al. <unk>) or due to interactions with medication (Moore, Whiteman and Ward <unk>). In addition, older adults with poor health may have fewer social opportunities to drink, and therefore drink less often. Partnership transitions were also associated with changes in drinking behaviour. The results show that women cut back on alcohol consumption if they do not have a partner, while presence of a female partner moderates the amount that men drink. Cross-national research on married couples' drinking behaviours has found that wives influence their husbands to drink less and our analysis would confirm this (Selin, Holmila and Knibbe <unk>). Yet for men, being in a partnership only influenced how much they drank, not how often. One possible explanation is that the moderation of how much partnered men drank occurred because consumption is shared, for example a bottle of wine (and wine was the favoured beverage <unk> Lifecourse transitions, gender and drinking in later life among the ELSA respondents), while single men may be less likely to share consumption. Our findings demonstrate that for women the end of a partnership in later life leads to a significant decline in drinking. For women, partnership did not influence drinking at the baseline, though change in partnership status did intensify the decline in alcohol consumption. In particular, women whose partnership ended experienced the steepest decline in frequency and quantity of alcohol consumption compared to all other partnership status. The analysis is not able to distinguish between different partnership transitions and cannot shed light on the impact of bereavement rather than separation or divorce. Yet what it does reveal is that the presence of a partner and the social dynamics of relationships are important for drinking behaviours and these are different for older men and women. For example, the importance of loss of a partner for women's alcohol consumption may reflect the fact that more older women experience widowhood compared to men (Arber, Davidson and Ginn <unk>). These findings for partnership contradict qualitative research that suggest being widowed is a causal factor for alcohol abuse (Wilson et al. <unk>), but correspond with analysis of stress events which has shown that stress reduces the risk of women's alcohol misuse in later life (Sacco, Bucholz and Harrington <unk>). This is an important area for further research to understand how and why partnership is an important, but different, determinant of both men's and women's drinking in later life, and how this might vary as the dynamics of relationships change in later life (Arber, Davidson and Ginn <unk>). For retirement, our analysis confirmed previous results that there is no definitive association between retirement and drinking behaviours (Brennan, Schutte and Moos <unk>b; Kuerbis and Sacco <unk>; Wang, Steier and Gallo <unk>). One of the limitations of the employment transition variable is that it is difficult to summarise the complexity of retirement transitions, particularly for women, as nearly one-third of women in the sample experienced 'other transitions' or had missing data for this variable. However, despite this limitation, being in employment was found to be associated with greater weekly consumption for women. What is rather surprising is that retirement was more significant for women's drinking compared to men. For men, any relationship between drinking and retirement was explained by other socio-economic characteristics. The analysis therefore confirms the heterogeneity of retirement transitions and that for men, in particular, it is individual circumstances that are related to drinking, rather than the transition to retirement. Analyses of baseline drinking confirmed that it was wealthier, better-educated men and women, as well as current and former smokers, who drank more and more often (Brennan, Schutte and Moos <unk>b; Grittner et al. <unk> Clare Holdsworth et al. --- <unk>). The importance of both economic capital (as measured by wealth) and cultural capital (as indicated by education) confirm that drinking was mediated by individual's social status. In particular, resources were strongly associated with the frequency of drinking. This suggests that material resources are important in enabling older people to drink on a daily basis, but are less important in influencing how much older people drink when they do consume alcohol. It would appear that the influence of socio-economic characteristics on drinking is not just a question of affordability, but that more affluent social groups cultivate daily drinking practices, while those with fewer resources limit the number of days on which they drink. Research to date has focused on the importance of social and cultural influences for younger people's drinking (van Wersch and Walker <unk>); our analysis would suggest that this continues in later life. --- Limitations There are a number of limitations to this study. First, the ELSA study is a household sample recruited from participants to the HSE, and this does not necessarily provide an appropriate sampling frame for identifying alcohol misuse. While there are individuals in the sample who consumed excessive quantities of alcohol (<unk> per cent of the sample, or <unk> cases, in wave <unk> reported heavy drinking, defined as drinking more than <unk> units a week for men and <unk> for women), the numbers were too small to provide detail on the risks of alcohol misuse in later life and our analysis has not considered these risks. Moreover, due to the number of different permutations we have had to simplify the lifecourse transition variables, and in particular have not distinguished between reasons for end of partnership. --- Conclusion The analyses of lifecourse transitions and alcohol consumption confirm the importance of declining health in influencing how often and how much older adults drink. These results suggest that among older adults, those who drink more often have better health, and regular alcohol consumption in later life is indicative of good health. Yet the findings for lifecourse transitions relating to partnership and employment are less consistent and vary by gender. In particular, the differences between men's and women's partnership status and the amount that they drink suggest that the social context of alcohol consumption is gendered. Lifecourse transitions that impact on older people's social connectivity may lead to less drinking (in the case of <unk> Lifecourse transitions, gender and drinking in later life health for men and women, and partnership and employment for women), but the results for partnership status for men hint at the possibility that some forms of social disconnection are associated with drinking more. Drinking in later life is just as much a social activity as it is at younger ages and one that older adults do because they enjoy a drink and derive some pleasure out of drinking with others. Our analysis has shown that as social relationships change in later life, these may impact on drinking behaviours, though the precise direction of this change varies according to the lifecourse event in question (health, employment or partnership) and by gender. These findings have implications for policies to support responsible drinking in later life, as they show that characteristics associated with improved wellbeing in later life (relating to resources and health) are also linked with drinking more and drinking more often. This highlights the importance of situating alcohol consumption within social and economic contexts and with reference to individual circumstances, as well as the possibility that policies to improve older people's wellbeing may have unintended consequences for alcohol consumption. --- Appendix --- T A B L E A <unk>. Distribution of variables --- Variable Men Women --- All respondents Drinkers in at least one wave --- All respondents Drinkers in at least one wave

Older people consume less alcohol than any other adult age group. However, in recent years survey data on alcohol consumption in the United Kingdom have shown that while younger age groups have experienced a decline in alcohol consumption, drinking behaviours among the elderly have not reduced in the same way. This paper uses data from the English Longitudinal Study of Ageing to analyse both the frequency and quantity of older adult's alcohol consumption using a lifecourse approach over a ten-year period. Overall drinking declined over time and the analysis examined how socio-economic characteristics, partnership, employment and health statuses were associated with differences in drinking behaviours and how these changed over time. Higher wealth and level of education were associated with drinking more and drinking more frequently for men and women. Poorer self-rated health was associated with less frequent consumption and older people with poor and deteriorating health reported a steeper decline in the frequency of alcohol consumption over time. Men who were not in a partnership drank more than other men. For women, loss of a partner was associated with a steeper decline in drinking behaviours. These findings have implications for programmes to promote responsible drinking among older adults as they suggest that, for the most part, characteristics associated with sustaining wellbeing in later life are also linked to consuming more alcohol.

Introduction Mental disorders constitute one of the greatest public health problems of our time. Several studies have shown that one out of four people in the world develop mental health problems at some point in their life [1]. Adolescence is a key period during which the incidence of externalizing and internalizing problems increases significantly [2][3][4]. The internalizing problems that most commonly occur during this developmental stage are depressive symptoms and suicidal ideation. Research has found a rapid increase in depressive symptoms and suicidal ideation from early to late adolescence [5,6], with their being occurrence greater in girls than boys [7][8][9]. In addition, depressive symptoms and suicidal ideation have been postulated to be the main predictors of adolescent suicide [10][11][12]. Thus, to identify the causes of these internalizing problems has become an international priority for suicide prevention among youths. With the focus on the prevention and understanding of these problems, in this study, we explored the association between certain demographic, social, and economic factors and the manifestation of depressive symptoms and suicidal ideation in two populations: referred adolescents, namely outpatients using mental health facilities, and non-referred adolescents, namely community adolescents. Furthermore, we aimed to clarify the moderating role that certain demographic, social, and economic factors may play in the relationship between depressive symptoms and suicidal ideation. --- Depressive Symptoms and Suicidal Ideation during Adolescence: Age and Gender Differences Adolescence is a developmental stage during which individuals are particularly vulnerable to the development of mental health problems [13,14], particularly internalizing problems [15]. Most studies report that the levels of some internalizing problems, such as depressive symptoms and suicidal ideation, increase from early to late adolescence [5,6]. However, other studies have found that the levels of these problems remain stable during this period [16,17]. In early adulthood, the levels of these problems begin to decline for some individuals, while they become chronic for others. Regarding gender differences, the available evidence also contains discrepancies. Many studies suggest that adolescent girls from referred and non-referred samples experience higher levels of depressive symptoms [18][19][20][21] and suicidal ideation [17,22] than boys. However, there are also studies that have not found gender differences in non-referred samples [23][24][25]. Likewise, other studies conducted simultaneously in referred and non-referred samples have found that levels of depressive symptoms were significantly higher in girls than in boys from the referred group, but that there were no gender differences in the non-referred group [26]. Although these results highlight the need to conduct studies that clarify gender differences based on the type of analyzed sample, few studies to date have simultaneously explored the existence of gender differences in samples of referred and non-referred adolescents. The comorbidity and stability of depressive symptoms and suicidal ideation have also been studied along with gender differences. In recent decades, ample research with the adolescent population has documented a strong and stable correlation between depressive symptoms and suicidal ideation [27,28]. In addition, both of these internalizing problems can persist over time, affecting mental health during adulthood [1,[29][30][31][32]. In fact, both depressive symptoms and suicidal ideation are well-established risk factors of suicide [10][11][12], which constitutes the second most frequent cause of death worldwide in individuals between 10 and 24 years old [33]. The high stability of these manifestations and their association with future health problems could explain the current interest of the scientific community in identifying the factors involved in their development during adolescence [34]. --- Social and Economic Factors of Depressive Symptoms and Suicidal Ideation It has been shown that depressive symptoms and suicidal ideation are not determined by a single cause. These complex phenomena have multiple determinants [35,36] and are caused by a conjunction of biological, psychological, social, economic, and cultural factors [37,38]. Regarding social factors, research has focused on the contribution of socioeconomic status and parents' educational level to depressive symptoms and suicidal ideation in adolescents. These studies indicate that adolescents who come from families with a low socioeconomic status are at a greater risk of developing depressive symptoms and suicidal ideation [39][40][41], both in girls and in boys [42]. Few studies, however, have analyzed the contribution of parents' educational level as a factor distinct from socioeconomic status. This research, although limited, has found a negative association between parents' educational level and the development of depressive symptoms [39] and suicidal ideation [36]. Therefore, despite the clear relationship between socioeconomic status and parents' educational level, the available literature highlights the importance of independently studying the influence of parents' educational level on the development of depressive symptoms and suicidal ideation in adolescents, in order to fill a gap in our knowledge that currently exists. Furthermore, most studies have focused on either a referred sample or a non-referred sample. Studies that use a combined design, in which both types of samples are integrated simultaneously, remain scarce. Thus, it seems necessary to know whether their contribution is similar or different in clinical and community adolescents. In addition, research should go one step further and analyze whether certain social and economic factors can modify the relationships that the scientific literature has shown to exist between depressive symptoms and suicidal ideation. To date, however, few studies have investigated the moderating role that such variables as adolescents' age and gender, parents' educational level, and family socioeconomic status play in the relationship between the two internalizing problems. These data may help us to identify vulnerable populations in order to develop intervention policies and programs for preventing or decreasing the manifestation of these two internalizing problems. --- The Present Study Considering the above-described background, the first objective of this study was to analyze the association between certain demographic (i.e., age and gender) and social and economic factors (i.e., mother's education level, father's education level, and family socioeconomic status) and depressive symptoms and suicidal ideation, in a mixed sample of referred and non-referred adolescents. Based on the existing literature, we expected to find higher levels of depressive symptoms and suicidal ideation in the group of referred adolescents, during middle and late adolescence, and among those adolescents-referred or non-referred-from families with a lower socioeconomic status and parental educational level. Regarding gender differences, we expected girls in the referred sample to exhibit higher levels of depressive symptoms and suicidal ideation than boys and that these differences, although smaller, would also be found in non-referred adolescents. The second objective was to explore the moderating role that these demographic, social, and economic factors play in the relationship between depressive symptoms and suicidal ideation in both samples. Due to the lack of available evidence, no expectations were set for this second objective. --- Materials and Methods --- Participants and Procedure This study used two groups of participants: a group of referred adolescents (n = 211) and a group of non-referred adolescents (n = 1401). The main demographic characteristics of the participants are summarized in Table 1. Referred adolescents were recruited from 12 Child and Adolescent Mental Health facilities of the public health system in the southern region of Spain according to a quota sampling procedure. The procedure had two stages. First, a total of 522 potential participants satisfying the selection criteria (aged between 12 and 17 years old, living with his/her family instead of residing in an institution, not enduring a psychotic episode or outbreak, and not being intellectually disabled) were identified. To represent the diverse range of youths who use mental health facilities, we divided the potential participants into 24 mutually exclusive subgroups based on their gender (boy or girl), age (12-14 or 15-17 years old), primary referral problem (externalizing, internalizing, or other problem), and parental educational attainment (basic education (primary school, secondary school and medium vocational training) or higher education (high-school degree, high vocational training, and university)). Each of the 24 subgroups (2 <unk> gender, 2 <unk> age, 3 <unk> primary referral problem, and 2 <unk> parental educational attainment) was assigned the same number of participants (n = 10). Second, researchers contacted 287 randomly chosen parents to obtain consent to the participation of youths in the study. Of the 24 subgroups, 7 had less than 10 eligible participants. The acceptance rate was high (75.96%). Non-referred adolescents were recruited from 12 high schools located in the same areas as the Mental Health facilities. High schools were selected according to the size of the municipality (<unk>30,000 or <unk>30,000 inhabitants for small and larger municipalities, respectively), the annual average per capita income of the school area (<unk>21,966 € for low-income areas and <unk>21,966 € for higher-income areas), and the type of school (public or private). Informed consent was obtained from adolescents and parents. Data on the referred sample were collected at the clinical facilities. Data on the non-referred sample were collected in the school setting during school hours. Participation was anonymous and voluntary. Trained researchers distributed the questionnaires and assisted the participants during data collection. This study was approved by the Biomedical Research Ethics Review Board of Andalusia (Spain). An ad hoc questionnaire was used to record information about the main demographic and social characteristics of the participants, including age (12-13, 14-15, or 16-17), gender (boy or girl), and parents' education level (no education, primary education, secondary education, or higher education). To evaluate participants' socioeconomic status (SES), we used the Family Affluence Scale II (FAS II) [43]. This is a brief measure of family wealth that includes four items regarding family holidays in the last 12 months, car ownership, bedroom occupancy, and home computers. The FAS II score, which ranged from 0 to 9, was computed by summing the responses to these four items. Following Currie et al. [44], this score was recoded into a three-point score for low (0-3), middle (4-6), and high (7-9) SES groups. --- Depressive Symptoms The anxious/depressed scale of the youth self-report (YSR) [45] was used to evaluate depressive symptoms. The YSR comprises 112 items that assess a range of adolescents' emotional and behavioral problems. The scale applied in this study includes 13 items (e.g., "I feel worthless or inferior") answered on a Likert scale ranging from 0 (not true) to 2 (very true or often true). Cronbach's alpha was 0.86 for the referred sample and 0.78 for the non-referred sample. --- Suicidal Ideation Suicidal ideation was evaluated using the negative subscale of the positive and negative suicide ideation inventory (PANSI) [46,47]. This questionnaire is based on the assumption that certain factors increase the risk of suicidal behavior (e.g., negative thoughts), whereas other factors modulate or buffer against this behavior (e.g., positive thoughts). It is comprised of 14 items grouped into two subscales that evaluate the presence of protective factors and risk factors associated with suicidal thoughts during the past two weeks. Items were answered on a Likert scale ranging from 1 (none of the time) to 5 (most of the time). For the purpose of this study, we only considered the negative subscale of risk factors (8 items; e.g., "During the past two weeks, including today, how often have you thought that your problems were so overwhelming that suicide was seen as the only option for you?"). Cronbach's alpha was 0.95 and 0.94 for the referred sample and the non-referred sample, respectively. --- Data Analysis Data analyses were performed using IBM SPSS Statistics (IBM Corp., Armonk, NY, USA) for MacOS, version 26. Little's [48] missing completely at random (MCAR) test was not significant in both samples (<unk> 2 /df = 0.27, p > 0.05, for the referred sample; <unk> 2 /df = 2.67, p > 0.05, for the non-referred sample), suggesting that missing data were completely at random. Most analyses were conducted separately for the referred and non-referred groups, since both samples were obtained through a quota sampling procedure that controlled for age, gender, parents' educational level, and socioeconomic status to facilitate subgroups comparisons within each sample and to systematize the variations produced by the demographic, social, and economic factors of interest [49]. As preliminary analyses, we calculated correlations between the demographic, social, and economic factors and depressive symptoms and suicidal ideation. Next, we carried out a multivariate analysis of variance (MANOVA) to determine the existence of differences in depressive symptoms and suicidal ideation between referred and non-referred samples. This analysis included mental health status (referred or non-referred) as a fixed factor and internalizing problems (i.e., depressive symptoms and suicidal ideation) as dependent variables. Then, we conducted two factorial MANOVAs-one per mental health status group-in order to examine the effects of demographic, social, and economic factors on internalizing problems in both samples, including age range (12-13, 14-15, or 16-17), gender (boy or girl), mother's education level (no education, primary education, secondary education, or higher education), father's education level (no education, primary education, secondary education, or higher education), and socioeconomic status (low, medium, or high) as independent variables, and depressive symptoms and suicidal ideation as dependent variables. Complementary Bonferroni tests were carried out to evaluate multiple comparisons. Finally, we conducted moderation analyses using the macro PROCESS 3.4 for SPSS Statistics. Five moderation models per sample were performed using depressive symptoms as an independent variable (X), suicidal ideation as a dependent variable (Y), and each demographic, social, or economic factor (i.e., age, gender, mother's education level, father's education level, and socioeconomic status) as a possible model moderator (W). Categorical moderators with more than two levels (i.e., age and SES) were coded as dummy variables (W 1 and W 2, respectively). All applications were based on 5000 bootstrap samples with a 95% confidence interval. --- Results --- Preliminary Analyses Bivariate correlations for the study variables are displayed in Table 2. In both samples, mother's education level, father's education level, and socioeconomic status were significantly and positively correlated to each other, and socioeconomic status was negatively corelated to depressive symptoms and suicidal ideation. The correlation between depressive symptoms and suicidal ideation was strong and positive. In the referred sample, age correlated positively to depressive symptoms and suicidal ideation, and mother's education level was negatively associated with suicidal ideation. In the non-referred sample, age and father's education level were negatively related to suicidal ideation. --- The Effect of Demographic, Social, and Economic Factors on Depressive Symptoms and Suicidal Ideation Before we performed the main analysis, we conducted a MANOVA to examine differences in depressive symptoms and suicidal ideation between samples. A significant main effect was found for mental health status (Wilks' Lambda = 0.07, F(2, 1556) = 55.45, p <unk> 0.001, <unk> 2 = 0.07). As expected, referred adolescents scored significantly higher in depressive symptoms (F(1, 1557) = 95.11, p <unk> 0.001, <unk> 2 = 0.06 (M = 0.74, SD = 0.47 for referred adolescents; M = 0.49, SD = 0.33, for non-referred adolescents)) and suicidal ideation (F(1, 1557) = 78.67, p <unk> 0.001, <unk> 2 = 0.05 (M = 1.78, SD = 1.15, for referred adolescents; M = 1.29, SD = 0.66, for non-referred adolescents)) than non-referred adolescents. Then, we carried out MANOVAs in both samples to examine the effect of demographic, social, and economic factors on depressive symptoms and suicidal ideation. For the referred group, no significant multivariate effects were obtained (see Table 3). The univariate analysis only showed significant differences in terms of socioeconomic status for suicidal ideation scores (F(2, 28) = 4.09, p = 0.02, <unk> 2 = 0.09). Bonferroni tests showed that adolescents from families with a low socioeconomic status scored higher in suicidal ideation than those from families with a medium or high socioeconomic status. No interaction effects were found among demographic, social, and economic factors. Table 4 displays the MANOVA results for the non-referred group. Significant multivariate effects were obtained for age (Wilks' Lambda = 0.01, F(4, 2240) = 3.17, p = 0.01, <unk> 2 = 0.01) and socioeconomic status (Wilks' Lambda = 0.01, F(4, 2240) = 2.54, p = 0.04, <unk> 2 = 0.005). Regarding depressive symptoms, the univariate analyses showed differences in terms of age (F(2, 1120) = 3.67, p = 0.02, <unk> 2 = 0.01) and father's education level (F(3, 1120) = 3.26, p = 0.02, <unk> 2 = 0.01). A post hoc analysis indicated that adolescents in the 14-15 age group showed a higher level of depressive symptoms than adolescents in the other age groups (i.e., 12-13 and 16-17). Likewise, the group of adolescents with fathers with no education scored higher in depressive symptoms than all other participants (i.e., fathers with a completed primary, secondary, or higher education). No interaction effects were observed among demographic, social, and economic factors. Regarding suicidal ideation, the univariate analyses showed differences in terms of age (F(2, 1120) = 5.50, p = 0.004, <unk> 2 = 0.01) and socioeconomic status (F(2, 1120) = 4.78, p = 0.009, <unk> 2 = 0.01). Bonferroni tests showed that youths in the 14-15 year-old age group displayed higher levels of suicidal ideation than youths in the other age groups (12-13 or 16-17 years old). Also, adolescents in the low socioeconomic status group showed a higher level of suicidal ideation than adolescents in the medium and high socioeconomic status groups. We observed interaction effects between age and father's education level (F(6, 1120) = 3.73, p <unk> 0.001, <unk> 2 = 0.02); gender, mother's level education, and socioeconomic status (F(5, 1120) = 2.52, p = 0.03, <unk> 2 = 0.01); and age, gender, father's level education, and socioeconomic status (F(9, 1120) = 2.06, p = 0.03, <unk> 2 = 0.02). Taken together, these results indicate a higher level of suicidal ideation among girls in the 14-15 year-old age group with parents with no education and whose family has a low socioeconomic status. --- The Moderating Role That Demographic, Social, and Economic Factors Play in the Depressive Symptoms-Suicidal Ideation Relationship Moderation analyses were conducted separately for referred and non-referred adolescents to determine the individual moderation role that each demographic, social, or economic factor (W) played in the relationship between depressive symptoms (X) and suicidal ideation (Y). In the referred sample, the results only showed moderation effects in the model that included age as a moderator (F(2, 197) = 8.08, p <unk> 0.001, <unk>R 2 = 0.03). As expected, depressive symptoms were significantly related to suicidal ideation (<unk> X = 1.07, p <unk> 0.001, 95% CI [0.68, 1.45]). However, although age was not associated with suicidal ideation (<unk> W 1 = 0.25, p = 0.08, 95% CI [-0.03, 0.53]; <unk> W 2 = 0.16, p = 0.23, 95% CI [-0.10, 0.42]), the interaction effect of depressive symptoms and age on suicidal ideation was statistically significant (<unk> X * W 1 = 0.81, p = 0.006, 95% CI [0.23, 1.40]; <unk> X * W 2 = 1.04, p <unk> 0.001, 95% CI [0.51, 1.57]). This indicates that the relationship between depressive symptoms and suicidal ideation was dissimilar among the three age groups. The plotted interaction effects (Figure 1a) indicate that the relationship between depressive symptoms and suicidal ideation was stronger in the 14-15 (B = 1.88, p <unk> 0.001) and 16-17 (B = 2.11, p <unk> 0.001) age groups than in the 12-13 age group (B = 1.07, p <unk> 0.001). In the non-referred sample, only socioeconomic status was found to moderate the relationship between depressive symptoms and suicidal ideation (F(2, 1350) = 3.06, p = 0.04, <unk>R 2 = 0.01). The results showed a significant simple effect of depressive symptoms (<unk> = 1.53, p <unk> 0.001, 95% CI [1.09, 1.97]) and socioeconomic status (<unk> = -0.17, p = 0.002, 95% CI [-0.28, -0.06]; <unk> = -0.16, p = 0.004, 95% CI [-0.27, -0.05]) on suicidal ideation. Likewise, we found a significant interaction effect of depressive symptoms and socioeconomic status on suicidal ideation (<unk> * = -0.43, p = 0.09, 95% CI [-0.94, 0.07]; <unk> * = -0.61, p <unk> 0.01, 95% CI [-1.10, -0.12]). As shown in Figure 1b, this interaction effect suggests that the relationship between depressive symptoms and suicidal ideation was stronger in those adolescents whose socioeconomic status was low (B = 1.53, p <unk> 0.001). In comparison, in adolescents with a medium (B = 1.10, p <unk> 0.001) or high socioeconomic status (B = 0.92, p <unk> 0.001), the strength of this association was similar. --- Discussion This study had two main objectives. The first objective was to examine the association between certain demographic, social, and economic factors and depressive symptoms and suicidal ideation in referred and non-referred adolescents. The second objective was to explore the moderating effect of these factors on the relationship between depressive symptoms and suicidal ideation in both groups. As expected, referred adolescents showed higher levels of depressive symptoms and suicidal ideation than non-referred adolescents. In addition, regarding age, we expected to find higher levels of internalizing problems in middle and late adolescence in both groups. As highlighted in the introduction, most studies found an increase in the levels of depressive symptoms and suicidal ideation throughout adolescence [5,6], although some studies have reported stability in these levels [16,17]. Our findings in referred adolescents showed stable levels of depressive symptoms and suicidal ideation during adolescence. In comparison, in non-referred adolescents, our data indicate an increase in the levels of depressive symptoms and suicidal ideation between 12-13 and 14-15 years old, and stability from 14-15 to 16-17 years old. These results suggest different developmental trends of internalizing problems among referred and non-referred adolescents. However, the cross-sectional nature of our data precludes the establishment of causality. Further longitudinal studies are required to clarify these trends. Another interesting finding is that when both socioeconomic status and parents' educational level were considered in the same analysis, suicidal ideation was significantly associated with socioeconomic status, but not with parents' educational level. As predicted, and consistent with In the non-referred sample, only socioeconomic status was found to moderate the relationship between depressive symptoms and suicidal ideation (F(2, 1350) = 3.06, p = 0.04, <unk>R 2 = 0.01). The results showed a significant simple effect of depressive symptoms (<unk> X = 1.53, p <unk> 0.001, 95% CI [1.09, 1.97]) and socioeconomic status (<unk> W 1 = -0.17, p = 0.002, 95% CI [-0.28, -0.06]; <unk> W 2 = -0.16, p = 0.004, 95% CI [-0.27, -0.05]) on suicidal ideation. Likewise, we found a significant interaction effect of depressive symptoms and socioeconomic status on suicidal ideation (<unk> X * W 1 = -0.43, p = 0.09, 95% CI [-0.94, 0.07]; <unk> X * W 2 = -0.61, p <unk> 0.01, 95% CI [-1.10, -0.12]). As shown in Figure 1b, this interaction effect suggests that the relationship between depressive symptoms and suicidal ideation was stronger in those adolescents whose socioeconomic status was low (B = 1.53, p <unk> 0.001). In comparison, in adolescents with a medium (B = 1.10, p <unk> 0.001) or high socioeconomic status (B = 0.92, p <unk> 0.001), the strength of this association was similar. --- Discussion This study had two main objectives. The first objective was to examine the association between certain demographic, social, and economic factors and depressive symptoms and suicidal ideation in referred and non-referred adolescents. The second objective was to explore the moderating effect of these factors on the relationship between depressive symptoms and suicidal ideation in both groups. As expected, referred adolescents showed higher levels of depressive symptoms and suicidal ideation than non-referred adolescents. In addition, regarding age, we expected to find higher levels of internalizing problems in middle and late adolescence in both groups. As highlighted in the introduction, most studies found an increase in the levels of depressive symptoms and suicidal ideation throughout adolescence [5,6], although some studies have reported stability in these levels [16,17]. Our findings in referred adolescents showed stable levels of depressive symptoms and suicidal ideation during adolescence. In comparison, in non-referred adolescents, our data indicate an increase in the levels of depressive symptoms and suicidal ideation between 12-13 and 14-15 years old, and stability from 14-15 to 16-17 years old. These results suggest different developmental trends of internalizing problems among referred and non-referred adolescents. However, the cross-sectional nature of our data precludes the establishment of causality. Further longitudinal studies are required to clarify these trends. Another interesting finding is that when both socioeconomic status and parents' educational level were considered in the same analysis, suicidal ideation was significantly associated with socioeconomic status, but not with parents' educational level. As predicted, and consistent with available evidence [40,41], we found higher levels of suicidal ideation among referred and non-referred adolescents from families with a low socioeconomic status. In contrast, the relationship between socioeconomic status and depressive symptoms was not significant when parents' educational level was partialled out. In line with previous research [39], our results also evidenced a significant and negative relationship between father's educational level and the development of depressive symptoms. The above findings support the convenience of analyzing the effect of socioeconomic status and parents' educational level based on differentiated variables. In addition, we must emphasize that our data indicate that socioeconomic status is related to suicidal ideation, in both referred and non-referred adolescents, but not to depressive symptoms. A possible explanation for this finding could be related to the factors that are involved in the increase of both problems during adolescence. While the development of depressive symptoms could be associated with the physical and psychological changes that come along with puberty [50], an increase in suicidal ideation may be linked to the occurrence of other mental health problems [34,51]. In view of this, socioeconomic status could be relevant to suicidal ideation only, and not to depressive symptoms, because suicidal ideation is more strongly associated with access to mental health services than depressive symptoms. Thus, belonging to a family with a low socioeconomic status, which limits access to resources for mental health promotion [40], may result in additional psychological problems and, consequently, increase the risk of developing higher levels of suicidal ideation. As opposed to our expectations, a low socioeconomic status does not seem to be a determining factor for depressive symptoms, perhaps because these symptoms are related to normative changes in adolescence [32] that do not always require specialized mental health care. The role that parents' educational level plays in depressive symptoms may be explained by previous studies showing that fathers with a higher educational level tend to be warmer and more communicative with their children and to possess better emotional skills [52,53]. These aspects may be, in turn, associated with lower levels of depressive symptoms. In fact, warmth and supportive parenting have been postulated to be protective factors against the development of depressive symptoms during adolescence [54,55]. Finally, regarding the first objective, we found no gender differences in the levels of depressive symptoms and suicidal ideation among referred and non-referred adolescents. Although the absence of gender differences in a non-referred sample has been observed in prior studies [23][24][25], most studies with referred samples have found higher levels of depressive symptoms and suicidal ideation among girls belonging to this group [21,22]. A possible explanation for our findings could be the reduction of differences in sex-type roles adopted by adolescent boys and girls [56]. Although our study did not confirm a direct effect of gender on the levels of depressive symptoms and suicidal ideation in referred and non-referred adolescents, we did find an interaction effect between the gender and age of adolescents, parents' educational level, and family socioeconomic status in the non-referred sample. In this adolescent group, girls who were 14-15 years old, had parents with no education, and belonged to families with a low socioeconomic status reported higher levels of suicidal ideation. These results suggest that, in addition to analyzing gender and age in the development of depressive symptoms and suicidal ideation, it is necessary to consider the effect of certain social and economic factors that may contribute to the identification of vulnerable groups that could be target populations of preventive programs. Regarding the second objective, we found a positive and strong association between depressive symptoms and suicidal ideation in both groups. This relationship, which was consistent with previous studies [28], was also found to be moderated by different factors. In the referred sample, this link was influenced by adolescents' age. Specifically, our results showed that the relationship between depressive symptoms and suicidal ideation was stronger in the 14-15 and 16-17 year-old age groups, and weaker in early adolescence (the 12-13 year-old age group). These results suggest that this relationship becomes stronger as adolescence progresses. A possible explanation could be related to the stages of cognitive development during adolescence and the acquisition of formal operational in middle adolescence, that is at 14-15 years old [57]. Boys and girls in mid-late adolescence have greater cognitive abilities and exhibit more complex thinking [58]. This may allow them to reflect and consider suicide to be an easy way out of a depressive situation, which in the referred sample sharpens significantly with age. Although adolescents in the 12-13 year-old age group do experience depressive symptoms, they may not have reached a sufficient level of cognitive maturation for the elaboration of a hypothetical plan to commit suicide or to think of suicide as a possible solution to their problems. In the non-referred sample, our results indicated that it was not age, but family socioeconomic status that moderated the relationship between depressive symptoms and suicidal ideation. Our results showed that the association between depressive symptoms and suicidal ideation was stronger in adolescents with a low socioeconomic status. These findings, in line with previous research [39][40][41], indicate that a low family socioeconomic status may be a risk factor for the emergence and maintenance of some internalizing problems. These results may help us to identify vulnerable populations and evidence the convenience of developing primary prevention programs designed to target non-referred adolescents who come from economically disadvantaged family contexts. In this context, it may be beneficial to implement in school programs that promote mental health. These programs could promote the competences and well-being that prevent the occurrence of internalizing problems in adolescents [59]. Although the present study yielded some interesting findings, these should be considered in light of the study's limitations. Firstly, this study had a cross-sectional design which allows us to examine the association between variables, but not to make inferences about causality. Further longitudinal studies are required to shed light on the directionality of relationships and to analyze the developmental trajectory of depressive symptoms and suicidal ideation in referred and non-referred adolescents. Second, assessments of depressive symptoms and suicidal ideation, as well as of social and economic factors, were based on adolescents' self-reports, which may have artificially inflated the strength of the observed associations through shared method variance. However, some authors, such as Andrew et al. [60], have found that adolescents report more realistically about their internal states than their parents or other external informants. Third, we developed this study by considering two samples that were obtained by a quota sampling procedure that controls for the variables age, gender, parents' educational level, and socioeconomic status. This type of sampling procedure was chosen to facilitate comparisons between subgroups and the systematization of the variations produced by the sociodemographic factors in each sample [49]. Nevertheless, this sampling procedure prevented us from offering prevalence rates and comparing samples in terms of some social and economic factors because their presence was controlled based on the established quotas. Finally, it is important that future research examines other social and economic factors intimately related to depressive symptoms and suicidal ideation, such as access to mental health services. The study of a more diverse variety of demographic, social, and economic factors could provide a more complete picture of the factors involved in the development of internalizing problems during adolescence. --- Conclusions The findings of this study emphasize the importance of examining the role that certain demographic, social, and economic factors play in the development of depressive symptoms and suicidal ideation during adolescence. This study also highlights the utility of employing a combined design in which referred and non-referred adolescents are assessed simultaneously. Our results suggest that it is possible that there is no single unique developmental trajectory for these internalizing problems, but a differentiated trend that is based on the individual's mental health status. Moreover, we did not find gender differences in depressive symptoms and suicidal ideation in referred and non-referred adolescents. However, the interaction effect demonstrated that, in non-referred adolescents, suicidal ideation levels were higher in girls who were 14-15 years old, with parents without studies, and belonging to families with a low socioeconomic status. These results emphasized the importance of analyzing, together with gender and age, other social and economic factors associated with the development of de

Depressive symptoms and suicidal ideation are common internalizing problems during adolescence. Numerous studies have explored the role of certain demographic, social, and economic factors in their development in referred or non-referred adolescents, but not simultaneously in both groups. In this study, we examined the association between age, gender, parents' educational level, and socioeconomic status (SES) and depressive symptoms and suicidal ideation in a referred group (n = 211) and a non-referred (n = 1401) group of adolescents. We also examined the moderating role that these factors play in the relationships between both internalizing problems. The results showed: higher levels of depressive symptoms and suicidal ideation in the referred group; an increase in both problems during early-to-middle adolescence in the non-referred group; an association between low SES and suicidal ideation in both groups; an association between low father's education level and depressive symptoms in the non-referred group; and no gender differences in either of these two internalizing problems. The moderation analyses showed that age, in referred adolescents, and SES, in non-referred adolescents, moderated the relationship between depressive symptoms and suicidal ideation. This study contributes to the identification of groups of vulnerable adolescents that could constitute the target populations of preventive programs.

diverse variety of demographic, social, and economic factors could provide a more complete picture of the factors involved in the development of internalizing problems during adolescence. --- Conclusions The findings of this study emphasize the importance of examining the role that certain demographic, social, and economic factors play in the development of depressive symptoms and suicidal ideation during adolescence. This study also highlights the utility of employing a combined design in which referred and non-referred adolescents are assessed simultaneously. Our results suggest that it is possible that there is no single unique developmental trajectory for these internalizing problems, but a differentiated trend that is based on the individual's mental health status. Moreover, we did not find gender differences in depressive symptoms and suicidal ideation in referred and non-referred adolescents. However, the interaction effect demonstrated that, in non-referred adolescents, suicidal ideation levels were higher in girls who were 14-15 years old, with parents without studies, and belonging to families with a low socioeconomic status. These results emphasized the importance of analyzing, together with gender and age, other social and economic factors associated with the development of depressive symptoms and suicidal ideation. In addition, our findings evidence the convenience of considering socioeconomic status and parents' educational level as differentiated variables. When both variables were studied together, it was found that socioeconomic status was related to suicidal ideation and that parental educational level was associated with depressive symptoms. Finally, our data show that demographic, social, and economic factors moderate the relationship between depressive symptoms and suicidal ideation. In the referred sample, our results show that the relationship between depressive symptoms and suicidal ideation was moderated by age and becomes stronger as adolescence progresses. In the non-referred sample, it was observed that family socioeconomic status moderated this relationship, finding that this relationship was stronger in those adolescents whose family socioeconomic status was low. As such, it is important to consider the moderating effect of these factors on the relationship between depressive symptoms and suicidal ideation with the objective of developing specific prevention strategies for each adolescent group. --- Conflicts of Interest: The authors declare no conflict of interest.

Depressive symptoms and suicidal ideation are common internalizing problems during adolescence. Numerous studies have explored the role of certain demographic, social, and economic factors in their development in referred or non-referred adolescents, but not simultaneously in both groups. In this study, we examined the association between age, gender, parents' educational level, and socioeconomic status (SES) and depressive symptoms and suicidal ideation in a referred group (n = 211) and a non-referred (n = 1401) group of adolescents. We also examined the moderating role that these factors play in the relationships between both internalizing problems. The results showed: higher levels of depressive symptoms and suicidal ideation in the referred group; an increase in both problems during early-to-middle adolescence in the non-referred group; an association between low SES and suicidal ideation in both groups; an association between low father's education level and depressive symptoms in the non-referred group; and no gender differences in either of these two internalizing problems. The moderation analyses showed that age, in referred adolescents, and SES, in non-referred adolescents, moderated the relationship between depressive symptoms and suicidal ideation. This study contributes to the identification of groups of vulnerable adolescents that could constitute the target populations of preventive programs.

Introduction Disinformation (or misinformation) is a human language phenomenon that has always existed based on a mechanism of spreading from mouth to ear [1,2]. However, with regard to the Internet and recent quantitative methods, we can investigate it with an up-to-date analysis. In the past, the spread of rumours could only be by word of mouth. The rise of social media provides an even better platform for spreading rumours. As Metaxas [3] explains massive amounts of data are being created and circulated, and often there are individuals or bots trying to manipulate this data to promote their own agenda. But sharing information with others after an emotionally powerful event can be cathartic. Understanding various rumour discussions could help to design and develop technologies to identify and track rumours, or reduce their impact on society. In psychology a rumour is a declaration that is generally plausible, associated with news, and is widespread without checking [2,4]. Some famous rumours are the urban legend "rue des Marmousets" in Paris where a barber and a pastry chef made cake trade based on human flesh in XVth century, or the disappearance of young girls in fitting rooms inside Jewish shops in the town of Orleans (France) in 1969 [5]. According to Gaildraud [6], a rumour is an informal noise that exists, persists, becomes evanescent and disappears as fast as it appeared. The definition of rumour is vague, such as one or several pieces of information that move around by individuals and/or the Internet. In the social sciences, rumouring behaviour is analysed as a social process of collective sense-making through which individuals can understand situations characterised by high levels of uncertainty, anxiety and a lack of official news. Classical social science research proposed two important ways of understanding rumour prevalence: (1) in terms of the amount of rumour-related information present in the environment, and (2) in terms of the number of individuals who have encountered or heard a particular piece of information. However, much of this very early work suffers from a lack of empirical support. Ongoing research on the spread of rumours online is roughly quantitative, including descriptive studies of trace data [7][8][9], theoretical research on network factors [10,11], and prescriptive studies that experiment with machine learning methods to classify rumours as true or false [12,13]. Kwon et al. [8] include a descriptive analysis of temporal characteristics; false rumours on Twitter have more spikes than true rumours. Quantitative understanding of rumours focuses on how people participated in the rumour discussions and how the rumour developed over time. For instance, it could lead to the extraction of patterns in the text content, or different user roles. Rumour analysis has gained from studies in the related fields of memetracking [14], diffusion [15,16] and virality [17,18] in social networks, measuring the influence in networks and information credibility estimation. Yet few studies provide significant insight into how and why rumours spread, and classification research has been limited to distinguishing between true and false information. Current studies work like outlier detection of a specific database. Hence, they learn a local model that is specific to a social media, not applicable to another platform, and they speculate that a rumour is a negative message, like'spam', which need to be rejected from the platform. One theory is nevertheless interesting in spreading rumor in a community [2]. They argue that transmission evolves in three steps: levelling, sharpening and assimilation. First step is deleting details, second step is keeping the main details, assimilation is transmission with noise. We can take advantage of social network datasets to test such theory. Taking the automatic content analysis and data mining processing of a message [19][20][21], we are interested in exploring the following research questions, summarised below: Q1: Which features are relevant? Q2: Can we model a rumourous event as a multi-spike event? Q3: How is a rumourous text different from a non-rumourous text? Q4: Can we observe levelling-sharpening-assimilation in datasets? In our article, part 1 is dedicated to an extensive review of literature of 80 papers on rumours. Among them, 58, written after 2010, were about rumour studies, revealing recent interest in rumour/credibility/misinformation issues, and specifically with social media platforms. We made a synthesis of principal features used to describe rumours in these quantitative approaches. Feature selection is a key question in quantitative and modelling investigation. Part 2 presents the datasets we used for spread and content analysis. We used not only ad-hoc corpora for our studies, but also external databases, such as hoaxes/disinformation repositories and language corpora. Part 3 presents our modelling approach for rumour spreading and a comparison with a standard approach such as epidemiological models. Finally, part 4 shows a comparison of rumour corpora and event corpora with n-gram and skip-gram studies. --- Material and methods --- Related studies Rumour theory. In psychology and sociology [1,2,22,23] were first attempts to study rumor and showing increase errors across the retellings. Rumours can be hoaxes, jokes, little stories or information leaks [24][25][26]. But it can be also early reports during breaking news lacking enough support or evidence. If we look at the classification proposed by [27], we observe seven categories of rumours: computer virus alerts, superstitious chains, solidarity chains, petitions, hoaxes, urban legends, fun stories and funny photos/pictures. But [28] imagined another classification with nine topics: urban legends, commercial disinformation, political attacks, commercial offer attacks, false commercial offers, financial disinformation, defamation, loss of credibility operations and panic alert to induce terror. Often a rumour is dedicated to disturb VIPs [6]. Recently, others [29] have suggested that rumours are a communication strategy similar to speech acts [30,31]. Rumour detection. Recently, more computing studies have investigated the emergence of rumours, but they stay at the level of a specific rumour, as in Fig 1 [32][33][34][35][36][37][38][39]. Contrary to these studies, our goal is to analyse any kind of rumour and a corpus of rumours. Some systems claim to detect rumours but they are based on the similarity between an unknown message (i.e. email) and a well-known database of hoaxes or rumours [41][42][43]; other kinds of systems are more of a surveillance system for interesting message detection from the Internet (that are possibly rumours), and in this sense, they are more like an approximate recommendation system [44]. Formulation of the problem: Microblog data can be modelled as a set of events = <unk>E i <unk>, and each event E i consists of relevant microblogs for which we can associate a value for being or not being a rumour <unk>m ij, y i <unk>. An event E i can be described by a set of k features from l different categories <unk>F kl <unk>. Hence, each message m ij can be described by some values of these features. The most difficult case is to discover, in an unsupervised way, the value y i for any message. In some cases we can know this value for a reduced amount of data from which we can learn a model (i.e. a profile), in a supervised way, and to detect similar messages. [45] makes a good survey in the field of rumor detection. Most of the existing research uses common supervised learning approaches such as a decision tree, random forest, Bayes networks and a support vector machine (SVM). [46] imagined of first rumour detection system for the Chinese language and the Weibo social network. Weibo has a service for collecting rumour microblogs [47]. Qazvinian et al. [13] used a tagged corpus of 10,000 tweets of about five rumours, five categories of features (1-grams, 2-grams, Part-of-speech, hashtags, URLs) to classify rumours using the log-likelihood approach with good results (95% of accuracy) but they cannot apply their method to new, incoming, emergent rumours. Rumour propagation. We can see rumour messages as a bag of documents, but also as a timeline with occurring messages. In that way, the formulation of the problem is a little different because it concerns the description of a discrete time series evolving over time [48]. Some previous work [49,50] focuses on rumour propagation through the social network. They try to use graph theory to detect rumours and find the source of rumours. Virality is a major concept in rumour propagation [51], using epidemiological models, and some current studies still try to improve the models [52]. Spiro et al. [9] also model the rate of posts over time in their exploration of rumouring during the Deepwater Horizon oil spill in 2011. [53] identified five kinds of rumour statements, coded posts accordingly, and presented a model of rumour progression with four stages characterised by different proportions of each statement type. The website TwitterTrails [54,55] is one of the rare tools that does not present only a database but also intelligent information exploration (timeline, propagators, negation, burst, originator, main actors) in 547 social media stories. [10] prove that minimising the spread of the misinformation (i.e. rumours) in social networks is an NP-hard problem and also provide a greedy approximate solution. Kwon et al [8] promoted uses of both temporal features, structural features and linguistic features. Linguistic features are related to the most words used in messages and taken from a sentiment dictionary (4,500 words stem). Network features are properties about the largest connected component (LCC). Temporal features point out periodicity of rumour phenomenon and give importance to an external shock that may incur not one but multiple impacts over time; here, the main feature is periodicity of an external shock. Fang et al. [56] describe a quantitative analysis of tweets during the Ebola crisis, which reveals that lies, half-truths and rumours can spread just like true news. They used epidemiological models. Fang et al. [56], studying 10 rumours about the Ebola crisis in 2014, claim that rumours propagate like news but they encourage quantitative analytics to distinguish news from rumours. Granovetter [57] explains with its seminal work about weak ties, that some nodes in social networks mediate between different communities. Acemoglu et al. [58] give importance to bridges in social networks to spread biased beliefs. Menczer [59], in a talk for a world-wide web conference, underlined the importance of misinformation detection and fact checking, with goods results from machine learning techniques. Social media and traditional media work together to spread misinformation. Structural, temporal, content, and user features can be used to detect astroturf and social bots. --- Rumour sources Disinformation sources. We are focusing on digital data that may be grabbed from the Internet. Others sources allow free access to misinformation like the website Emergent [60]. It monitors and evaluates the propagation of a rumour that has recently received a lot of attention. A new web service, emergent.info, developed by journalist Craig Silverman, is using journalists to evaluate online claims and deem them as true/false/unverified. They track the number of shares a rumour has on Facebook, Twitter and Google+ and report the numbers along with links to articles that support or counter the rumour. We identified at least seven websites containing curated databases and serve as a reference to inform and to provide reassurance about rumours and disinformation on the web. These databases contain not only rumours but also hoaxes and jokes that may propagate on the Internet. 'Snopes' is the biggest, but with 'hoaxkiller', it is impossible to know how many articles it contains because the interface requires query function by keywords (Table 1). 'Hoaxkiller', 'hoax-slayer' and 'dehoaxwijzersite' are databases that display a list of hoaxes to show hoaxes and frauds. 'Debunkersdehoax' is a website that helps to invalidate rumours and disinformation from nationalists. 'Hoaxes.org' is a website that explores disinformation throughout history. 'Snopes' covers urban legends, rumours on the Internet and email, and other doubtful stories. We made a crawler (robot in perl language) to collect automatically the content of each website. The famous and open encyclopaedia, Wikipedia, gives 220 as the number of existing social networks on Internet. These social media play as web 2.0 platforms with thousands till millions of active users where information as rumours can propagate quickly and easily. Twitter is one of them, and probably the most famous microblogging platform where 500 million tweets are published each day and 600 million users are registered, with 117 million active accounts publishing at least one tweet per month. Such a social platform is an ideal dissemination'relais' for rumours. Two API (application programming interface) allows any computing programme to query the twitter database. Twitter Search API can index more than tweets but only from the previous seven days. Twitter Streaming API can retrieve more messages, but no more than 1% of the content per day. From the database cited in Table 1, we compiled a corpus of 1,612 rumours (DIS-corpus) and disinformation texts among with 1,459 in English and 153 in French (81,216 tokens; 6,499 words). Part 2 presents information sources and datasets. Part 3 is related to propagation. Part 4 addresses issues about information patterns in messages. We used R as the computing framework for modelling [61]. --- Text data collections: Social media corpora and reference corpora From Table 1, it is possible to see a sample of texts that is more related to rumours and disinformation because texts from databases are classified with categories. Hence, we were able to grab 1,612 texts discussing rumours (1,010 texts) and disinformation (602 texts). The size of the texts is relatively small, such as the news. But it is quite difficult to automatically select lexical information (by one or two words) that is typical from a given text. So we have manually chosen four texts and built a lexical query with two or three words to grab tweets from the Twitter social network (S2 Appendix). From data collected in an open-access web database, we made a manual query to grab tweets from Twitter [62], and we built eight corpora to compare with the rumour corpora (Table 2). The first rumour, 'Hollande rumour', is about the French political leader Franc <unk>ois Hollande. The rumour started in 2002 in private parties and in editorial offices. According the rumour scenario, the president of France-at that time he was deputy of the Correze region and first secretary of the labour party-was the father of one of Anne Hidalgo's children, at that time, the First Executive Assistant of the Paris governor. Wikipedia's description of Anne Hidalgo highlights that she had two children from a previous relationship. A black hole of information is sufficient to excite the web. The following query induced the retrieval of data: --- (hollande AND hidalgo AND fils) lang:fr The 'lemon rumour' pointed out that a lemon could cure cancer, saying it exceeds the power of chemotherapy by 10,000. The origin of this rumour is a Reuters news article in 2003, 'An Orange a Day May Keep Some Cancers Away'. The following query induced the retrieval of data: --- (citron AND cancer) -femme-campagne-musique-arabes-punk-branché-limonade-Kickstarter-gato-Crowdfunding-Baptême-court-CM-tittytuesday-morito-nestea-bracelet-aluminium-déodorant-déodorants-agrumes-pucepoils-tropic-art-astrologie-bouteille-crame-coude-photo-tartes-bronzage-olive-horoscope-bonbons-google-jeu-hypocrisie-rose-malboro-Ananas-Bronzage-quantitatif-Tropiques-Téflon lang:fr The 'PIN rumour' claimed that in New York, entering your personal identification number (PIN) backwards will automatically send a message to the police that you are in trouble and that they will respond to the machine. This rumour seems to have appeared in 2006. The reverse PIN system was first imagined in 1994 and patented in 1998 by Joseph Zingher but never adopted by the banking industry. The following query induced the retrieval of data: (pin AND atm AND police) lang:en 'Swine flu rumour', related to the swine flu virus or officially called the H1N1 flu virus, mentioned that thousands of people were sent to the hospital during the soccer championship in 2009 in South Africa. The following query induced the retrieval of data: ("swine flu"AND "South Africa") lang:en There are two kinds of reference corpora. The first group is random corpora made on Twitter with a stopword. We chose the first 1000 tweets for each operation, repeated two times and for both French and English. The second group is related to events, and we also collected data from Twitter in April 2016. First event is a concert in France in August 2016 by Rihanna. The second event is the UEFA Europe football championship in France in 2016. For both events, data was collected in French and English and we kept no more than 1000 tweets. We used two reference corpora for comparison with common language and for each language (Table 3). FR-corpus is an open database that contains 500 literary works from the 18 th to 20 th century. It is a free sample of the Frantext online database containing 248 million words [63]. ER-corpus is a collection of news from the French local newspaper East-Republican ('L'Est Re <unk>publicain') about 1999, 2002 and 2003 [64]. BNC-corpus is a collection of samples of written and spoken language of British English from the latter part of the 20 th century. The written part consists of extracts from regional and national newspapers, specialist periodicals and journals for all ages and interests, academic books and popular fiction, published and unpublished letters and memoranda, school and university essays, among many other kinds of text. The spoken part (10%) consists of orthographic transcriptions of unscripted informal conversations and spoken language collected in different contexts, ranging from formal business or government meetings to radio shows and phone-ins [65]. The COCA-corpus contains spoken texts, fiction, popular magazines, newspapers, and academic texts produced between 1990 and 2015. It is a free sample of the 520 million word original corpus [66]. --- Information propagation Classical epidemiological models. In the Internet era, many studies about rumours have shown that that rumours disseminate as a disease contagion like a Poisson distribution. We tried to confirm this hypothesis. We made two displays of propagation with our four rumours corpora. First, visualisation is obvious, and we can plot the occurrence of tweets as on a timeline in a histogram plot. We do not know the IP number of senders of a tweet but we can know if a tweet is a retweet, hence, if a tweet has been transmitted. More generally, we can study the natural language content of each tweet. Hence, the second visualisation concerns tweet grouping by similarity to explore their distribution over time. A rumour can be seen as a disease propagating over a population of sane individuals becoming infected over time. Several models are possible. Let be S the sensible population that is likely to be infected, E the population that is exposed, I the population that is infected and R the population that is cured. Eq (1) to Eq (18) summarise main models (Fig 2 shows the respective infected output for each model). The most simple is the SI (sensible-infected) model created by Hamer in 1906. In this model no individual can be cured. <unk> Parameter is valued between 0 and 1. <unk>*P(S$I)<unk>P(S!I), where P(S$I) is the probability that a sensible dS dt 1<unk>4 <unk> bSI N <unk> gI dI dt 1<unk>4 bSI N <unk> gI Eq<unk>3<unk> (d). SIRS model dS dt 1<unk>4 <unk> bSI N <unk> fR dI dt 1<unk>4 bSI N <unk> gI dR dt 1<unk>4 gI <unk> fR Eq<unk>4<unk> (e). SEI model dS dt 1<unk>4 <unk> bSI N dE dt 1<unk>4 bSI N <unk> <unk>E dI dt 1<unk>4 <unk>E Eq<unk>5<unk> (f). SEIR model dS dt 1<unk>4 <unk> bSI N dE dt 1<unk>4 bSI N <unk> <unk>E dI dt 1<unk>4 <unk>E <unk> gI dR dt 1<unk>4 gIEq<unk>6<unk> (g). SEIS model dS dt 1<unk>4 <unk> bSI N <unk> gI dE dt 1<unk>4 bSI N <unk> <unk>E dI dt 1<unk>4 <unk>E <unk> gI Eq<unk>7<unk> (h). SEIRS model dS dt 1<unk>4 <unk> bSI N <unk> fR dE dt 1<unk>4 bSI N <unk> <unk>E dI dt 1<unk>4 <unk>E <unk> gI dR dt 1<unk>4 gI <unk> fR --- Eq<unk>8<unk> Harmonic modelling. A harmonic oscillator is an ideal oscillator that evolves over time by a sinusoid, with a frequency independent of the systems properties, and the amplitude is constant. Oscillations can be damped, and the equation is hence written as follows: d 2 s dt 2 <unk> 2 t ds dt <unk> o 2 0 x t <unk> <unk> 1<unk>4 0Eq<unk>9<unk> If o 0 > 1 t state is sub-critical, solution is a damped oscillation with such pulsation: o 1<unk>4 2pf 1<unk>4 o 0 : ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi 1 <unk> 1 t 2 o 2 0 s Eq<unk>10<unk> s t <unk> <unk> 1<unk>4 A:e <unk> 1 t :cos ot <unk> <unk> 0 <unk> <unk>Eq<unk>11<unk> where A is the amplitude, f is the frequency, <unk> 0 the phase to origin, <unk> the pulsation, <unk> the relation time. Models implementation. Epidemiological model displays were done with R with the basic plot function. Experimental implementation of harmonic modelling was done by fast Fourier transform using fft function and least-square in R using function nls (stats package) [61]. --- Rumour lexical content Frequent syntagmatic extraction. In this part we try to understand what kind of combinations can be typical of a rumour or a set of messages about a specific rumour. We can set two main kinds of combinations. The first ones are lexical n-grams. A lexical ngram is a sequence of n contiguous words separated by a blank. If n = 1, it is a simple word (as we can see in any dictionary entries for instance) if n>1, it is what it is named in linguistics 'collocations'. Some collocations can be paradigmatic and then they are named 'phrases' (if they do not contain verbs, they are named 'noun phrases'). The second kind of combination is a set of 1-gram separated by an n-gram not included in the combination. In case such a combination consists of two n-grams, it is named 'co-occurrence'; in the cases where it is several ngrams, it is called a 'frequent itemset'. We can also find the word'skipgram', by analogy of ngram. Rare syntagmatic extraction. We tested the capacity of a rumour text to involve a nonstandard combination of words. For such studies we used common languages corpora. The first experiment is an extraction of cleaned n-grams, and we checked presence/absence in reference corpora. The second experiment is a check of frequent skipgrams consisting of most frequent simple words. In the first experiment we measured originality of a given corpus by the ratio MW c of ngrams not included in a reference corpus by the number of total segments. We used 12 corpus among those four rumours corpus, but also randomly constituted corpora, and corpora based on recent real-world events in French and in English (in the present case: Rihanna concert in Europe in summer 2016, and UEFA Euro 2016). The measure MW c is expressed as follows: MW c 1<unk>4 NMW c <unk>no<unk> NS cEq<unk>12<unk> where NMW c = NMW c (no)+ NMW c (yes) with NMW c is the number of multiwords in the corpus c and NMW c (no) is the number of multiwords not contained in a language reference corpus (for instance COCA-corpus for English). Syntagmatic combination analysis. Finally, the next step after analyzing lists of features of 2 or 3 words is to measure the incidence of content with vector of words. For that, we cannot use the DIS-corpus because each rumour is unique and a set of ten or twenty words could not show similarity with other rumours. But if we take the Twitter rumours, we can observe how people talk about a rumour and compare the specificity of rumour discourse with ordinary messages. We would like now get an overview of words importance in the rumorous content over time. Recall that (Allport, and Postman, 51) specifies a rumor mechanisms in three different mechanisms applicable in any situation. The first mechanism is a selection of main features (leveling, or loss of details). The second mechanism is sharpening refers to is an emphasis of some details during the transmission. Finally the last mechanism, assimilation refers to a distortion in the transmission of information. Linguistic assimilation usually consisted of inserting the words "is," "is as," "as," or "it's" or noise. Let suppose a rumor starts with nine details and ends with three, they would say that six were leveled and three were sharpened. Our empirical studies is done in four steps: • first step is lexical preprocessing of the dataset-splitting data into elementary words. • second step is time preprocessing of the dataset-splitting dataset into 7 timestamps (getting enough data in each chunk at least 50 messages). • third step is subset preprocessing of the dataset-splitting word features into three box according Zipf law saying that lexical distribution is always distributed into a small set of high frequency, medium frequency set words, and big set of low frequency. • fourth step is computation of transitions. • fifth step is plotting transitions. We implemented the scripting in R platform, using regular expression for lexical splitting, 'intersect' function for calculation of transitions and GMisc'package 'transitionplot' for display of transitions. Another angle to capture association is machine learning algorithms. Why, because machine learning algorithms use features, often within non-linear techniques indirectly taking into account combination of features. In summary, it captures correlation of features to make a good prediction without specifying association between features. We used four famous algorithms to make prediction: 'Maxent', 'Random Forest' (regression tree), 'SVM' and 'SLDA' (topic model). The first question that arises, due to sensitivity of algorithms to the feature space, is to define the dimensionality of the feature space. We can take the whole set of words (between 3,000 and 4,000 words) but it can be time consuming for some techniques or noise generation. We make a documents x terms matrix using different samples, i.e. the 10, 50, 100, 150, 200 and 300 most frequent words. We consider that rumorous messages starting by the same 70 characters (half of the message) are the same and we delete them for building the dataset. Hence the dataset consists of 1,678 messages containing all the four rumors messages, the pool of message to predict. We mixed this subset with 9,818 non-rumor messages. As training dataset we chose all the rumor subset and 2,000 nonrumor messages. As test dataset we take the 1,648 rumorous messages (17%) and 8,170 nonrumorous messages (83%). As baseline for comparison of techniques we consider the random assignment. A message can be assigned randomly as rumorous or non-rumorous. So the success rate is 50% percent of accuracy. Let suppose we classify all messages as nonrumorous we get 83% of accuracy but we lost all rumorous prediction because accuracy for rumorous will be 0%. Hence for each classification method we compute two indicators that are the global accuracy that we want enough high better than random for a stream of both rumorous and non-rumorous messages, and accuracy specific for rumorous messages that we expect also close to random score. In the next experiment we keep the same matrix as before with 100 most frequent feature space but we change the document space. We make three submatrix: the first submatrix is 100% of the document space (1,618 rumorous messages), the second submatrix is the first 30% over time (498 rumorous messages), the last 30% over time (524 rumorous messages). Amount of non-rumorous messages in test set is always about 8,000 messages, and for the train set we keep the same amount than the rumorous set (about 500 or 2,000 messages). Models implementation. The experimental implementation was done in R. The syntagmatic extraction is a function using regular expression analysis with gsub function (base package), multi-word extraction with ngram function (ngram package), and data cleaning using a stopwords list. Classification models were created using train_model function (RTextTools package) [61]. An advantage of the oscillator model is that it produces several local maxima (see Fig 5), whereas epidemiological models produce only one or two local maxima. --- Results --- Spreading modelling Fig 4 shows us a fit of Hidalgo-corpus with a damped oscillator model. It fits quite well, and better than any epidemiological model. But it seems that amplitude is not stable. s t <unk> <unk> 1<unk>4 P n i1<unk>41 A 1i :e <unk> t A 2i :cos A 3i t <unk> A 4i <unk> <unk>Eq<unk>13<unk> A --- Frequent syntagmatic extraction Table 4 shows us a list of frequent n-grams for each corpus of rumours: Hidalgo-corpus, Lemon-corpus, Pin-corpus and swine-corpus. 'Counting' is the number of occurrences in terms of documents about cleaned n-grams. We cleaned n-grams by subtracting the prefix or suffix matching with stopwords. Processing is done in both languages. In Table 4 no information appears to make sense for a rumour in general. We mostly distinguish lexical patterns clearly related a given rumour like 'flu death', 'h1n1', 'Africa swine', 'flu cases' for swine corpus. If we look at Table 4's top four lexical strings, we see that only simple words appear; it is a general observation that stopwords are more frequent than simple words, and simple words are more frequent that multi-words. Next we tried to extract the most frequent simple words over the 1,612 rumourous texts (1,459 in English, 153 in French). Table 5 shows the most frequent words in the database by decreasing order of occurrences or documents. If we set a threshold such as 10% of documents (146 in English, 15 in French) and if we consider the number of occurrences, we observe that only 20 simple words are significant. Among these words we can see only two words about a specific topic (cancer, Obama) and no word very typical for a rumourous alert. If we consider the number of documents, 160 words are relevant (64 in French, 96 in English). Most of words are very short (two or three characters). We cannot see any named entity in these lists (person's name, organisation, product names). Many words seem to be tool words such as: pro, ex, hey, side, app, etc. Another big cluster of words are general verbs such as go, use, eat, see, etc. Some general meaning words seems recurrent too such as men, one, day, king, war, ease, etc. We cannot extract any global argumentative structure of a rumour that is redundant across a large set of documents. Table 6 represents another view of word frequency in the text database. It points out the distribution of lexical units (1-grams) over each database (French, English). We kept only words occurring in more than 10% of the documents, and we are displaying the list of words by decreasing order of coverage per cent. More French words are involved because 10% of a small sample covers only 15 documents. For English documents only three words cover more than 25% of the corpus: one, people, know. These words are not informative about a rumour's general representation. We can also find prepositions or adverbs such as like, now, us. For French, 17 words cover 25% of documents,

Rumour is an old social phenomenon used in politics and other public spaces. It has been studied for only hundred years by sociologists and psychologists by qualitative means. Social media platforms open new opportunities to improve quantitative analyses. We scanned all scientific literature to find relevant features. We made a quantitative screening of some specific rumours (in French and in English). Firstly, we identified some sources of information to find them. Secondly, we compiled different reference, rumouring and event datasets. Thirdly, we considered two facets of a rumour: the way it can spread to other users, and the syntagmatic content that may or may not be specific for a rumour. We found 53 features, clustered into six categories, which are able to describe a rumour message. The spread of a rumour is multi-harmonic having different frequencies and spikes, and can survive several years. Combinations of words (n-grams and skip-grams) are not typical of expressivity between rumours and news but study of lexical transition from a time period to the next goes in the sense of transmission pattern as described by Allport theory of transmission. A rumour can be interpreted as a speech act but with transmission patterns.

name, organisation, product names). Many words seem to be tool words such as: pro, ex, hey, side, app, etc. Another big cluster of words are general verbs such as go, use, eat, see, etc. Some general meaning words seems recurrent too such as men, one, day, king, war, ease, etc. We cannot extract any global argumentative structure of a rumour that is redundant across a large set of documents. Table 6 represents another view of word frequency in the text database. It points out the distribution of lexical units (1-grams) over each database (French, English). We kept only words occurring in more than 10% of the documents, and we are displaying the list of words by decreasing order of coverage per cent. More French words are involved because 10% of a small sample covers only 15 documents. For English documents only three words cover more than 25% of the corpus: one, people, know. These words are not informative about a rumour's general representation. We can also find prepositions or adverbs such as like, now, us. For French, 17 words cover 25% of documents, and among those, only two words are semantically significant-France, pays-but very general in any case. Other significant words are logical and argumentative such as: si, donc; but they still have a very global meaning for a consequence or condition. Other less frequent words deal with different topics such as people and domestic policy. An interesting fact is that the word true is often used in a message claiming a falsehood. We would like now get an overview of words importance in the rumorous content over time. Rumorous datasets were initiated before creation of twitter platform except for'swine flu' that emerged in 2009. About 'lemon', 'hidalgo', and 'pin' we can not observe the levelling step. About'swine flu' we do not observe any loss of lexical information at beginning of the rumour propagation (see Fig 6). Sharpening in a transition point of view can be seen as frequent words that can become more frequent. Assimilation can be seen as noise words that come in and out. Our transition diagram can differentiate growing in frequency details (transfer from low and medium boxes to high frequency box)-i.e. sharpening-and capturing noise (transfer from low to medium The second experiment is based on simple words shown in Tables 5 and6 from which we made a file of 144 simple English words; we computed all combinations between two words (2-skipgrams) and three words (3-skipgrams). Hence, we checked the presence or absence of each skipgram in the corpora of common language in English (COCA-corpus). --- Rare syntagmatic extraction In Table 8 we see that only five 3-skipgrams are not inside the common language corpus: • obama please thing • alert obama sh • number obama please • alert info obama • don obama please Specificity of these combinations is clearly related to the Obama name and cannot provide information about rumour structure in general. --- Syntagmatic combination analysis On Fig 6 we can see different groups of similar messages for Hidalgo-corpus over time. At the beginning are two distinct groups of messages in bright blue and red, and at the end, a cluster in green. This figure shows us that during a flow of messages for a specific rumour, groups of similar messages can emerge in the same time window. Fig 7 shows that bursts of similar messages occur over time, and leads us to think that indeed the content of rumour discourse is not heterogeneous. We can suppose that a rumour discourse consists of local grammar and typical vocabulary in Twitter but also in the primitive short text. We plotted a timeline occurrence of rumours sorted (y-axis) by message similarity. Another angle to capture association is machine learning algorithms that use features, often within non-linear techniques taking into account combination of indirectly correlated features. Fig 8 shows four plot for each classification methods. On each plot we have three curves: random (in black), rumorous accuracy (in red), global accuracy (in blue). We see that scores are not so good for a small amount of features (less than 50,) and scores degrade when they are more than 200 features. So we decide to keep the solution of 100 features. Fig 9 shows the results. We can observe that the behaviour of predication is almost the same for Random Forest, SVM and SLDA and we see that there is a change between the overall dataset prediction behavior and the first 30% dataset, and the overall dataset keep the same behaviour as the 30% last dataset but with a degradation of performance in prediction. It means an impact of the lexical composition over time that changed. Maxent seems to have a bad behaviori with low score of prediction. If we filter the number of prediction with more than 60% of certainty, we get only about 3,727 values, when other methods have about 9,500 values. When using the whole set of features (3,336, instead of 100 most frequent), the amount of values with high confidence raises to 7,351 but we still get only 9,2% for accuracy about the rumorous set when other methods get more than 33%. Maxent seems to work better with a highest dimensional space, but keeping a lower performance. --- Discussion Our results show the complexity of rumour description and tracking in its diverse facets. Rumour analysis, being a psycho-social phenomenon, has regained interest because of social media platforms that relay news efficiently and widely, as well as events and information about important persons or organisations. Relevant studies have proven that the integration of specific features for automatic detection gives interesting results for case studies. Globally, there is no comparison of the difference between news and rumours. Furthermore, relevant features involved in models reveal that some misinformation lacks specific features or have more specific features, but each social media space can generate its own properties and because of this, rumours can spread with a combination of features that are not found in existing platforms (like Weibo or Wikipedia). Indeed we observed 53 features involved in models, but the combination of these features is high and it is not realistic to imagine a unique set of features to anticipate the shape of a rumour in a given digital context. Globally detecting rumours can be implemented locally in the context in which it is spread for a specific category of users. Can we imagine a connected world without rumours? Language evolves in any social world, and a rumour is in itself a marker of the language at a rhetorical level. So rumours can evolve in the same way that language evolves. For instance, a series of hashtags in a microblog can be a new kind of message, but in the same way a new kind of rumour construction. A rumour lifecycle evolves naturally like a scientific hypothesis, requiring confirmation or denial by other publications; in this sense, the majority of people socially accept this rhetorical process. --- Conclusion To complete rumour and disinformation studies widely explored by qualitative means, we decided to investigate quantitative issues across any data sources. We studied several rumour datasets leading to a disinformation corpus of 1,612 rumourous texts (in French and English) from which we chose four rumours (French Hidalgo politician, lemon and cancer, ATM PIN code and swine flu in South Africa). We manually built two or three keyword queries to get tweets data about these four corpora. About the propagation of each rumour over time, we highlighted different profiles that may be either epidemiological-based but multi-harmonicbased. Focusing on the disinformation corpus we found that the intrinsic lexical content of rumours themselves has no specific content in term of lexical patterns when we compared them with reference corpora for the English or French common language, or to the corpora of event-based tweets. We tried also to highlight some previous theory of rumor argueing a transmission in three steps: levelling-sharpening-assimiliation. Taken this as a basis, we consider social network data as an empirical framework to provide data for validation of such theory. We can only confirm the assimilation part; we guess that levelling and sharpening occur enough early in dissemination and we do not observed it under the scope of 4 given rumors. So we distinguish two properties of rumors, largely disseminated in natural language (as a speech act) whereby they seem to have lexically no specific genre, and have a propagation with a certain resilience and assimilation process. --- Competing interests: The author has declared that no competing interests exist. --- Supporting information S1

Rumour is an old social phenomenon used in politics and other public spaces. It has been studied for only hundred years by sociologists and psychologists by qualitative means. Social media platforms open new opportunities to improve quantitative analyses. We scanned all scientific literature to find relevant features. We made a quantitative screening of some specific rumours (in French and in English). Firstly, we identified some sources of information to find them. Secondly, we compiled different reference, rumouring and event datasets. Thirdly, we considered two facets of a rumour: the way it can spread to other users, and the syntagmatic content that may or may not be specific for a rumour. We found 53 features, clustered into six categories, which are able to describe a rumour message. The spread of a rumour is multi-harmonic having different frequencies and spikes, and can survive several years. Combinations of words (n-grams and skip-grams) are not typical of expressivity between rumours and news but study of lexical transition from a time period to the next goes in the sense of transmission pattern as described by Allport theory of transmission. A rumour can be interpreted as a speech act but with transmission patterns.

Background Historically, population-based studies of mental health disorders in Non-Latino Blacks have found that the rates of disorders are lower than or equal to those of Non-Latino Whites [1][2][3]. We have learned in the last couple of decades predominantly from the National Survey of American Life that these disorders may vary considerably by nativity and ethnic subpopulation membership [4]. While in general rates of mental health disorders are low in Blacks as a group particularly for Black men, Caribbean Black men have been found to show an elevated burden of mood, anxiety and substance use disorders compared to African American men [5][6][7]. Despite this finding of lower or equal rate of disorders when Blacks are diagnosed with mental health disorders, they are also more likely than Whites to rate their disorders as more severe, disabling and persistent in nature [1,6,7]. Due to the low rates of help-seeking and use of mental health treatment by Black men in the United States, our knowledge [8] about the chronicity of their disorders is nascent and in need of additional study. Population-based studies of mental health needs of Black men have examined rates of psychiatric disorders among Black fathers [9], correlates of depression among African American men [10][11][12], and barriers to professional help-seeking in this population [13,14]. Moreover, we are increasingly gaining an appreciation of ethnic and nativity variations in mental health needs in the U.S. The Patient Protection and Affordable Care Act (ACA) presented an opportunity to improve the treatment of psychiatric disorders for racial and ethnic minorities. Expansions of health insurance and investments in the primary care and safety net systems are likely to provide a new pathway to treatment for African American and Caribbean Black men [15][16][17][18]. With these opportunities, improved understandings of ethnic and nativity variations in mental health service needs are required to inform the development of health services interventions that will ultimately reduce racial/ethnic disparities and improve the health and well-being of Black men in the U.S regardless of whether it is the ACA or some other approach to integrated mental health services. --- Study Aim The current study aimed to investigate ethnic and nativity differences in the chronicity and treatment of psychiatric disorders among Black men in the U.S. including Bipolar I and II. We used data from the National Survey of American Life (NSAL) to provide nationally-representative estimates of the prevalence and chronicity of psychiatric disorders, as well as mental health services use for African American and U.S. and foreign born Caribbean Black men with DSM-IV mood, anxiety, and substance use disorders. Epidemiological information regarding the prevalence and chronicity of psychiatric disorders across immigrant and non-immigrant groups of Black men is important for identifying cultural and environmental factors that contribute to Black men's mental health. Moreover, such information is needed to develop targeted interventions to reduce the burden of untreated mental disorders in African American and Caribbean Black men. --- Methods --- Study Design We used data from the National Survey of American Life (NSAL) a cross sectional survey of DSM-IV psychiatric disorders and mental health service utilization among Black Americans [19]. The sampling frame for the survey used a four-stage probability sampling design [20]. The core sample is nationally representative of individuals living in households located in the contiguous United States where at least one Black adult (age 18 or older) resides. To recruit persons of Caribbean descent, geographic areas with a high density (at least 10%) of Caribbean Blacks was also included in the sampling frame. Participants were interviewed between January 2001 and March 2003 with response rates of 70.7% for African Americans and 77.7% for Caribbean Blacks. Population-based weights were created to account for the unequal probability of selection, nonresponse, and post-stratification [21]. --- Analytical Sample The NSAL includes 6072 individuals including 1271 African American men and 643 Black men of Caribbean descent. For the current study, we excluded 33 African American men who were born outside of the U.S. because the sample size was too small for analyses. We also excluded 16 African American men and 6 Caribbean Black men because their nativity status was missing. The analytical sample in this study included 1222 African American men born within the U.S., 176 Caribbean Black men born within the U.S., and 461 Caribbean Black men born outside of the U.S. --- Measures --- Psychiatric Disorders The NSAL assessed the presence of probable lifetime and 12-month psychiatric disorders using the World Mental Health version of the World Health Organization's Composite International Diagnostic Interview (WHO-CIDI) [22,23]. The WHO-CIDI is a fully structured interview used to assess the prevalence of DSM-IV psychiatric disorders. In the current study, we examined patterns of five anxiety disorders (General Anxiety Disorder, Panic Disorder, Agoraphobia, Social Phobia, Post-Traumatic Stress Disorder; PTSD), four mood disorders (Major Depressive Disorder, Dysthymia, Bipolar Disorder I and II), and four substance use disorders (Alcohol Abuse, Alcohol Dependence, Drug Abuse, and Drug Dependence). In addition, we investigated lifetime and twelve-month prevalence of a panic attack. --- Race and Ethnicity Participants were grouped by race and ethnicity based on the racial/ethnic classification in the U.S. Census. For the current study, we use the term, Black, to refer to those participants who by self-report labeled themselves as Black. We also use Caribbean Black to refer to participants who both identified as Black and reported ancestral ties to a Caribbean country. African American refers to participants who identified as Black, were born in the U.S., and who did not report ancestral ties to the Caribbean. --- Chronicity Disorder chronicity was defined as the proportion of adults with a lifetime psychiatric disorder that met criteria in the twelve months prior to the interview [6,24]. The sample was restricted to the 486 Black men who met criteria for a lifetime psychiatric disorder and who reported an age of onset at least two years prior to the interviews. This measure is a proxy for identifying those with a disorder lasting more than twelve months [6,7]. We coded participants as positive for lifetime mental health service utilization if they indicated ever seeking help for nerves, emotions or mental health, or for use of alcohol or drugs from a medical or mental health provider. In this study we included formal health service providers including general medical providers (general practitioners, medical specialist (e.g., cardiologist), medical professionals (e.g., nurse), and specialty mental health providers (e.g., psychologist, psychiatrist, counselor, or social worker in a mental health setting). --- Statistical Analysis The analyses were conducted in four stages. First, we used cross-tabulations to investigate prevalence estimates of lifetime and prior twelve-month disorders, chronicity, and mental health services use across the three groups of men. We present weighted study proportions and standard errors adjusted for the complex survey design. Second, we used multivariate logistic regression models to compare the prevalence of psychiatric disorders (lifetime and twelve-month disorders) among men varying in U.S. nativity and Caribbean heritage. In these analyses, we compared the adjusted odds of meeting the DSM-IV disorder criteria among Caribbean groups compared to African American men. We then evaluated nativity differences in adjusted odds of meeting the DSM-IV criteria among Caribbean Black men (the African Americans were excluded). The lifetime and twelve-month prevalence analyses controlled for participant characteristics known to vary in risk for a psychiatric disorder [3]: age, household income, poverty status (a ratio of family income to the U.S. census poverty threshold in 2001), education, employment status, marital status, and geographic region of the U.S. Third, we used logistic regression analyses to examine group differences in the chronicity of each measured disorder. The sample was restricted to those participants whose disorder began at least two years prior to the interview. In these analyses we compared the odds of meeting the twelve-month criteria for a disorder among the Caribbean groups compared to U.S. born African American men. We then evaluated nativity differences among Caribbean Black men (the African Americans were excluded). These analyses controlled for sociodemographic characteristics identified as being associated with chronicity [25][26][27][28][29]: age (18-30, 31-50, 51+), education (high school or less vs. other), marital status (married vs. other), and poverty status (household income below 100% federal poverty level vs. other). Additionally, the analyses controlled for the age of disorder onset [6]. Lastly, we used logistic regression models to examine group differences in mental health services use among those who met lifetime and twelve-month criteria for each anxiety, mood, and substance use disorders. These models controlled for predisposing, enabling, and need factors which are commonly associated with access and use of mental health services [30][31][32]: age (18-30, 31-50, 51+), education (high school or less vs. other), marital status (married vs. other), poverty status (household income below 100% federal poverty level vs. other), and insurance status at the time of the interview. All analyses were conducted using SAS, a statistical program that uses the Taylor expansion approximation technique to estimate variances given the complex sampling design of the NSAL [33]. The survey weights were employed to account for the unequal probability of selection and nonresponse, and for post-stratification. Because the Caribbean Black sample is smaller and more clustered than the African American sample, the standard errors for Caribbean Black men are often larger than those for African American men. --- Results There were notable differences in the distribution of sociodemographic characteristics of African American and U.S. and foreign born Caribbean Black men (see Table 1). Caribbean Black men born in the U.S. tended to be younger on average than African American men. Further, both U.S. and foreign born Caribbean Black men, as compared to African American men, tended to live in households with higher levels of income. Rates of employment were highest among foreign born Caribbean Black men and lowest among the African American men. Foreign born Caribbean Black men were also most likely to be married or cohabiting at the time of the NSAL interview. In contrast, U.S. born Caribbean Black men reported the highest rates of being single or never married and African American men reported the highest rates of being previously married (separated, divorced, or widowed). Compared to African Americans, both groups of Caribbean Black men tended to report higher levels of education (some college or greater). Finally, consistent with residential patterns in the United States [34], a greater proportion of African American men resided in the South when compared to both groups of Caribbean Black men. The majority of participants were covered by health insurance at the time of the interview. --- Lifetime DSM IV Psychiatric Disorders Approximately 30% of Black men overall met criteria for at least one of the measured lifetime psychiatric disorders (see Table 2). This differed among the three groups of men. Specifically, U.S. born Caribbean Black men had significantly higher rates of meeting criteria for any of the measured lifetime disorders compared to both the African American and the foreign born Caribbean Black men. In addition, U.S. born Caribbean Black men were also most likely to meet criteria for two or more lifetime disorders. For the most part, this pattern of findings was repeated when anxiety, mood, and substance use disorders were considered separately. Overall, 14% of men met criteria for a lifetime anxiety disorder, 10% for a lifetime mood disorder, and 18% for a lifetime substance use disorder. Rates were significantly higher for any anxiety disorder and any substance use disorder among U.S. born Caribbean Black men as compared to either the U.S. born African Americans or foreign born Caribbean Black men. U.S. born Caribbean Black men had significantly higher rates of Major Depressive Disorder and significantly lower rates of Bipolar Disorder compared to African American men. The rates of Post-Traumatic Stress Disorder and Panic Disorder appeared particularly high among the U.S. born Caribbean Black men, though these ethnic differences were not statically significant after adjusting for sociodemographic characteristics. Rates of specific psychiatric disorders for the foreign born Caribbean Black men were generally comparable to those of African Americans with the exception of substance use disorders. Foreign born Caribbean Black men showed the lowest rate of substance use disorders with just 7% meeting the criteria for any lifetime substance use disorder compared with one in three U.S. born Caribbean Black men and one in five U.S. born African American men. Subscripts (a, b, c) indicate pairwise comparison results using multivariate logistic regressions. Groups with different subscripts are significantly different from each other. For example, African American men have significantly lower rates of Major Depressive Disorder compared to U.S. born Caribbean Black men as they do not share a subscript. However, both groups share a subscript with the foreign born Caribbean Black men indicating that the rates are not significantly different at the p <unk> 0.05 level. No subscript indicates that there are no significant differences in the rate of disorder across the three groups. -unable to estimate prevalence estimate because there were zero cases that met the disorder criteria. <unk> Multivariate pairwise comparisons of ethnic differences in the specific twelve-month prevalence rates were not conducted due to the small samples of Caribbean Black men. SE-standard error. --- Twelve-Month DSM IV Psychiatric Disorders Approximately one out of eight Black men (12%) met twelve-month criteria for a psychiatric disorder. U.S. born Caribbean Black men were the most likely to meet criteria for the presence of at least one recent disorder, and this group was also significantly more likely than foreign born Caribbean Black men to meet criteria for two or more recent disorders. For the sample as a whole, prevalence of recent psychiatric disorders was 7% for any anxiety disorder, 5% for any mood disorder, and 4% for any substance use disorder. U.S. born Caribbean Black men were significantly more likely than both of the other two groups to meet criteria for any twelve-month anxiety or mood disorder. Significant differences were specifically observed in Post-Traumatic Stress Disorder (PTSD), Panic Disorder, Panic Attack, and Major Depressive Disorder. Within the Caribbean samples, U.S. born men had higher rates of twelve-month Dysthymia and any substance use disorder compared to foreign born Caribbean Black men. We were unable to obtain adjusted group differences in the rates of alcohol abuse, alcohol dependence, drug abuse, and drug dependence disorders between the U.S. and foreign born Caribbean Black men due to the sparse rates of substance use in these groups. None of the foreign born Caribbean Black men in the current study met criteria for alcohol abuse or drug dependence in the twelve months prior to the interview, and the rates of alcohol dependence and drug abuse in this group were less than one percent. --- Chronicity of DSM IV Psychiatric Disorders Four of ten Black men with a lifetime psychiatric disorder continued to meet DSM-IV criteria in the twelve months prior to the interview. Among the entire sample, 49% of those with a lifetime anxiety disorder, 51% of those with a lifetime mood disorder, and 22% of those with a lifetime substance use disorder continued to meet criteria in the twelve months prior to the interview. Caribbean Black men born in the U.S. more frequently evidenced chronicity among the measured disorders compared to African American men. However, the pattern of ethnic and nativity differences in disorder chronicity varied for specific disorders (see Table 3). U.S. born Caribbean Black men evidenced chronic Panic Disorder more frequently than foreign born Caribbean Black men. In contrast, the foreign born Caribbean Black men more frequently evidenced chronic Social Phobia compared to both U.S. born groups. African American men tended to have lower rates of chronic anxiety disorders and higher rates of chronic substance use disorders compared to foreign born Caribbean Black men. The bivariate analyses indicated ethnic differences in the chronicity of Agoraphobia or Bipolar Disorder (p <unk> 0.01). However, small samples precluded the use of multivariate models to identify adjusted group differences in chronicity for these disorders. Finally, we were unable to estimate ethnic variations in the chronicity of specific substance use disorders since none of the foreign born Caribbean Black men with lifetime alcohol abuse or lifetime drug dependence disorder continued to meet criteria for these conditions in the twelve months prior to the interview. Analyses restricted to those that first met the DSM-IV criteria at least two years prior to the interview. <unk> Multivariate pairwise comparisons of ethnic differences in the chronicity specific disorders were not conducted due to the small samples. -indicates that weighted estimates could not be obtained because none of the lifetime cases continued to meet criteria in the twelve months prior to the interview. Subscripts indicate pairwise comparison results using multivariate logistic regressions. Groups with different subscripts are significantly different from each other. For example, African American men had lower rates of persistent Generalized Anxiety Disorder compared to both groups of Caribbean Black men ( a vs. b ). However, the rates of persistent Generalized Anxiety Disorder were comparable between the U.S. and foreign born Caribbean groups ( b = b ). No subscript indicates that there are no significant differences in the rate of persistence across the three groups. SE-standard error. --- Lifetime Mental Health Services Utilization Approximately half (52%) of all Black men who met the criteria for any psychiatric disorder during their lifetime also reported ever speaking with a medical or mental health provider about mental health concerns (see Table 4). Rates of treatment contact for the full sample were 59%, 59%, and 52% among those with any mood disorder, any anxiety disorder, or any substance use disorder, respectively. For example, African American men with a twelve-month mood disorder have higher rates of service utilization compared to foreign born Caribbean Black men ( a vs. b ). However, both U.S. born Caribbean Black men with a twelve-month mood disorder share a subscript ( a or b ) with African American and foreign born Caribbean Black men, indicating they had equal rates of mental health service use. No subscript indicates that there are no significant differences in the rate of service utilization across the three groups. N-number of participants that met the criteria for a disorder during the lifetime. <unk> Pairwise comparisons of service use between the U.S. and foreign born Caribbean Black men were not estimated due to the small samples. na-Rates of service use could not be estimated because none of the U.S. born Caribbean Black men met the DSM-IV criteria for Agoraphobia during the lifetime. SE-standard error. The rates of service use among Black men with any lifetime psychiatric disorder were 65% for African American men, 76% for Caribbean Black men born in the U.S., and 81% among foreign born Caribbean Black men. While these rates appear comparable, there were significant ethnic variations in mental health services use among Black men with specific psychiatric disorders. Of those who met the DSM-IV criteria for Post-Traumatic Stress Disorder or Generalized Anxiety Disorder during their lifetime, U.S. born Caribbean Black men were more likely than the U.S. born African American men to report speaking with a provider about mental health concerns. For those men with a positive lifetime history of a Panic Disorder, U.S. born African American men had significantly higher rates of service use compared to foreign born Caribbean Black men. Lastly, among men who met criteria for a substance use disorder during their lifetime, foreign born Caribbean Black men were more likely than U.S. born Caribbean Black men to report speaking to a provider regarding mental health or substance use issues. --- Twelve-Month Mental Health Service Use Finally, we investigated rates of mental health services use of Black men who met criteria for a mood, anxiety, or substance use disorder in the twelve months prior to the interview. Just one quarter (27%) of Black men with a recent disorder had sought mental health care from a medical or mental health provider in the past year. There were significant differences in twelve-month services use. Seven percent of foreign born Caribbean Black men with a disorder reported services use in the past year compared to 48% of U.S. born Caribbean Black men and 27% of U.S. born African Americans, p <unk> 0.05. --- Discussion The current study sought to determine ethnicity and nativity differences in mental health experiences of African American and Caribbean Black men in the U.S. To achieve this goal, we provided nationally representative estimates of the prevalence and chronicity of DSM-IV mood, anxiety, and substance use disorders, and mental health services use among African American and U.S. and foreign born Caribbean Black men. The mental health needs of Black men have not always been visible as this population is less likely than others to seek treatment in traditional primary care and specialty mental health settings [8,13,35]. Our study found that, although the prevalence of most DSM-IV disorders is low in community-dwelling samples of Black men, disorders in this population are often chronic and untreated. Moreover, there were significant ethnic and nativity variations in the prevalence, chronicity, and treatment of DSM-IV mood, anxiety, and substance use disorders which may provide clues towards the etiology and persistence of disorders in Black men. --- Extra Risk of U.S. Born Caribbean Black Men U.S. Born Caribbean Black men experienced elevated prevalence of PTSD, Panic Disorder, MDD, and Alcohol and Drug Use Disorders compared to African American Men. While prior studies have reported that approximately one in 12 Caribbean Black men are likely to meet the criteria for PTSD [36], our study found that the lifetime prevalence of this disorder is almost twice as high with 1 in 7 Caribbean Black men born in the U.S. meeting criteria for PTSD. Our findings correspond with others which have found elevated rates of DSM-IV disorders [9,37], suicide attempts [38] and psychiatric hospitalization [39] among Caribbean Black men born in the U.S. For example, Doyle and colleagues found that Caribbean fathers in the U.S. had elevated rates of psychiatric disorders compared to African American fathers [9]. This study expands upon prior investigations of Black men's mental health by providing nationally representative estimates of chronicity, and mental health service use as well as the prevalence of Bipolar Disorder I and Dysthymia for Black men. Although rare, the lifetime and twelve-month prevalence of Bipolar Disorder I was higher among African American men compared with Caribbean Black men. Because there is a history of misdiagnosis of affective disorders and over-diagnosis of schizophrenia in African Americans [40,41] particularly men, it is important to provide clinicians with the training and resources needed to accurately identify affective disorders including Bipolar I disorders in African American populations. --- Most Disorders Are Chronic This study extends prior investigations of the persistence of psychiatric disorders among Non-Latino Blacks [6,7] and provides new findings of ethnic and nativity variations in the chronicity of specific psychiatric disorders among African American and Caribbean Black men. Results of our study finds Caribbean Black men born in the U.S. more frequently evidenced chronic Generalized Anxiety Disorder (GAD), Panic Disorder, and Substance Use Disorders than the other groups of Black men. Interestingly, U.S. born Caribbean Black men with a lifetime history of GAD and Alcohol Abuse were more likely than African American men with the same disorders to report lifetime mental health services use. Results from other literature into Black men's mental health may help to explain the patterns of ethnic and nativity differences in chronicity observed here. In an early study that was among the first to notice persistence (i.e., chronicity) of mental disorders in Black men, it was thought that this chronicity is potentiated by factors that are concurrent with or subsequent to the onset of disorders [1]. Some have suggested that the experience of race-based discrimination could be a contributing factor in mental health outcomes [42][43][44]. In a study we conducted of the relationship between perceived discrimination and mental health services in the primary care setting we found that, for Black Americans, when they perceived discrimination within the context of these services they were more likely to drop out of treatment prematurely [45]. While there has been much research documenting that race-based discrimination is associated with mental health consequences, we may need to look beyond just the association with the disorders themselves and starting look at how these experiences of discrimination may serve to complicate the recovery process. In the Mays et al. study, Blacks were more likely to drop out of treatment than Whites, which clearly would serve to support persistence of their mental health problems. In a study by Mereish et al. [46], the association of discrimination with depressive symptoms was mediated by the self-esteem in African American men, but not in Caribbean men. The authors postulate that differences between African American and Caribbean men in their responses to discriminatory experiences may be a function of racial socialization as well as the appraisal of what constitutes race-based discrimination. While the current study did not assess race-based discrimination or racial socialization, these may be important areas of inquiry in future studies focused on elucidating ethnic and nativity differences in the chronicity of disorders among Black men. Our findings raise important questions about the need to explore factors that may drive U.S. born Caribbean Black men to seek mental health services more readily than other Black men. Given the potentially elevated psychiatric morbidity experienced by U.S. born Caribbean Black men, future studies are needed to elucidate what contributes to the persistence of psychiatric disorders in this population as it delays the recovery and resiliency process. --- Mental Health Services Use Was Subpar for All Black Men Only half of Black men who met the criteria for a psychiatric disorder in this study had ever spoken with a healthcare provider about mental health or substance use concerns. The rates were not much higher (62%) among those with two or more comorbid disorders. Prior studies have indicated that the low rates of mental health services use among Black men may be attributed to economic barriers, discrimination in the healthcare setting, provider mistrust, and low perceptions of need [35,45,[47][48][49]. It is possible that the low rates of mental health services use observed in the current study are due, in part, to discontinuous insurance coverage or lack of a usual source of care [50][51][52]. As noted above, when Black men use general medical services, they are more likely than Whites to perceive experiences of discrimination in the healthcare encounter. Thus, mental health service use may also be as a function of how Black men perceive systems of care to work for them, as well as the extent to which their racial socialization plays a role in what they perceive to be the likelihood and consequences of discrimination in receiving mental health services [45]. While masculinity norms could prevent some men from utilizing mental health services [13], literature into Black men's help-seeking found that Black men will reach out for help with emotional problems; they are likely to use informal systems of care [48,49]. Thus, some of the Black men with a psychiatric disorder in the current study are likely to have sought mental health assistance from clergy, family or friends, or other informal sources of support [14,[53][54][55][56][57][58][59][60]. Given the chronic and persisting nature of psychiatric disorders, which has been reported for Blacks more than others [6,7], our findings indicate that Black men's mental health needs are not being met adequately through these alternate pathways to care. Gender and culturally specific interventions may be required to improve access, use and effectiveness of mental health treatment for African American and Caribbean Black men. --- Limitations The study findings should be interpreted with four limitations in mind. First, the sample sizes were small when rates of treatment were broken down by ethnicity, nativity status, and specific disorders. This restricted our power to only detect differences which may have been present with larger samples. Second, our estimates of the lifetime and twelve-month prevalence of alcohol and drug dependence disorders may be underestimated as the WHO-CIDI used diagnostic criteria that did not strictly adhere to those provided by the DSM-IV [61,62]. Third, the estimates here only generalize to non-institutionalized Black men. Individuals that are homeless, incarcerated, or living on a military base were excluded from the study. Fourth, we were unable to examine group differences in past year services use among those with specific twelve-month disorders due to the small number of men meeting criteria for a recent disorder and the low proportion of those men with services use. As an example, none of the 18 foreign born Caribbean Black men with a recent panic attack had spoken with a provider about mental health or substance use concerns in the twelve months prior to the interview. Similarly, none of the ten U.S. born Caribbean Black men with a recent drug use disorder had spoken with a provider in the twelve months prior to the interview. --- Specific Considerations and Future Research As much of our information about the mental health status, needs and service utilization of African American and Caribbean men comes from non-institutionalized populations it is important to remember that we have little to no data for the mental health needs of the estimated over 1 million Black men in jails and incarcerated settings [63,64]. African American men are more than six times as likely as White men to be incarcerated [65], and rates of psychiatric disorders are higher among incarcerated Black men compared to those in the community [66,67]. Moreover, the leading cause of death among Black men in jails is suicide [68]. These men are often released into communities who are unprepared to care for their mental health problems, under-resourced in general for mental health providers, and lacking information on the ways in which a history of incarcerations adds to their mental health recovery [63]. It is important to attend to these sampling considerations when interpreting the low prevalence rates of disorders found in non-institutionalized samples of African American men and when planning for population level service needs. A balanced research agenda that includes institutionalized Black men will address some of the myths of low prevalence of mental health disorders and underscore the high levels of unmet service needs for Black and Caribbean males. Additional research is needed to isolate the underlying factors, such as acculturative stress or exposure to discrimination, that contribute to elevated rates of PTSD, panic disorder, and MDD among Caribbean Black men born in the U.S. There is also a need to study whether integrated care and its' treatment approach result in increased or decreased likelihood of treatment and whether that treatment approach given the system of care in place contributed to reducing chronicity of mental disorders in Black men in the United States. --- Conclusions Our study provides new findings on the prevalence of Bipolar I and Dysthymia, the chronicity of psychiatric disorders, and patterns of mental services use for DSM-IV mood, anxiety, and substance use disorders among African American and U.S. and foreign born Caribbean Black men. The integrated care approach presents a unique opportunity to expand access to mental health services particularly for Black men in the United States [15]. In light of these opportunities, improved understandings of ethnic and nativity variations in mental health service needs are necessary [69,70]. Mental health screening efforts in the primary care setting must be gender and culturally specific to improve the early identification and treatment of disorders for African American and Caribbean Black men and to mitigate the chronic course of their disorders [71]. --- Author Contributions: V.M.M. and R.J.T. conceived the study. A.L.J., S.D.C., and V.M.M. designed the study and refined the conceptualization. R.J.T. obtained the data. A.L.J. and S.D.C. oversaw statistical analyses with the help of J.R. who also assisted with the creation of variables and conducted the statistical analyses. V.M.M. wrote the first draft of the manuscript. A.L.J., S.D.C., R.J.T. and J.S.J. participated in the interpretation of results and their discussion. All authors participated in manuscript revisions and approved the final version. --- Conflicts of Interest: The authors declare no conflicts of interest.

This study investigated ethnic and nativity differences in the chronicity and treatment of psychiatric disorders of African American and Caribbean Black men in the U.S. Data were analyzed from the National Survey of American Life, a population-based study which included 1859 self-identified Black men (1222 African American, 176 Caribbean Black men born within the U.S., and 461 Caribbean Black men born outside the U.S.). Lifetime and twelve-month prevalence of DSM-IV mood, anxiety, and substance use disorders (including Bipolar I and Dysthmia), disorder chronicity, and rate of mental health services use among those meeting criteria for a lifetime psychiatric disorder were examined. Logistic regression models were employed to determine ethnic differences in chronicity, and treatment utilization for disorders. While rates of DSM-IV disorders were generally low in this community sample of Black men, their disorders were chronic and remained untreated. Caribbean Black men born in the U.S. had higher prevalence of Post-Traumatic Stress Disorder, Major Depressive Disorder, and Alcohol Abuse Disorder compared with African American men. Foreign born Caribbean Black men experienced greater chronicity in Social Phobia and Generalized Anxiety Disorder compared to other Black Men. Utilization of mental health service was low for all groups of Black Men, but lowest for the foreign born Caribbean Black men. Results underscore the large unmet needs of both African American and Caribbean Black men in the United States. Results also highlight the role of ethnicity and nativity in mental disorder chronicity and mental health service utilization patterns of Black men.

Introduction What governs the masses? What are the reasons that make social actors gather and fight for their rights? What movements would these be that interfere in the democratic game by altering the functioning of party institutions? Do street protests carry the influence of ideologies? What role has the left and right played in social movements since the 1970s? Faced with these issues and the importance of social movements, a more detailed analysis of their origins, characteristics and influences in the Brazilian democratic process from the beginning of the 1970s to the present day is necessary. This research seeks to demonstrate the characteristics of these movements and their influence for the consolidation of democracy and its relationship with party organizations, institutions necessary for the exercise of the democratic game. It also seeks to draw a parallel between left and right movements, passing through the struggle for democracy in the 1970s and 1980s until its polarization today. More than carrying out an analysis of social movements in Brazil in the last fifty years, this research aims to explain the current political situation, justifying, therefore, the Copyright<unk> Ricardo Braga de CR. Ricardo Braga de CR. Social Movements in Brazil: From the 1970's to the Present Day the Influence of Social Movements and their Relationship with Democracy and Party Organizations in the Last Fifty Years. Anthropol Ethnol Open Acc J 2023, 6(1): 000204. strengthening or stagnation of Brazilian democracy. In this article, we analyze the consequences of social movements for the democratization process that took place in Brazil and their relationship with party institutions in the period between 1970 and 2020. In this sense, we highlight, in this research, the conceptualization and historical reference of social movements in the Brazil in the last five decades, as well as its typology and chronological location. Regarding the historical context, we identify its origins and characteristics, also highlighting its evolution in the history of social movements in Brazil at different times. The main point being analyzed in this article is the influence of ideologies within social movements and their relationship with party organizations in Brazil. Our hypothesis is that there are elements that prove the influence of ideologies, both left and right, in social movements. And this is the main focus analyzed in this article, which began in the 1970s with the institutional support of sectors of the Church, and the consequent consolidation of the CEBs (Base Ecclesiastical Communities) as a possible practice of opposition to the dictatorial regime. This decade was marked by movements and demonstrations for political rights lost during the military dictatorship and the search for social rights. In the 1980s, we will analyze the movements for redemocratization and popular participation, mentioning the demonstrations by Diretas Já. It is a fact that social movements raise issues of relevance to society as a whole and influence governments in decision-making, as well as generating demands for public policies that put pressure on actors in political society and government institutions, in addition to strengthening their individual identity and collective. --- Social Movements: Concepts and Historical Background Gohn points out that there is no "proper theoretical paradigm" about Latin American social movements. The Marxist theoretical model was the initial classic model that aimed to formulate an explanatory reference and a project of action for social movements [1]. For Gohn an important category of analysis of social movements forged by Marx is that of solidarity. In the words of Gohn, by Müller [1], solidarity cited by Marx refers to a social relationship, with the same interests and should be focused on a given objective: the emancipation of workers. And it is this relationship of solidarity that we will find in the demonstrations that took place in Brazil between the years 2013 and 2020; this period clearly showed the influence of political ideologies in these social movements. It is noted, in0020 The CEBs are ecclesiastical communities that emerged in Brazil and Latin America in the 1970s and 1980s. They are inclusive communities mainly linked to the Catholic Church that spread through the encouragement of Liberation Theology. Diretas Já was a political movement that aimed to restore direct presidential elections in Brazil. The movement began in May 1983 and lasted until 1984, having mobilized millions of people in rallies and marches. The movement brought together diverse elements of Brazilian society. The rallies for Diretas Já were the largest public demonstrations in Brazil since the end of the military dictatorship. these movements, not only the thought of Marx explained by Gohn with regard to solidarity in relation to common interests, but also the ideas of Touraine [1,2] who sees in the social subject the transforming agent of the society in which he lives. It is emphasized that it is in this period, due to the ideological influence; and it could not be different; that polarization in Brazil reaches its peak. --- Social Movements from the 1970's to the 1990's According to Gohn [3] the understanding of participation processes, both in civil society and in public policies, leads us to the process of democratization of society; the rescue of participation processes takes us to the social struggles that have been waged by society to gain access to social rights and citizenship. In this sense, participation is also a struggle for better living conditions and the benefits of civilization. The participatory scenario in Brazil from the 1960s to 2010 is subdivided into four cycles of political protests, Brazilian social movements and struggles, and forms of civil society participation [3]. The first cycle will be analyzed briefly with just a few notes, because the 1960s are not part of the analysis period covered by this work, but will serve as the basis for the protests that occurred in the 1970s and subsequent decades. It is known that the 1960s had three distinct moments in Brazil. First, the intense participation of students, workers and other social groups in strikes and demonstrations that preceded the 1964 military coup; second, after the coup, specific sectors of civil society participated in resistance struggles such as the strikes in Osasco and Contagem, the student movement in 1968, cultural resistance in the arts, the focal action of numerous leftist groups, etc. [3]. During the military dictatorship in Brazil of populations "marginalized" from economic development. The current term was civic participation, also the subject of a mandatory subject in the school curriculum, Moral and Civic Education [3]. From the 1970s onwards, a new cycle of social struggles and resistance to the military regime began. This resistance intensifies from the second half of the 1970s onwards, looking for loopholes to restore democracy. Movements for Amnesty for exiles and political prisoners, movements coming from the CEBs (Base Ecclesiastical Communities), movements of the new unionism of ABCD S<unk>o Paulo and other regions, popular movements of women in peripheral neighborhoods for day care centers, or of residents for collective transport, or for health centers, all of this created the context of a new cycle of struggles in Brazilian society [3]. In this cycle, another category of participation emerges, completely opposite to that which was stimulated by public policies in the 1960s, in the field of civil associations, which was popular participation, in neighborhoods and regions on the outskirts of large cities, in search of minimum conditions of survival in the urban environment, fighting for basic social rights, in a totally unfavorable political situation [3]. Gohn [3] points out that, in Brazil, given the specificity of the military regime and the glaring survival conditions of the popular strata, the popular movements in the 1970s were analyzed more from a Marxist perspective, relying on Manuel Castells, these theories are described in this article. The two cycles, from the 1970s and the beginning of the 1980s, have in common the vision of the struggle for basic rights, with an axis focused on the issue of equality or the recovery of seized rights, through the strengthening of civil society, through the construction /reconstruction of citizenship, recovering democracy. That's why the fight for Diretas Já was so important and symbolic. Social movements, with their numerous fronts and demands, were also civil struggle movements [3]. The third cycle of movements in Brazil takes place from 1980 onwards in the period of transition to democracy, with changes in the political situation and the return of elections for state governments. In the first years of the 1980s in Brazil, the great highlight in the field of civil participation was the growth of urban popular movements, through grassroots organizations, which also included the CEBs (Base Ecclesiastical Communities), of the progressive wing of the Catholic Church, and the advancement in organization in the world of work, with the creation of union centrals [3]. The ABCD is an industrial region in Greater S<unk>o Paulo, Brazil. It is formed by four cities: Santo André (A), S<unk>o Bernardo do Campo (B), S<unk>o Caetano do Sul (C) and Diadema (D). This region is considered one of the most important economic regions in the state of S<unk>o Paulo and one of the most important industrial centers in the country. The region gained national prestige since the 1950s, a period of considerable industrial expansion in Brazil and the installation of automobile manufacturers in the region. This conjuncture resulted in the emergence of numerous gender movements, of Afro-descendants, trade union centers and social movements, movements in the countryside with the creation of the MST (Movement of Landless Workers), the struggle for Diretas Já, in 1984, and the constituent phase of 1985-1988. In the Brazilian scenario, specific struggles were articulated -from neighborhoods, localized, and national struggles that called for the end of the military regime and the return of presidential elections. The Diretas Já movement is the culmination of this phase [3]. The fourth cycle of participation occurs after the approval of the new Constitution, in 1988, when the focus of many investigations shifts from the movement of civil society actors to the performance of multiple actors in public policies, and to the construction of new channels of participation institutionalized, provided for in the new Constitution [3]. Throughout the 1990s, the field of civil society expanded in practice and in the discourses about it. The decentralization of the subject and the emergence of a plurality of actors gave another concept, that of citizenship, the same relevance that the concept of autonomy had in the 1980s [3]. The issue of citizenship was already raised in the 1980s, both in the struggles for redemocratization, which led to the Diretas Já movement, the Constituent Assembly and the new Constitutional Charter of 1988, highlighting the issue of civil and political rights, and in popular struggles for improvements in the quality of urban life, where citizenship has gained a new outline as collective citizenship, and extrapolates the demand for civil rights to include other rights such as basic, elementary, first-generation social rights, already equated since the French Revolution, contained in the demands for home, shelter and food; as modern social rights, related to working conditions, education, health, etc. [3]. In the new scenario, civil society expands to intertwine with political society, contributing to the new contradictory and fragmented character that the State takes on in the 1990s [3]. The transition from the 1980s to the 1990s was the moment when different political forces sought to develop proposals for a new standard of relationship between State and society, each one of them stating how democratic construction should be in Brazil [3]. Copyright<unk> Ricardo Braga de CR. Ricardo Braga de CR. Social Movements in Brazil: From the 1970's to the Present Day the Influence of Social Movements and their Relationship with Democracy and Party Organizations in the Last Fifty Years. Anthropol Ethnol Open Acc J 2023, 6(1): 000204. Gohn [4,5] points out that the new social actors that emerged in Brazilian civil society after the 1970s, in spite of the State, and against it at first, configured new spaces and formats of participation and relationships social. For Gohn [4,5], these new spaces were basically built by social movements, popular or not, in the 1970s and 1980s and in the 1990s by a special type of NGOs that we previously called citizens, that is, non-profit entities that are oriented towards the promotion and development of needy communities based on relationships based on rights and duties of citizenship [4,5] points out that citizen movements and NGOs have proved to be structures capable of playing roles that formal, substantive structures have not been able to play as state, official structures, created with the objective and purpose of serving social area. According to Gohn, [4,5] the scenario of social struggles in Brazil, in the 1990s, was redefined. The urban popular social movements of the 1970s and 1980s changed substantially. Some entered an internal crisis: militancy, mobilization, and daily participation in organized activities, credibility in public policies and reliability and legitimacy with the population itself. --- Social Movements: From Impeachment to --- Struggles for Education From June 2013, demonstrations began to take place throughout Brazil, in protest movements against social policies in force in the areas of transport, education, health, priority given to spending on the World Cup, among other reasons. The crowd, absent from major demonstrations since the early 1990s, returns to the streets [3]. It is estimated that more than one million people took to the streets in Brazil during the month of June 2013 [6]. June 2013 is a milestone in Brazilian political and sociocultural life. This is when a new cycle of participation in Brazilian society begins, made up of different collectives and social movements, with different projects and proposals classified into three distinct types, namely: classics, new and brand new. The classics include unions, the landless, students, popular community movements in neighborhoods, the homeless, etc. The new ones include movements fighting for rights, identities, etc. created from the end of the 1970s; and the newest include current movements, most created or affirmed in the public scene in the 2010s, such as the Passe Livre Movement, on the one hand; and on the other, Vem Pra Rua and the Brasil Livre Movements [3]. The 2013 transformations generated new statements and new groups emerged. Conservative, reactionary and neoliberal groups gained space in social media calls for street protests [3]. Therefore, there are profound differences between the young people who initiated the first calls for demonstrations in June 2013, and other groups of young people who, from 2014 onwards, created other types of mobilizations and other repertoires, focused on the political level against the federal government and the process that led to the impeachment of Dilma Rousseff [3]. In 2015, the focus of the main demand on the streets changes -no longer urban mobility, but the federal government, with the impeachment of President Dilma Rousseff as a goal. The largest public demonstrations in Brazilian history take place. Polarization is established in politics [3]. In 2016, demonstrations in the streets, for and against the president's removal, dominated the repertoire of demands [6]. In 2016, the classic social movements and the old "new" social movements came together, forming "fronts" to retake the streets as a space of protest, after losing the hegemony of these territories to the countercurrents created from 2014 onwards [3]. In 2017, there was a reactivation of social struggles, which had been reduced after the impeachment, in August 2016. Trade union demonstrations with the support of social movements returned to the context of struggles. In this scenario, the struggles for education that were part of the context of social movements in the 1970s, 1980s and 1990s, resumed, with emphasis, their participation in mobilizations and demonstrations from the year 2017. --- Social Movements and Identity Construction The Passe Livre (Free Pass) is a social movement that fights for free public transport in Brazil. Vem Pra Rua is a right-wing political movement that emerged in 2014 during the protests against the Dilma Roussef government. Brasil Livre is a libertarian movement that advocates for free markets and limited government. According to Gohn [7], the current context of the main social movements in Latin America is one of the backdrops for discussing their forms, demands, identity they build, networks that structure them, cultural and political manifestations social to which they are articulated. Gohn [7] observes, in this scenario, the educational character of the movements, their role in the public scene regarding the theme of social inclusion, political culture and its manifestations in the area of Education. based on a project of life and society. Today, social movements are different from the movements that took place in the late 1970s and early 1980s, which opposed the military regime, these movements fighting for social rights supported by the Catholic Church, based on the Liberation Theology and movements that sought the redemocratization of the country. It should be noted that many of the current movements have characteristics of the social struggles of the 1980s. However, what differentiates the current movements from those of the 1980s is that, in this decade, the movements fought to have the right to have rights. The social struggles of that decade aimed at universal rights, they did not look only at themselves, and they looked at the whole, even to be able to build their identity. For Gohn [7] the social mobilization category, in previous decades, has always been subordinated to other larger categories in the analysis of the social, such as process, change, etc. In this new century, according to Gohn [7], the concept of social mobilization gains life, meaning and its own dynamics. There was a definition of identity, since the 1980s and which lasted until the end of the 1990s, which was based on the positioning of individuals in a given social group, not only in relation to the antagonism between capital and work, but also in relation to in relation to income and acquired status, which were collective representations accepted by society. Identity is a tool under construction. Given this, Gohn [7] highlights that it is not about the identity built throughout a movement, but a modeled, granted identity, in which certain socio-political and cultural subjects are mobilized to be included. Gohn [7] states that the recognition of political identity is done in the struggle process, before civil and political society; this is not a bestowed, donated recognition, a top-down inclusion. There is an appropriation of the other's image, which serves as a mirror. And once reflected, little by little the identity is being built. --- Social Movements in the Marxist View For Gohn [4,5], the analysis of social movements from the perspective of Marxism refers to processes of social struggles aimed at transforming existing conditions in social reality, economic shortages and/or socio-political oppression and cultural. Müller [1] points out that, when addressing social movements, Marxist theories have not completely abandoned the concept of social classes, which ends up placing the actors involved in the context of movements in the face of the social structure and inserted in it. For Müller [1] the social class, and the elements that compose it, end up defining parameters about the social conscience of the actor involved in the struggle for conquests of the class itself. It should be noted that the history of all society until now has been the history of class struggles [8]. According to Gohn, an important category coined by the author is that of social praxis, which is, for Marx, the capacity that the working social classes and dominated groups have to transform society through theoretical, political and cultural activities productive [1]. Gohn, quoted by Müller [1], states that another important category of analysis of social movements forged by Marx is that of solidarity. In Marxist thought, solidarity has a double meaning: on the one hand, it is a factor of group cohesion, indicating a consolidated level of commitment among those who share the same ideology, and, on the other hand, it brings together people who seek and share an emancipator proposal. Marx, quoted by Gohn [5,7] refers to the issue of social movements as follows: "It should not be said that the social movement excludes the political movement. There will never be a political movement that is not at the same time social". In a letter addressed to F.Bolt, cited by Gohn [5,7], he also refers to the issue of social movements in the following terms: • Thus, in addition to the different economic movements of the workers, political movements, that is, class movements, arise everywhere with the aim of imposing their interests in a general way, in a way that has a general social-coercive force. If these movements presuppose a certain degree of prior organization, on the other hand they also mean means of developing this organization. Marxist ideas make it clear that social movements are confused with political movements, developing both their own class ideology and an individual and collective identity within these movements. --- Network Social Movements and the Search for Legitimation In the words of Fernando Henrique Cardoso [9], it may be insufficient and imprecise to speak of citizenship or even a civic culture, nowadays, with the internet and signs of the formation of a "planetary civil society". Castells [10] points out that, the real world of our time is a hybrid world, not a virtual world or a segregated world that will be separated from online from offline interaction. And it is in this world that social movements are born in a natural transition for many people, who go from sharing Copyright<unk> Ricardo Braga de CR. Ricardo Braga de CR. Social Movements in Brazil: From the 1970's to the Present Day the Influence of Social Movements and their Relationship with Democracy and Party Organizations in the Last Fifty Years. Anthropol Ethnol Open Acc J 2023, 6(1): 000204. their sociability to sharing their indignation, their hope and their struggle. For Castells [10], networked social movements, like all social movements in history, bear the imprint of their society. In large part, they are formed by individuals who easily coexist with digital technologies in the hybrid world of real virtuality. Its values, objectives and organizational style make direct reference to the culture of autonomy that characterizes the young generations of a young century. They couldn't exist without the internet. But its importance goes much deeper. They are adapted to their role as agents of change in the network society, in sharp contrast to obsolete political institutions inherited from a historically outdated social structure. --- The Social Subject as a Transforming Agent of Society In the 1970s, Touraine states that social movements are always, ultimately, the expression of a class conflict [2,4,5]. Touraine also states that every Fernando Henrique Cardoso is a Brazilian sociologist, professor and politician who served as the 34th president of Brazil from 1 January 1995 to 31 December 2002. He was the first Brazilian president to be reelected for a subsequent term. Social movement is geared towards critical action, which rests on contradiction and not on conflict [4,5]. According to Alain Touraine [2], a social movement is not just a set of goals; it also assumes the participation of individuals in a collective action. Touraine [2] points out that, The formation of movements, at the same time weak, because they are dispersed, and very strong, because they are determined to manage themselves, to define their ends and means by themselves, without being subordinated to parties or theorists, must create new forms of collective action. In the words of Touraine [1,2], society is a set of rules, customs and privileges against which individual and collective creative efforts must continue to fight. For Touraine, the social subject is the transforming agent of history and the society in which he is inserted, shifting the axis of attention from State actions aimed at certain objectives. In the words of Touraine [2,11], The new contestations do not aim to create a new type of society and, even less, to free the forces of progress and the future, but rather to "change life", to defend human rights, both the right to life of those to whom hunger or extermination threatens the right to free expression or the free choice of a personal style or life story. According to Touraine's theory, postmodern society represents the opportunity for the emergence of the subject, for the transformation of the subject into an agent of society, that is, one who acts and transforms the society in which he lives. It is now about transforming the subject into an agent [2,11]. --- The Construction of Social Movements from Ideologies It is a fact that every social movement has an ideology. However, certain ideologies build, over time, a utopia, an ideal, a goal, a purpose that mobilizes social actors for the struggle; they are endowed with a set of beliefs that deny the instituted and restore a new paradigm for action and thought [4,5]. In general, we know that ideologies helped shape the history of the modern world and, even today, are part of the imagination of some societies, which create utopias and ideals that serve as the basis for their movements and mobilize people to fight for their rights, purposes and also against the rights of third parties. Socialism, social democracy, Nazism, fascism and liberalism are some examples of ideologies that influence and mobilize social actors for the fight, based on the ideals acquired over time. Left ideologies are associated with the principles of equality and social justice and right ideologies are inclined to the ideas of freedom and the guarantee of private property. For Gohn [4,5], these ideologies were, and still are, capable of sensitizing thousands of people. They generated specific social movements; fed, as a reference paradigm, several social movements. --- Final Considerations This article demonstrates the evolution of social struggles that have taken place in Brazil since the 1970s, including demonstrations against the military regime and movements fighting for social rights with a strong influence from the progressive wing of the Catholic Church and which took shape in the 1980s, and 1990 with the movements for redemocratization and popular participation, in favor of Diretas Já. In this sense, one can see a clear and progressive influence of ideology on social movements and party institutions, promoting a strong and growing demand for public policies, observed in the twelve consecutive years of leftist governments in Brazil. It is clear and evident that there is a significant growth in polarization in Brazil due to the ideological influence of the right, which began in 2013, with demonstrations in favor of the impeachment of President Dilma Rousseff. This polarization reached its peak in 2018 with the election of And what draws attention, and which is somewhat impressive, is that today we have a strong influence from conservative sectors of a religious nature, while in the 1970s the influences coming from the Catholic Church originated in the progressive wing in favor of the social and collective rights. The distance between the 1970s and the year 2020 is expressively noted, not temporally, but in opposite directions and polarized mainly by religious sectors of Brazilian society. And the distance is so significant that we started the 1970s with demonstrations against the military regime and in favor of collective social rights, with full support from the progressive wing of the Catholic Church, through the CEBs (Base Ecclesiastical Communities) and we reached 2020 with the influence of conservative sectors of a religious nature, aiming at the right not to have rights, which goes completely against what is expressed in the Federal Constitution. Social struggles in Brazil have been marked by the influence of party ideologies from the 1970s to the present day, sometimes passing through progressive ideologies, sometimes more to the right, with a more conservative bias. It is noticed that the ideological influence more to the right, with more conservative characteristics, clearly leads to polarization; and this, in turn; to the hatred and manifestations of intolerance seen recently in the Brazilian streets. It is important to remember that the construction of social movements over time, based on ideologies, carries with it not only the search for legitimacy, but mainly the construction of individual and collective identities. It should be noted that social movements are essential for the construction of collective identities, as well as for the demand and formulation of public policies, as well as for strengthening the democratic rule of law and the principle of freedom of expression, basic assumptions of a consolidated democracy. The divergence of points of view, different interests and conflicts between social movements are indispensable to the democratic process and the construction of public policies. Finally, we reinforce the importance of social movements for the Brazilian democratic process in recent decades, as well as we address the divergences of point of view and the different interests essential to democracy and freedom of expression, while we consider as a negative point the high degree of polarization verified in the street demonstrations between the years 2013 and 2020.

This article discusses the relationship between social movements and democracy in the last fifty years in Brazil. This research aims to analyze the influence of demonstrations and street protests, which took place between 1970 and 2020, on the Brazilian democratic process and party institutions. The analysis starts from the social struggles against the dictatorship and Catholic Church through the CEBs (Base Ecclesiastical Communities) at the end of the 1970s. The struggles for redemocratization and for popular participation in the 1980s are also analyzed. The networked social movements and the new social movements, according to Alain Touraine, are also objects of this research. It is observed that social movements influence not only institutions and the democratic process but also individual and collective identities that share their ideologies, which expressively altered the political game and institutions' functioning.

Introduction Currently, social networks are being adopted rapidly by millions of users most of whom are young people with a great number of purposes in mind (Lenhart & Madden, 2007). There are hundreds of social networks with various technological applications serving to a wide range of interests most of which support the maintenance of pre-existing social relations, however, many others help strangers to connect to others' profiles with shared interests, needs, political views etc. (Boyd & Ellison, 2010). Social networks include various people both as authors as well as readers, nonetheless personalized content, information sharing and collaboration are the socializing dimensions of these tools ( Bartlett-Bragg, 2006). As Lenhart and Madden suggested, 55% of teenagers use social networks in their daily lives most of whose purposes are limited to communicating with friends, making new friends and sharing personal information and materials (Lenhart & Madden, 2007). Social networking websites, such as Facebook and RenRen are member-based Internet communities that allow users to post profile information, such as a username and photograph, and to communicate with others in innovative ways such as sending public or private online messages or sharing photos online. In the spring of 2006, Nielsen//Net Ratings reported that the top 10 social networking sites in the U.S. grew in number of users from 46.8 million to 68.8 million during the previous year (Nielsen, 2006). Nowadays Facebook has 630 million daily visitors. Started 15 years ago and now it's ranked second based on daily view (Wolfram|Alpha, 2011). Social networking websites reveal important information about how adolescents and young adults are interacting with one another in the information age. Social Networking Sites (SNSs) such as Facebook are one of the latest examples of communications technologies. Millions of contemporary young adults use social networking sites. However, little is known about how much, why, and how they use these sites. The purpose of this study was to provide descriptive information about the use of social networking sites by both Chinese and German SNS users. Social networking sites persuade millions of users each day to adopt specific behaviors. To understand this phenomenon in the context of persuasive technology, the present study analyzed how persuasion takes place in leading social networking sites from two different countries: Facebook in the Germany and RenRen in china. The present study also compared the two services on some specific tasks. Our analysis reveals the differences and similarities in how Facebook and RenRen are designed to influence users toward the achievement of these tasks. These persuasion styles seem to map generally cultural differences between the Germany and China. --- Literature Review In the 'Online Persuasion in Facebook and Mixi: A Cross-Cultural Comparison' B.J. Fogg and Daisuke Iizawa from Stanford analyzed the cultural difference between Facebook and Mixi users (Fogg & Iizawa, 2008). About creating a personal profile page, they found out, that Facebook profile creation is a one long step without any initial trust-building period, whereas Mixi's approach is a two stage model. Kim deals with the comparison of American and Korean college students and the way they use social networks (Kim, Sohn, & Choi, 2011). The result was that the motives for using SNSs were identical: seeking friends, social support, information, entertainment and convenience. But the patterns of usage were very different. Acquistiand and Grossdeals's research suggest that age and student status are the most significant factors in determining a Facebook membership (Acquisti & Gross, 2006). The study shows that there is a dichotomy between privacy concerns and willingness to share private information. Wilson indicated that the emphasis was set on user interactions and to show that users tend to interact most of the time only with a small group of friends, thus they have basically no interactions with almost 50 % of their other Facebook friends (Wilson, Boe, Sala, Puttaswamy, & Zhao, 2009). As a result it is shown that interaction activity on Facebook is significantly skewed towards a small portion of each user's social links. This is contrary to the general assumption, that all social friendships have the same meaning. Therefore they introduced a so called 'interaction graph' as a more accurate representation of meaningful friendships on social networks. Kwon and Wen's reserach show the extended TAM based on the conventional TAM with the new perceived constructs and external variables to test the social network service (SNS) (Kwon & Wen, 2010). It chooses seven variables including three individual characteristics which are social identity, telepresence, and altruism; three external determinants which are perceived ease of use, perceived usefulness, and perceived encouragement; and the final variable, actual use. In L. Li's 'Case Study on Facebook and Renren Network' the main focus lies on the comparison between Social Network Sites in the United States and China. One of the result shows that Facebook users diversify their sources of information, while Renren network users have only single sources (Li, 2011). Facebook users are mainly young people, but Renren users are aged between 18 and 34. Facebook has also more widely user groups than Renren. The last result is that on Facebook the men-to-women ratio is pretty much equal whereas Renren users are mostly women. This may be due to the fact that the two sites provided different application whose attractiveness is different between men and women. --- Research Framework The present study concentrated about three principal arguments, invitation, knowledge sharing and trust. The present study concentrated to the typologies of friends have a typical user in the different SNS. The result about the discussion was interesting because the present study realized that Facebook users usually have more friends who are not friends in the real society. This phenomenon is called cyber-friends and in the last 15 years it is in continuo increase in the USA. So our questions are: do typical Renren user have cyber friends? And if he has not, why does not this phenomenon start in Renren SNS? Hypothesis1: Facebook users have more net friends who are not friends in the real society, while Renren users have more friends who are already friends in the real world. The second topic about this study was knowledge sharing. The result was that the goal to use SNS for Renren users is to share emotional things and talk about himself, but for Facebook users that is uneasily. In addition Facebook users share more useless things than Renren users. So the research questions are: do typical Renren user share more simply emotional things than typical Facebook user? And are the topics shared more serious in Renren than Facebook? Hypothesis2: Renren users share more emotional things and thinking about the society than Facebook users. The last topic about this study was trust in the two different SNSs. For Chinese people, one interesting argument was that Renren users usually share their argument in a fixed group but for Facebook users that is not usually true. So our questions are: do Renren users usually share their topic in a fixed group? And do Facebook users share their topic in the same way? If not, why? Hypothesis3: The topics of Renren users are more restrict to a fixed group. --- Method To test the hypotheses more accurately, the present study use the method of experiment to collect data. Besides a small questionnaire survey is included in the experiment. --- Experiment Design This is a controlled laboratory environment, where distraction or external stimuli is avoided. 8 Facebook users from Germany and 8 Renren users from China with comprehensive network (at least 50 friends) were invited to the experiment. The tasks included adding new friend, grouping new friends and answering some questions about their current friends (in a 7-point Likert scale survey, as shown in Table 1). Total experiment time was last for around 120 seconds. --- Results and Discussions In a study, a population is the set of all the elements of interest and a sample is a subset of the population. A primary purpose of statistical inference is to develop estimators and test hypothesis about population parameters using information contained in a sample. Our hypotheses are universal hypotheses and therefore the participants of our experiment and survey have to meet several requirements to ensure a valid scientific analysis. First each participant selected has to be from the population which includes all the social network users worldwide. This criterion is easily met for our experiment because the present study have Chinese and other international students here in Tsinghua University. A second criterion is that the participants are selected independently. Since Facebook is used throughout the whole world, there can be a lot of differences between Facebook users from different cultural backgrounds. So to be able to compare RenRen users with a specific homogenous representative group of Facebook users the present study choose a group of German users. Besides, in order to meet this criterion the present study had to make an assumption. The assumption is that Chinese students are representative for RenRen users and German Students are representative for one user group of Facebook. --- Analysis of Hypothesis1 The that if the friend is just on the list of the participant's social network, but they meet each other seldom, then The present study consider they are not friends in the real society but net friends. In order to verify our Hypothesis1, The present study should pay attention to the results of all the three tasks. First, Chinese participates add more new friends than German participates; --- Analysis of Hypothesis2 To analyze Hypothesis 2, it had to split the hypotheses into two parts: First whether Renren users share more emotional things and second if they share more thoughts about the society. The first part relates to the questions number four and five of our questionnaire. The second part of hypothesis 2 relates to questions number one, two and three of the questionnaire. Mean rating of Chinese participants is 4.625(SD=3.41), and mean rating of German participants is 6.625 (SD=0.268). The ttest result indicate the difference in the answers for all the questions on emotionrelated sharing is significantly different between Chinese and German participants (t=2.949, p-value= 0.009). Here German participants chose a higher value which means that they want to stay more in touch with friends. Statistically it can be proven, that both means (<unk>) are different. This can be done with the rejection of H(0) with a probability of 95% because the p-value for this question's answers is less than <unk>(0,05).Very peculiar is the fact that the answers of the German for question 4 have a very small variance. This may be due to the fact that all Germans understood this question in the same way, whereas Chinese maybe didn't understand the question in the same way. Since this statement can be seen as a motto of SNS the German participants maybe were influence by the media, which often states that SNS's main purpose is 'to stay in touch with friends'. Maybe the variance in the answers of Chinese people is due to the fact that they are not yet influenced by the media and by publicity concerning SNS. This result maybe can also be explained by the fact, that Facebook is not limited to Germany and that Facebook users can have a lot of international friends. Most people build up a more or less international network of friends in Facebook. RenRen users don't have this possibility, because RenRen is pretty much restricted to China and to Chinese users. So the fact, 'to stay in touch with friends over physical boundaries of countries', is an issue for Facebook users, but not for RenRen users. This is the only significant difference that could be found during the analysis of the data. Otherwise it cannot be verified, that RenRen users share different things than Facebook users. This counts for emotional topics as well as serious conversation about society and politics. Both SNS users have the same attitude and share the same values, except for the international fact. Fig. 5. Answers to Questionnaire --- Analysis of Hypothesis3 To analyze Hypothesis3 the present study used the data about task 2 in the experiment. The experiment gave us how many friends each participant added to each group. For analyzing this data the present study summed for each participant the friends added in all groups. In average, Chinese participants add 9.25 (SD=12.78) friends in one group and German participant add 2.13(SD=4.98) in one group. The ttest result indicates there is a significant different between Chinese and German participants (t-value = 1.782 and p-value = 0.000) We are looking for evidence to conclude that <unk> (Chinese) is greater than <unk> (German) so the difference between the two population means <unk>(c)-<unk>(g) will be greater than zero. The present study can see that The present study can prove our hypothesis because The present study can reject H(0) with 95% of probability because p-value is less than <unk>(0,05). The p-value is very small because as The present study can see the two mean estimator values are very different. --- Answers to Questionnaire Chinese German I Totally Agree I Don't Agree At All Question No. --- Fig. 6. Group of friends It is also shown that Renren users added more friends than Facebook users in groups like university and hometown but this is not true for other group and the results for groups like work or hobbies are the same. --- Conclusions Although the present study conducted the experiment, questionnaire and following statistical analysis very thoroughly it is almost impossible to make a statistically safe statement, because the number of participants was very scarce. Still it is possible to see some very interesting tendencies and to draw some conclusions from that. To sum up, the present study observed one can say that these Chinese SNS users have more net friends than German SNS users, but German users see their friends more regularly in real life. There is no significant difference between Chinese and German SNS users in terms of communicating with their friends and sharing information. For our research participants it can be seen that the Chinese participants are more restricted to people from university -this being a cultural difference -and to people from Chinathis would be a structural aspect, because Renren is only in Chinese.

Social network service is very popular all around the world. Global social network service assists Chinese users to have more friends from different countries. Friend-making on social network service is not systematically documented and reported for Chinese. Making friends in social network service and making friends in real life include similar activities such as invitation, knowledge sharing and communication. Almost everything users can do in real life can be done through internet. For example, in social network service E websites such as Facebook users can share a photo with their friends as they do in real life. The difference between making friends in social network service and in real life exits on the way how these activities are done. For example, in real life photo sharing is usually accompanied with a face-to-face talking while in social network service it is mainly the online comments on the photo. Therefore it is interesting to find if Chinese user behavior on social network service is consistent with that in real life. Thus it is necessary to design social network service with consideration of Chinese usage pattern. For example, as Chinese tend to have more virtual friend social network service websites can recommend. Social network website can use online activity such as "campus star" to require Chinese user to explore other's home page or use "explore" button to have user randomly go to strangers' homepage. More virtual communication style can be added to increase ease of use. For example voice and video message can be applied in social network service in addition to text message. The result indicates that the Chinese participants have more virtual friends than German participants while German users contact their friends more often in real life. There is no big difference between Chinese and German participants on communicating with their friends and sharing information. The result shed light on future research on Chinese friend-making behavior on social network service and user interface design for Chinese.

INTRODUCTION The decline of a nation's morale can be attributed to a variety of factors, such as economic instability, political corruption, social unrest, and lack of trust in institutions (Dewi & Rakhimahwati, 2021). These factors can lead to feelings of hopelessness, frustration, and disillusionment among citizens, which can, in turn, contribute to a decline in morale. Social unrest, such as protests, riots, and violence, can also decrease confidence (Jejimut, 2019). Citizens may feel unsafe and unsettled, leading to a sense of hopelessness and despair. Lack of trust in institutions, including the media, education, and the justice system, can also contribute to a decline in morale. Citizens may feel that these institutions are biased, untrustworthy, or ineffective, leading to frustration and disillusionment (Irma, 2021). One example of moral decline that occurs in children in Indonesia, especially as cheating in exams (Sari et al., 2020). A child with low morale may tend to cheat on exams without guilt. This is because the child does not understand the importance of honesty and integrity (Setyowati & Ningrum, 2020). Then there is theft; a child with low morale can steal things from the store or his friends without guilt (Concha-Salgado et al., 2022). Another case is violence, a child who has low morale may become aggressive or violent toward others for no apparent reason. This could be bullying, beatings, or even more severe acts of violence (Dewi & Rakhimahwati, 2021). Parents with low morals may engage in pornographic activities, such as watching pornographic videos or engaging in inappropriate sexual acts, even if they are very young and emotionally immature (Rahmatullah, 2019). It is important to remember that low morale can be caused by various factors, including environmental influences, parenting, or mental health problems (Erzad, 2018). Therefore, it is necessary to conduct appropriate evaluations and interventions to help the child improve his behavior through the moral education of the child. Several ways can help overcome declining morale in children, including 1) Good example, Children learn through examples given by parents and the surrounding environment. Therefore, as a parent or caregiver, it is essential to set an excellent example in terms of morals and expected behavior; 2) Appropriate monitoring, especially at crucial times such as when playing with peers, accessing the internet or television, or in an unfamiliar environment; 3) Involve children in suitable activities: Getting children involved in social activities, such as social action, religious activities, or other volunteer activities, can help them understand moral values and practice them in life every day (Rusilowati & Wahyudi, 2020). In overcoming declining morale in children, consistency and perseverance in applying expected moral values are very important. In addition, parents also need to provide positive direction and guidance and support the development of children's social and emotional skills (Mawarti, Diah Ayu, & Wibowo, 2022). Moral education refers to the process of developing an individual's understanding of right and wrong behavior, as well as their ability to make ethical decisions. It is a lifelong process that begins in childhood and continues throughout one's life (Widodo, 2019). Moral education aims to develop a strong sense of moral character in individuals, including qualities such as honesty, compassion, respect, responsibility, and fairness (Kusnan et al., 2022). This is accomplished through a variety of methods, including modeling positive behavior, teaching moral values and principles, and providing opportunities for individuals to practice ethical decision-making (Wongarso et al., 2022). Moral education can occur in many settings, including schools, communities, and families. Each family has its own culture in building interaction relationships that make families different. In addition, the background of each family also affects the culture in a family, such as the rules contained in the family, family habit patterns, to parenting culture. The family environment, especially parents, has a big role in children's development related to their parenting styles (Mawarti, Diah Ayu, & Wibowo, 2022). It can be taught explicitly through lessons and discussions and implicitly through role modeling and reinforcement of positive behavior. The benefits of moral education include developing a sense of personal and social responsibility, improved decision-making skills, increased empathy and understanding of others, and greater social harmony and cohesion. One moral education in the family is taught to families in the Minahasa Tribe, where they teach moral education based on local culture. Riyanti et al. (2022) stated that the Minahasa Tribe community teaches about moral education, such as not allowing free association, not having social jealousy, quickly adjusting, preparing about constancy, tolerance, nationalism, love, caring, toughness, hard work and discipline (Berdame et al., 2013). The values of character education contained in the teachings of the Minahasa Tribe in the Book "Education in the Minahasa Tribe" include; (1) Being patient and not discouraged; (2) Religiosity; (3) Honesty; (4) Hard Work; (5) Independent and Creative; (6) the spirit of liberation; (7) Togetherness and Fraternity; (8) Equal Rights; (9) Peace loving; (10) Care for the environment; and (11) Responsibility (Handayani et al., 2022). --- Theoretical Framework According to Yusuf (2014), the term moral comes from the Latin word "mos," which means customs, customs, rules/values, or ways of life. At the same time, morality is the willingness to accept and implement regulations, discounts, or moral principles. Hurlock (2003) suggests that an honest attitude means an attitude that matches the moral code of a social group. Moral comes from another word, mores, which means ordinances, routines, and customs. The cultivation of children's moral values began to be introduced from an early age. From birth, each child starts to be decorated by the colors of life so that throughout the growth process will develop an understanding of love as a bestowed nature. The definition of moral education itself is the existence of a learning process with which learners can understand themselves and the world around them (J. Julia et al., 2020). The importance of moral education is none other than its essential meaning in life. Moral education is ethical education so that learners can follow good principles in life (Croce, 2019). The content of this education is in the form of the main principles needed to support the permanence of life, such as honesty, truth, sympathy for goodness, and so on. Learners need these good teachings because, in living life, the principles of morality become tools to live this life properly so that we can all become citizens of society who play an active role in encouraging the survival of life itself (Wilson, 2000). Moral education is an awareness a student possesses in processing feelings and thoughts in social life to distinguish between bad and good (Adesaputra et al., 2019). Morals become responsive standards for how humans can fulfill their obligations by empathizing with people or other living things. Moral education aims to help the younger generation achieve better life satisfaction and form a moral society based on a sense of care and affection (Riyanti et al., 2022). Moral education will succeed if the values and behaviors transmitted can be received by students well by understanding, living, and practicing them in various aspects of life (Wahyuni, 2019). Moral education is intended to be how the child can understand and understand the meaning of moral values taught according to religious values, moral traditions of local communities, and cultures that will concretely foster a sense of justice, kindness, politeness, and the concept of right-wrong (Mukarromah et al., 2020). Moral learning in children can be done through religious views, traditions, and community culture. According to Likona, morals consist of (1) consciousness is a person's moral standard in good ethics, (2) self-control is how we direct our emotions by doing good, (3) humility is to behave rationally towards events, (4) moral habits are accustom to various virtues in various activities, (5) will is a commitment that drives us to do something good and right (Mabruri &;Musnandar, 2020). Moral values are divided into morals towards Khaliq (who creates) and ethical morals towards creatures (who are made). Ethical/moral education aims to develop students' personality by providing knowledge, appreciation, and experience about Islamic creed and morals so that they can become qualified Muslims, and have noble morals personally in society, nation, and state (Parinduri, 2020). Moral education's main purpose is to lead people to the straight path to get pleasure and shade from Allah SWT. The hope is that human life will be happy, not only when in the world but also the hereafter (Amen, 2018). The main milestone in cultivating moral values is the family because it is the first place to carry out social interactions that influence children's behavior. Therefore, families must instill noble morals, including truth, honesty, sincerity, patience, affection, love, kindness, generosity, and so on (Juwita, 2018). This study's discussion theme certainly has similarities with previous research, including research from (Kusnan et al., 2022) entitled character education in Building the Morals of Minahasa Children. Furthermore, the study was titled parenting style in educating children with local wisdom of the Minahasa Tribeby (Djakaria, 2018). Finally, the writing of (Manoppo et al., 2022) took the title moral cultivation of children based on Minahasa Culture. The three previous studies have something in common: moral education for Minahasa children. While the location of the difference and, at the same time, the novelty value of this paper is the moral education of Muslim families in the Minahasa Tribe, Manado, where Islam in Manado is a minority religion. So, this study aims to analyze the role of Muslim parents of the Minahasa Tribe, Manado, in educating children's morals in Islamic minority communities with a local cultural approach. --- METHODS This research uses phenomenological study methods with a qualitative approach. Creswell (2018) Phenomenological studies are a research method that focuses more on individual subjective experience. In educating children's morals, phenomenological studies are used with the personal experience of Muslim parents of Minahasa, Manado. The location of the study was conducted in Tomohon District, Minahasa Regency, Manado. The study used snowball sampling techniques. Research data were obtained from observations, and interviews with research subjects using purposive sampling techniques. The instrument research uses a deep interview with the subject. The data obtained are then analyzed reductively, phenomenologically, and editically during data collection. Some things done are reducing data, presenting data, drawing conclusions, and carrying out data verification (Miles et al., 2014). Reducing data is an activity of collecting data from the field, which is then sorted and choosing which data is essential and non-essential. --- FINDINGS AND DISCUSSIONS --- Background of Minahasa Muslim Community, Manado Minahasa etymologically comes from the word Mina-Esa (Minaesa) or Maesa, which means to become one or unite, meaning the hope to unite various Minahasa sub-ethnic groups consisting of Totemboan, Tombulu, Tonsea, Tolour (Tondano), Tonsawang, Ponosokan, Pasan, and Bantik. The name "Minahasa" itself was only used later (Umaternate et al., 2022). "Minahasa" is generally interpreted as "having become one." Palar noted, based on several historical documents, it was said that the first time to use the word "minahasa" was J.D Schierstein, Resident of Manado, in his report to the Governor of Maluku on October 8, 1789 (Rubama et al., 2023). Observation Deep Interviews Data Analysis Figure 2. Research Location Minahasa Regency is an area of the Manado region. People who embrace Islam only have a distribution of 5%, amounting to 13,301 people of the entire population, with a total population of 279,257 people in the area. Christians dominate with a total of 264,806 people, or 95% of the total population (Ratuwalangon, 2021). Therefore, the most likely thing to do to describe the Muslim population in the heart of Batak land is to look at its distribution. The largest Muslim populations in North Tapanuli can be found in the Tarutung, Simangumban, and Pahae Jae areas, with a distribution of over 200 people. In regions such as Purbatua, Pahae Julu, Pangaribuan, and Garoga, the distribution of Muslims is below 2000-1200 inhabitants. A smaller distribution is found in Adian Koting and Siatas Barita, with numbers below 600 inhabitants. Meanwhile, the distribution of Muslims in areas such as Sipahutar, Parmonangan, Pagaran, and Muara is only under 200 people, even none (Killen, M., & Smetana, 2015). One research site is in Taratara Village in Tomohon City, West Tomohon District. The distance from the capital city of North Sulawesi Province (Manado) to Taratara Village is 19 km. The travel time is 45-60 minutes by two-or four-wheeled motorized vehicle. Taratara is a village called Wanua (house) in the Tombulu language, located between Mount Lokon and Rawingkolo Hill (Wawona). The Makalesung River and Meras River flank the town in the north and the Ranowangko River in the east and south. Based on data collected in 2018 by the village government, the population of Taratara Village is 1,385 people, consisting of 1,384 Christians and one Muslim. The number of households is 403 (KK) (Tampake, 2021). Taratara Village is divided into eight wards, each led by a Head of Environment (Pala) who is in charge of conveying suburban orders to the community and is responsible for the security and order of the community within his neighborhood (Umaternate et al., 2022). The residents of Taratara Village still adhere to the culture and traditions of the Minahasa tribe. This can be seen from the people's houses which are still Minahasa-style stilt houses. The daily social language in the community is Minahasa Tombulu and Indonesian, which is more widely used by youth and children. Although still very close to Minahasa culture and customs, the people of Taratara Village are open and friendly towards migrants. This is the capital for maintaining a harmonious life and developing a respectful attitude toward other cultures (Tampake, 2021). Looking at the data compiled by the Bandan Statistics Center of Muinahsa Regency has recorded the number of houses of worship in the Minahasa Regency area from 2020-2022, which can be seen in the following table: Based on the number of religious adherents based on the total population of the Minahasa Muli Regency in 2015-2018, according to the Central Bureau of Statistics, Minahasa Regency has gathered the number of residents with the largest religious adherents in order of Khatolik, followed by Christianity, then Islam, Hinduism, and finally Buddhism, while more details are explained in the table below: For Muslim communities or families in Minahasa Regency who have a residence that is predominantly based on Muslims, the cultivation of moral values or morals is a must so that the continuity of generations can be appropriately maintained. According to Abdul Munir Aritonang, moral education remains a priority for Muslim families here. Even though relatives have different beliefs in their families, respect for differences can still be well maintained (Mu'ti, 2016). Moral education in question can also be said to be a process of change or self-development of students in all aspects of their lives so that it is expected to form individual and social personalities that can interact with community life (Yaqin, 2016). --- Characteristics of the Minahasa Tribe Community This research was conducted by interviewing 4 resource persons with the condition that they have lived in Minahasa for more than 2 years and have a family and children. The identity of the source is detailed in the table below: First family, researchers conducted interviews with RW families living in Minahasa for 15 years and working as private office employees. RW's family has 2 children aged 15 years and 13 years, respectively. The RW family said that educating children's morals is carried out in various ways, such as reciting at home, participating in religious activities at Islamic study centers (mosques), or participating in studies at schools or Islamic organizations on campus. Kelaurga RW prefers to use the model of moral education of children in a democratic way, where children are free to seek education freely. This is so that children can learn directly how the education process that occurs in Muslim minority countries creates a tolerant child's soul. The RW family is a Challenger resident in Minahasa and felt some culture shock there. For example, calling your older sibling directly by name without "frills" calling pak, mom, aunt, etc. In addition, enter the house without saying greetings, and when eating still soy sauce and the sound of a spoon meets the plate. But this is a common thing that happens there. However, the RW family believes that due to environmental factors that influence, it will be fixed; the RW family binds to respect each other and behave politely. The background of the Javanese tribe, the RW family finally educated morals that were slightly different from the surrounding environment. "Children here call people directly by name or by the word ngana (you), and that's a common thing. This affected my son who ended up entering the house without greetings or soy sauce when eating" (RW Family Interview, 2023). In addition, the language used uses a high tone because of the influence of geographical location, so it is accustomed to it. So that the RW Family educates in a lower manner, more comfortable communicating between children and families. Because this RW family applies cross-cultural parenting, thus comparing Minahasa and Javanese cultures. However, it is not impossible to use cross-culture in the RW family. --- b. GC Family The researcher conducted an interview with the GC family, who have lived in Minahasa for more than 30 years as lecturers at private universities in Manado City. This GC family has been blessed with 2 children aged 18 years and 15 years. As a result of the interviews conducted, researchers found that GC families educate their children like what parents used to teach these GC families. The GC family mentioned that Manadonese people generally communicate as they are or in "blunt" terms, so it differs from parenting in the Javanese tribe, for example. The GC family applies mutual openness betw een parents and children; if the child is wrong, the parent must remind them, and vice versa. If the parent is bad, the child may improve, and the parents are not offended by the matter. As revealed by the GC Family in the interview, as follows: "We are very open with children because we implement a family system. So, we communicate in two directions, parents can criticize children or children criticize parents. We want children to think of us as friends so that whatever communication in the family is all involved to give their opinions" (GC Family Interview, 2023). In addition, eastern Indonesians are famous for speaking in high tones, so when communicating, there is no difference between high tones, which means impolite. However, different words are chosen to speak more politely to the elders. Because a high tone does not mean that you are angry with people. Therefore, researchers found a phenomenology of parents or children who speak harsh words such as "dog" or "bastard" thrown by children to older people. But these are common words in Manado and are not a moral violation of decency. Thus, creating a nurturing that has no boundaries between the old and the young because it results from open parenting. However, the above only happens in some coastal parts of Manado, and people who speak are also seen from the background of their environment and education. --- Figure 4. GC Family In educating children, GC families always apply rewards and punishments. This is to create children's discipline; for example, when they go out, they have to say goodbye and tell them when they go home. If you violate the agreement, there will be punishment for the child. The penalty is carried out not on physical organization but on understanding children by instilling moral values or awareness in children. In addition, rewards will be given to children if children get children's achievements, but this reward is by the child's developmental age. However, we teach not because of the prize for the successes achieved but because of the premium for passion and hard work that results in success in children. "There will definitely be punishments and rewards to children for violations or achievements. Because this is a form of moral education for children to always obey the rules and motivate them to always work hard and be enthusiastic about learning" (GC Family Interview, 2023). --- c. DS Family The third family that the researcher interviewed was the family of DS. They have lived in Tomohon, Minahasa, for more than 10 years by working as high school teachers. The DS family has been given 2 children, who are 15 years old and 8 years old. Interviews with DS's family say that in educating their children morally, they use methods that lead to authoritarian models. The Tomohon region is where most Minahasa tribes and Christians are indeed very few to find a central institution of Islamic studies as an educator of children's morals. Therefore, DS families try to educate their children in an authoritarian way, that is, in a way that parents emphasize supervising children to be obedient. Some methods are done, one of which is by telling parents' experiences to children, Suri tauladan, namely imitating the behavior of the Prophet Muhammad, direct learning by practicing directly to children and giving rewards and punishments. "We are more dominant in educating children, especially Islamic-based children's morals. Although in the Netherlands Islam is a minority religion, we educate children as much as possible according to the knowledge we have." In addition, this DS family often holds meetings with Muslim families from other sub-districts or districts to hold joint Islamic studies. This event is held in rotation once a month and invites ustadz to lecture. This is done to strengthen brotherhood among Muslims in the Tomohon region and tell each other problems to educate good children's morals by Islamic guidance. --- Educating children in Islamic minority areas such as Tomohon is challenging for DS families. Where facilities regarding Islamic study centers are still very minimal. Thus, the DS family made several alternatives for educating children's morals. One is by learning lectures about the stories of prophets or the story of the struggle of ancient Islamic figures in the period of the spread of Islam. This lecture activity is done after praying or during the time to gather with family. The content of the lectures led to the values of the spirit of defending Islam and tolerance between religions so that they can also adapt to other non-Islamic children. In essence, although DS families instill moral values in their children authoritarian, they still uphold the value of tolerance between religious communities so that their children have a blessed soul, good manners, obedient worship, obedience to parents, and piousness. In contrast, children live both in the majority and Muslim minority areas. --- d. SA Family The SA family was the second family that researchers met to conduct interviews in Tomohon, Minahasa. This SA family has lived in Tomohon, Minahasa, for over 15 years as Minahasa regional office workers. The SA family has 3 sons who are 18 years old, 15 years old, and 12 years old. Based on the results of interviews, it is known that in educating children's morals, SA families apply a democratic and massive model, which is constantly communicating both ways with children about what they want to learn and who to learn with. The SA family basically frees the child to choose what he wants, and the parent pays attention to the child's will by deliberation whether or not the child and parent can respect each other's decisions. This is done to create children's morals to tolerate each other and respect each other between family members. This was revealed by the YG family as follows: "Educating children's morals, we are free to want to learn what and to anyone, we as parents only give advice, but our decisions are free to children so that children can be responsible for what they choose" (SA Family Interview, 2023). The model of educating children's morals democratically means freeing children to learn Islam with sources from anywhere, which can be from their parents, from the ustad in the mosque, or learn from internet sources. Furthermore, educating children's morals permissively is that parents do not force children to choose what they want, so parents are only limited to monitoring and providing input because they want to train their children to be responsible for their choices. This is revealed in the interview as follows: "Our children are free to study religion from anywhere, whether at home, at school, in the mosque, or sourced from the internet. We only supervise children to learn the religion from the right source, so we as parents are only limited to supervising them. We want to teach children to have morals that are responsible for what they have chosen" (SA Family Interview, 2023). --- Discussions Instilling Moral Values in Minahasa Tribe Children The cultivation and development of moral aspects for children in the Minahasa area must be distinct from the parenting style of parents and educators. Parenting is a relationship between parents and their children to meet various children's needs. Meeting children's needs is centered on the needs of children in the family environment and the needs of children outside the family environment, such as in the school and community environment. Parenting children prepares the child's personality to adjust to various values and orders that apply both in the family environment and the community (Massang et al., 2022). Kohlberg suggests that the development of a child's moral values does not focus on ethical behavior, meaning that a child does not focus on what he does. According to Kohlberg, moral reasoning is the focus of his studies. Observation of behavior does not represent the maturity of the moral aspect. For example, adults may have similarities in behavior but do not reflect similarities in aspects of moral maturity (Safitri et al., 2019). This shows the complexity of the maturity aspect of everyone, including a child. We cannot guarantee that the behavior of an early childhood reflects the maturity of its moral aspects. Thus, a valid measuring tool must be used to assess aspects of a child's moral maturity. Instilling moral and religious values in early childhood requires appropriate learning media relevant to children's growth and development stages (Manoppo et al., 2022). Based on the findings in the field collected from observation data and interviews with 4 families in Minahasa, researchers found several general parenting methods applied by the Minahasa community. This upbringing indeed leads to parenting to educate the morals of Minahasa children. There are several parenting styles for the moral education of Minahasa tribal children, as follows: a. Authoritarian Parenting Authoritarian parenting involves the use of coercion, strict rules, and harshness to control children's behavior (Bun et al., 2020). Parents expect their children to comply with their wishes and commands, and disobedience is met with punishment or sanctions. The authoritarian parenting style can have negative effects on a child's psychological development. It can lead to difficulties in self-control and emotional regulation when interacting with others. Additionally, children raised under authoritarian parenting tend to lack creativity, confidence, and independence. This parenting approach can result in stress, depression, and trauma for the children. Consequently, it is not advisable to practice authoritarian parenting (Puspita Sari, 2020). --- b. Permissive Parenting Permissive parenting involves granting children a high degree of freedom, allowing them to do as they please (Nasution, 2018). Parents exhibit minimal concern for the development of their children. The primary source of upbringing for these children tends to be formal institutions or schools. This style of parenting can lead to selfishness in children as parents often indulge them with material possessions. This selfishness, in turn, hampers the child's ability to form healthy relationships with others (Nuryatmawati & Fauziah, 2020). The permissive parenting approach results in children lacking social competence due to a lack of self-control. --- c. Democratic Parenting This parenting approach involves parents granting both freedom and guidance to their children, allowing them to develop naturally and establish harmonious relationships with their parents (Rahman, 2008). Through open and two-way communication, children become more receptive and wiser. Meanwhile, parents remain objective, and attentive, and provide positive encouragement to their children. This democratic parenting style promotes independence, problem-solving abilities, emotional well-being, positive behavior toward the environment, and overall competence in children. Thus, this parenting pattern is highly recommended for parents (Primayana & Dewi, 2020). --- Children's Moral Education Based on the Local Culture of Minahasa Tribe Cultural reality or local wisdom in society is hereditary knowledge. Therefore, it must be given to children early because they live in an environment and culture. The introduction and teachings of local culture should be introduced in the family and school environments from an early age (Windi Wahyuni & Muazimah, 2020). Massage explained that local cultural values closely related to environmental conservation must be preserved and even passed on to the next generation. Muzakki & Fauziah also argue that children cannot be separated from their identity, culture, and environment. Thus, it is very important to have a learning model for children that reflects local culture (Muzakki & Fauziah, 2015). Based on this, researchers argue that in facing problems that will be faced in the future, children need to have life skills through local wisdom in the context of problem-solving. Through the provision of education or learning based on local culture, children are prepared to face and solve challenges and even problems with knowledge of local culture. The following are the principles of learning in moral education based on local culture, namely: a. The implementation of local culture-based learning activities is based on the stages of child development. b. The learning process based on local culture in its implementation is based on the principle of learning through play. c. Although there is no permanent building, the implementation of local culture-based learning for early childhood must be actualized in a safe state and full of innovative things. d. Implementing learning based on local culture for early childhood must be relevant to integrative thematic learning. e. Learning based on local culture for early childhood is applied to develop potential comprehensively and integrated (Muzakki & Fauziah, 2015). The values contained in local wisdom given in learning can influence children, namely how children respond to their culture. Printing stated that adding knowledge based on the Indonesian nation's culture and local wisdom was carried out to instill a positive attitude in children and appreciation of their cultural richness (Massang et al., 2022). Berdasarkan hasil temuan lapangan, budaya masyarakat Minahasa sangatlah banyak dan kental. However, in this study, the analysis of local wisdom of the Minahasa community, especially the Tondano/Toulour community, limits it to the cultural values of Maesa-esa'an, Mangenang-puddle, Masawang-Trawangan, Matombo-tombola, Malinga-lingam, Mapalus, and Maleo-loan. The explanation is as follows: a. Maesa-esa'an Maesa-esaan means unity. This idea became a symbol for the Minahasa Community in strengthening the unifying spirit among others. This is found in every village that is the location of the study. Rerer Village, for example, whose community consists of several tribes: Javanese, Batak, Minahasa, and others. Although diverse, still one that is a manifestation of the form of masa -Asian. It can also take the form of creating neighborhood pillars, social gatherings, and village pillars; For when there was grief, all community members in the village participated in the event. Not much different from other villages, Makalonsouw Village embodies these Maesa-Asian values through the organization of family harmony (If there is grief, family pillars hold a love meal, which is an aftersorrow meal to comfort the bereaved family). --- b. Mangenang-ngenangan, Puddles are the inner bond / standard inga. Like when there is grief, if he doesn't go to the funeral event, he will feel very embarrassed and dare not leave the house. It can also be manifested in the form of praying for each other, sending souvenirs to each other, visiting each other, in the form of advice, reminding each other; When rich, remember people who are in trouble. This condition is slightly different in the village of Kampung Java, Magenang-puddles are realized through Cultural Preservation Arisan (used as a place to remind each other with cultural reinforcement, consisting of young and old); In the form of tambourine art activities, Javanese visits (every Tuesday night and Sunday night) as well as through art festivals. But the meaning of puddles remains the same, distinguishing them only in their organizations' application and form. --- c. Masawang-sawangan Masawang-Trawangan has the meaning of mutual help, which is manifested in the form of village social gathering activities to help ethics; some joys and sorrows involve almost all residents in Kampung Jawa Village, as well as other areas in Minahasa Regency. Other villages realize this idea through kumaus/weekly activities are activities held on the Sunday after the day of the death/burial of one of the community members. In this activity, each head of the family cooks in his own home and is taken to the funeral home to eat together. This is a manifestation of help. --- d. Matombo-tombola Matombo-tombola has the meaning of supporting each other, which is manifested in everyday life; namely, if any of us is an achiever or has a higher position than us, we must support or take care of each other, lest he falls from his position. If any of us become village ships or school ships, we must keep or support those concerned. Please don't do it the other way around; we try to knock him down. This is the characteristic of the Minahasans. If there are our children who excel and then want to go to a higher level, it must be supported by helping to increase costs for their school needs so that these children do not experience difficulties in school fees. Surely the school process will be completed quickly. The excellent help the less clever, and the rich help the less fortunate. The buttons can be likened to a

The moral degradation of children has recently been found to reflect the quality of upbringing from low-income families. Furthermore, wrong parenting can have an impact on causing understanding in children to be negative, so they tend not to be enthusiastic about pursuing their goals. This incident can also occur as a result of the parenting style applied by his parents not paying attention to moral decay taken from the local culture. The family is the first environment that impacts various elements of children's growth and development, especially their social growth. Procedures and conditions in the family are suitable environments for social children. The purpose of this study is to analyze how the role of Muslim parents of the Minahasa Tribe, Manado, in educating children's morals in Islamic minority communities with a local cultural approach. The research method uses phenomenological study methods with a qualitative approach. Based on the results of the study, researchers found that local culture-based moral education in the formation of children's character/morals in the Tomohon area, Minahasa became one of the concerns of more than the government and even the community and parents to be able to maintain and preserve a valuable character education process from ancient times so that the current generation can understand the value of manners, mutual respect, respect for parents is critical to forming attitudes and moral. Moral education based on local cultural values in the formation of children's character/morals in Tomohon, Minahasa has a very drastic value from parents past and present to educate their children. This is what makes children open to each other by considering parents as friends in discussing problems in the family.

the Sunday after the day of the death/burial of one of the community members. In this activity, each head of the family cooks in his own home and is taken to the funeral home to eat together. This is a manifestation of help. --- d. Matombo-tombola Matombo-tombola has the meaning of supporting each other, which is manifested in everyday life; namely, if any of us is an achiever or has a higher position than us, we must support or take care of each other, lest he falls from his position. If any of us become village ships or school ships, we must keep or support those concerned. Please don't do it the other way around; we try to knock him down. This is the characteristic of the Minahasans. If there are our children who excel and then want to go to a higher level, it must be supported by helping to increase costs for their school needs so that these children do not experience difficulties in school fees. Surely the school process will be completed quickly. The excellent help the less clever, and the rich help the less fortunate. The buttons can be likened to a house pole that stands upright, supporting the house's roof. Matombo-tombola today is still alive in the activities of the people, but it does not exist as in the past. --- e. Mapalus Mapalus is still applied by Minahasa communities such as in Makalonsouw Village, although it is rare, this is due to the fewer malus members due to the number of people who have changed their livelihoods to become motorcycle taxi drivers, many of whom are starting to care about the world of education and others. Mapalus is a group of 20 members who take turns helping each other in terms of agriculture. Each Mapalus group has its own rules, as the Mapalus group in Makalonsouw has rules, including a helping work system; 3 days are provided to help fellow Mapalus group members, and another 3 days to earn a living for each Mapalus member. In a malus group, a person's obligation in membership as his malus group and is reciprocal. The help he gives will be reciprocated similarly by other members. This group has a rule: if Member A does not help Member B, then when Member A has activities, Member B will not help him either. f. --- Malinga-lingaan Meanwhile, the ideas contained in Malinga-lingam or often also called mutual listening, Listen to each other so that harmony is created. Children listen to parents; younger siblings listen to older siblings, and so on. If there is in a monastery, let alone giving advice and the like; then others must listen first; if given the opportunity, they can speak. They must be mutually respectful and respect people's opinions. If anyone older than us is talking, the younger ones should listen. What's more, it is our leader who speaks. With thieves, people will live in order, and a harmonious society can be realized. --- g. Maleo-leosan Maloe-loan is an attitude and behavior that loves and is kind to each other. If all love each other, then you enjoy safe and peaceful living conditions. If maleo-loan is done correctly, then easily other aspects of local wisdom can be carried out properly, such as masa-Asian (we must unite), imaginingpuddles (we must remember each other), masawang-sawangan (We must help each other), matombotombolan (We must carry burdens with each other), Malinga-lingam (We hear each other) and malus (we must help each other in working especially in the garden). Because these maleo -leosan have the highest position of the different rules. The form of linkage between the local wisdom of the Minahasa community, especially the Tondano/Toulour community, can be seen in the following chart: From the diagram above, Maleo-lesson is the core of the Minahasa people's local wisdom, especially the Tondano / Toulour community. Maleo-loan is an attitude and behavior that loves and is mutually kind to each other. Thus, if maleo-loan is done correctly in everyday life, then all other aspects of local wisdom can be carried out properly, such as masa-Asian (we must unite), imagining-puddles (we must remember each other), masawang-sawangan (we must help each other), matombo-tombolan (We must carry the burden with each other), Malinga-lingam (We hear each other) and malus (we must help each other in work mainly in the garden). This is to the explanation of Koentjaraningrat ( 2004), which affirms the value or ideas of unity (maesa-esa'an), inner bonds (magenang-puddles), and cooperation (mapalus). This idea is an effort that can concretely be seen at the level of kinship groups and rural communities. At other, broader levels, these ideas serve more as the basis of organizational ideals, such as in the organizations of former traditional (Pakistan or w leak) and ethnic (Kawana and masa) government units in Minahasa Society (Massang et al., 2022). --- The Influence of parental parenting style on the Morale of Minahasa Children Parenting encompasses various forms of interaction and processes between parents and children within a family, which significantly influence the development of a child's personality (Rahmatullah & Diana, 2022). It is undeniable that the nature of adolescents is influenced by the cultural values instilled by their parents, which are rooted in the society they live in. Every parent aspires to provide the best for their children, and this aspiration shapes the unique parenting style that is passed down to their children. The transmission of values through parenting has a profound impact on a child's personality as they transition into adolescence and adulthood. This influence is attributed to the inherent characteristics and components of an individual's disposition that are established during their early childhood years (Oktariani, 2021). In the community of Tomohon Village, Minahasa Regency, parents long ago educated their children "hard" even though they did not have high education. Still, they had good morals and behavior, one of which was about behaving politely in everyday life such as teaching to apologize when making mistakes, asking for help when needing help, and saying thank you when receiving help from others (Djakaria, 2018). The younger generation appears to be gradually losing their ability and creativity in comprehending the fundamental principles of local cultural values and traditions. For instance, parents from the Pasan tribal community recall that children used to perform the "ya'sumambing" gesture (walking while lowering their hands) when passing in front of their parents. However, this customary practice has gradually faded away and been replaced by a more casual and disrespectful attitude towards their parents. Even the "ya'sumambing" gesture has been substituted with greetings like "hello" or "hi" and so on. Additionally, there are cultural values inherited from previous generations, such as "sihadatalud" (courtesy), "muhahormat" (mutual respect), "tarangtarang" (firm and consistent), "mature karekenan" (honesty), among others. These values were traditionally imparted by parents during shared dinner gatherings, where they also provided guidance when their children made mistakes. However, as time has passed, the current generation has gradually lost their understanding of the fundamental principles underlying local cultural values and traditions. This can be attributed to a lack of socialization and the process of internalization within children from early childhood to adolescence, as parents have not effectively transmitted these values (Tangkudung, 2016). The inheritance process in the formation of children's moral character by the Tomohon community of Minahasa Regency certainly has different processes or ways, even though the goal is the same to form a good character because of the changes that occur in each inheritance in the family, but what kind of process can maintain and educate so that they become children with good moral character. The process of inheriting cultural values to shape a child's character has changed between how parents are educated then and now. This is what makes the loss of cultural values of shade and politeness of children to parents, friends, and even schoolteachers because of different parenting styles, some educate with a foolish attitude, some educate firmly, but from the answers of children some are educated firmly, but the child does not want to listen. This is what drives many changes in the past and present (Tampinongkol, 2021). In the formation of moral education for children in Tomohon, Minahasa government figures also play an essential role in providing direction to the community about the pattern of inheritance of local cultural values in the formation of children's character in Tomohon, Minahasa. The government always reminds the public about the importance of the moral formation of children, and it is good to form good attitudes and behaviors for those who will not become naughty. Not only reminds the community to hold socialization, but it is also one of the programs as a government to sensitize not only children but parents who are also obliged to listen so that they can wisely educate children because the government plays a vital role in constantly reminding and running the program. The role of religious leaders also participates in the formation of character education in children in Tomohon, Minahasa, constantly reminding the community not to educate children so that they always do good things such as inviting them to participate and go to worship, both worship in church and worship in their own homes. This is to be a prayer for them so that they become God-fearing children, and it also makes them strengthen their faith stance so as not to do and avoid things that they do not want; religious leaders play an essential role in reminding their people to direct children to join fellowship such as participating in Sunday school services, Youth worship, youth worship, as well as other organizations and if the Muslim religion is obliged to hold prayers 5 times. This is to form their attitude of faith, which is first faith and fear of God Almighty. --- CONCLUSION Culture-based moral education is a relevant approach to shaping the character and behavior of Minahasa tribal children. Using local cultural values, moral education can become more contextual and effective in the face of social and cultural change. This research is expected to provide a deeper understanding of the influence of culture-based moral education on Minahasa tribal children and contribute to the development of broader moral education in Indonesia. Moral education based on local cultural values in the formation of children's character/morals in Tomohon, Minahasa has a very drastic value from parents past and present to educate their children. This makes children open to each other by considering parents as friends in discussing problems in the family. Although there is a bad habit of using harsh words, high tones, or politeness in calling older people, it is a common thing to happen, but it should be considered always to remind children to respect their elders. Democratic, permissive, and authoritarian moral education in caring for children in the Minahasa tribe is a way for parents to educate intensely by paying attention to local culture.

The moral degradation of children has recently been found to reflect the quality of upbringing from low-income families. Furthermore, wrong parenting can have an impact on causing understanding in children to be negative, so they tend not to be enthusiastic about pursuing their goals. This incident can also occur as a result of the parenting style applied by his parents not paying attention to moral decay taken from the local culture. The family is the first environment that impacts various elements of children's growth and development, especially their social growth. Procedures and conditions in the family are suitable environments for social children. The purpose of this study is to analyze how the role of Muslim parents of the Minahasa Tribe, Manado, in educating children's morals in Islamic minority communities with a local cultural approach. The research method uses phenomenological study methods with a qualitative approach. Based on the results of the study, researchers found that local culture-based moral education in the formation of children's character/morals in the Tomohon area, Minahasa became one of the concerns of more than the government and even the community and parents to be able to maintain and preserve a valuable character education process from ancient times so that the current generation can understand the value of manners, mutual respect, respect for parents is critical to forming attitudes and moral. Moral education based on local cultural values in the formation of children's character/morals in Tomohon, Minahasa has a very drastic value from parents past and present to educate their children. This is what makes children open to each other by considering parents as friends in discussing problems in the family.

Background El Salvador is among the most violent countries worldwide. International data show El Salvador's homicide rate at 52.2 per 100,000 over the 2000-2012 period, second only to Honduras (63.4) [1]. Figure 1 shows the crude homicide rate increasing from 37.0 in 2002 to 69.9 in 2011, then dropping to 41.2 per 100,000 in 2012 [1]. In 2015, the homicide rate reached 104.6 per 100,000, its highest rate during the last 20 years or so. Also, El Salvador is part of the group of countries with high suicide rates. International data show this nation with an age-standardized rate of 13.6 suicides per 100,000 in 2012, ranking 40th in the world and the highest in Latin America [2]. Our own calculations using official data show a declining trend in the crude suicide rate from 13.5 in 2002 to 7.7 per 100,000 in 2011 (Fig. 1). The same official data show that suicide is more prevalent among the youth (15-29 years) than among the elderly (60 years and over) and that there is a convergence of the age-specific suicide rates. The ratio of the youth to elderly rates has dropped from 1.91 in 2002 to 1.26 in 2012. A similar trend was observed among people in the 30-59 year age-group (Fig. 2). Overall, males are nearly four-times as likely to die by suicide as females, except for young people in the 15-19 age group where the females' suicide rate was higher than the males' suicide rate [3]. El Salvador differs from other Central American countries in that most suicides between 2005 and 2009 were poisoning related (81.6%) compared to firearm related (1.9%) [4]. Depression, alcohol and drug abuse have been identified as the main factors affecting the rate of death by suicide in El Salvador [5]. Though suicide is a major social and mental health issue in El Salvador, little is known about its prevalence, incidence, etiology and spatio-temporal behavior. No published research on suicide in El Salvador in major academic or professional journals was found at the time of this study. From here forward, total violence is defined as all homicides and completed suicides. This definition excludes instances of homicide followed by suicide. Figure 3 shows both the internally standardized rate of total violence, and the ratio of the internally standardized rate of suicide to the internally standardized rate of total violence for Salvadorian municipalities in 2002, 2008 and 2012. 1 Henceforth we will use the terms internally standardized rate and relative risk interchangeably. The data suggest that while the regional distribution of the relative risk of total violence has remained quite stable over time, the regional pattern for the ratio of suicide to total violence has both varied across municipalities and clustered around smaller numbers of places. Also, the maps indicate that in some municipalities, mostly located along the Pacific coast and the north-east side of the country, the relative risk of total violence has been consistently dominated by suicide. --- Conceptual approaches to regional variations of suicide Geographic variations in suicide are driven by cultural, social and economic factors. Some of them such as divorce and unemployment, have a stable influence on suicide mortality, independently of context and time. There are other variables such as religion, fertility and female participation in the workforce whose association with suicide rates varies across temporal and contextual factors [6]. Marked changes to any of these factors can modify social roles, values and relationships, and therefore are associated with higher suicide rates [7]. Durkheim's socio-environmental theory predicts high suicide rates in societies with low levels of either social integration or social regulation with the former leading to egoistic suicide and the latter to anomic suicide. The concept of social integration is also referred to as social isolation [8], social cohesion [9], or social support [10]. The frustration-aggression theory [11], status integration theory [12] and social imitation (contagion) [13][14][15][16] have been offered as alternative explanations for the clustering of suicides. According to the former, suicide and homicide result from a process by which increased aggression leads to increases in lethal violence, therefore suicide occurs when individuals perceive themselves as the source of frustration, predictable by social class and status. Integration theory considers the conflict between social roles related to sex, age, occupation and marital status as the crucial driver for suicide. A number of studies provide evidence for suicide contagion, particularly among the youth [17] and suggest that social networks may play an important role for the presence or absence of strong bonding social relationships. From an economic perspective, suicide rates increase with unemployment, in particular among older individuals, and decline with permanent incomes among all but the youngest age groups [18]. Unemployment increases suicide by lowering both current income level and future income expectations through human capital depreciation, an effect that is strongest among people in the middle-aged group [19]. These findings are consistent with the hypothesis that the loss of status due to business contraction may drive lethal violence [11,20]. For countries with such high levels of violence as El Salvador, a link between suicide and homicide rates remains a distinct possibility. For the United States, there is a positive relationship between the two [21]. The so called stream analogy of lethal violence views homicide and suicide as two forms of aggression arising from common sources, mostly frustrations generated by economic factors such as unemployment, economic wellbeing or income inequality, and mediated by relational systems. Similar sets of factors determine whether lethal violence takes the form of homicide or suicide [11]. The strength of relational systems is affected by events such as divorce, birth rates, migration, education and population growth and distribution [22]. Among these factors, the likely impact of migration on both total violence and suicide is particularly relevant for this study. In 2013, the number of Salvadorians living in the U.S. was assessed at 1.2 million [23], equivalent to 19% of El Salvador's current population. It is estimated that daily, at least 276 Salvadorians try to migrate illegally to the U.S. [24]. Migration causes the separation of families, friends and co-workers, and the weakening of ties in the localities of origin. The families and children left behind by migrants may be initially exposed to stress and other mental pain, which may result in suicide. On the other hand, emigration may improve the quality of life among those left behind in the communities of origin, and might relate to lower suicide rates. Little is known about the impacts of emigration on homicide and suicide in the communities of origin. Several studies have investigated the effect of alcohol use disorder (AUD) on suicide, but there are discrepancies across studies. A meta-analysis of 31 studies addressing the association between AUD and suicide concluded that AUD significantly increases the risk of suicide [25]. --- Findings from empirical studies of suicide Both the spatial concentration and the temporal trends in suicide have been of interest to researchers and practitioners in many disciplines. In the U.S., suicide rates correspond closely to social correlates. Within states, temporal variations in suicide rates are associated with larger numbers of foreign-born as well as with fewer numbers of Episcopalians. Across U.S. states, variations in suicide rates are related to demographic (percentage male), economic (per capita income), social (percentage divorced) and cultural (alcohol consumption and gun ownership) factors [26]. In the case of Mexico over the period from 1950 to 2008, despite having suicide rates lower than other countries with similar socioeconomic conditions, these rates have increased amongst young and elderly males. Factors such as demographic changes, migrations, divorce and low access to education, as well as decreasing per capita income from recurrent economic crises might explain the observed increase in suicide; eight times more likely in adult men than in adult women [27,28]. Clusters of suicide in the municipality of Sao Paulo, Brazil are associated with socio-economic and cultural characteristics. Suicide concentrates in areas with high levels of social isolation (percentage of single persons, percentage of migrants and percentage of Catholics) [29]. For Australia, high suicide risk has been found to cluster in geographical areas having lower median socioeconomic status, higher unemployment rates, and larger proportion of aboriginal population. Also, temperature correlated strongly with suicide rates in Australian Local Government Areas (equivalent to municipalities) [30,31]. In Canada, there is no common set of social and economic factors lying behind variations in total, male and female suicide across provinces, some of which drive the results for some socioeconomic variables [32]. A joint temporal analysis of male and female suicide mortality in the 33 London boroughs showed a growth in the impact of deprivation for both genders combined with a small waning in the role of social fragmentation [33]. The findings in ecological studies suggest the likely presence of spatially clustered unobserved influences on suicide [34][35][36]. The possibility might exist, as it has been found for homicide [37,38], that spatial effects in suicide be of greater magnitude than local characteristics and that general suicide patterns be similar at all spatial scales. Other studies have found a negative relationship between lithium in drinking water and suicide mortality, that seems to be moderated by altitude [39][40][41]. Studies based on the frustration/aggression thesis and the attribution thesis provide evidence in favor of the stream analogy as an explanation to both total violence and suicide. An analysis of cross-national data showed income inequality and economic development as the key predictors for suicide. Divorce increased total violence but did not influence suicide [20]. The observed stability of the spatio-temporal pattern for the relative risk of total violence (see Fig. 3) is somehow consistent with a constant (relative) amount of lethal violence (suicide plus homicide) in a society, one of the key assumptions of the stream analogy to total violence. Also, the temporal variation in the spatial patterns of the suicide to total-violence ratio is consistent with the risks of homicide and suicide moving in opposite directions, another assumption of the stream analogy. --- About this research This research uses the stream analogy of lethal violence to explain geographical concentrations of suicide over time among municipalities in El Salvador to test the two main hypothesis that, after controlling for the effects of both local economic conditions and the weakening of the local relational system, (1) the concentration of inward violence among well identified clusters of municipalities has remained stable over time, and (2) that there is a process by which new spatial clusters of suicide are forming over time. Understanding both the geographic stability and spatial expansion of suicide have important theoretical and policy implications. Time-stable concentrations of suicide death reveal areas sharing social, economic and other conditions driving residents to turn violence to themselves. In the same way, the identification of patterns of geographic displacement of suicide over time helps authorities in designing programs to contain the spread of these effects. --- Methods --- Data sources and measures Counts of suicides and homicides over the period from 2002 to 2012 were obtained from IML [42] for each of the 262 Salvadorian municipalities. Population projections at the municipal level were obtained from DIGESTYC [43]. No data on municipal projected populations by sex and age were available, so it was not possible to conduct a separate analysis for male and female suicide rates or working with age standardized rates. For each municipality, the ratio of observed to expected rates of lethal violence (suicide, homicide and homicide plus suicide) was used as a measure of risk. A program for the periodic production and dissemination of social and economic statistics for municipalities does not exist in El Salvador. Municipal data often are produced for specific studies such as the set of municipal Gini coefficients used for this research [44]. We measured the strength of the relational system in each municipality with its migration rate per 10,000 population. These rates were computed from unit record data for the latest census of population and housing conducted in 2007 [45]. No municipality-based data on AUD were available at the time of this study. --- Data analysis Bayesian spatial and temporal models are widely used in public health and crime research [30,31,33,34,36,[46][47][48][49][50][51][52][53][54]. Two separate hierarchical models were fitted to the data, one for total violence and another for suicide as a proportion of total violence. Let the i index designate a municipality,(i = 1, 2,..., 262), the k index, a specific year,(k = 1, 2,..., 12), S ik, the number of completed suicides and n ik, the number of cases of total violence in municipality i during year k. --- Model for total violence The first level of the hierarchical model for total violence assumes that n ik follows a Poisson distribution with mean E ik <unk> ik, given by n ik j <unk> ik <unk>Poisson E ik <unk> ik <unk> <unk>1<unk> In this expression, <unk> ik is the unknown relative risk and E ik, the expected numbers of cases of total violence in municipality i and year k. Relative risks are estimated by (indirectly) standardized mortality rates (SMRs). For rare events such as violent deaths, extra Poisson variation may arise either from heterogeneity of individual risk levels within municipalities, or from the clustering of incidences in either space or time, or both. Allowing area-specific risks to depend on latent variables (spatial random effects) is a standard approach to accommodate over-dispersion in counts in the epidemiological literature [49,50]. At the second level of the hierarchy, it is assumed that the logarithm of the relative risk of total violence in municipality i and year k, varies over space and time around an overall rate according to. log <unk> ik <unk> <unk> 1<unk>4 <unk> <unk> <unk>t k <unk> X j <unk> j X ikj <unk> V i <unk> U i <unk> <unk> i t k ;<unk>2<unk> where U i and V i represent correlated and uncorrelated spatial components, <unk>t k is a trend that is linear in time t k, <unk> i t k represents an area-specific trend, <unk> is an intercept representing an overall relative risk, X ikj represents municipality-specific covariates (i.e. Gini coefficient, migration rate, etc.), and <unk> j, their respective coefficients [55]. The Deviance Information Criterion (DIC) [55] was used to assess the fit and identification of the model in ( 1) -(2) relative to a simpler model without spatio-temporal interaction terms and covariates. Models with smaller DIC are considered better. The DIC for the selected model (including the area-specific temporal trend, the Gini coefficient and the migration rate) was 16,367 much lower than the DIC of 17,033 for a model excluding these terms. --- Model for the suicide to total-violence ratio The number of suicides in municipality i during year k, S ik, follows a Poisson distribution. The distribution of S ik conditional on n ik, the number of cases of total violence in municipality i during year k, is distributed as Binomial with parameter p ik, 2 given by S ik j p ik <unk>Binomial p ik ; ; n ik <unk> <unk> ;<unk>3<unk> with p ik denoting suicide as a proportion of total violence [56]. At the second level, a model for the association of the log-odds of suicide, <unk> ik 1<unk>4 log p ik 1-p ik, with spatial and temporal variation [46,47,50], is specified as <unk> ik 1<unk>4 <unk> <unk> <unk>t k <unk> X j <unk> j X ikj <unk> V i <unk> U i <unk> <unk> i t k;<unk>4<unk> with U i, V i, <unk>t k, <unk> i t k, <unk>, and <unk> as defined for (2). The DIC for the model in ( 3) and (4) was 10,678 compared to a value of 11,896 for the model not including a space-time interaction, Gini coefficient and migration rate. In both models (2) and ( 4), the <unk> parameter was assigned an improper uniform prior on the whole real line and the mean time trend (<unk>) was given a vague prior normal distribution with a zero mean and a variance of 1000. The choice of these priors expressed the absence of genuine prior expectations on the parameter values. A normal prior with zero mean and a variance <unk> 2 v was given to the unstructured random effects (V i ). The regression-like coefficients <unk> j were given normal priors with zero means and <unk> 2 <unk> variances. Similarly, in both models, conditionally autoregressive (CAR) priors [57] were used for the spatially structured random effects (U i ) and the spatio-temporal interaction terms (<unk> i ). Under the CAR specification and for a given municipality, the mean of U i and <unk> i depends upon the U i ′ s and <unk> i ′ s of its neighboring municipalities. Variance parameters <unk> 2 u and <unk> 2 <unk> control the variability of the random effects U i and <unk> i conditional upon the random effects in the neighboring municipalities, respectively. Uniform distributions (i.e. U(0, 10)) were chosen as hyperpriors for all variance parameters(<unk> 2 <unk>, <unk> 2 v, <unk> 2 u, <unk> 2 <unk>, <unk> 2 <unk> ). Both models were fitted using WinBUGS [58], a programming language based software implementing MCMC algorithms to generate random samples from their respective posterior distributions. The WinBUGS code for models ( 2) and ( 4) is available on request from the author. In each case, two chains were run and convergence was achieved by 20,000 iterations. A further 20,000 samples were run for each chain to obtain the desired posteriors with Monte Carlo errors lower than 5% of the posterior standard deviation. --- Results and discussion Table 1 shows the posterior means of the overall log-relative risk (<unk>), the time trend (<unk>) and variance components (<unk> u, <unk> v and <unk> <unk> ) for both the model for total violence and the model for the ratio of suicide-risk to total violence-risk. Additionally, it includes the lower and upper credible interval limits for each posterior mean. --- Rate of death by total violence Municipal-level SMRs of total violence varied around a national average of 0.247 (1<unk>4 exp -1:397 <unk> <unk>). With a 95% chance, the temporal trend (<unk>) took on any value between -0.001 and 0.010, including zero. This means that with a large probability, the national relative risk of total violence might have remained constant over most of the 2002-2012 period. This finding supports the key assumption of the stream analogy to total violence that a constant (relative) amount of lethal violence (completed suicide plus homicide) exists in a society. Variation around the baseline SMR for specific municipalities was made up of spatial effects, unstructured heterogeneity, and a space-time interaction. Consistent with expectations, spatial variation dominated the total variance of municipal rates of total violence around the national average, indicating the presence of time persistent regional clusters of (inward plus outward) violence. Data in Table 1 show that all the variances for the posterior distributions of area-random effects had 95% credible intervals excluding zero, with variation due to spatial correlation (U i ) being greater than variation due to heterogeneity (V i ) and variation due to spatio-temporal interaction(<unk> i ). Figure 4 shows Salvadorian municipalities according to whether their posterior relative risk of total violence increased, decreased or remained unchanged over the 11 years under study. In 148 out of the 262 municipalities (shown as red on the map), the risk of total violence (homicide plus completed suicide) increased with most of this increase arising from spatial correlation. This is confirmed by the map in Fig. 5 that displays the posterior probability of having a positive spatial random component within each municipality. The spatial clustering of total violence is similar to the one identified for homicide [52]. With 2 exceptions, clusters of high-incidence municipalities, including the capital city of San Salvador, were located in the western side of the country. In these places, the risk of total violence was affected by the risk of total violence in neighboring municipalities, with this spatial pattern remaining stable over time. In the eastern municipalities where pure spatial effects were weaker, increases to the risk of total violence were due to the formation of new clusters in the west-east direction. Figure 6 shows strong area-specific differential trends for many eastern municipalities. Combining these results with those in [52] suggests that in most of these places, the geographic spread of the risk of total violence was due to an increasing risk of homicide, but there were a few municipalities where the source of this spread was an increase in the risk of suicide. The data in Table 1 show that consistent with the stream analogy, total violence increased with income inequality and decreased with a weakened relational system. --- Suicide Completed suicide as a proportion of total violence varied around a national average of 2.25% (1<unk>4 100 <unk> exp 1:052 <unk> <unk> 1<unk> exp 1:052 <unk> <unk> ). Fluctuation around this baseline arose from a temporal trend (<unk>) which with a 95% chance took on any value between -0.061 and -0.042, excluding zero. This result suggests that with a large probability and over the 2002-2012 period, suicide as a proportion of total violence has declined by an average 5.1% a year. Figure 7 shows municipalities where the contribution of suicide to the SMR of total violence depended upon its proportion in neighboring localities. The clustering was stronger in the north-east side and the center-south sides of the country. Figure 8 shows maps of the probability of an increasing (left) or decreasing (right) spatiotemporal parameter. These results suggest the presence of time-persistent clusters of suicide risk among north-eastern and centersouthern municipalities. In general, the municipalities where clustering was strongest (red) (Fig. 7) had stable On the other hand, there is a process of formation of suicide clusters in the north-west, south-west and center-west sides of the country (Fig. 8, left panel). Coincidentally, the municipalities where suicide is increasing its contribution to total violence, were also identified as forming stable clusters of homicide [52]. Table 1 above shows that the greater the Gini coefficient within a municipality, the lower is the contribution that suicide makes to total violence. On the other hand, larger migration rates were associated with increased suicide relative to total violence. These findings were consistent with the predictions of the stream analogy of total violence and somewhat similar to those reported in previous studies [20]. --- Conclusions This research used a Bayesian model based on the stream analogy of lethal violence to examine the spatiotemporal evolution of suicide for the 262 municipalities of El Salvador. Our findings provide some empirical evidence for this conceptual approach that considers suicide and homicide as different manifestations of the same event, particularly that the rate of death by total violence (homicide plus completed suicide) has remained stable across municipalities over time. The rate of death by suicide has declined over the 2002-2012 period but the decrease was not uniform across municipalities. Consistent with the stream analogy, a high level of income inequality increased the rate of total violence but was related to a lower tendency of completed suicide. On the other hand, while increased social isolation was related to higher rates of suicide, it was associated with a reduced risk of outward violence. Spatial variation dominated variation due to other sources for both total lethal violence and suicide over homicide. For total violence, spatial correlations were stronger for municipalities located in the western-side of the country, whereas for suicide, spatial effects were stronger among municipalities located in the north-eastern and center-south sides of the country. Clusters of municipalities with either a high SMR of total violence or a heavy tendency of death by suicide over homicide were stable Fig. 6 Total violence, posterior probability of area-specific differential trends (<unk> i ) greater than cero Fig. 7 Suicide as a proportion of total violence, posterior probability of spatial random effects (U i ) greater than cero over time suggesting a sharing of social, economic and other conditions that drive residents to either turn to violence on others or to turn it against themselves. New clusters of high SMRs of total violence are forming in the west-east direction. Dynamic clustering of suicide is affecting municipalities located in the north-west, south-west and center-south sides of the country. Previous research identified these localities as being part of timestable clusters of homicide [52]. Prevention efforts to reduce income inequality and mitigate the negative effects of a weak relational system should focus upon municipalities belonging to the timepersistent clusters of high suicide mortality. Income support programs, education and employment programs, and initiatives aimed at strengthening social capital and the functionality of families have the potential to revert suicidal tendencies among local residents. Given the relatively large numbers of poisoning related suicides in El Salvador, interventions aimed at legislating to remove locally dangerous pesticides from agricultural practice; enforcing regulations on the sale of pesticides; reducing access to pesticides; and reducing their toxicity may prove effective in containing the geographic spread of suicide mortality. Reducing the poisoning related suicides may require interventions for individual risk factors such as alcohol consumption or mental disorder. Continual and repeated training of health workers in the assessment and management of mental and substance use disorders is crucial in suicide prevention [59]. Municipalities being part of newly-formed suicide clusters, in particular those located in areas identified as belonging to time-stable clusters of homicide may require quite different approaches. In these areas, suicides might be caused by factors related to stress of long-run exposure to violence; emotional stress and depression due to trauma or abuse; and social isolation among those left behind in highmigration regions. Interventions such as gatekeeper training programs for identifying individuals at risk might prove effective. Key potential gatekeepers include health providers, teachers and school staff, community leaders, police officers and firefighters, military officers, social welfare workers, religious leaders, human resource staff and managers [60]. Interrupting the process of timeincreasing concentrations of suicide may require the implementation of both public health and public safety interventions. --- Endnotes 1 Internally standardized mortality rates were obtained from the ratio of the observed number of deaths to the expected number of deaths in each municipality and year. Expected numbers of deaths were computed from the product of the average mortality rate over municipalities and years, times total population for each municipality and year. 2 Assuming that the numbers of homicides and suicides within municipality i during year k, both are Poisson random variables with rates <unk> H and <unk> S, respectively, the conditional distribution of S ik given n ik is Binomial with p ik 1<unk>4 <unk> S <unk> S <unk> H. Not applicable. --- Availability of data and materials The suicide and homicide datasets supporting the conclusions of this article were derived from data obtained in PDF format from the Unidad de Acceso a la Información P<unk>blica, Corte Suprema de Justicia, through the procedures established by the Ley de Acceso a la Información P<unk>blica de la Rep<unk>blica de El Salvador.Population projection datasets are available from DIGESTYC at http://www.digestyc.gob.sv/index.php/temas/ des/ehpm/publicaciones-ehpm.html?download=517%3Aestimacionesy-proyecciones-de-poblacion-municipal-2005-2025. --- Funding --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background: In 2012, international statistics showed El Salvador's suicide rate as 40th in the world and the highest in Latin America. Over the last 15 years, national statistics show the suicide death rate declining as opposed to an increasing rate of homicide. Though completed suicide is an important social and health issue, little is known about its prevalence, incidence, etiology and spatio-temporal behavior. The primary objective of this study was to examine completed suicide and homicide using the stream analogy to lethal violence within a spatio-temporal framework. Methods: A Bayesian model was applied to examine the spatio-temporal evolution of the tendency of completed suicide over homicide in El Salvador. Data on numbers of suicides and homicides at the municipal level were obtained from the Instituto de Medicina Legal (IML) and population counts, from the Dirección General de Estadística y Censos (DIGESTYC), for the period of 2002 to 2012. Data on migration were derived from the 2007 Population Census, and inequality data were obtained from a study by Damianović, Valenzuela and Vera. Results: The data reveal a stable standardized rate of total lethal violence (completed suicide plus homicide) across municipalities over time; a decline in suicide; and a standardized suicide rate decreasing with income inequality but increasing with social isolation. Municipalities clustered in terms of both total lethal violence and suicide standardized rates. Conclusions: Spatial effects for suicide were stronger among municipalities located in the north-east and center-south sides of the country. New clusters of municipalities with large suicide standardized rates were detected in the north-west, south-west and center-south regions, all of which are part of time-stable clusters of homicide. Prevention efforts to reduce income inequality and mitigate the negative effects of weak relational systems should focus upon municipalities forming time-persistent clusters with a large rate of death by suicide. In municipalities that are part of newly-formed suicide clusters and also are located in areas with a large rate of homicide, interrupting the expansion of spatial concentrations of suicide over time may require the implementation of both public health and public safety interventions.

INTRODUCTION In high-income western countries such as Switzerland, cardiovascular (CV) disease is the leading cause of death. CV disease has its origin early in life due to genetic predisposition, environmental influence as well as socioeconomic and lifestyleassociated status. Over the last decades, the assessment of retinal vessel diameters and arterial stiffness by aortic pulse wave velocity (PWV) have become well-established and recommended vascular biomarkers as a surrogate end point for CV risk assessment (1)(2)(3)(4). We recently demonstrated that childhood obesity, high blood pressure (BP) and low physical fitness are associated with retinal micro-and macrovascular alterations in young children (5). Children with obesity and elevated BP had higher arterial stiffness and narrower retinal arteriolar diameters. Cardiorespiratory fitness was associated with a favorable higher retinal arteriolar-to-venular ratio (AVR) as well as lower arterial stiffness (5). Environmental conditions such as socioeconomic status (SES), migration background and parental lifestyle seem to play a major role in the development of CV risk factors in children and adolescence. Evidence suggests that the prevalence of CV risk factors is higher in children with low SES. For example, an inverse correlation has previously been described between household income and childhood overweight and obesity (6,7). The International Study of Childhood Obesity, Lifestyle and the Environment in 12 countries around the world found that parental educational level is also inversely associated with childhood overweight and obesity (8). Parental educational level seems to be contributed to a higher prevalence of childhood obesity (9), elevated BP (10,11) and sedentary behavior (7,12). Passive smoking seems to affect development of childhood obesity (13) and vascular health (14,15). Migration background has also been associated with higher prevalence of children with obesity and physical inactivity (7,9,12). Nonetheless, evidence for associations between socioeconomic status, parental smoking and physical activity (PA) behavior with childhood macroand microvascular health are scarce. This study, for the first time, aimed to investigate the association of SES, migration background and parental lifestyle with large artery stiffness and retinal microvascular health in young school children. --- METHOD Study Design and Participants This study is part of the large-scale, cross-sectional EXAMIN YOUTH study (16). Children were included if they were between 6 and 8 years old and had an informed consent from their parents to participate. Briefly, children with written informed consent were screened for large artery stiffness and retinal vessel diameters to assess macro-and microvascular health. Anthropometric parameters and blood pressure were determined according to standardized procedures for children. In addition, physical fitness was assessed on a separate day. The medical screening was performed in the morning on-site the regular school setting and children had to remain fastened before anthropometric and CV assessments. Parents were requested to complete a questionnaire on lifestyle behavior, level of education and SES (16). The questionnaires were handed out and recollected by the teachers. If the questionnaire was not returned to the teachers by the parents, an additional reminder was sent to the parents asking them to return the completed questionnaire. Approval from the Ethics Committee of the University of Basel (EKBB, Basel, No. 258/12) was conferred. The study was designed according to the Guidelines for Good Clinical Practice of the Declaration of Helsinki (17) and the manuscript conforms to The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Guidelines (18). --- Measurements --- Retinal Microcirculation Retinal vessels were assessed using a fundus camera (Topcon) and an advanced image processing unit (Visualis 3.1, Imedos Systems, Jena, Germany). This technique allows a non-invasive and semi-automated measurement of retinal vessel diameters. The method has been described elsewhere (16). Briefly, two valid images from the retina of the left and the right eye were taken at an angle of 45 • with the optic disc in the center. Retinal vessel diameters were estimated to central retinal arteriolar (CRAE) and central retinal venular diameter equivalents (CRVE) applying the Parr-Hubbard formula (19). The AVR was calculated from CRAE and CRVE. CRAE and CRVE were presented in <unk>m. One measuring unit assigns one <unk>m in the Gullstrand's normal eye. Reproducibility for retinal vessel analysis is high with an interclass coefficient for CRAE of r = 0.94 and a coefficient of variation of about 2% in young children (20). --- Large Artery Stiffness PWV serves as a well-established and validated indicator of arterial stiffness (21). The assessment of PWV was performed using the non-invasive and validated oscillometric Mobil-O-Graph Monitor (I.E.M. GmbH, Germany) with integrated ARCSolver software (22)(23)(24). PWV was achieved in a sitting position and appropriate small-sized cuffs were settled on the left upper arm. After a resting period of 5 min, a blood pressure measurement was performed, which ensured a calibration with systolic BP. Two measurements of pulse wave analysis followed. After checking every measurement for erroneous values, the mean of two valid measurements was used for further analysis. Data on the validity for use of the Mobil-O-Graph in children has previously been published (5,25). --- Anthropometric Parameters and Physical Fitness Anthropometric measurements and physical fitness assessments have been explained in detail elsewhere (16). Briefly, body height and weight (InBody 170 Biospace device; InBody Co., Seoul, Korea) were measured in light sport clothes without wearing shoes, and BMI was calculated. The InBody device has been validated in school-aged children and correlates strongly with the measurement of dual-energy X-ray absorptiometry (26). BP was assessed five times using an automated oscillograph (Oscillomate, CAS Medical Systems, Branford, CT, USA) after 5 min of rest and based on the recommendations of the American Heart Association. Physical fitness was objectively measured by the validated 20-m shuttle run test (27,28). --- Questionnaire Parents were asked to fill in a questionnaire regarding their SES, migration background and parental lifestyle. Questionnaire items were included from our previous study (7) and translated into the seven most spoken languages in Switzerland. --- Socioeconomic Status SES included household income and parental educational level. Household income was determined as low (under CHF 5000/month), medium (CHF 5000-9000/month) and high (over CHF 9000/month) income. Parental educational level referred to the highest school level completed by at least one parent. A low educational level was defined as both parents having/ not having completed compulsory school education, whereas neither of them absolved a vocational training. A medium educational level was determined as at least one parent having completed compulsory school education and vocational training. A high educational level was defined as at least one parent having completed high-school education with or without tertiary education. --- Migration Background Children were categorized into children with two-sided non-European migration background (two non-European parents), migrants with one-sided European migration background (only one non-European parent) or without migration background (both parents are European). --- Parental Lifestyle Parental lifestyle consisted of parental PA level and smoking status. Parental PA was categorized into three groups. Low PA was determined as both parents never being physically active or no more than once a week. Medium PA was defined as at least one parent being physically active twice a week. High PA was defined as at least one parent being physically active several times per week or on a daily basis. Additionally, parents were asked if they were smoker, or non-smoker. --- Statistical Analysis Univariate analysis of covariate (ANCOVA) was used to analyse the association of household income, educational level, parental PA level and smoking status. Different models were fitted to adjust for age and sex as well as household income, educational level and parental lifestyle. Bonferroni post hoc testing was conducted to reveal the direction of the results. Additionally, we applied sensitivity analysis to adjust for further potential confounders. For quantitative analysis of parental PA categories and smoking status, bivariate analysis was performed. To indicate the amount of uncertainty the measurement of effect presents 95% confidence intervals (CI) and a two-sided level of significance of p = 0.05 denotes statistical significance. All analyses were performed using an up-to-date version of Stata 15 (StataCorp LP, College Station, TX, USA). The sample size of the study was given by the expected large number of children and parents giving their consent and was based on calculations of our previous smaller scale study (7,20). In this study conducted by Imhof et al., we assumed a moderate effect size for the influence of a physical fitness performance on retinal vessel diameters in children. The sample size calculated with the software G-Power using F-tests (f = 0.25, power = 0.90% and 5% level of significance) was estimated to be <unk>290 children in total. --- RESULTS --- Population characteristics are presented in --- Analysis of Covariance The results for differences in group-categories are shown in Table 2. --- Socioeconomic Status Thirty percentage (n = 250) of children were in the group with low, 33% (n = 279) with medium, and 37% (n = 304) with high household income. Eleven percentage (n = 83) were children with low, 22% (n = 181) with medium, and 67% (n = 569) with FIGURE 1 | Flow diagram. highly educated parents. Children with low household income parents had a higher PWV compared to children with high household income parents, also after adjustment for educational level (p = 0.033). Retinal vessel diameters were not associated with household income. Lower parental educational level was associated with a lower AVR (p = 0.007) and higher PWV (p = 0.041), but not independent of household income. --- Migration Background In our cohort, 9% (n = 73) of children had two-sided non-European parents, 13% (n = 106) had one-sided non-European parents and in 78% (n = 399) of children both parents were European. Children with a European background showed narrower CRVE (p = 0.012) and higher AVR (p = 0.034) compared to non-European children, independent of household income and parental education. European migration background was not associated with arterial stiffness. --- Parental Lifestyle Based on parental PA level, 25% (n = 211) were classified as parents with a low PA level, 21% (n = 177) as medium and 54% (n = 445) as parents with a high PA level. Based on parental smoking status, 67% (n = 555) were non-smoking parents. Higher parental PA level was related to higher AVR (p = 0.020) and lower PWV (p = 0.035), but not after adjustment for parental smoking status. Parental smoking status was associated with a lower AVR (p = 0.013), but not independent of parental PA level. Children with smoking parents had a higher PWV compared to children with non-smoking parents (p = 0.001), also after adjustment for potential confounders. Bivariate analysis illustrated that microvascular AVR increased according to increasing parental PA level and a non-smoking status (p <unk> 0.001) (Figure 2). --- Association With Parental Gender Mother's and Father's educational level, migration background, smoking and PA behavior are shown separately in Supplementary Table 2. As a main result, children with a mother of high education had a higher AVR (p = 0.001) due to wider arterioles, independent of household income. The educational level of the father was not associated with childhood vascular health. Children with a European mother had a higher AVR (<unk>0.001) due to narrower venules (p = 0.001) compared to children with a two-or one-sided non-European mother independent of household income and educational level. The migration background of the father was not independently associated with childhood vascular health. In children with high physically active mothers, retinal microvascular AVR was higher (p = 0.020) independent of parental smoking status. The PA level of the father was not associated with childhood vascular health. However, smoking status of the father was associated with an unfavorable lower AVR (p = 0.044) and a higher PWV (p = 0.001) independent of parental PA levels. --- DISCUSSION Our study demonstrated several detrimental associations of socioeconomic and migration status as well as parental lifestyle with macro-and microvascular health in young children. The influence of parental lifestyle on vascular phenotype was evident examining parental PA and smoking status with both, retinal vessel diameters and PWV. Children with smoking parents had higher arterial stiffness, independent of parental PA level. The analyses of the interrelation between parental PA categories and smoking status with microvascular AVR demonstrate the additive deleterious effects of parental physical inactivity and smoking on the microvascular phenotype of the children (Figure 2). Children with smoking and low physically active parents showed the lowest AVR in retinal microcirculation. Parental PA levels and smoking status affect the vascular phenotype in young children. It has previously been shown that parental PA and perceptions of competence are associated with parental support for children's PA and therefore, for children's PA at home (29,30). This could be one explanation for the association of parental PA level with vascular health in our study. In addition, there is evidence that childhood exposure of environmental tobacco smoke is associated with vascular functional impairments in adulthood (14,15). In line with our results, a previous study demonstrated that passive smoking is associated to attenuated endothelial dysfunction in 11-year old children (31). Vascular alterations are related to impaired arterial vasodilatation. Arterial dilatation is mediated in large part by the release and bioavailability of nitric oxide. Our findings suggest that the lower AVR and higher arterial stiffness in smokingexposed children may be due to the interaction of smoke particles with the nitric oxide pathway. In our cohort of young children, higher household income was independently associated with lower arterial stiffness. The Young Finns study previously reported that childhood SES, defined as annual household income and years of parental education, is inversely associated with PWV in adulthood (32). Previous studies have shown that parental educational level seems to contribute to a higher prevalence of childhood obesity (9), elevated BP (10,11) and sedentary behavior (7,12). Lower maternal education has been associated with higher blood pressure in 6 year old children (11). However, no relationship between maternal education and arterial stiffness was found (11). In line with previous findings, parental education was not independently associated with PWV in our cohort. Our results show that high maternal education is related to favorable microvascular health (wider CRAE and higher AVR). Interestingly, maternal SES and lifestyle behavior was primarily associated with vascular health of children. An exception was seen for smoking status of the father which was independently association with micro-and macrovascular alterations of children. In adults, it has been shown that narrower CRAE and wider CRVE as well as higher arterial stiffness are predictors for increased risk of CV disease (1,33). Therefore, disadvantaged SES may be an important influencing factor for the development of CV disease later in life. With respect to migration background, European children had narrower CRVE and a higher AVR compared to non-European children. Similar results were found in a previous smaller-sized study on migration status and retinal microcirculation (7). In our cohort, no association of migration background and PWV was observed. There is evidence that migration background is associated with CV risk factors such as childhood obesity and sedentary behavior (7,9,12). We recently showed that childhood obesity and elevated blood pressure are related to retinal vessel alterations and arterial stiffness in this cohort of children (5). Based on our findings, we assumed that both micro-and macrovascular beds give separate clinically relevant information on health disadvantaged of children with migration background. Some limitations have to be discussed. Our cross-sectional study is associative in nature and thus no differentiation in terms of causality can be made. Causal associations of vascular alterations with the development of CV disease in adulthood have to be verified by a longer-term follow up study. Our study was designed in a school setting and 45% of parents did not give consent for participation. Oftentimes parents find it difficult to follow instructions and to fully comprehend the content of the questionnaire and may therefore have decided not to participate (34). The low participation rates in questionnairebased surveys are common limitations (34,35). We were not able to compare participants with non-participants with respect to parental lifestyle and SES. However, children who did not participate in the medical screening and did not have a completed questionnaire had a similar BMI, but a lower physical fitness compared to those participating. The risk of a selection bias was therefore considered to be moderate. The oscillometric measurement with the Mobil-O-Graph device to estimate PWV is based on a calibration with systolic blood pressure. The assessment is thus associated with level of current systolic blood pressure. The assessment of PWV in large cohort of school children has been shown to be a validated approach (5). We refrained from additionally adjusting for systolic blood pressure in our statistical models to avoid over-adjustment. Our findings are related to a predominant Caucasian population with a small percentage of other ethnical groups. We have therefore chosen to characterize our population by looking at non-European migration background. Future studies have to investigate the association of SES, migration background and parental lifestyle with vascular health in other ethnic groups and populations from different countries. One strength of our study is the large sample size and the limited age range of young children. During childhood, mirco-and macrovascular function and structure continuously develop, and especially during puberty, age adaptations occur rapidly. Investigating different vascular beds in a large sample of children at the same age therefore reduces a developmental impact of our findings. Future studies may still investigate the association of SES, migration background and parental lifestyle with vascular health in a wider age range from infancy to adolescence to account for the dynamic pattern of small and large vessel development during childhood. --- CONCLUSION Parental PA has been found to be associated with better microand macrovascular health but not independent of parental smoking status. It appears that parental smoking may mitigate the positive association of parental PA on childhood vascular health. Parental smoking was itself independently associated with impaired large artery stiffness in children. Cessation of parental smoking therefore seems key to improve childhood vascular health. Prospective follow-up studies will have to prove if cessation of parental smoking can improve CV outcome in offspring long-term. SES and migration background also affect micro-and macrovascular impairments early in life. Primary treatment strategies will have to address reduction of socioeconomic barriers. Family-based interventions targeting cessation of parental smoking, increase of parental PA, and reduction of socioeconomic barriers may be effective means to counteract the development of CV risk and disease in offspring and during lifespan. --- DATA AVAILABILITY STATEMENT The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. --- ETHICS STATEMENT The studies involving human participants were reviewed and approved by Ethics Committee of the University of Basel (EKBB, Basel, No. 258/12). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin. --- AUTHOR CONTRIBUTIONS SK planned and conducted the study, collected data, performed the statistical analysis, prepared and revised the manuscript. KE designed the study and revised the manuscript. JG collected data and revised the manuscript. LS, GL, and CH interpreted data and revised the manuscript. LZ designed the study and revised the manuscript. HH conceptualized and designed the study, discussed the statistical analysis, prepared and critically reviewed the manuscript. All authors approved the final manuscript as submitted and agreed to be accountable for all aspects of the work. --- SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpubh. 2021.610268/full#supplementary-material --- Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Background/Aims: Socioeconomic barriers and lifestyle conditions affect development of cardiovascular disease in adults, but little is known about the association of parental lifestyle and education with childhood health. We aimed to investigate the association of socioeconomic status (SES), migration background, parental physical activity (PA) and smoking status with micro-and macrovascular health in children.In 2016/2017, 833 school children (aged 7.2 ± 0.4 years) in Basel (Switzerland) were screened for retinal arteriolar-to-venular ratio (AVR), pulse wave velocity (PWV), SES, migration background and parental PA as well as smoking status. Results: High parental PA levels were associated with a favorable higher AVR (p = 0.020) and lower PWV (p = 0.035), but not independent of parental smoking status. Children with parents who smoked had a higher m/s] compared to children with non-smoking parents [4.32 (4.29-4.34) m/s, p = 0.001]. Children of parents with a low household income had a higher PWV [4.36 (4.32-4.41) m/s] compared to children of parents with a high household income [4.30 (4.26-4.34) m/s, p = 0.033]. Low parental educational level was associated with a lower AVR [0.86 (0.85-0.88)] compared to children with highly educated parents [AVR:0.88 (0.87-0.88), p = 0.007; PWV: 4.33 (4.30-4.35) m/s, p = 0.041]. Children with a European background showed a higher AVR [0.88 (0.87-0.88)] compared to non-European children [AVR: (0.86 (0.85-0.87), p = 0.034].Parental PA is associated with better macro-and microvascular childhood health. However, the positive association is lost when parental smoking is considered in the analysis. Socioeconomic factors seem to associate with subclinical vascular alterations in children. Primary prevention programs should focus on including parental lifestyle interventions and educational programs to reduce the burden of lifestyle-associated barriers in order to improve cardiovascular health during lifespan. Clinical Trial Registration: ClinicalTrials.gov Exercise and Arterial Modulation in Youth, https://clinicaltrials.gov/ct2/show/NCT02853747, NCT02853747.

INTRODUCTION Overweight and obesity are associated with increased mortality and morbidity (Kopelman, 2007), placing a significant burden on healthcare resources, with direct annual costs to the UK National Health Service (NHS) estimated at £4.2bn (Butland et al, 2007). Survey data estimates that, in England, 61% of adults are overweight (Body Mass Index [BMI]<unk>25) or obese (BMI<unk>30) (Tabassum, 2010). In the UK, overweight and obesity is associated with lower socioeconomic status (SES; National Obesity Observatory, 2010) and, among women, with Black African and Black Caribbean groups (Hirani & Stamatakis, 2006). Ethnic differences in England are likely to reflect SES inequalities, because Black and Minority Ethnic people are more likely to live in economically deprived communities (Tinsley & Jacobs, 2006). These disparities are also reflected in health-related behaviours. Whereas 32% of men and 37% of women in the highest income quintile of the UK population consume the Government-recommended five or more fruits and vegetables (F&V) per day, only 18% and 19% respectively of those in the lowest quintile do so (Ogunbadejo & Nicholson, 2010). Of those in the highest income quintile, 42% of men and 34% of women undertake at least five 30-minute sessions of moderate or intensive physical activity (PA) per week, compared to 31% and 26% respectively in the lowest quintile (Roth, 2009). Inequalities are also found in the success rates of interventions to increase healthy eating (HE) and PA. People from disadvantaged communities (i.e. economically deprived and minority ethnic groups) are typically harder to recruit to, and retain in, such interventions (Chinn, White, Howel, Harland & Drinkwater, 2006). This does not necessarily imply less motivation among disadvantaged groups; evidence from the smoking literature suggests that lower SES smokers want, and try, to quit as much as other smokers, but are less likely to succeed (Kotz & West, 2009). Disadvantaged groups have reduced access to the material and social resources needed to sustain behaviour change over time. There is a paucity of HE or PA interventions specifically targeted at low-SES groups: recent reviews identified only seven interventions aimed at low-income adults (published between 1996 and2006;Michie, Jochelson, Markham & Bridle, 2009), but 122 among the general population (Michie, Abraham, Whittington, McAteer & Gupta, 2009). (Michie, Rumsey et al, 2008). The role of HTs is to assess health and lifestyle risks, reinforce motivation, build confidence and skills, and collaborate with clients to set goals (e.g. weight loss of 5kg) and make specific, achievable and realistic action plans (e.g. eating two less fried snacks per day on at least 4 days next week). While most HTs (58%) operate in a paid role, a considerable proportion of HTs (42%) are volunteers (DH, 2010). The aim of recruiting HTs from local communities is to increase understanding of clients' circumstances to help them choose more realistic goals and specific plans to attain them (DH, 2004). HTs that reflect characteristics of clients from local communities (e.g. those matched in ethnicity to clients) are more likely to be effective role models, an important facilitator of behaviour change (Bandura, 1977). Clients can refer themselves or may be referred by others working in health or social care. Most are seen individually and the number of sessions varies according to client need or engagement. The HTS is, to our knowledge, the only national public health service based on health psychology theory and evidence. The service is thus based on evidence that sustained behaviour change requires not only motivation, but also self-regulatory skills (goal-setting, action and coping planning, awareness of environmental triggers to unwanted actions) and self-efficacy for initiating and maintaining change (DH, 2004). These principles are central to evidence-based theories, such as Control Theory (Carver & Scheier, 1982;Michie, Abraham et al, 2009), the Health Action Process Approach (Schwarzer, 1992(Schwarzer,, 2008)), and Social Cognitive Theory (Bandura, 1997). HTs receive a handbook which lists behaviour change techniques known to bolster motivation, skills and confidence (e.g. action planning, barrier identification, general encouragement, using contingent rewards; Abraham & Michie, 2008;Michie, Ashford et al, 2011;Michie, Rumsey et al, 2008). The HTS is notable for the extent of input from health psychologists in its design, implementation and evaluation. A team of health psychologists wrote the HT training handbook, and advised the DH on the development of a set of competencies that form the basis of a recognised vocational qualification for HTs. The team negotiated the creation of a database to detect changes in behaviour and cognition using validated measures, procured DH support for use of the database across all HTSs, and successfully argued for the ringfencing of intervention content, change techniques and outcome measures to prevent the service shifting from its evidence base. Additionally, the team provided long-term support to the HTS, the DH, and regional HT managers and practitioners, and led and evaluated an initial Londonbased pilot of the service (see Wilkinson, Sniehotta & Michie, 2011). Since its launch in 2007, the HTS has grown rapidly, with an estimated 1900 HTs in employment in 2010 (Gowers, 2010;D Smith, Gardner & Michie, 2008;Wilkinson, Jain & Hyland, 2007). Baseline data indicated that a pilot HTS based in inner London successfully reached disadvantaged and typically hard-to-reach clients who engage in unhealthy behaviours and typically have no plans to change their behaviour prior to using the service (Wilkinson et al, 2011). Local evaluation data from other HTSs have shown the service to be valued by stakeholders, HTs and clients (Ball & Nasr, 2011;South, Woodward & Lowcock, 2007). This study uses a national dataset to address the following questions: (1) Is the HTS used by members of disadvantaged communities (those in deprived areas and minority ethnic groups)? (2) Do BMI, associated behaviours (dietary consumption, physical activity), and predictive cognitions change among clients who set HE and PA goals? (3) Does change vary according to deprivation and ethnicity, and whether HTs are from similar ethnic backgrounds to clients? --- METHOD --- Procedure and design Data were recorded in a national database by HTs. This is one of several sources of national HT data, but is the only database requiring systematic reporting of behaviour data. In 2008-09, when the present data were collected, HTSs were encouraged but not compelled to use this database. Cross-referencing with other sources suggests that the 75 local health service organisations (Primary Care Trusts) that provided data for the current analyses represent 62% of the service as at 2009 (J Smith, Gardner & Michie, 2009). Data were collected at sequential timepoints: (i) at first contact, when client demographics (age, gender, ethnicity, deprivation), behaviour (HE and/or PA), height and weight, self-efficacy, and perceived health and wellbeing were recorded; (ii) when clients set their goals and plans; and (iii) at a final assessment when all baseline measurements (except demographics) were repeated (Figure 1). Many aspects of the HT-client consultation process are however personally tailored to clients' needs and circumstances, including the timing of goal-setting and follow-up, the duration of each consultation, and the number of follow-up reviews. 'Post-HTS' data reported herein relate to the final post-baseline assessment for which data were recorded for each client. Due to expected poor literacy levels among many clients, HTs were encouraged to read questions aloud to clients and recorded their verbal responses. An uncontrolled before-after design was used. --- INSERT FIGURE 1 HERE --- Sample Of 27,670 first-time users of the HTS between 1 st April 2008 and 31 st March 2009, 140 were ineligible for the service (reasons for ineligibility were not recorded), 2,544 were under 18 or did not provide age data, 9,265 did not set goals and there were missing progress data for 8,829 clients. From Service Manager meeting reports, many HTs appeared not to perceive data collection to be an important part of the service and so did not comply with the central request to do so. 4,418 clients (16% of all first-time users, 18% of eligible first-time adult users) set primary goals1 relating to HE or PA (rather than, e.g., smoking) and had data on goal achievement and behaviour change available, so were entered into analyses. The sample represents 96.0% of all clients who set goals and for whom goal achievement and pre-post behaviour data were available. The remaining 4% of clients set goals relating to decreasing or ceasing alcohol consumption or smoking, or unspecified 'other' changes. --- Measures Ethnic group was based on UK census classifications, and HT-client ethnicity similarity, i.e. whether HT and client were of the same ethnic group, was recorded. Age was calculated from self-reported date of birth and date of baseline assessment. Deprivation was assessed by mapping client postcode to national 'Super Output Areas' (SOAs; geographical units with an average population of 1,500). Deprivation status for each SOA is determined by the Index of Multiple Deprivation, which summarises seven deprivation indicators (income; employment; health and disability; education, skills and training; housing and services; crime; and living environment; Department for Communities and Local Government, 2011). Each SOA is sorted into quintiles to denote deprivation status within the population. BMI was calculated from self-reported or objectively measured height and weight. Records of whether data were self-reported or objective were not kept by HTs. BMI data were used to categorise participants as underweight (BMI<unk>18.5), normal weight (BMI<unk>18.5<unk>25), overweight (BMI<unk>25<unk>30), or obese (BMI<unk>30). Health behaviours. Clients with PA goals reported the number of 30-minute moderate and/or 20-minute intensive PA sessions undertaken over the previous week. Moderate activity was defined as activities that "make you breathe a bit harder than normal", e.g. "carrying light loads, cycling or walking at a reasonable pace". Intensive activity was defined as activities that "make you breathe much harder than normal", e.g. "heavy lifting, digging, aerobics or fast cycling". Clients with HE goals reported the number of F&V portions and/or quantity of fried snack foods consumed daily over the past week. General self-efficacy was assessed using eight items (e.g. "When facing difficult tasks, I am certain that I will accomplish them", strongly disagree [0]strongly agree [4]; Chen, Gully & Eden, 2001). Higher scores on the summed scale (range 0-32) indicate higher self-efficacy (<unk> =.89). Perceived general health was measured using a visual analogue scale, with clients indicating their current health status from 0 ('poor') to 100 ('perfect'; Lorig, Stewart, Ritter, Gonzalez, Laurent & Lynch, 1996). Wellbeing was assessed using the five-item WHO Wellbeing Index (World Health Organisation, 1998), which records the frequency of positive experiences over the previous two weeks (e.g. "Over the last two weeks I have felt calm and relaxed"; --- At no time [0] -All of the time [5] ). Higher scores on the summed scale (0-25) indicate better wellbeing (<unk> =.76). Since only HT and client demographics were mandatorily recorded, there was variation between HTSs in whether data was recorded (e.g. height and weight, behaviours). --- Data analysis Data cleaning. Prior to analysis, outlier BMI or behaviour scores (values with a relative z-score outside of <unk>3.0), and cases where HTS attendance was less than one week (<unk>7 days) in duration were removed from analysis as they were judged to be either data reporting pilot runs or data entry errors. This excluded less than 2.1% of cases. Biased loss to follow-up. Potential differences were explored between (a) clients who set any goals versus clients who did not set goals, and (b) of clients who set HE or PA goals, those for whom baseline and follow-up data were available versus those for whom such data were unavailable. Chi-square analyses investigated whether there were differences in ethnicity and deprivation between these groups, and t-tests assessed BMI differences. BMI, behaviour and cognition changes were assessed using paired sample ttests for pre-post differences. Effect size d was calculated based on raw means and standard deviations (SDs; Dunlop, Cortina, Vaslow & Burke, 1996) and d <unk>.20,.50 and.80 were interpreted as small, medium and large effects respectively (Cohen, 1992). The consistency of change across demographic groups was investigated by entering client ethnicity, deprivation, and HT-client ethnicity similarity into three ANCOVAs, controlling for baseline scores on the outcome variable, duration of service use and number of progress reviews2. To achieve sufficient analytic power, the small number of clients of'mixed' and 'other' ethnicity (see Table 1) were removed from analyses, and Asian and Black ethnicities were represented by two dummy variables ('Asian vs White and Black', 'Black vs White and Asian')3. Comparison of coefficients for the two variables permits estimation of effects observed between all three ethnicity clusters. To investigate whether the most socially disadvantaged clients derived equal benefit to others, the most deprived 20% of the population was compared with the remaining 80%. Significant effects within the ANCOVA model were explored via comparison of estimated marginal subgroup means (EMM). Predicting BMI and behaviour change. The unique contribution of demographic variables to predicting BMI and behaviour change was assessed using multiple linear regressions to control for other relevant variables. A four-step model was employed, with baseline behaviour values entered at the first step, duration of service use and number of reviews at the second, a duration x number of reviews interaction term at the third, and demographic variables at the fourth. Significant interaction terms were examined using simple slope analyses (Aiken & West, 1991), to model effects of service use duration at the mean, one SD below, and one SD above the mean number of reviews. We report only regression models which significantly improved the explanation of variance offered by the model at a previous step, but were not further improved at subsequent steps. --- RESULTS Unless otherwise stated, all effects were statistically significant at p<unk>.001. The available sample size (N) varies due to missing data. --- Who uses the HTS? Of 4418 clients, 3503 (79%) were female and 660 (17%) were of Asian or Black ethnicities (Table 1). Nearly half (1836 clients; 43.2%) were from the most deprived quintile of the UK population, and a further quarter (1093 clients; 25.7%) were from the second most deprived quintile. Of 3759 clients for whom BMI data were available, 2717 clients (72.3%) were obese and 824 (22.4%) were overweight at baseline 4. 3346 clients (75.7%) set HE goals and 1072 (24.3%) set PA goals. Overweight and obesity prevalence was higher at baseline among clients setting HE goals (95.4%) than clients setting PA goals (88.8%; <unk> 2 [3] = 85.39), but there were negligible other demographic differences between these two groups. The mean duration of HTS attendance was 6.14 months (SD = 4.10; N = 4347), and clients typically had between one and two progress reviews (M = 1.51, SD = 2.07; N = 3298). Clients with PA goals attended the service for longer (M = 6.77 months; SD = 4.49; d =.16) and had more reviews (M = 2.03, SD = 2.28; d =.29) than clients with HE goals (M = 5.98, SD = 3.97 and M = 1.34, SD = 1.97 respectively). --- INSERT TABLE 1 HERE --- Is the sample affected by selection bias? A greater proportion of White (35%) and Asian (30%) clients set goals than did Black clients (25%) (<unk> 2 [2] = 97.15). More goals were set in the least deprived quintile than in the rest (42% vs 35-38%; <unk> 2 [4] = 26.62). Of clients who set HE or PA goals, a greater proportion of White clients (35%) had pre-post data than did Asian (30%) or Black clients (27%) (<unk> 2 [2] = 27.17). Pre-post data were available for 45% of clients in the least deprived quintile, but for only 29% in the most deprived group (<unk> 2 [4] = 4 Overweight and obesity prevalence (94.7%) was higher at baseline among clients setting HE or PA goals than those with goals unrelated to HE or PA (63.4%; <unk> 2 [3] = 265.04). 149.21). Clients with pre-post data had a higher BMI (33.36) than clients without these data (28.69) (t [945.11] = 18.99). --- Do behaviours and cognitions change following use of the HTS? Among all clients, BMI reduced by 1.77 points, from 34.03 to 32.26 (N = 3759; d =.28) (Table 2). Overweight/obesity prevalence decreased from 94.7% to 91.0%. The proportion of obese clients decreased from 72.3% to 60.1%. Self-efficacy, perceived general health and wellbeing all increased markedly (d =.47,.94, and.92 respectively). Among clients with HE goals, BMI reduced by 1.88 points, from 34.33 to 32.45 (5.5% decrease; N = 3164; d =.30), overweight/obesity prevalence decreased from 95.6% to 92.3%, and obesity from 74.3% to 61.0%. Clients with HE goals reported a 41% increase in F&V intake, from 3.08 to 5.23 daily portions (d =.97). The proportion of clients consuming five or more portions of F&V increased from 24.8% to 60.5%. Mean fried snack consumption decreased by 60%, from 1.99 to 0.79 servings per day (d =.85). There were considerable increases in self-efficacy (d =.46), perceived general health (d =.95) and wellbeing (d =.85). Among clients with PA goals, BMI reduced by 1.24 points, from 32.46 to 31.24 (3.8% decrease; N = 595; d =.18), overweight/obesity prevalence decreased from 89.7% to 84.2%, and obesity from 61.7% to 55.6%. Clients with PA goals reported a 63% increase in number of intensive PA sessions per week, from 0.63 to 1.71 (d =.46), and the proportion of clients engaging in 5 or more intensive sessions weekly increased from 3.9% to 11.6%. The number of moderate PA sessions per week increased by 36% from 3.06 to 4.77 (d =.42) and the proportion undertaking 5 or more moderate sessions weekly increased from 24.4% to 45.5%. There were considerable increases in self-efficacy (d =.53), perceived general health (d =.88) and wellbeing (d = 1.06). --- INSERT TABLE 2 HERE --- Do all clients experience behaviour changes equally? Ethnicity. There were no differences in BMI change between ethnic groups for the whole sample. Among clients with HE goals, Asian clients achieved less BMI change than did White clients (0.55 BMI points; F[1,3050] = 21.62, Eta 2 =.007; p =.006). BMI change among Black clients did not differ from Asian or White clients. Ethnicity was not associated with F&V, nor with fried snack consumption. Among clients with PA goals, ethnicity was not associated with BMI, nor behaviour. Deprivation. Clients from the least deprived 80% of the population achieved more BMI change than did the most deprived clients (0.28 BMI points; F[1,3572] = 12.13, Eta 2 =.003; p=.001). Among clients with HE goals, less deprived clients achieved more BMI reduction (0.24 BMI points; F[1,3010] =7.77, Eta 2 =.003, p=.005). Deprivation was not associated with F&V nor fried snack consumption changes. Among clients with PA goals, deprivation was not associated with BMI, nor behaviour change. HT-client ethnicity similarity. Clients of similar ethnicity to HTs achieved greater BMI reduction than did those who were not (0.28 BMI points; F[1,3708] = 10.14, Eta 2 =.003; p=.001). Among clients with HE goals, those of similar ethnicity to HTs achieved higher BMI reductions (0.31 BMI points; F[1,3123] = 10.75; Eta 2 =.003; p=.001), increases in F&V (0.33 more portions; F[1,2328] = 13.80; Eta 2 =.004), and decreases in fried snack consumption (0.13 fewer snacks; F[1,1834] = 11.52; Eta 2 =.001; p=.001). Among clients with PA goals, ethnicity similarity was not associated with BMI nor intensive PA, but those of similar ethnicity to HTs showed greater increases in moderate PA (0.70 more sessions; F[1,894] = 10.05; Eta 2 =.007; p=.002). --- Predicting behaviour change Baseline BMI (<unk> =.30), duration of contact (<unk> =.08), client ethnicity (Asian; <unk> = -.06), and an interaction between duration and number of reviews (<unk> = -.05) predicted 11% of variance in BMI change (Table 3). Longer duration of service use had greater positive impact on BMI change among clients who had fewer reviews (<unk> =.12) relative to those who had more reviews (<unk> =.08). Clients with higher baseline BMI and of non-Asian ethnicities experienced greater BMI change. Among clients with HE goals, baseline BMI (<unk> =. Among clients with PA goals, baseline scores predicted 15% of the variance of BMI, and 21% and 40% of the variance of intensive and moderate activity, respectively. There was greater change among clients with higher baseline BMIs and fewer baseline intensive and moderate activity sessions (<unk>s =.38, -.64 and -.45, respectively). --- INSERT TABLE 3 HERE In sum, health behaviour change was consistently predicted by baseline scores, such that those clients with most room for improvement achieved greater change. --- BMI reduction among clients with HE goals was also associated with ethnicity, with Asian clients experiencing less change. Clients who attended fewer progress reviews benefitted more from longer service use duration. --- DISCUSSION This study suggests that the HTS is achieving its aim of promoting health via behaviour change among disadvantaged groups. The HTS successfully targeted disadvantaged clients: nearly half were from the most economically disadvantaged 20% of the population, and almost all clients were overweight or obese. Clients experienced changes in targeted behaviour (dietary consumption or physical activity), BMI, self-efficacy, perceived health and wellbeing. Caution is however needed in interpreting these results, in light of inconsistencies between HTs and services in intervention delivery and data collection methods, and the absence of a control group. Greater change was observed among clients with higher BMI, poorer diet or lower levels of physical activity prior to using the service. Clients from the most deprived quintile and of Asian ethnicities were less likely to set behaviour change goals and achieved less BMI change than others. It is unclear whether this reflects lower motivation to change, restricted opportunities for healthy eating or physical activity, lower self-efficacy, and/or poorer quality HTSs serving these clients. Nonetheless, the HTS appears to be reaching and engaging these hard-to-reach clients in behaviour change attempts, which many health interventions have failed to do (e.g. Bernal & Sharron-del-Rio, 2001). More evidence is needed regarding the strategies most likely to translate this engagement into BMI change for those from the most deprived communities, or those of Asian ethnicity. A framework is available which specifies discrete psychological domains that underpin and explain the behaviour change process (e.g. knowledge, skills, environmental context and resources; Michie, Johnston, Abraham, Lawton, Parker, & Walker, 2005). Applying this framework within qualitative interview or quantitative survey designs may help to pinpoint reasons for lack of change among some client groups. There were significant gains in the number of clients meeting UK Government HE and PA recommendations. Such behaviour changes among disadvantaged groups have the potential to reduce social inequalities in health, overweight and obesity (Michie, Jochelson et al, 2009). BMI reduction of over 5%, achieved amongst clients with HE goals, is considered to significantly lower the risk of cardiovascular disease amongst obese people (Blackburn, 1995). Baseline scores reliably predicted greater change in BMI and behaviour, indicating that clients with poorer diet or who were less physically active, and so most likely to benefit from the service, typically achieved greatest change. This could however reflect methodological artifice, such as regression to the mean, whereby atypical baseline values naturally move closer to true scores at follow-up, rather than specific intervention effects. Clients also reported enhanced self-efficacy following use of the HTS. This can boost motivation, so increasing the likelihood of clients pursuing further health targets after disengaging from the HTS (Bandura, 1997), thereby sustaining HTS impact over time. Longitudinal research, with objective behaviour measures and procedures in place to ensure high levels of data collection and retention, is needed to establish whether changes in self-efficacy precede behaviour change, whether behaviour change leads to increased self-efficacy, or whether there is a more complex relationship between HTS use, self-efficacy and behaviour change. Although HTs received training in evidence-based behaviour change techniques, including a bespoke handbook (Michie, Rumsey et al, 2008), no data were available to assess the extent to which these techniques were used in practice (i.e. intervention fidelity), and whether this was associated with outcome. Tape-recording a sample of HT-client consultations would permit coding of the behaviour change techniques used in practice (Hardeman, Michie, Fanshawe, Prevost, McLoughlin & Kinmonth, 2008). This would facilitate not only investigation of fidelity, but also of process-outcome associations, and whether responses to change attempts differ across client groups. Collecting a sufficiently large sample of HT-client consultations would be time-consuming (Hardeman et al, 2008), but such work is necessary to identify mechanisms underpinning observed change and to build evidence to inform future interventions. Since data were obtained mostly via client self-report and there was no control group, the magnitude of observed changes and causality must be interpreted with caution. Self-report may overestimate behaviour change due to influences such as social desirability and cognitive dissonance (Chaudhury & Esliger, 2009;Festinger, 1957;Paulhus, 1986). Self-reports should be complemented by objective measures, such as biochemical verification of HE and PA and measuring weight in the clinic. Our data used a pre-post design; a more rigorous research design could establish whether observed associations can be reliably attributed to the HTS. We echo calls for systematic investment in large-scale prospective assessments of population-level interventions (Boerma & de Zoysa, 2011). Stepped wedge designs should be routinely considered in guiding the rollout of national intervention programmes, because they involve staggered implementation across constituent regions in a random order and so permit comparison between intervention and 'waiting-list' regions (Craig, Dieppe, Macintyre, Michie, Nazareth & Petticrew, 2008). Behaviour change will only impact on public health when change is maintained in the long-term. Extended follow-up data are needed to assess the stability of observed changes. We sought to capture a national picture of behaviour change among HTS users over a 12-month period. Yet, comparison with other national data sources suggests that our data captured the activity of only 62% of HTSs in this period, and just 16% of first-time clients recorded as accessing the service within these HTSs were eligible for analysis. The database that we drew upon is the only available source of cognition and behaviour data, and so we could not assess the generalizability of the behaviour or beliefs of clients recorded on the database to HTSs more widely. The quantity of missing data attests to the challenges faced in implementing and evaluating a national behaviour change service within the NHS. We have no reason to suspect that clients or HTs were unwilling to provide data. Rather, missing data problems may stem from organisational variations in the history of each service, levels of support from local Primary Care Trusts, and the commitment of local HTS managers to collecting and recording behaviour change and related data within a centralised and standardised database. Although use of the standardised central database was optional in 2008-09, HTSs are now required by the DH to log relevant data in this database, and so missing data problems are likely to be reduced for any future analyses. Even where data were collected, there were differences between HTs and services in the methods used to assess key outcome variables (behaviour and BMI); discussions between the BPS consultancy team and regional HT managers suggested that, even within services, there was variation in whether height and weight were measured objectively, or via self-report. Systematic records of measurement methods were not kept, and so we cannot assess the impact of assessment modes on our findings. Additionally, no data were available regarding whether clients pursued both HE and PA goals, and the potential inclusion of clients who modified both energy intake and expenditure may have resulted in greater observed reductions in BMI than can be attributed to either HE or PA change in isolation. Reducing problems of missing or inconsistent data is likely to depend on both raising awareness among service managers and HTs of the importance of robust behavioural data for intervention evaluation purposes, and incentivising data collection. Notwithstanding these limitations, this study provides the first investigation of health behaviour changes following a nationwide behavioural intervention designed for socially disadvantaged groups based closely on health psychology theory and evidence. The results suggest that the HTS has the potential to improve health among disadvantaged groups, an approach which may help to reduce health inequalities. Replication of these findings is needed, using more objective measures and a more rigorous study design.

Our thanks to Fiona Adshead, who devised the Health Trainer scheme as part of Choosing Health whilst Deputy Chief Medical Officer and offered helpful comments on a draft of the manuscript, and to Raj Bhopal, Aziz Sheikh, and Falko Sniehotta for helpful comments on an earlier draft of the manuscript. We also thank Janet Andelin,

In the past few decades, premarital sex, extramarital sex, and homosexuality have become increasingly visible, if not more widespread, in China. This has led many scholars to claim that a national "sex liberation" movement is under way (Farrer, 2002;Farrer & Sun, 2003), and that China's "sexual revolution" has reached "a point of no return" (Burger, 2012). Indeed, according to a report on sexual health in China, the proportion of Chinese who have had premarital sex increased from 15% in 1989 to 40% in 1994 and 71% in 2012 (Ruan, 2013). Analyzing data from China's 2006 Sexuality Survey, Zhang, Parish, Huang, and Pan (2012) reported that 4.5% of women and 16.5% of men have had extramarital sex. Estimates indicate that approximately 3% to 5% of the Chinese population are homosexual, and that around 11% of unmarried men and 5.8% of married men have had sex with another man (Neilands, Steward, & Choi, 2008). The emergence of "pride" campaigns has also made homosexuality more visible (Cao & Lu, 2014). Ample empirical evidence has been provided of changes in the practice and visibility of nonconventional sexual behavior in China. Nevertheless, it would be problematic to infer ideology from behavior. Due to a lack of upto-date research that directly examines sex ideologies in China, this research was motivated first by the need to identify the current state of sex ideologies in China. Despite much speculation, the macro-level social forces that help to shape China's ideational landscape of sex have yet to be identified. Modernization theorists, such as Goode (1970) and Inglehart and Norris (2003), have predicted a global convergence toward less traditional family and gender values due to socioeconomic development. Scholars focusing on China have attributed the country's changing sexual ideologies to Westernization in the wake of the open-door policy in late 1970s. For example, Parish, Laumann, and Mojola (2007) argued that "a sexual revolution is occurring in China, reflected by the fact that the sexual behavior of the country's post open-door generation differs significantly from that of earlier generations" (p. 730). Other scholars have argued that China's economic reform-its deindustrialization, its privatization, and the development of its service sector-has resulted in more liberal views of sex (Clifford, 2013). It is widely believed that modernization, Westernization, and deindustrialization have not only transformed China's sex ideologies over time but produced an even more conspicuous internal variation in sex ideologies across China's vast geographic span (Parish et al., 2007;Zhang et al., 2012). Accordingly, scholars have found that people from more modernized, Westernized, and deindustrialized provinces hold less traditional attitudes toward patrilineality, filial piety, and conjugality (Cheung & Kwan, 2009;Therborn, 2004;Yan, 2009). Nevertheless, no empirical efforts have yet been made to use nationally representative data to test the extent to which modernization, Westernization and deindustrialization are associated with the differences observed in sex ideologies among Chinese provinces. Because past research has identified an "ideational inconsistency" that distinct domains of sociocultural values may not interact with China's social changes in a unidirectional manner (Hu & Scott, 2014), the need to test explicitly the link among modernization, Westernization, and deindustrialization and inter-province differences in sex ideologies was the second source of motivation for this research. In addition, these theories foreground recent social processes and are thus predicated on the problematic assumption that Chinese provinces share a homogeneous socioeconomic and cultural origin. Although recent social events may have led to the differences observed in sex ideologies across China, a priori geographic differences may also be responsible for divergent social attitudes (Nisbett, 2010). In a recent publication in Science, Talhelm et al. (2014) tested the "rice theory" and reported that the distinction between rice and wheat agriculture explains large-scale sociopsychological differences in people's social attitudes in China. They found that distinct modes of agricultural production require varying levels of cooperation and thus cultivate differentiated levels of social tolerance of behaviors such as divorce. As tolerance of nonconventional sexual behavior is central to "sex liberation" (Farrer & Sun, 2003;Parish et al., 2007;Scott, 1998), it is plausible to attribute China's interprovince differences in sex ideologies to preexisting geographic and sociopsychological differences, as specified by the rice theory (Henrich, 2014;Talhelm et al., 2014). Therefore, the third objective of this research was to assess the association between the rice theory and variations in sex ideologies between Chinese provinces. Why are sex ideologies important? Premarital sex, extra-marital sex, and homosexual sex are increasingly cited in discussion of the risks of teen pregnancy (Wang et al., 2007), sexually transmitted diseases (STDs), and human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) (Neilands et al., 2008). Suffice it to say that social acceptance is closely associated with the ways in which sexual behavior is practiced as well as the degree to which sex education and medical intervention are formalized (Gao, Lu, Shi, Sun, & Cai, 2001;Lou, Zhao, Gao, & Shah, 2006). As noted by Scott (1998), "Attitudes are important indicators of people's tendencies to respond to the opportunities and constraints that are posed by the structural conditions of life" (p. 818). On issues as sensitive as sex, there is likely to be a gulf between attitudes and behavior. However, the importance of understanding sex ideologies does not lie in the ability of ideology to predict behavior. Rather, social attitudes help constitute the normative opinions against which behavior is judged (Farrer & Sun, 2003;Neilands et al., 2008;Tian, Merli, & Qian, 2013). In this research, I drew on nationally representative data from the 2010 China General Social Survey and the 2011 Chinese Statistics Yearbook to investigate the extent to which modernization, Westernization, deindustrialization, and the rice theory are associated with China's inter-province differences in sex ideologies. I first introduce these four major theories in relation to ideologies pertaining to pre-marital, extramarital, and homosexual sex across Chinese provinces. Next, I describe the characteristics, measurement, and analysis of the data. Finally, I report the results and discuss their implications for future research. --- Theorizing Inter-Province Differences in Sex Ideologies Sex ideologies in China have evolved over thousands of years and have been influenced by ancient philosophies as well as contemporary political and sociocultural developments (So & Cheung, 2005). Sex was a taboo subject for centuries in feudal and imperial China. Before the founding of the People's Republic of China (PRC) in 1949, Chinese families were governed by pronatalism (Parish et al., 2007). As sex was coupled with marriage and reproduction, marriage was regarded as "the pole around which sexual culture is organized" (Cara<unk>l, 1995, p. 6), and premarital sex was strictly forbidden (Zhang, Gao, Dong, Tan, & Wu, 2002). At the same time, Confucianism has long underlined the importance of loyalty and chastity. Despite the prevalence of male polygamy in China's feudal and imperial eras, extra-marital sex was considered immoral, particularly for females, and was subject to legal punishment (Farrer & Sun, 2003). The emphasis placed by Taoism on a balance between yin (female) and yang (male) has also had a significant effect on Chinese people's views of homosexuality as an "incompatible" form of sexual behavior (Neilands et al., 2008). In the past few decades, however, premarital sex, extra-marital sex, and homosexual sex have all been shown to be on the rise as a result of social changes in China (Burger, 2012;Higgins, Zheng, Liu, & Sun, 2002;Ruan, 2013). Central to the discourse of a "sexual revolution" is the decoupling of sex from the institution of marriage, as well as the emphasis on individual pleasure over the reproductive function of sex (Yan, 2002). Attempting to identify the social forces that detraditionalize sex ideologies, Bongaarts (1980) highlighted the lower age of puberty and thus the lower age of sexual maturity as responses to socioeconomic development in general and improved nutrition in particular. Meanwhile, modernization theorists have suggested that socioeconomic modernization detraditionalizes values that are core to the formulation of attitudes toward premarital and extramarital sex (Goode, 1970;Inglehart & Norris, 2003). According to Inglehart (1990), attitudes toward homosexuality are closely linked with postmaterialist values that emphasize individual autonomy over and selfexpression of sexual and gender identities. Indeed, socioeconomic modernization is closely associated with the detraditionalization of values pertaining to filial piety, patrilineality, and conjugality in China (Therborn, 2004;Yan, 2009). Therefore, it seems likely that modernization theory may predict that socioeconomic development detraditionalizes sex ideologies. As China's 1978 reform prioritized the economic development of urban centers and coastal areas, the extent of China's modernization varies among provinces. --- H1 (Modernization): People from provinces that are more socioeconomically developed have less traditional attitudes toward premarital sex, extramarital sex, and homosexuality than people from less developed provinces. In addition to socioeconomic modernization, China has under-gone considerable deindustrialization in the past three decades. The early history of the PRC featured a centralized economy composed of heavy industries (Bramall, 2006). In the post-reform era, the mass dissolution of state-owned industries was accompanied by a notable ideological shift from collectivism to individualism (Yan, 2009). The de-anchorage of individuals from collective work units (danwei) fostered a sense of individualism, which has detraditionalized conventional gender roles and enabled the explicit expression of intimacy in public (Yan, 2002). The rise of the (tertiary) service sector in areas such as domestic outsourcing and sexrelated services provided a wide array of alternatives to traditional sexual, marital, and familial configurations (Barthel, 1976;Kongar, 2008). As an emphasis on individual pleasure is core to any sexual liberation, the deindustrialization thesis predicts that the development of the (tertiary) service sector 1 results in increasingly liberal sexual attitudes (Barthel, 1976). As deindustrialization has taken place unevenly across China, the following hypothesis can be derived. --- H2 (Deindustrialization): People from more deindustrialized provinces have less traditional sex ideologies than people from less deindustrialized provinces. Alongside China's economic reform, the open-door policy was instated in the late 1970s, introducing China to the global market after decades of closure to the world. The open-door policy was implemented to boost the country's profile of international trade and commerce. Inevitably, cultural exchange between China and the West has taken place through economic exchange. Based on the widespread assumptions that a dichotomy exists between China and the West, and that the West is more open to sex than China, it has been speculated that the detraditionalization of sex ideologies in China is due to the Westernization of Chinese culture (Farrer, 2002). Like modernization and deindustrialization, the influence of Western culture varies between Chinese provinces, which have been opened up to international trade and commerce to varying degrees. --- H3 (Westernization): People from provinces with greater international trade and commerce have less traditional sex ideologies than people from provinces with less international trade and commerce. Although the proponents of the modernization, deindustrialization, and Westernization theories have attempted to explain the observed geographic variation in sex ideologies in China by focusing on recent social events, the assumption that distinct Chinese provinces share a homogeneous socio-cultural origin is problematic. Indeed, ideological differences may result from a priori differences between regions as well as from recent contingencies (Henrich, 2014;Nisbett, 2010;Talhelm et al., 2014). Based on large-scale sociopsychological experiments, Talhelm et al. (2014) proposed the rice theory, indicating that agricultural mode may have a significant impact on the way people think. The authors found that because growing rice requires more irrigation than growing wheat, people from rice-growing regions demand a higher level of cooperation than their counterparts from wheat-growing regions. It is crucial to note that the need for functional interdependence is substantively different from the ideological collectivism imposed by Chinese traditions. Whereas the former fosters social tolerance of deviant ideologies for the sake of subsistence (Henrich, 2014;Talhelm et al., 2014), ideological collectivism does just the opposite, marginalizing deviant ideations that pose a potential threat to collective ideological aims (Lás<unk>ló, 2013). As a result, Talhelm et al. (2014) found that Chinese people from rice-planting regions held less traditional attitudes toward divorce than their counterparts from wheat-growing regions. As social tolerance is also crucial to social attitudes toward nonconventional sexual behavior, the following hypothesis can be derived: 2 H4 (Rice Theory): People from rice-growing provinces have less traditional sex ideologies than people from wheat-growing provinces. University of Science and Technology, the CGSS is one of China's most prominent nationally representative social surveys. In the 2010 CGSS, multistage, stratified, probabilityproportional-to-size sampling was used to survey 12,000 individuals randomly selected from their respective households from 480 street areas in 125 cities, towns, and villages across mainland China. The response rate to the 2010 CGSS was 73.2%. The survey had a wide coverage, ranging from major cities to remote villages, and from the eastern coast to the western hinterland. To date, the 2010 CGSS is the most up-to-date and nationally representative source of information on Chinese people's sex ideologies. --- Method --- Data and Sample To construct the analytical sample, I excluded 437 respondents who provided no valid information for key variables (3.6% of the whole 2010 CGSS sample). Statistical tests were conducted, and no significant correlations between missing information and major demographic characteristics such as age and gender were found. The final analytical sample contained 11,563 randomly selected individuals from 30 Chinese provinces and municipalities. 3 --- Measures --- Dependent Variable Sex Ideologies. In the 2010 CGSS, respondents were asked the following questions in a selfcompletion module: "What is your general opinion on a man or a woman having sexual relations before marriage?"; "What is your general opinion on a married man/woman's having sexual relations with a woman/man other than his/her spouse?"; and "What is your general opinion on sexual relations between two adults of the same sex?" The answers for each question were recorded on a 5-point Likert-scale ranging from 1 (Always wrong) to 5 (Always right). As the responses were highly left-skewed, I recoded the responses to each of the three questions into a binomial categorical measure, combining 1 (Always wrong) and 2 (Mostly wrong) into the first category, "disapproval," and combining 3 (Neither right nor wrong), 4 (Sometimes right), and 5 (Always right) into the second category, "approval." Alternative analyses were conducted using the skewed 5-point scales for sex ideologies, and the results were consistent with those using the categorical measures reported in this article. --- Province-Level Predictors As presented in Table 1, a wide variety of indicators were used to measure the levels of modernization, deindustrialization, and Westernization and agricultural mode at province level in China. As the province-level indicators listed in Table 1 may correlate with one another, exploratory factor analysis was conducted to extract distinct factors from the measures listed in Table 1. As presented in Table 2, four distinct factors were identified based on the number of the eigenvalues exceeding 1. These factors were able to explain a majority of the variance observed in the province-level data. Next, Bartlett factor scores were calculated to yield four scales for the four province-level factors. (See the appendix for detailed factor scores for each province.) [Insert Table 1 andTable 2 Here] Modernization. According to previous research (Inglehart, 1990;Talhelm et al., 2014), the degree of modernization is associated with socioeconomic development, measured by indicators such as per capita gross domestic product (GDP) and level of income. As urbanization is a key part of China's state-guided modernization program (Tian et al., 2013), I also included a measure of the percentage of the population residing in urban areas in each province. As shown in Table 2, the modernization factor accounted for 34.63% of the variance in the original variables. Highly modernized provinces are characterized by a large urban as opposed to rural population, a high level of individual income, and a high GDP per capita. Deindustrialization. As previous research indicates that the development of the (tertiary) service sector is associated with sociocultural change in Chinese society (Barthel, 1976), I included measures of the proportion of GDP in each province contributed by the industrial sector and the tertiary sector, respectively. The deindustrialization factor accounted for 17.64% of the variance in the original variables. In a more deindustrialized province, the tertiary sector contributes a higher share while the industrial sector contributes a lower share to the GDP than in a less deindustrialized province. Westernization. China's exposure to Western cultural influences can be attributed mainly to the international trade and commerce enabled by the 1978 open-door policy. As the degree of Westernization is closely associated with the volume of international trade and commerce, I included measures of the gross value of international imports, international exports, and foreign direct investment (FDI) in each province. The Westernization factor accounted for 25.49% of the variance in the original variables. A more Westernized province has higher levels of importation, exportation, and FDI than a less Westernized province. Mode of Agriculture. The distinction between rice and wheat agriculture in China is closely related to regional variation in thinking patterns and social tolerance. The percentages of farmland dedicated to rice paddy and wheat, respectively, were used to determine the major mode of agriculture in each province. The agricultural-mode factor accounted for 15.06% of the variance in the original variables. A higher factor score indicates a larger share of farmland dedicated to rice as opposed to wheat. --- Individual-Level Covariates Although macro-level factors are the focus of this research, sex ideologies are also subject to a variety of individual-level influences. Therefore, it is important to control for individuallevel factors to assess the four hypotheses. Table 3 presents the descriptive statistics of the individual-level factors. [Insert Table 3 Here] Birth Cohort. As cohort shift is a major mechanism of ideational change (Scott, 1998) Gender. As men and women in China have distinct roles and different levels of access to power (Therborn, 2004), they are likely to relate to premarital sex, extramarital sex, and homosexuality in different ways. A dummy variable was used to indicate the respondents' gender; 51.9% of the respondents were female. Ethnicity. As China's numerous ethnic minorities may have different cultural customs that affect their sex ideologies, I used a dummy variable to control for the respondents' ethnicity; in other words, whether they belonged to the majority Han ethnic group or an ethnic minority group (8.7% of the respondents). Level of Education. I also included a categorical variable to control for the respondents' level of education, distinguishing between No education (13.6%), Primary school (22.0%), Middle school (29.6%), and High school and above (34.8%). Communist Party Membership. As the Chinese Communist Party (CCP) plays an important role in enforcing sex morality, I included a dummy variable indicating whether the respondents were or were not CCP members. Around 17.3% of the respondents were CCP members. Rural or Urban Residence. Given China's entrenched rural-urban division in sociocultural values, I controlled for the respondents' residential location. Areas under the jurisdiction of an urban neighborhood committee (juweihui) were designated as urban, and those under the jurisdiction of a rural village committee (cunweihui) were designated as rural; 61.1% of the respondents lived in urban areas at the time of survey. Migration Status. As China's internal migration exerts considerable impacts on people's sociocultural values, I created a dummy variable indicating whether each respondent was a rural-to-urban migrant by determining whether their hukou status (as conferred by China's household registration system, which distinguishes between rural and urban hukou) matched their residential location. Around 15% of the respondents were rural-to-urban migrants. Socioeconomic Status. Socioeconomic status was measured using a quartile ranking of annual household income. Missing responses to questions on household income were coded as an additional category separate to the quartile ranking. Marital Status. The respondents' marital status was measured using a categorical variable distinguishing between Never married (9.7%), Currently married (80.2%), and Previously married (10.1%). Widowed and divorced respondents were included in the latter category due to the small number of responses for each. --- Modeling Strategy As the data were hierarchically structured and the main purpose of the research was to explore province-level variance in sex ideologies, and the dependent variables were binomial categorical (Snijders & Bosker, 2011), binomial multilevel random-intercept logistic regression models were used. As there were only 30 provinces for the higher-level units, I conducted Bayesian Markov-Chain-Monte-Carlo (MCMC) estimation, which uses the actual posterior distributions of model parameters to construct credible intervals without relying on a hypothetical sample distribution posited by conventional estimates based on maximum likelihood (Stegmueller, 2013). Separate models were fitted for ideologies pertaining to premarital sex, extramarital sex, and homosexuality. In each set of models, Model A comprised the fixed effects of individuallevel variables and the random variance parameter at the province level. Model B also included the four province-level factors, namely modernization, deindustrialization, Westernization, and agricultural mode. Comparing the province-level random variance parameters in Model A and Model B allows one to estimate the extent to which provincelevel variance is explained by the four province-level factors (Snijders & Bosker, 2011). To estimate the explanatory power of each of these factors, I calculated the partial effect size of each province-level factor K by comparing the difference in the province-level random variance parameters between full models with and without K. The variance inflation factor (VIF) tests were conducted to ensure that there was no considerable multiple collinearity among the four province-level factors; the results showed that the VIF values for the four factors were low (i.e., around 1 for each factor, which are well below the 10 threshold), indicating an extremely low degree of multiple collinearity among the four factors (O'Brien, 2007). --- Results --- Descriptive Results Figure 1 presents the levels of modernization, deindustrialization, and Westernization and differences in agricultural mode across Chinese provinces. To help visualize the data, the factor scores were ranked as quintiles, with darker colors indicating higher quintile ranks. The results are consistent with the findings of past research showing that China's eastern and coastal provinces are more socioeconomically modernized than its western and hinterland provinces. This can be attributed to the fact that China's 1978 economic-reform policy prioritized economic development in eastern coastal areas, as vividly expressed in a quotation from China's former leader Deng Xiaoping: "Let part of the people become rich first." Compared with southern provinces, provinces around China's capital city, Beijing, such as Shanxi and Shandong, remain highly industrial. The results indicate that provinces in central China are least Westernized. Unsurprisingly, given their location on China's borders with other Central and South Asian countries, western and southern provinces such as Xinjiang and Yunnan experience a larger volume of international trade and commerce than central provinces such as Guizhou and Hunan. Whereas rice-growing provinces are concentrated in southern and eastern China, most of the provinces north of the Yellow River grow wheatwith the exception of Liaoning Province in northern China. [Insert Figure 1 and Figure 2 Here] Figure 2 presents the inter-province differences in sex ideologies in China. The percentages of approving attitudes toward premarital, extramarital sex and homosexuality were ranked in quintiles, with darker colors indicating higher rates of approval of each of the three types of sexual behavior (see Appendix for detailed statistics). For premarital sex, the rate of "approval" at the province level ranged from 7.0% (in Xinjiang Province) to 53.4% (in Guangdong Province), with a national mean of 27.4% and a median of 28.1%. As shown in the first panel of Figure 2, people from China's southeastern and coastal provinces were found to hold more liberal attitudes toward premarital sex than those from northwestern provinces in the hinterland. Chinese people were found to hold more traditional attitudes toward extramarital sex than toward premarital sex. On balance, only 8% of Chinese people approved of extramarital sex (with a national median of 6.5%). In the most liberal provinces, such as Hainan Province, 15.1% of people approved of extramarital sex, whereas almost no one in Ningxia Province approved of extramarital sex. The second panel of Figure 2 shows that, similar to the pattern for premarital sex, people from China's coastal and southern provinces held less traditional views on extramarital sex than their counterparts in the hinterland and northern provinces. However, it is worth noting that people from some provinces in central China, such as Sichuan and Guizhou, held considerably less traditional views on extramarital sex than people from other hinterland provinces. Chinese people were found to hold more traditional attitudes toward homosexuality than toward premarital sex, but somewhat less traditional attitudes toward homosexuality than toward extramarital sex. On balance, only 12.2% of Chinese people approved of homosexuality, with a national median of 11.7%. Similar to the pattern for premarital sex, people from Guangdong Province held the most liberal attitudes to homosexuality, with an approval rate of 24.1% at the province level, whereas only 2% of people from Xinjiang Province approved of homosexuality. Notably, attitudes were found to vary across the three domains of sexual behavior. This "ideational inconsistency" between attitudes toward premarital, extramarital sex, and homosexuality supports Widmer, Treas, and Newcomb's (1998) recommendation that researchers treat different sexual practices in their own contexts, as the explanations for changing sexual attitudes may well differ between types of sexual behavior. This recommendation is particularly important to the current research, as modernization, deindustrialization, Westernization, and agricultural mode cannot be assumed to shape attitudes toward premarital sex, extramarital sex, and homosexuality in the same ways or to the same degrees. --- Multilevel Random-Intercept Models Table 4 presents the results for the binomial multilevel random-intercept logistic regression models for attitudes toward each of the three types of sexual behavior examined in the research. As province-level factors are the main focus of the article, I first report the province-level results, followed by a summary of the individual-level results. [Insert Table 4 Here] Province-Level Results. The results confirm that attitudes toward premarital sex, extramarital sex, and homosexuality vary considerably among Chinese provinces. In terms of attitudes toward each of the three types of sexual behavior, comparing Model A and Model B reveals that the inclusion of the four province-level factors significantly reduced province-level variance (i.e., the parameter in the random part of the multilevel models). This indicates that the four factors contributed to China's inter-province differences in sex ideologies. Figure 3 more intuitively depicts the percentages of province-level variance in each of the three sex ideologies explained by the four factors for modernization, deindustrialization, Westernization, and agricultural mode, respectively. Hypothesis 1, which states that socioeconomic modernization predicts less traditional sex ideologies, was unsupported by the results. The findings suggest that modernization is not significantly associated with people's attitudes toward premarital sex, extramarital sex, and homosexuality at province level. As shown in Figure 3, modernization explained only 1.3%, 4.4%, and 0.7% of the province-level variance in attitudes to premarital sex, extramarital sex, and homosexuality, respectively. Hypothesis 2, which states that deindustrialization is positively associated with liberal sex ideologies, was partly sup-ported. People from provinces with a higher level of development of the tertiary service sector were more likely to approve of premarital sex (p <unk>.05) and extramarital sex (p <unk>.01) than those from provinces with a lower level of deindustrialization. Meanwhile, deindustrialization explained 8.8% and 27.8% of the province-level variance for premarital and extramarital sex, respectively. Nevertheless, the association between deindustrialization and attitudes toward homo-sexuality was not statistically significant. The level of deindustrialization explained only 0.9% of the province-level variance in attitudes to homosexuality. The results only partly support Hypothesis 3, which states that a higher level of Westernization is associated with less traditional sex ideologies. On one hand, people from provinces with a higher level of Westernization (as reflected by a larger volume of international trade and commerce) were more likely to approve of premarital sex (p <unk>.05) than those from provinces that are less Westernized. As shown in Figure 3, Westernization explained 8.9% of province-level variance in attitudes toward premarital sex. On the other hand, however, Westernization was not found to be significantly associated with attitudes toward either extramarital sex or homosexuality; it explained only 4.7% and 7.2% of the province-level variance in attitudes toward extramarital sex and homosexuality, respectively. Hypothesis 4, the rice theory, was consistently supported by the results, as the distinction between rice and wheat agriculture was significantly associated with ideologies pertaining to premarital sex, extramarital sex, and homosexuality. The results confirm that people from provinces where a larger share of farmland is devoted to rice as opposed to wheat were more likely to approve of all three types of sexual behavior. Rice theory explained 14.1%, 15.3%, and 29.4% of the province-level variance in attitudes toward premarital sex, extramarital sex, and homosexuality, respectively. I discuss the implications of these results in the Discussion section. Individual-Level Results. The multivariate results also revealed significant associations between various individual-level factors and sex ideologies. Compared with men, women in China were less likely to approve of premarital and extramarital sex. This is not surprising, because the traditional Chinese values of chastity and fidelity are significantly gendered; they tend to regulate the sexual behavior of women rather than men (Burger, 2012;Ruan, 2013). Meanwhile, traditional ideals of femininity and masculinity in China position women as more "vulnerable" than men in relation to nonconventional sexual behavior (Farrer & Sun, 2003). Not until the founding of the PRC did China's Marriage Law entitle women to a larger share of property in divorce disputes when their husbands' extramarital sex was the cause of divorce. Consistent with the findings of studies conducted in the West (e.g., Scott, 1998;Widmer et al., 1998), a significant cohort shift has been observed in attitudes toward sex in China: Younger as opposed to older cohorts have consistently been found to be more likely to approve of premarital sex, extramarital sex, and homosexuality. The results-that bettereducated Chinese were less likely to approve of all three types of sexual behavior than those who received no or little education-are different from findings from the United Kingdom and the United States (Scott, 1998). This may be due to the great emphasis placed on moral education in Chinese academic syllabi, which fosters a sense of strict sexual morality among those who are highly educated (Zhang, Li, & Shah, 2007). Similarly, given the role played by the CCP in enforcing sex-related morality, it is unsurprising that CCP members were less likely to approve of extramarital sex than non-CCP members. The results also indicate that urban Chinese were more likely to approve of premarital sex, extramarital sex, and homosexuality than their rural counterparts, and that married people were less likely to approve of all three sex behaviors than never-married respondents. --- Conclusions and Discussion To date, nationally representative research on sex ideologies in China has been scarce. Most previous researchers have focused either on specific segments of the Chinese population, such as university students (Ruan, 2013;Zhang et al., 2002), or on specific places, such as Shanghai (Farrer, 2002;Farrer & Sun, 2003) or rural villages (Yan, 2002). The analysis of data from the 2010 CGSS and the 2011 Chinese Statistics Yearbook clearly revealed that most Chinese people disapprove of extra-marital sex and homosexuality. Although attitudes toward pre-marital sex were somewhat less traditional, as many as 71% of Chinese people disapproved of premarital sex. The clear evidence of considerable internal variation in sex ideologies across China also calls into question the generalization that a national "sexual liberation" is under way in China, which is largely based on behavioral observations (Parish et al., 2007;Zhang et al., 2012). Nevertheless, the importance of studying sex ideologies does not lie in the accuracy with which we can infer behavior from ideology. Ideologies not only guide behavior but also provide the moral touchstone against which premarital

In recent decades, premarital sex, extramarital sex, and homosexuality have become increasingly visible in China, leading scholars to claim that a national "sex revolution" is under way. However, China's internal sociocultural diversity calls this nation-level generalization into question. How do sex ideologies vary across China's distinct provinces? To what extent are inter-province variations in sex ideologies associated with distinct macrolevel social factors in China? In this research, data from the 2010 China General Social Survey and the 2011 Chinese Statistics Yearbook were analyzed using multilevel models to test four contending theories of inter-province differences in sex ideologies in China: modernization, Westernization, deindustrialization, and the "rice theory." The modernization theory was unsupported by the results, as socioeconomic development is not significantly associated with sex ideologies. Higher levels of deindustrialization and Westernization were associated with less traditional sex ideologies, but the strength of association varied across the domains of premarital sex, extramarital sex, and homosexuality. The rice theory was consistently supported, as the distinction between rice and wheat agriculture explained up to 30% of the province-level variance in sex ideologies. The findings underline the roles of both long-standing geographic differences and recent social changes in shaping China's ideational landscape of sex.

previous researchers have focused either on specific segments of the Chinese population, such as university students (Ruan, 2013;Zhang et al., 2002), or on specific places, such as Shanghai (Farrer, 2002;Farrer & Sun, 2003) or rural villages (Yan, 2002). The analysis of data from the 2010 CGSS and the 2011 Chinese Statistics Yearbook clearly revealed that most Chinese people disapprove of extra-marital sex and homosexuality. Although attitudes toward pre-marital sex were somewhat less traditional, as many as 71% of Chinese people disapproved of premarital sex. The clear evidence of considerable internal variation in sex ideologies across China also calls into question the generalization that a national "sexual liberation" is under way in China, which is largely based on behavioral observations (Parish et al., 2007;Zhang et al., 2012). Nevertheless, the importance of studying sex ideologies does not lie in the accuracy with which we can infer behavior from ideology. Ideologies not only guide behavior but also provide the moral touchstone against which premarital sex, extramarital sex, and homosexuality are judged. China's inter-province differences in sex ideologies have often been cited as evidence of the efficacy of China's social changes in recent decades (e.g., Siu-Lun, 1986). This is one of the first studies to systematically explore the macro-level social factors associated with China's inter-province differences in sex ideologies. The results emphasize the importance of both recent social, cultural, and political events and long-standing regional traditions in helping shape China's landscape of sex ideologies. The assumption that China's provinces are socioculturally homogeneous is called into question. The results also problematize modernization theory, which predicts a global convergence of values pertaining to gender and sexuality (Goode, 1970;Inglehart & Baker, 2000;Inglehart & Norris, 2003). Although it has been cited to explain changing values pertaining to patrilineality, filial piety, and conjugality in China (Siu-Lun, 1986;Yan, 2009), socioeconomic modernization is not significantly associated with sex ideologies. In contrast, deindustrialization was found to be significantly associated with sex ideologies at province level, particularly ideologies relating to extramarital sex. This is understandable, because China's deindustrialization entailed a significant ideational shift from the collectivism of the pre-1978 centralized state-controlled economy to the individualism of the post-1978 market-guided economy. In the socialist era, the collective organization of the danwei ("work unit") system also played a major role in enforcing an emphasis on marital fidelity (Burger, 2012;Farrer, 2002). As sexual dissatisfaction within marriage and the pursuit of individual sexual pleasure provide the major rationale for extramarital sex (Farrer & Sun, 2003), deindustrialization is likely to lead to less traditional attitudes to extramarital sex. Although scholars have generally ascribed China's alleged "sex liberation" to the country's exposure to the West after the 1978 open-door policy (Bond & King, 1985;Parish et al., 2007), the results of this study indicate that the significant association between Westernization and less traditional attitudes toward sex applies to only one type of sexual behavior: premarital sex. A key argument for premarital sex is the decoupling of sex and the institution of marriage. Despite ongoing debate about the "deinstitutionalization" and "privatization" of marriage in Western and Chinese societies alike (Cherlin, 2004;Davis, 2014), it seems that Westernization does little to explain inter-province differences in attitudes toward extramarital sex. Nor are people from provinces that are supposedly more Westernized more likely to approve of homosexuality. An important contribution of this research to the existing scholarship lies in its attempt to explore China's inter-province differences in sex ideologies in terms of a priori geographic differences in addition to recent social trends. The results lend support to rice theory, as the distinction between rice and wheat agriculture was found to be closely associated with people's attitudes toward premarital sex, extramarital sex, and homosexuality. The results concur with Talhelm et al. (2014), that rice plantation and wheat-based agriculture foster differentiated levels of social tolerance of behaviors such as divorce in China, and that this differentiation is long-lasting and applies not only to people who actually engage in agricultural production but to those who are acculturated in regions that traditionally grow rice or wheat. Therefore, this research supports Nisbett's (2010) argument that observed geographic differences in ideologies should not be understood merely as the outcome of divergent social processes. Rather, geographic differences could be responsible for the production of the observed inter-province differences in sex ideologies. Major alternative explanations that might operate in parallel to the rice theory were also tested early in this research. For example, because agricultural mode may also be associated with weather and temperature, the pathogen-prevalence theory suggests that diseases are more likely to spread in warmer areas, and people from warmer areas are less likely to engage in activities that pose potential health risks. The inclusion of annual average temperature for each province did not affect the results for sex ideologies, which may be because STDs are spread in different ways from airborne or other direct-contact diseases. Other forms of subsistence theory distinguish between farming and herding agriculture. I experimented with controlling for whether the major agriculture in a given province is herding or farming. The inclusion of this variable neither had a significant influence on the other parameters in the models nor increased model fit, which may be because only a very small number of provinces in China feature herding agriculture. In this research, cross-sectional data were analyzed, which means the results should be interpreted in terms of association rather than causality. Therefore, it is important to collect com-parable data to examine the over-time change of sex ideologies in China. Efforts should also be made to collect richer longitudinal data to enable researchers to systematically unpack the intricate causal mechanisms of how modernization, deindustrialization, Westernization, and rice theory may operate to transform sex ideologies in China at an individual level. As the CGSS data do not contain information to allow for the test of social desirability, another limitation of this research is that the results may in part reflect the respondents' engagement with perceived social norms pertaining to sexual conduct. Whereas most surveys, such as the CGSS, collect data on either ideology or behavior but not both, it is important to collect data on both aspects to enable future assessment of the consistency and discrepancy between sexual ideologies and behaviors in China. Despite its limitations, this research underlines the importance of examining and understanding China's internal variation in sex ideologies, which is crucial to the development of region-specific policies for sexual health interventions and sex education syllabi. --- Notes 1. Whereas the term tertiary sector is used to refer to the educational and cultural sectors in Western societies, its coverage is wider in China, including all nonagricultural and nonindustrial economies such as education, catering, service provision, and cultural enterprises. 2. Talhelm et al. (2014) confirmed that rice theory applies to border areas between ricegrowing and wheat-growing regions. 3. Tibet, Hong Kong, and Macau were not included as they are not covered by the CGSS. These three regions host less than 1% of the Chinese population.

In recent decades, premarital sex, extramarital sex, and homosexuality have become increasingly visible in China, leading scholars to claim that a national "sex revolution" is under way. However, China's internal sociocultural diversity calls this nation-level generalization into question. How do sex ideologies vary across China's distinct provinces? To what extent are inter-province variations in sex ideologies associated with distinct macrolevel social factors in China? In this research, data from the 2010 China General Social Survey and the 2011 Chinese Statistics Yearbook were analyzed using multilevel models to test four contending theories of inter-province differences in sex ideologies in China: modernization, Westernization, deindustrialization, and the "rice theory." The modernization theory was unsupported by the results, as socioeconomic development is not significantly associated with sex ideologies. Higher levels of deindustrialization and Westernization were associated with less traditional sex ideologies, but the strength of association varied across the domains of premarital sex, extramarital sex, and homosexuality. The rice theory was consistently supported, as the distinction between rice and wheat agriculture explained up to 30% of the province-level variance in sex ideologies. The findings underline the roles of both long-standing geographic differences and recent social changes in shaping China's ideational landscape of sex.

Introduction One interesting phenomenon that emerges from the typical structure of social networks is the friendship paradox [6]. It states that, on average, your friends have more friends than you do. This paradox basically exists because of the discrepancy on node degree values in typical social networks [2], in which individuals with a high number of friends are over-represented when averaging over them [10]. As a consequence, the friendship paradox can dramatically skew an individual's local observation, making such an observation appear far more common than it is in reality [4,14]. In this context, identifying variations of this paradox in different ecosystems has been the topic of some important recent research efforts [5,9,12]. For instance, two new paradoxes have been verified on Twitter [9]: (1) the virality paradox that states that your friends receive more viral content than you do, and (2) the activity paradox that states that your friends post more frequently than you do. More recently, the friendship paradox was generalized to any complex network [5] and its origins are highly correlated with the skewed distribution of node degree (i.e., the number of network friends) [10]. In a nutshell, these efforts suggest that any attribute that is highly correlated with node degree is likely to produce this kind of paradox [5]. Thus, in this article we take the case of scientific collaborations to investigate whether a similar paradox also arises in terms of a researcher's scientific productivity as measured by her H-index. The H-index [8] is a metric originally proposed to measure a researcher's scientific output. Its calculation is quite simple as it is based on the researcher's set of most cited publications and the number of citations they have received. More specifically, a researcher has an H-index h if she has at least h publications that have received at least h citations. Thus, if a researcher has at least 10 publications with at least 10 citations, her H-index is 10. Like any metric that attempts to summarize a complex and subjective evaluation in a single number, the H-index has its limitations, including being biased towards the researchers' scientific lifetime, not accounting for the number of coauthors in the publications and ignoring the distinct citation patterns across different areas [3]. Nevertheless, the H-index became popular as it provides a notion of both quality and quantity of a researcher's scientific output in a simple and easy-to-compute metric. As a consequence, researchers are often tempted to evaluate themselves based on the H-index. Systems like Google Scholar1 and ArnetMiner2 help researchers track their publication impact and coauthors, as well as to maintain their profiles, where the H-index is clearly stamped. Thus, it is natural to assume that researchers may use their coauthors' H-indexes as a way to estimate whether their own H-index is adequate in their respective research areas or within a department or university. Despite recent efforts to generalize the friendship paradox [5], it is still unclear whether a similar paradox actually happens when we consider the Hindex in a coauthorship network. However, we have been able to show that the average H-index of a researcher's coauthors is usually higher than her own H-index. Next, we briefly discuss how we have estimated the H-index for researchers from distinct Computer Science research communities, and then provide empirical results that corroborates the existence of the H-index paradox. --- Estimating H-index In order to provide evidence of the H-index paradox, we need to be able to (1) identify the coauthors of a large set of researchers and (2) estimate the H-index of these researchers as well as of their respective coauthors. We focus on constructing the coauthorship network of Computer Science researchers from different areas. To do that, we gathered data from DBLP3, as it offers its entire database in XML format for download. We gathered this data for those researchers who published in the flagship conferences of 10 major ACM SIGs (Special Interest Groups) 4 : SIGCHI, SIGCOMM, SIGCSE, SIGDOC, SIGGRAPH, SIGIR, SIGKDD, SIGMETRICS, SIGMOD and SIG-PLAN. There are several tools that measure the H-index of researchers, of which Google Scholar is today the most prominent one. However, in order to have a profile in this system, a researcher needs to sign up and explicitly create it. In a preliminary collection of part of the profiles of the DBLP authors, we found that less than 30% of these authors had a profile at Google Scholar [1]. Thus, this strategy would largely reduce our dataset. To overcome this limitation, we used data from the SHINE (Simple HINdex Estimation) project 5 to estimate the researchers' H-index. SHINE provides a website that shows the H-index of almost 1,800 Computer Science conferences. It was created based on a large scale crawl of Google Scholar. Its strategy consisted of searching for the title of all papers published in such conferences, thus effectively estimating their H-index based on the citations computed by Google Scholar. Although SHINE only allows searching for the H-index of conferences, their developers kindly allowed us to use its dataset to infer the Hindex of researchers based on the citations received by their conference papers. However, besides covering only conferences, SHINE does not track all existing conferences in Computer Science, which might cause the researchers' Hindex to be underestimated when computed with this data. To investigate this issue, we compared the H-index of researchers with a profile on Google Scholar with their estimated H-index based on the SHINE data. For this, we randomly selected 10 researchers for each of the ACM SIG's flagship conferences and extracted their H-indexes from their respective Google Scholar profiles. In comparison with the H-index we estimated from SHINE, the Google Scholar values are, on average, 50% higher. Figure 1 shows the scatterplot for the two H-index measures. We can note, however, that although the SHINE H-index is lower, the two measures are highly correlated. The Pearson's correlation coefficient is 0.85, indicating that the H-index estimations are proportional in both systems. Table 1 summarizes the collected data, including the SIG, the conference acronym, the period considered (some conferences had their period reduced to avoid a hiatus in the data), the conference's SHINE H-index and the total number of authors, publications and editions. This dataset is useful to our purposes, since it allows us to investigate the H-index paradox on real Computer Science communities, in which researchers might tend to compare themselves with their peers. --- Comparing the H-index of a Researcher with her Coauthors' Having estimated the H-index of each researcher, we can compare it with her coauthors'. Figure 2 shows the fraction of authors with an H-index that is lower than the average of their coauthors for the 10 conferences we have considered. We note that even focusing on authors that have published in flagship conferences of ACM SIGs, the fraction of authors that are below average is quite high for all research communities analyzed, varying from 69% (POPL) to 81% (SIGDOC). When we look at the percentage of authors with at least one coauthor with a higher H-index than theirs, the numbers are higher than 90% for most of the conferences. These results confirm the H-index paradox since one's coauthors in a research community have, on average, a higher H-index than hers. The reasons behind the H-index paradox might be explained by the high correlation between node degree and H-index in a research community. Usually, high degree nodes tend to be senior researchers that not only advise a large number of students but also establish more collaborations, often with different groups along their career [1]. To further investigate this issue, Figure 3 shows the distribution of the number of authors as a function of the H-index. It clearly resembles a long tail distribution, thus suggesting that some authors disproportionally contribute to the average H-index. This disproportion on the average H-index might be even sharper with the typical structural properties of coauthorship networks, which are similar to many social networks [11,13], i.e., they have a long tail degree distribution, in which highly connected authors create bridges across multiple highly connected components, leading to the properties of high clustering coefficient and short diameter. Finally, we measured the Pearson's coefficient correlation between a researcher's H-index and her degree. Such correlation is 0.36 a value that, although not very high, is positive, thus sug-gesting that a small number of researchers simultaneously have a high H-index and a large number of connections in the network. --- Conclusions In this article we have analyzed a variation of the well-known friendship paradox. By analyzing the average H-index of a researcher's coauthors for different Computer Science research communities, we show that the H-index paradox arises because the H-index is positively correlated with node degree. One of the implications of the friendship paradox is the fact that it leads to systematic biases in our perceptions. Thus, similarly, the H-index paradox induces researchers to feel that they rank below average in comparison with their coauthors. This phenomenon is an instantiation of a sensation that occurs in different scenarios and is popularly captured by an expression that is common to many languages and cultures: the grass is always greener on the other side of the fence [7].

One interesting phenomenon that emerges from the typical structure of social networks is the friendship paradox. It states that your friends have on average more friends than you do. Recent efforts have explored variations of it, with numerous implications for the dynamics of social networks. However, the friendship paradox and its variations consider only the topological structure of the networks and neglect many other characteristics that are correlated with node degree. In this article, we take the case of scientific collaborations to investigate whether a similar paradox also arises in terms of a researcher's scientific productivity as measured by her H-index. The H-index is a widely used metric in academia to capture both the quality and the quantity of a researcher's scientific output. It is likely that a researcher may use her coauthors' H-indexes as a way to infer whether her own H-index is adequate in her research area. Nevertheless, in this article, we show that the average H-index of a researcher's coauthors is usually higher than her own H-index. We present empirical evidence of this paradox and discuss some of its potential consequences.

During this coronavirus disease 2019 (COVID-19) pandemic, the world over, we hear that these are 'unprecedented times' in which nations, communities, families and individuals must dig deep into resources and strengths that they may not realize they have. 1 Social workers are frontliners in the crisis that attend to urgent vulnerabilities-social constructs that inform of inequalities in society. These must be contextualized to be relevant to 'local moral worlds' 2,3 such as those in extreme situations. In the case of persons deprived of liberty (PDLs), a correspondence in this journal 4 can be followed up by explaining further that, after conversing with them through immersion in a certain jail, the condition is worse when they even take turns in going to sleep due to unequal access to the floors and benches-spaces they consider for sleeping. Urgent needs of vulnerable groups are immediately responded to by social workers. This article reports an investigation of the social workers' interventions done in a Philippine province, which is vital due to the different circumstances not only of the vulnerable groups attended to but also to the different public health mandates such as quarantine measures in each province. Apart from some who were not available during the study, the participants of the study are 18 social workers from the 18 municipalities and 1 city in the province of Southern Leyte, who provide key information on their social work interventions since the lockdown last year up to date. Social work interventions done in the province do not focus on the one field of intervention. Instead, social workers carry out many tasks from different major fields of intervention. Before COVID-19, 16 (88.9%) participants worked on elderly people; 15 (83.3%) worked on women, disabilities and gender violence; 6 (33.3%) worked on primary care and justice; 5 (27.8%) worked on health; 4 (22.2%) worked on mental health; 3 (16.7%) worked on homelessness; 2 (11.1%) worked on convicts or ex-convicts and 1 (5.6%) worked on the youth and children. During COVID-19, the interventions are the same with the elderly, women, disabilities, primary care and justice. The notable differences are that interventions on gender violence (77.8%), homelessness (11.1%) and convicts or ex-convicts (5.6%) are reduced, but interventions on health (38.9%), mental health (38.9%) and the youth (11.1%) are raised. There is also the glaring concern of individuals in crises, especially those with medical concerns. Solo parents are emphasized as vulnerable during the crisis as well. Finally, key informants raised the concern about rebel returnees or former rebels who have surrendered during the pandemic. In considering the most vulnerable in the pandemic, the informants highlighted their major fields of intervention, mostly on the elderly or the senior citizens and those with disabilities because, as some participants note, 'they don't have the stamina and stability against the virus unlike the other sectors' and'most of them belong to below food thresholds with low resistances'. What is emphasized in this follow-up even if the category on youth and children has 11.1% in focus is that they are considered next as the most vulnerable, with some claiming that they are on par in status with the elderly because they'most likely can't protect their wellbeing from any harm'. An enduring material that the author attended virtually for continuing education at Harvard Medical School ascertained that the most likely determinant of the number Downloaded from https://academic.oup.com/jpubhealth/article/43/3/e563/6290573 by guest on 10 May 2024 of infected children in a community is socioeconomic status. One participant notes that the most vulnerable are the 'indigent families having an insufficient income, derived only from small scale farming, fishing, being hired laborers, and being a habal-habal (motorcycle taxi) driver'. Another idea is that while children may have fewer angiotensin-converting enzyme 2 (ACE2) receptors for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), they can still spread the virus silently and this depends on many factors, not just on distance, 5 which one of the editors rightly pointed out, but also on the indoor ventilation system. There are also problems in working out their field interventions. Further contextualizing the study on social work and sustainability, 6 the participants share some resonating responses. Sixteen (88.9%) reported on the lack of preparedness of social workers about pandemic consequences and the increase in basic demands (e.g. food, housing, medical attention, moral support, etc.) of vulnerable groups, thirteen (72.2%) reported on the scarcity of personal protective equipment (PPE) for social workers, twelve (66.7%) reported on the problems of changes in the official guidelines and protocols to work in the COVID-19 context, nine (50%) reported on lacking in financial aid and one (5.6%) reported each on the problematizing the closure of services for the homeless and the delay of regional services due to restrictions. There are new sets and/or region-specific interventions. Apart from the lockdown, curfews, isolation facility accreditations and quarantine measures, there is extended financial assistance through the social amelioration program (SAP), the establishment of a municipal interagency task force (MIATF) as 'a special body to impose health protocols and strengthen advocacy campaign for wide information dissemination on COVID-19', QR-coding system and transparent barriers applied in all establishments and for the latter also in public vehicles with limited passengers. Creative proactive coping programs are also introduced such as the Bag-ong Abilidad sa Bag-ong Normal (New Skills in the New Normal) that promote and teach activities like the Intensified Kitchen Garden Project, Bangus (milkfish) production, hog raising and fruit tree growing programs (e.g. jackfruit production). Laptop provisions are also made available to some local government units (LGUs) for online transaction purposes such as work from home and virtual learnings or seminars. Finally, the social workers responding to various vulnerable sectors appeal to improve the COVID-19 interventions in their areas to (i) increase budget allocation for social services and manpower demand to add more workforce (as'some are burned out due to plenty of responsibilities'), (ii) provide more financial aid for vulnerable sectors like the elderly, persons with disabilities and solo parents who cannot work dur-ing the pandemic, (iii) make social preparations to encourage people to accept the vaccine, (iv) heighten people's awareness more by spreading information about following health protocols and (v) strictly implement health protocols. These are key information that presents 'diverse social attention tools and their contribution to the sustainability of social services with a long-term impact'. 6 --- Conflict of interest The author certifies that the study was objectively steered and reports that no financial interests or nonfinancial interests influenced whatsoever the conduct of the study. Any relationship with the participants-personal or professional-is treated incognito to ensure objectivity.

Intending to contextualize vulnerabilities at the local level during COVID-19, this article reports an investigation of the social workers' interventions done in a Philippine province. Social workers as frontliners report on (i) the shift of interventions before and during COVID-19, (ii) the most vulnerable sector in their assigned areas, (iii) the problems in working out their field interventions, (iv) the emergence of creative programs and (v) key information to better improve or to have a more sustainable set of interventions in the long run.

Introduction Creativity as a problem attracts attention of scientists, who study ads, because it demonstrates dynamics of values of consumers. It has always been of interest to specialists from different spheres such as sociologists, psychologists and economists. In recent decades, the experience economy has been researched by many copywriters and has become popular and in demand. The experience economy is aimed at actively using values, emotions, emotional creativity to select a creative idea that can interest consumers. The creative economy is based on the information about values and is viewed by scientists and practitioners as a worldwide reality. "New model of the economy combines culture, society, technology and economyboth at the micro and macro levels. Central to this system is the recognition of creativity, knowledge and access to information. And here we have something new: creativity is finally appreciated. Improving things across the company can now depend on one idea. The profitability of creative contributions can reach incredible levels. Creativity and the ability to implement innovations, or in other words, to apply ideas in a business context, are becoming a major component of the newly formed creative economy of the 21st century" [1, p. 46]. Creative economy is: "1) a concept that evolves and is based on creative resources; 2) can potentially contribute to economic growth and development; 3) has the potential to stimulate income generation, job creation and export earnings, while promoting social integration, cultural diversity and personal development of a human being; 4) covers economic and social aspects together with technology, intellectual property and tourism; 5) is a set of innovative economic processes that involve the aspect of development and have intersectoral connections with the general economy at the macro -and micro levels" [1, p. 47]. --- Literature review J. Howkins notes, that neither creativity nor economy is new to us. The only fact, that deserves attention of scientists and practitioners is the extent of the relationship between creativity and economy and how they combine to create extraordinary value and wealth [2]. R. Florida, an American economist, developed the theory of the creative class and wrote that creative services, such as design, advertising and entertainment, not only add value to each product, but also drive innovation in virtually all industries [3]. The presence of creative abilities, which was previously only a supplement in the modern global economy, becomes decisive for further cultural, economic and social development [1]. The problems of the charismatic leader were the subject of analysis by M. Weber, W. Mommsen, and S. Eisenstadt. The desire to interpret human creativity on the basis of a relationship with one of the three ways of connecting with the world was also proposed by J. Habermas. The idea of expression describes creativity mainly in relation to the subjective world of the acting individual. The idea of production refers creativity to the objective world, the world of material objects, which are the conditions and means of action. Finally, the idea of revolution assumes the possibility of human creativity in relation to the social world [4]. H. Joas criticizes the theory of communicative action of J. Habermas because the characteristics of creativity in this case are limited by the concept of rationality, ie emotional and creative actions are ignored [5]. Thus, the concept of communicative rationality can be considered convincing only when it comes to the theory of rationality. To correct the mistake, H. Joas proposes to add a third model to the dominant models of rational and normative-oriented action, in relation to which we should talk about the creative nature of human action. Moreover, the researcher claims that this third model covers the first two [5], but also takes into account emotional actions. However, despite the active study of the problem of creativity by representatives of different sciences, some aspects of its use, for example, in advertising, are still insufficiently studied, which determines the topicality of the problem. Existing works devoted to the study of creativity in advertising confirm the relevance of the study of this phenomenon and determine the main directions of this study. Thus the work of S Rosengren, M. Eisend, S. Koslow [6] emphasizes the need for empirical study of the mechanism of creativity in advertising through both originality and appropriateness. The works of M. Dahlén, S. Rosengren, F. Törn, T Modig, R Smith, J. Chen, X. Yang [7,8,9] are devoted to the question of the influence of creativity on the effectiveness of advertising. The study of the national features of creativity in advertising were devoted the works of B. N. Kim, S. Han, S. Yoon, D. West, A. Kover, A. Caruana, L. Howe-Walsh, S Turnbull, P. Budhwar [10,11,12]. Further areas of the study of creativity in advertising, such as digital advertising, targeting advertising in social media were presented in the works of D. West, S. Koslow, M. Kslgour, L. P. Wulandari, G. S. Darma [13,14]. The aim of the article is to identify the features of formation of advertising creativity. To achieve the aim, the following tasks were solved: 1) the types of creativity were considered; 2) features of emotional creativity and creative criteria were analyzed; 3) types of creativity in advertising were defined; 4) conditions of functional and dysfunctional creativity were considered; 5) examples of the use of creative advertising tools in Ukrainian and foreign advertising were shown. --- Results E.P. Torrance tried to describe creative thinking as the process of sensing difficulties, problems, gaps in information, missing elements [15] The Torrance test is one of the most famous because it reflects the diversity of the concept of creativity: Fluency takes into account the speed of producing ideas, Flexibility as the ability to put forward various ideas, move from one aspect of the problem to another, uses various strategies of solution, Originality as non-standard thinking and Elaboration as amount of detail in the responses. Creativity as non-standard, the ability to differ from others can be represented in advertising / PR in two main dimensions: creativity of design, contained in visual images and copywriting creativity, contained in a verbal message -slogans, appeals, accompanying text. In our opinion, we could single out a third separate dimensionemotional creativity as the ability of an advertising / PR product to evoke an emotional response in the audience. It is clear that emotions are closely related to both visual images and verbal accompaniment of advertising / PR, but at the same time they are a separate component of influence. Both creativity of design and creativity of copywriting do not guarantee a positive perception of the product and positive consumer actions. The measurement creativity of Flexibility is important for advertisers and PR people, because different target audiences at the expense of the specifics of age, gender, lifestyle, values require a differentiated approach to creating an advertising message. A common mistake of advertisers is to try to make one universal message for everyone, but since the concept is very abstract, the probability of reaching the target audience becomes problematic, which once again emphasizes the importance of consumer segmentation and personal marketing. The personal marketing approach was implemented by Google through the contextual advertising service Google Ad, which not only demonstrates advertising that is relevant to the search query of customers, but also takes into account their demographic, geographical, social characteristics. Measuring creativity of Elaboration can be important for advertisers to create a realistic, recognizable advertising image. However, it should be remembered that this requires experience in monitoring the audience of consumers, to monitor how customer values in relation to the product of advertising change. Like P. Torrance, another well-known researcher of creativity, G. Guilford, linked creativity to divergent thinking, but emphasized that they were not identical. In his opinion, creativity consists of four components [16]: 1) originality; 2) semantic flexibility -the ability to suggest a new use of the object; 3) figurative flexibility -the ability to see new features in the object; 4) spontaneous flexibility -the ability to produce a variety of ideas in unregulated situations. The use of semantic flexibility, figurative flexibility and spontaneous flexibility are quite common methods of producing creative advertising content. For example, inventions such as tea bags in the form of the pillow "Pillow tea bag" or the Mitsubishi Pajero advertisement where instead of comparing this car to another one, this advertisement is comparing this car to a rhinoceros are examples of semantic flexibility that uses familiar things to build familiar associations. A good example of figurative flexibility can be the advertising campaign-winner of the "Cannes Lions" -"See sound", which presents a mobile application combined with a smart watch for deaf people in order to better adapt them to life and create a safer space. The program identifies and transmits appropriate signals regarding loud sounds -a child's cry, a siren signal, etc. An example of spontaneous flexibility in advertising is a commercial where a woman's high-heeled shoes break, she takes a chewing gum, tears off her second heel and continues her journey. Features of emotional creativity deserve a separate analysis. The most famous researcher of emotional creativity is R.J. Averill, who notes that creativity is highly classified in cognitive processes, while emotions haven't been considered cognitive for a long time [17]. According to R. J. Averill emotional creativity should be considered as a consequence of the socio-constructivist view of emotions [17]. That is, emotions are considered as social constructs, because the expression of emotions occurs in accordance with existing social norms, which the individual learns in the process of socialization. Emotional creativity can begin with a change in the rules according to which it is customary to express emotions. However, a non-traditional form of expression is not necessarily considered creative. Therefore, R.J. Averill emphasizes that: emotional creativity may involve the development or modification of standard emotions to better suit the needs of an individual / group or the development of new forms of emotion expression. An important issue in the study of creativity is its criteria. R. J. Averill considers novelty, efficiency, authenticity as criteria of creativity [17]. Moreover, all three of these criteria work only together. That is, novelty that is not effective is not considered creative. There is a question about the effectiveness of advertising creativity. The most wellknown formula for influencing advertising product the AIDA provides answers to questions about the functions of creativity in advertising -to attract attention, create an emotional response, create a need, stimulate to action. AIDA also focuses on the emotional component, emphasizing that without emotion (it can be not only emotions of interest, but any other) the chain of effectiveness of the advertising message will not work. At the same time, it should be noted that at each stage of the impact of advertising, creativity can be both functional and dysfunctional. Creativity may be functional, ie one that contributes to the ultimate goal of advertising -to sell a product / change the model of behavior (for social advertising) under certain conditions. The first and most important condition of a functional creativity is the relevance of the creativity to the advertised product. There are many cases when bright, non-standard advertising attracted attention and was remembered, at the same time consumers could not remember which product was advertised, ie creativity in advertising existed separately and did not contain associations with the advertised product. However, there are also successful advertising messages, when the connection with the advertising product is traced and the stages of formation of the need for the product during the period of primary and secondary socialization are demonstrated. Thus, the combination of firstorder knowledge and second-order knowledge allowed us to trace the etymology of interest or motivation to buy a product and was appreciated as a memory of a happy childhood, successful deals, authoritative people who helped or appreciated consumer actions. The confirmation of this is the advertising of cheese for children, for example. The second condition of functional creativity is conformity with consumer values. The point is that advertising can be creative, it can create associations with the product, but not coincide in value with consumer values. Value Mismatch is caused by a lack of understanding of the needs and interests of consumers. It can manifest itself as a generation gap (age gap), gender gap (mismatch with models of gender identities in advertising), behavioral gap (instead of praising the consumer those who are not consumers are humiliated), emotional gap (mismatch of emotional response to the needs of the audience). The emotional gap can be observed, for example, in the use of negative emotions in social advertising: an attempt to intimidate consumers to make them fasten belts in the car, instead of creating a positive emotional response to a positive pattern of behavior. As a result, dysfunctional creativity is formed. By dysfunctional creativity we mean one that does not contribute to the goals of advertising as communication and business indicators: not relevant to the product, does not influence the purchase of the product, does not form loyalty (or even offends, repulses consumers), etc. The third condition of functional creativity is the heterogeneity of advertising, ie its diversity depending on: 1) the target audience; 2) product life cycle; 3) geolocations, etc. Creating several variants of advertising images and messages is the best solution and illustrates the principle of heterogeneity of advertising. To reach the target audience, understanding the needs, interests and values of a particular group requires constant study of the audience, its behavioral patterns, habits, and allows you to create consumer-oriented advertising. It is no coincidence that such super-successful companies as Google, Coca-Cola, Procter & Gamble, Microsoft, General Motors create commercials that raise important social issues for their audiences, such as responding to and meeting the values of inclusion and diversity. Yes, Coca-Cola draws attention of consumers to the diversity of nations with the slogan "Together is beautiful." The advertising campaign Google Pixel 2 "The Picture Perfect Life", on the one hand, shows the diversity of people of different genders, ages, nations. On the other hand -draws attention to the problem of perfect life in the photo, which behind the scenes may not be perfect and may tell an incomplete story because perfect photos and pages on social networks caused suicide of their owners, when their lives did not "meet" the set standards, were at odds with reality. The advertising campaign Always #LikeAGirl was incredibly successful, which showed the insults of gender stereotypes and showed them unrealistic, that the phrase "like a girl" should get rid of offensive connotations and simply reflect their achievements, choices, life style. The fourth condition of functional creativity is the focus on ethical norms, because the boundary between creativity and insults, discrimination can be sensitive, but at the same time quite measurable. Although there are no administrative sanctions for violating ethical norms, unlike violations of the law, their observance is important for building and maintaining the brand image and reputation. A striking example of the change in the advertising model under the pressure of public condemnation for non-compliance with ethical standards are the fashion shows of the Victoria's Secret brand. For several years, the public has criticized the company's policy of selecting super-skinny models, demanding that they lose more weight before shows, subsequent disorders, and treating these models for actual violence against their eating behavior. The brand continued to cultivate ethically outdated standards of beauty for women: to present women as sexual objects. As a result, the brand changed its policy: the face of the brand became women who gained popularity due to their achievements, rather than appearance. An example of creativity as a dysfunction in advertising is sexism. Discrimination on the grounds of sex, origin, social and property status, race and nationality, education, political views, religion, language, gender and occupation, place of residence, ie advertising that violates the standards of non-discriminatory advertising. Unfortunately, in the advertising space of Ukraine, cases of discriminatory advertising are not uncommon, for example in advertising where the image of a dark-skinned man with the inscription "Black Friday" is used. The fifth condition of functional creativity -emotional perception is a key factor in the communicative effectiveness of advertising. Both "positive" and "negative" emotions in advertising can be functional, provided that they correspond to the audience and the product. The correspondence of emotions is provided by the situational context, which acts as a driver of emotion. According to P. Thoits [18] emotion consists of situational context, physiological changes, gestures of expression and emotional labels. There is a problem of the mechanism of creation of advertising creativity. American researcher J. Goldenberg analyzed more than 200 award-winning creative advertisements and discovered common structural patterns among almost 90% of them. The results of this study and, accordingly, 8 tools of advertising creativity are presented in the book Cracking the Ad Code [19]. Let's consider these tools and try to give relevant examples of Ukrainian and foreign advertising practices. 1. The unification tool: This refers to delivering the message using perhaps unconsidered existing elements of the medium or those indicated by the message. An example of this creative tool is a public PR campaign aimed at the problem of bad roads: when activists planted flowers in potholes on the roads, placed photos of politicians in these potholes. 2. The activation tool: Here the viewer of the message is a necessary resource for proper delivery. Physical activation involves an immediate action by a viewer directly in front of the ad to release its message. Examples of this tool are flavored pages in cosmetic catalogs, QR codes in flyers, brochures. 3. The metaphor tool: A single visual image is used to connect the message and product to known cultural symbols or cognitive frameworks. The symbol is visually manipulated in a way that subtly fuses it to the product and message. For example, tea "Lipton" in the form of pyramids. Or the advertisement of the tourist complex: "I washed the sky, ironed the roads, warmed up the sea. I'm waiting. Odessa Mama". 4. The subtraction tool: Removing an essential element yet maintaining a strong message and product identification. Eliminating the text, image, or name of the product, with potentially varying emphases depending on the item removed, helps the message stand out. One instance could be the removal of a brand's labeled name from an ad while displaying its visually familiar package shape. For example, in the Toyota # stay at home ad, we see the subtraction of half the image of a car -a semi-enclosed garage, which is used to increase advertising appeal. 5. The extreme consequence tool: The promise is replaced by the depiction of an often very negative or unexpected result of using the product and enjoying the benefit. For example, a promotional video for "Evian" mineral water under the slogan "Live young" demonstrates the super properties of water due to the extreme consequence: excessive rejuvenation until childhood, as a result of which only kids live on the island. 6. The absurd alternative tool: Highlights an obtainable yet clearly undesirable circumstance yielding a benefit that derives from a positive product attribute. For example, you can usually use a fan, but what for, when there is a refreshing chewing gum. 7. The inversion tool: In place of a focus on how great the world is due to a specific product, the message is inverted to instead show how bad it would be without. For example, advertising "Dirol" "Charge the summer": without the advertised product, the summer will be gloomy and sad, but with the product it is colorful, funny, bright. 8. The extreme effort tool: The group's most common pattern, useful with unremarkable or generic products, it has two different categories of application. First, the company offering the product makes absurd efforts to please the customer. Second, a customer irrationally seeks to get hold of or protect the advertised product. As in the "John West" fish commercial, when a fisherman takes a fish from a bear so that it reaches the consumer. --- Conclusion Thus, the tools for ensuring creativity of advertising content were considered and illustrated with examples of Ukrainian and foreign advertising practices. Analysis of the conditions of functionality of advertising creativity allowed us to highlight the following features: 1) relevance of the advertised product; 2) adaptation to the needs of consumers, namely: conformity with consumer values; 3) heterogeneity: the type of creativity may differ depending on the target audience, the stage of the product life cycle and the advertising message; 4) focus on ethical norms, as the line between creativity and insult, discrimination can be sensitive; 5) emotional perception is a key factor in the communicative effectiveness of the advertising message. Emotions which arise in the process of perception of the information of the ad may be the cornerstone for the creativity of the advertising message. In order to make consumers trust the advertising message one should appeal to positive experience of consumers, associated with the product of the advertising message. One more way to attract attention of consumers to the advertised product and suggest creative idea is to mention in the ad the perspectives of the use of the product in personal and professional activity. It may influence the consumer's final decision making directly (when the quality of the product is perfect) or indirectly (when, for example, the image is improved when consumers buy the advertised product). In the future, it would be useful to consider the history of current trends of advertising creativity, for example, by analyzing the ads of the winners of prestigious awards of recent years. A separate area of analysis may be the study of the specifics of advertising creativity depending on the portrait of the target audience, which types of creativity are better perceived by women / men, people of different ages, education. Received by the editorial office 10.05.2021 --- <unk> <unk> <unk> <unk> <unk>тат<unk> <unk>осв<unk>ена и<unk>у<unk>ени<unk> осо<unk>енносте<unk> креативности реклам<unk>. <unk>реативност<unk> как <unk>ро<unk>лема дл<unk> анали<unk>а все<unk>да <unk>ла интересна со<unk>иоло<unk>ам, <unk>си<unk>оло<unk>ам, <unk>кономистам. <unk>оследние дес<unk>тилети<unk> <unk>о<unk>ул<unk>рно<unk> и востре<unk>ованно<unk> <unk>вл<unk>етс<unk> <unk>кономика в<unk>е<unk>атлени<unk>. <unk>кономика в<unk>е<unk>атлени<unk> <unk>ред<unk>ола<unk>ает активное ис<unk>ол<unk>ование <unk>мо<unk>и<unk>, <unk>мо<unk>ионал<unk>но<unk> креативности, котора<unk> <unk>ри<unk>вана с<unk>ормироват<unk> рекламное соо<unk>ение интересное дл<unk> <unk>отре<unk>ителе<unk>. <unk>днако, несмотр<unk> на мно<unk>о<unk>исленн<unk>е исследовани<unk> <unk>ро<unk>лем креативности <unk>редставител<unk>ми ра<unk>н<unk> наук, отдел<unk>н<unk>е ас<unk>ект<unk> ее ис<unk>ол<unk>овани<unk>, на<unk>ример, в рекламе, е<unk>е недостато<unk>но и<unk>у<unk>ен<unk>, <unk>то о<unk>условливает актуал<unk>ност<unk> ра<unk>от<unk>. <unk>ел<unk> стат<unk>и -иденти<unk>и<unk>ироват<unk> осо<unk>енности <unk>ормировани<unk> креативности реклам<unk>. <unk>ассмотрен<unk> и<unk>мерени<unk> креативности (<unk>е<unk>лост<unk>, <unk>и<unk>кост<unk>, ори<unk>инал<unk>ност<unk>, ра<unk>ра<unk>отанност<unk>, семанти<unk>еска<unk> <unk>и<unk>кост<unk>, о<unk>ра<unk>на<unk> <unk>и<unk>кост<unk>, с<unk>онтанна<unk> <unk>и<unk>кост<unk>) и и<unk> <unk>ример<unk> в рекламе. <unk>роанали<unk>ирован<unk> осо<unk>енности <unk>мо<unk>ионал<unk>но<unk> креативности, котора<unk> мо<unk>ет вкл<unk>ат<unk> ра<unk>витие или моди<unk>ика<unk>и<unk> стандартн<unk> <unk>мо<unk>и<unk> дл<unk> лу<unk>е<unk>о <unk>рис<unk>осо<unk>лени<unk> к <unk>отре<unk>ност<unk>м индивида / <unk>ру<unk> или ра<unk>витие нов<unk> <unk>орм в<unk>ра<unk>ени<unk> <unk>мо<unk>и<unk>. <unk>риведен<unk> <unk>ример<unk> украински<unk> и <unk>ару<unk>е<unk>н<unk> рекламн<unk> <unk>рактик, илл<unk>стриру<unk>ие ис<unk>ол<unk>ование инструментов со<unk>дани<unk> креативно<unk>о контента со<unk>ласно <unk>. <unk>олд<unk>ер<unk>у: инструмент о<unk>единени<unk>, инструмент актива<unk>ии, инструмент мета<unk>ор, инструмент в<unk>итани<unk>, инструмент <unk>кстремал<unk>н<unk> <unk>оследстви<unk>, инструмент а<unk>сурдно<unk> ал<unk>тернатив<unk>, инструмент инверсии, инструмент <unk>кстремал<unk>н<unk> усили<unk>. <unk>ка<unk>ан<unk> критерии креативности: нови<unk>на, <unk>ективност<unk>, аутенти<unk>ност<unk>, <unk>ри<unk>ем, все три ука<unk>анн<unk> критери<unk> ра<unk>ота<unk>т тол<unk>ко вместе. <unk>ассмотрен<unk> во<unk>рос<unk> <unk>ективности креативности реклам<unk>: и<unk>вестна<unk> <unk>ормула рекламно<unk>о вли<unk>ни<unk> <unk>D<unk> дает ответ на во<unk>рос о <unk>унк<unk>и<unk> креативности в рекламе -<unk>ривле<unk> внимание, со<unk>дат<unk> <unk>мо<unk>ионал<unk>н<unk> отклик и, как следствие, интерес, с<unk>ормироват<unk> <unk>отре<unk>ност<unk>, <unk>одтолкнут<unk> к де<unk>стви<unk>. <unk>то <unk>окусирует внимание и на <unk>мо<unk>ионал<unk>но<unk> составл<unk>е<unk>, <unk>од<unk>еркива<unk>, <unk>то <unk>е<unk> <unk>мо<unk>ии <unk>е<unk> <unk>ективности рекламно<unk>о соо<unk>ени<unk> не <unk>удет ра<unk>отат<unk>. <unk>о<unk>на<unk>ено, <unk>то на ка<unk>дом <unk>та<unk>е во<unk>де<unk>стви<unk> реклам<unk> креативност<unk> мо<unk>ет <unk>т<unk> как <unk>унк<unk>ионал<unk>но<unk>, так и дис<unk>унк<unk>ионал<unk>но<unk>. <unk>унк<unk>ионал<unk>но<unk>, то ест<unk> тако<unk>, котора<unk> с<unk>осо<unk>ствует дости<unk>ени<unk> коне<unk>но<unk> <unk>ели реклам<unk> -<unk>родат<unk> <unk>родукт / и<unk>менит<unk> модел<unk> <unk>оведени<unk> (дл<unk> со<unk>иал<unk>но<unk> реклам<unk>) креативност<unk> мо<unk>ет <unk>т<unk> <unk>ри о<unk>ределенн<unk> услови<unk>. <unk>тме<unk>ен<unk> услови<unk> <unk>унк<unk>ионал<unk>но<unk> и, дис<unk>унк<unk>ионал<unk>но<unk> креативности. <unk>ассмотрение услови<unk> <unk>унк<unk>ионал<unk>ности креативности реклам<unk> <unk>о<unk>волило в<unk>делит<unk> следу<unk>ие е<unk>о осо<unk>енности. <unk>реативн<unk>ст<unk> <unk>к <unk>ро<unk>лема дл<unk> анал<unk>у <unk>ав<unk>ди <unk>ула <unk>кава со<unk>оло<unk>ам, <unk>си<unk>оло<unk>ам, економ<unk>стам. <unk>станн<unk> дес<unk>тил<unk>тт<unk> <unk>о<unk>ул<unk>рно<unk> та <unk>атре<unk>увано<unk> <unk> економ<unk>ка вра<unk>ен<unk>. <unk>коном<unk>ка вра<unk>ен<unk> <unk>еред<unk>а<unk>а<unk> активне використанн<unk> емо<unk>, емо<unk>но<unk> креативност<unk>, <unk>ка <unk>окликана с<unk>ормувати рекламне <unk>ов<unk>домленн<unk>, <unk>о <unk> <unk>кавим дл<unk> с<unk>о<unk>ива<unk>в. <unk>днак, не<unk>ва<unk>а<unk>и на <unk>исленн<unk> досл<unk>д<unk>енн<unk> <unk>ро<unk>лем креативност<unk> <unk>редставниками р<unk>ни<unk> наук, окрем<unk> ас<unk>екти <unk> використанн<unk>, на<unk>риклад, у реклам<unk>, <unk>е недостатн<unk>о вив<unk>ен<unk>, <unk>о о<unk>умовл<unk> актуал<unk>н<unk>ст<unk> ро<unk>оти. <unk>ета статт<unk> -<unk>денти<unk>кувати осо<unk>ливост<unk> <unk>ормуванн<unk> креативност<unk> реклами. <unk>о<unk>л<unk>нуто вим<unk>ри креативност<unk> (<unk>видк<unk>ст<unk>, <unk>ну<unk>к<unk>ст<unk>, ори<unk>нал<unk>н<unk>ст<unk>, детал<unk>а<unk>, семанти<unk>ну <unk>ну<unk>к<unk>ст<unk>, о<unk>ра<unk>ну <unk>ну<unk>к<unk>ст<unk>, с<unk>онтанну <unk>ну<unk>к<unk>ст<unk>) та <unk> <unk>риклади у реклам<unk>. <unk>роанал<unk>ован<unk> осо<unk>ливост<unk> емо<unk>но<unk> креативност<unk>, <unk>о мо<unk>е вкл<unk>ати ро<unk>виток <unk>и моди<unk>ка<unk> стандартни<unk> емо<unk> дл<unk> л<unk>о<unk>о <unk>астосуванн<unk> до <unk>отре<unk> <unk>ндив<unk>да /<unk>ру<unk>и а<unk>о ро<unk>виток нови<unk> <unk>орм вира<unk>енн<unk> емо<unk>. <unk>аведено <unk>риклади укра<unk>нс<unk>ки<unk> та <unk>но<unk>емни<unk> рекламни<unk> <unk>рактик, <unk>о <unk>л<unk>стру<unk>т<unk> використанн<unk> <unk>нструмент<unk>в створенн<unk> креативно<unk>о контенту <unk>а <unk>. <unk>олд<unk>ер<unk>ом: <unk>нструмент о<unk>'<unk>днанн<unk>, <unk>нструмент актива<unk>, <unk>нструмент мета<unk>ор, <unk>нструмент в<unk>дн<unk>манн<unk>, <unk>нструмент екстремал<unk>ни<unk> насл<unk>дк<unk>в, <unk>нструмент а<unk>сурдно<unk> ал<unk>тернативи, <unk>нструмент <unk>нверс<unk>, <unk>нструмент екстремал<unk>ни<unk> <unk>усил<unk>. <unk>а<unk>на<unk>ен<unk> критер<unk> креативност<unk>: нови<unk>на, е<unk>ективн<unk>ст<unk>, автенти<unk>н<unk>ст<unk>, <unk>ри <unk>ому вс<unk> три <unk>а<unk>на<unk>ени<unk> критер<unk> <unk>ра<unk>т<unk> т<unk>л<unk>ки ра<unk>ом. <unk>о<unk>л<unk>нуто <unk>итанн<unk> <unk>одо е<unk>ективност<unk> креативност<unk> реклами: на<unk>в<unk>дом<unk>а <unk>ормула рекламно<unk>о в<unk>ливу <unk>D<unk> да<unk> в<unk>д<unk>ов<unk>д<unk> на <unk>итанн<unk> <unk>одо <unk>унк<unk> креативност<unk> у реклам<unk> -<unk>ривернути ува<unk>у, створити емо<unk>ни<unk> в<unk>д<unk>ук <unk>, <unk>к насл<unk>док, <unk>нтерес, с<unk>ормувати <unk>отре<unk>у, <unk>д<unk>тов<unk>нути до д<unk>. <unk>е <unk>окусу<unk> ува<unk>у <unk> на емо<unk>н<unk> складов<unk>, <unk>дкресл<unk>и, <unk>о <unk>е<unk> емо<unk> лан<unk> е<unk>ективност<unk> рекламно<unk>о <unk>ов<unk>домленн<unk> не <unk>уде <unk>ра<unk>вати. <unk>а<unk>на<unk>а<unk>т<unk>с<unk>, <unk>о на ко<unk>ному ета<unk> в<unk>ливу реклами креативн<unk>ст<unk> мо<unk>е <unk>ути <unk>к <unk>унк<unk>онал<unk>но<unk>, так <unk> дис<unk>унк<unk>онал<unk>но<unk>. <unk>унк<unk>онал<unk>но<unk>, то<unk>то тако<unk>, <unk>о с<unk>ри<unk> дос<unk>ненн<unk> к<unk>н<unk>ево<unk> мети реклами -<unk>родати <unk>родукт/<unk>м<unk>нити модел<unk> <unk>овед<unk>нки (дл<unk> со<unk>ал<unk>но<unk> реклами) креативн<unk>ст<unk> мо<unk>е <unk>ути <unk>а <unk>евни<unk> умов. <unk>а<unk>на<unk>ено умови <unk>унк<unk>онал

The article is devoted to the study of the peculiarities of advertising creativity. Creativity as a problem for analysis has always been of interest to sociologists, psychologists, economists. In recent decades, the economy of experience has been popular and has been in demand. The economy of experience involves the active use of emotions, emotional creativity, which is called to form an advertising message that is interesting to consumers. However, despite numerous studies of the problems of creativity by representatives of various sciences, some aspects of its use, for example, in advertising, are still insufficiently studied, which determines the topicality of the work. The aim of the article is to identify the features of formation advertising creativity. Measurements (indicators) of creativity (fluency, flexibility, originality, elaboration, semantic flexibility, figurative flexibility, spontaneous flexibility) and their examples in advertising are considered. Features of emotional creativity are analyzed, which may include the development or modification of standard emotions to better adapt to the needs of the individual / group or the development of new forms of expression of emotions. Examples of Ukrainian advertising practices illustrating the use of tools for formation of creative content according to J. Goldberg are given: unification tool, activation tool, metaphor tool, subtraction tool, extreme consequence tool, absurd alternative tool, inversion tool, extreme effort tool. The criteria of creativity are noted: novelty, efficiency, authenticity, and all three of these criteria work only together. The question of the efficiency of advertising creativity is considered: the most famous formula for the impact of advertising product AIDA provides answers to questions about the functions of creativity in advertising -to attract attention, create an emotional response and as a result interest, create a need, stimulate to action. This focuses on the emotional component, emphasizing that without emotion, the chain of effectiveness of the advertising message will not work. It is noted that at each stage of the impact of advertising, creativity can be both functional and dysfunctional. Functional, ie one that contributes to the ultimate goal of advertising -to sell a product / change the pattern of behavior (for social advertising) creativity may be under certain conditions. The conditions of functional and dysfunctional creativity are indicated. Consideration of the conditions of functionality of the advertising creativity allowed us to highlight its features. These are: relevance of the advertised product; adaptation to consumer needs, namely: conformity with consumer values; heterogeneity: the type of creativity may differ depending on the target audience, the stage of the product life cycle and the advertising message; focus on ethical norms. The line between creativity and insults, discrimination can be sensitive. Emotional perception is a key factor in the communicative efficiency of the advertising message.

енн<unk> не <unk>уде <unk>ра<unk>вати. <unk>а<unk>на<unk>а<unk>т<unk>с<unk>, <unk>о на ко<unk>ному ета<unk> в<unk>ливу реклами креативн<unk>ст<unk> мо<unk>е <unk>ути <unk>к <unk>унк<unk>онал<unk>но<unk>, так <unk> дис<unk>унк<unk>онал<unk>но<unk>. <unk>унк<unk>онал<unk>но<unk>, то<unk>то тако<unk>, <unk>о с<unk>ри<unk> дос<unk>ненн<unk> к<unk>н<unk>ево<unk> мети реклами -<unk>родати <unk>родукт/<unk>м<unk>нити модел<unk> <unk>овед<unk>нки (дл<unk> со<unk>ал<unk>но<unk> реклами) креативн<unk>ст<unk> мо<unk>е <unk>ути <unk>а <unk>евни<unk> умов. <unk>а<unk>на<unk>ено умови <unk>унк<unk>онал<unk>но<unk> <unk> дис<unk>унк<unk>онал<unk>но<unk> креативност<unk>. <unk>о<unk>л<unk>д умов <unk>унк<unk>онал<unk>ност<unk> креативност<unk> реклами до<unk>волив вид<unk>лити насту<unk>н<unk> <unk>о<unk>о осо<unk>ливост<unk>. <unk>е: релевантн<unk>ст<unk> рекламованому <unk>родукту; ада<unk>та<unk> до <unk>отре<unk> с<unk>о<unk>ива<unk>в, а саме: в<unk>д<unk>ов<unk>дн<unk>ст<unk> <unk>нност<unk>м с<unk>о<unk>ива<unk>в; <unk>етеро<unk>енн<unk>ст<unk>: ти<unk> креативност<unk> мо<unk>е в<unk>др<unk>н<unk>тис<unk> в <unk>але<unk>ност<unk> в<unk>д <unk>л<unk>ово<unk> аудитор<unk>, ета<unk>у <unk>итт<unk>во<unk>о <unk>иклу <unk>родукту та рекламно<unk>о <unk>ов<unk>домленн<unk>; ор<unk>нта<unk> на ети<unk>н<unk> норми, оск<unk>л<unk>ки ме<unk>а м<unk> креативн<unk>ст<unk> та о<unk>ра<unk>ами, дискрим<unk>на<unk> мо<unk>е <unk>ути <unk>утливо<unk>. <unk>мо<unk>н<unk>ст<unk> с<unk>ри<unk>н<unk>тт<unk> <unk> кл<unk>овим <unk>актором комун<unk>кативно<unk> е<unk>ективност<unk> рекламно<unk>о <unk>ов<unk>домленн<unk>. <unk>л<unk>ов<unk> слова: осо<unk>ливост<unk> креативност<unk> реклами, реклама, емо<unk>на креативн<unk>ст<unk>, <unk>нструменти креативност<unk>, укра<unk>нс<unk>к<unk> та <unk>но<unk>емн<unk> рекламн<unk> <unk>рактики.

The article is devoted to the study of the peculiarities of advertising creativity. Creativity as a problem for analysis has always been of interest to sociologists, psychologists, economists. In recent decades, the economy of experience has been popular and has been in demand. The economy of experience involves the active use of emotions, emotional creativity, which is called to form an advertising message that is interesting to consumers. However, despite numerous studies of the problems of creativity by representatives of various sciences, some aspects of its use, for example, in advertising, are still insufficiently studied, which determines the topicality of the work. The aim of the article is to identify the features of formation advertising creativity. Measurements (indicators) of creativity (fluency, flexibility, originality, elaboration, semantic flexibility, figurative flexibility, spontaneous flexibility) and their examples in advertising are considered. Features of emotional creativity are analyzed, which may include the development or modification of standard emotions to better adapt to the needs of the individual / group or the development of new forms of expression of emotions. Examples of Ukrainian advertising practices illustrating the use of tools for formation of creative content according to J. Goldberg are given: unification tool, activation tool, metaphor tool, subtraction tool, extreme consequence tool, absurd alternative tool, inversion tool, extreme effort tool. The criteria of creativity are noted: novelty, efficiency, authenticity, and all three of these criteria work only together. The question of the efficiency of advertising creativity is considered: the most famous formula for the impact of advertising product AIDA provides answers to questions about the functions of creativity in advertising -to attract attention, create an emotional response and as a result interest, create a need, stimulate to action. This focuses on the emotional component, emphasizing that without emotion, the chain of effectiveness of the advertising message will not work. It is noted that at each stage of the impact of advertising, creativity can be both functional and dysfunctional. Functional, ie one that contributes to the ultimate goal of advertising -to sell a product / change the pattern of behavior (for social advertising) creativity may be under certain conditions. The conditions of functional and dysfunctional creativity are indicated. Consideration of the conditions of functionality of the advertising creativity allowed us to highlight its features. These are: relevance of the advertised product; adaptation to consumer needs, namely: conformity with consumer values; heterogeneity: the type of creativity may differ depending on the target audience, the stage of the product life cycle and the advertising message; focus on ethical norms. The line between creativity and insults, discrimination can be sensitive. Emotional perception is a key factor in the communicative efficiency of the advertising message.

Introduction A considerable literature studies the impact of disruptive events on children's outcomes. Prior studies suggest that shocks are on average detrimental for children's academic progress and social-psychological well-being, though effects vary depending on the type of shock. Local violence and family instability have been shown to impact children's cognitive performance and educational achievement negatively [1][2][3][4][5]. In contrast, relocation to procure better school quality or policies that subsidize college education are generally achievement-enhancing [6,7]. Some types of US immigration enforcement have been conceptualized as shocks to the wider Hispanic community [8,9]. A number of recent studies show that immigration enforcement not only affects undocumented migrants through the direct risk of deportation [10][11][12], but also has unintended negative consequences for documented and US-born Hispanics [13][14][15][16][17]. While family ties to undocumented migrants have been linked to anxiety that a parent or sibling will be deported [16], statistical discrimination can lead to difficulties in securing work or stigmatization of the wider Hispanic community [18,19]. As many as 5.1 million children under age 18 lived in mixed status households in the US during the period 2009-13 [20]. Considering the size of the population as well as dramatic increases in deportations observed over recent decades, it is important to gain a better understanding of how Hispanic children respond to immigration enforcement. Prior studies have indicated that immigration enforcement can have negative consequences for Hispanic students' educational outcomes [21,22]. Bellows examined the relationship between Secure Communities (SC) and student achievement at the county level [23]. Analyses based on data from the Stanford Education Data Archive showed that the activation of SC was associated with decreases in Hispanic students' English language arts achievement and black students' English language arts and math achievement. In contrast, Kirksey and co-authors' study using the same data found an association between deportations and the Hispanic-white educational achievement gap, but no association between deportations and the black-white achievement gap [24]. These divergent findings may be in part due to different samples and study designs, where the first study assesses county level achievement across the US [23], and the second study assesses school district level achievement gaps in cities which are within 100 miles of deportation sites [24]. How students across racial groups are impacted by immigration enforcement, and how they respond to them in their educational outcomes, requires further investigation. This study builds on the prior literature in two ways. First, it exploits the staggered rollout of SC to estimate dynamic difference in differences on the impact of SC on school district level achievement (SDLA) of Hispanic, white, and black students using detailed information on the implementation and geographic scope of SC and data from the Stanford Education Data Archive (SEDA). It also runs a placebo test on sanctuary jurisdictions, which are cities and counties that ordained laws or regulations to block immigration enforcement. The previously described study by Bellows assessed educational outcomes at the county level [23]. While SC was implemented at the county level, there is considerable heterogeneity within counties. The US has about 3,240 counties, whereas there are more than 13,000 geographically defined public school districts. School districts, thus, provide a more disaggregated measure of test scores, as well as students' demographic characteristics and socioeconomic background. Most school districts also have authority over public schools that are within their geographic boundaries and have been shown to differ in their approaches to school improvement, which can impact student learning through school district-wide reforms [25][26][27]. Second, this study assesses heterogeneity within the relationship between SC and Hispanic students' educational achievement across different contexts. Even though SC was activated nation-wide, regional variation in attitudes toward immigrants and the size of the undocumented population likely shape how disruptive the policy is for Hispanics. For instance, roughly half of children under age 18 are estimated to be children of migrants in California, whereas the national share lies at 23 percent [28]. I therefore incorporate Current Population Census (CPS) data, which allow me to assess the likely undocumented population in the state. Using this information, I analyze differences in the impact of SC across areas with different shares of likely undocumented migrants, as well as census regions, and rural and urban areas. This can provide indirect evidence on the likely mechanisms underlying the impact of SC on Hispanic students' educational outcomes. Common perceptions about the implications of disruptive events, such as changed policy settings and sudden economic distress, differ by the underlying perspective. On the one hand, fear that a family member will be deported as well as reduced interaction with the wider society can lead to stress and impede English progress, negatively impacting educational performance [16,29,30]. On the other hand, children's outcomes can also be positively affected. Considering that undocumented migrants face barriers to participating in the formal labor market and Hispanic students are more likely to encounter employer discrimination than other students, education can provide an important alternative to entering the labor market [7,31,32]. This may encourage Hispanic families to invest in their children's human capital with the aim of facilitating the transition to upper secondary education. Hispanic children, who are frequently subjected to shocks, evoke particular concern. Studies show that return and circular migration have become less common among Mexican migrants in the US, at least in part, because of rising deportations [33,34]. During their longer stays in the US, many undocumented migrants form families north of the border [17,35,36]. Even though most children of undocumented migrants hold a US citizenship, they risk facing a combination of economic disadvantage, school instability, and discrimination [30,37]. Early life adversities may accrue over the life course, for example due to group-level differences in the returns to socioeconomic resources, as hypothesized by the theory of cumulative disadvantage [38], or when certain outcomes are used as markers for subsequent opportunities, as the theory of cumulative advantage posits [39,40]. These mechanisms have the potential to impede future labor market outcomes and can exacerbate social differences in the US [41]. In the following sections, I discuss the implementation of SC. Thereafter, I review the theoretical background and previous literature. Then, I describe the data, followed by the empirical strategy, and findings. Finally, I conclude and outline next steps for research. --- Secure Communities-A policy background Federal initiatives, such as the SC program, have drawn state, county, and municipal police forces into the active enforcement of federal immigration laws. The SC program sought to increase the detention and deportation of immigrants who commit crimes in the US. Beginning in October 2008, SC created a screening process wherein every person arrested by local law enforcement officials in the US would automatically be cross-referenced with federal authorities for immigration status and deportation eligibility [42]. Previous programs required police officers or Immigration and Customs Enforcement (ICE) agents to interview each person individually to assess whether they were immigration violators. By automating the process, SC greatly increased the likelihood of detention. The fiscal year of 2012 marked a peak of more than 464,000 migrants cycling through detention facilities [43]. Prior studies show that the program has not impacted overall crime rates [44], but instead increased the likelihood that Hispanics are arrested and undocumented migrants deported [45]. Even though SC was directed at all migrants, the policy primarily targeted the Hispanic community, at least in part, due to differences in the share of the undocumented population across groups. Estimates based on the CPS indicate that about 23 percent of Hispanics age 20 and above are likely undocumented migrants. Among white and black persons, the respective shares are 5 and 2 percent. Although SC marks the largest expansion of interior immigration enforcement in the US to date, deportations have risen consistently in bipartisan fashion since the early 1990s. Immigration enforcement previously focused on border crossings but has shifted to the US interior in more recent years. This is largely due to the decreasing number of undocumented migrants attempting to cross the border [33,34]. Besides SC, a number of other immigration enforcement policies were implemented at the state and county-level during the 2000s. These included E-verify, Omnibus Immigration Laws, and 287(g) agreements, which enable employers, as well as local and state officials, to check employees' migration status and to apprehend undocumented migrants. These policies have been implemented selectively and are not nearly as widespread as SC. Local values and political views have been found to influence which states and counties implement these policies [46]. The scope of SC, resource bottlenecks, and technological constraints meant that the program could not be activated in the entire country at once. Consequently, the rollout was staggered over four years. The federal government determined the sequence of the rollout and implemented it on a county-by-county basis. Fig 1 shows that the program was first activated in counties located by the Mexico-US border. Prior work also reveals that the implementation of the program mirrored federal enforcement priorities for immigration [47]. By 2011, SC was implemented in roughly half the country and by January 2013 it was completely activated nation-wide. The federal government prohibited local governments from formally opting out of SC even though some counties did not want to participate. Once the screening system was in place, informal noncompliance was practically impossible [44]. Still, several cities and counties ordained laws or regulations that to this day block immigration enforcement and protect undocumented migrants from ICE. These are called sanctuary jurisdictions. A report based on data from the Transactional Records Access Clearinghouse (TRAC) at Syracuse University shows that sanctuary jurisdictions reduce the odds that undocumented migrants are arrested and deported by ICE. Despite official statements that ICE compensates for this by allocating more resources to communities where local jurisdictions fail to cooperate, the report does not find that the odds of other arrests, i.e., that are not linked to SC, are greater in sanctuary versus non-sanctuary jurisdictions [49]. Strong critique of SC started in Los Angeles and spread between 2012 and 2014. In November 2014, the program was officially discontinued. However, the Priority Enforcement Program directly replaced SC. The Priority Enforcement Program is in the same spirit as SC and relies on the same automated screening. Still, under this new program, ICE was instructed to transfer only information on severe criminal offenders and to request a detainer only if the person in custody was subject to a final order of removal or if there was other sufficient probable cause that the person was removable. The Trump administration subsequently re-labelled the Priority Enforcement Program as SC [50]. There are a number of educational reforms that were enacted simultaneously. The Deferred Action for Childhood Arrivals program was passed in 2012 and has been instrumental in providing undocumented migrants relief from deportation and work authorization if they are in education. Dream Acts have been passed in a number of states and allow undocumented students to pay in-state tuition at public higher education institutions [7]. In some states, including Texas, undocumented students can also apply for financial aid. Lastly, the No Child Left Behind Act administers standardized testing and tracks the progress of vulnerable student groups, such as migrant and minority groups. These reforms have provided undocumented migrants with opportunities to continue their education and increased the monitoring of lowperforming groups. --- Previous research Immigration enforcement influences the context that migrants live in and how they interact with society. The concept of racialized legal status has been used to describe social positions that are based on race-neutral legal classifications, such as documentation status, but which disproportionately impact racial minorities [8]. Racialized legal status directly affects undocumented migrants through economic distress [51,52], detrimental health outcomes [10,11,13], and by exacerbating the psychosocial consequences of perceived stigma related to undocumented status in the US [53]. Racialized legal status has also been argued to impact documented migrants and US-born Hispanics indirectly. Such indirect effects are termed chilling effects referring to the fact that they arise from an icy policy climate rather than from a direct risk to the individual. In more general terms, chilling is used to describe a "discouraging or deterring effect, especially one resulting from a restrictive law or regulation" [54,55]. In the context of welfare reform, chilling has been cited as a potential explanation for declines in program participation beyond what would be predicted due to eligibility changes alone [14,15]. Chilling effects may operate through two mechanisms. The first mechanism hinges on social proximity to undocumented persons. For instance, children, who are born to undocumented parents in the US and are thus US citizens, fear that immigration enforcement will dissolve their families [16]. This increases the group's anxiety levels and may have longer-term consequences for their academic progress [56]. The second mechanism operates through statistical discrimination. This impacts the wider Hispanic community, as employers or police officers racialize Hispanics by assuming that Hispanics, as a rule, are undocumented [19,57,58]. Investigatory traffic stops have been shown to disproportionately put Hispanics at risk of arrest and citation, independent of their legal status [9,59]. Immigration policies may also incite anti-immigrant sentiments. For example, the enactment of anti-immigrant legislation in Arizona increased the frequency of negative tweets about migrants in that state [60]. These situations and the stress they trigger likely have negative consequences for educational performance. Prior research further reveals that the way in which immigration enforcement impacts the wider Hispanic community depends on features of the local context. Considering regional differences in political values and attitudes toward immigrants, there is likely heterogeneity in the consequences of perceived stigma and discrimination of Hispanics across the US [61]. For instance, republican views have been shown to correlate with higher levels of deportation [46], which are overrepresented in more conservative Southern and mid-Western red states. In contrast, Northern, coastal states, and urban areas tend to be more politically liberal [62]. Similarly, residents in rural areas often have more restrictive views on immigration than residents in urban areas [63,64]. While nearly 60 percent of residents in rural areas believe that the growing number of immigrants threatens traditional American customs and values, 35 percent of urban residents share this view [61]. In rural areas and small towns that have become new destinations to growing numbers of immigrants, Hispanics are also often met by negative stereotypes when speaking Spanish or gathering in public places [19]. Another feature of the local context that shapes the extent to which children are exposed to immigration enforcement is the size of the undocumented population in the area. For Hispanic children, who live in areas with large undocumented populations, immigration enforcement has a strong presence. In New Jersey, where the Mexican community is relatively large and established, deportations directly affect children's neighbors and other community members. Seeing this, children try to predict the risk that their parents' lack of documentation will lead to deportation. In contrast, in settings where the Mexican community is small, documentation status does not affect Hispanic children's daily lives, unless the family experienced deportation directly [16]. Children are, however, aware that for many Americans 'Mexican' is synonymous with 'illegal' and describe being teased and excluded because of their racial background [16]. This suggests that Hispanics are impacted in both settings, but that the level of exposure to immigration enforcement is likely higher in areas with large undocumented populations. Depending on the underlying perspective, shocks resulting from immigration enforcement may have different impacts on children's and young adults' educational outcomes. Immigration enforcement has been shown to augment economic insecurity [51,65] and the frequency of relocation [66,67], which can influence Hispanic students' cognitive performance negatively [29,30]. Reduced interaction with native-English speakers may, in particular, impact Hispanic students' English language development [19,53]. In line with this perspective, evidence based on the CPS reveals that intensified enforcement raises the probability of repeating a grade by 14 percent for children aged 6-13, whose parents are assumed to be undocumented. The likelihood of dropping out of school also increases by 18 percent for those aged 14-17 [21,22]. The previously mentioned study by Bellows reveals a negative relationship between SC and county level achievement among Hispanic and black students [23]. Whereas, Kirksey and co-authors find that deportations are related to the Hispanic-white school district level achievement gap, but not to the black-white achievement gap [24]. However, immigration enforcement could potentially also have a positive impact on educational outcomes. In response to policy shifts, undocumented migrants' labor market position has become increasingly tenuous since the mid-1980s [32,68]. For many undocumented students, upper secondary education has therefore come to represent an escape from menial jobs when finishing school [7,31]. Policy shifts have also changed how employers treat Hispanic workers in general [69]. Seeing that greater enforcement means more penalties related to the hiring of undocumented migrants, employers increasingly outsource the hiring of migrant workers to avoid federal sanctions. Employers have also started to verify the legal documents of all persons who they suspect to be foreign-born [57]. In other words, immigration enforcement has increased employer screening and in turn often augmented racial discrimination. Lower employment opportunities and discrimination resulting from immigration enforcement may then incentivize Hispanic families to invest in the human capital of their children, independent of their documentation status. This can lead Hispanic students to perform better in school, as they aspire to stay in education rather than entering the labor market where they face fewer opportunities. This study contributes to the literature by assessing the potential positive and negative impacts of SC on Hispanic, white, and black students' educational outcomes at the school district level, and by showing how these vary across census regions, rural and urban areas, and areas with different proportions of likely undocumented migrants. Since the policy implementation targeted the Hispanic community and white and black students are less likely to be in close social proximity to undocumented migrants, I do not expect to observe sizeable changes in the educational outcomes of white and black students. --- Data First, I obtain publicly available data from the Department of Homeland Security (DHS) that provide the month and year in which SC was implemented for each county 2008-12 [70,71]. I use these data to define the treatment status, coding the academic year in which SC was implemented in the county and subsequent years when the policy was in place as one, otherwise the indicator variable is coded as zero. Second, I link these data using state ID, county name, and academic year to the SEDA data, which are publicly available aggregate data [72]. Specifically, the data provide annual measures of English language arts and math achievement for Hispanic, white, and black students over a set of different geographic units (geographically defined school districts, counties, metropolitan areas, commuting zones, and states) for each year 2009-18 and for each grade 3-8. Year refers to the spring of each academic year throughout the paper, so 2009 represents the 2008-9 academic year. Given that the DHS provides the month and year of the implementation of SC, the indicator variable shifts to one in the academic year of the policy activation. As an example, the indicator shifts to one in the academic year of 2010, if SC was implemented between July 2009 and June 2010. The SEDA data uniquely identify schools by defining stable school identifiers and assign schools to larger geographic units. The stable school identifiers are based on Longitudinal ID Crosswalks provided in the Common Core of Data and enable tracking schools as they change school ID or school district. School district changes can be due to school districts splitting, merging, or some other administrative change [72,73]. To define school districts, the SEDA data use the 2019 EDGE Unified and Elementary School District Boundaries [74]. The US has more than 13,000 geographically defined public school districts. The structure of school districts varies by state and region. For instance, school districts in the Mid-Atlantic and New England states tend to follow county, township, or city boundaries, while school districts in the Midwest and Western states are generally independent of other municipal boundaries [75]. The total population size also varies across school districts. While the total population of the Berkeley Unified School District in California amounts to over 120,000 individuals, the Cochise Elementary District in Arizona comprises about 300 [76]. Most public-school districts have administrative control over the public schools that fall within their specific geographic boundaries. For schools that are physically located within a school district but not under its administrative control, such as charter and magnet schools, the SEDA data use information on location (latitude and longitude coordinates), school type, and school status to assign them to school districts [72]. This assignment ensures that school district level achievement reflects most of the students living in the school district. It also provides better alignment between achievement at the school district level and students' demographic characteristics and socioeconomic background. These are constructed using information from the American Community Survey and reported for all families living in the geographic boundaries of the school district. To assign schools to counties, the SEDA data use the county code provided in the Common Core of Data in the most recent year the school was observed, meaning that it is also stable over time. While most school districts are in one county, five percent of school districts belong to multiple counties. I assign each school district, which is in multiple counties, to the first recorded county. This is documented in S1 shows that 76 percent of school districts, which are in multiple counties, are in counties that activated SC in the same year (indicated by zero). The treatment assignment may therefore be incorrect for no more than 164 school districts. I ran sensitivity analyses dropping these school districts finding similar patterns to those presented in the main analysis. The results and are provided in S1 Fig. I analyze school district level achievement aggregated across grades 3-8 by race and subject. The SEDA data are constructed using information on standardized tests that each state in the US is required to administer in public schools. States have flexibility in designing and administering the test. States also set their own standards regarding the proficiency level in each grade and subject. In order to improve comparability, the SEDA data transform SDLA into a common national scale based on overall state performance and on school district performance on the state exam relative to other school districts in the state [77]. Specifically, the cohort standardized scale is standardized within subject and grade, relative to the average of the four cohorts in the data who were in 4th grade in 2009, 2011, 2013, and 2015. Standardized school district means have an overall average near zero and a standard deviation of one. For example, a school district with a mean of 0.5 on the cohort standardized scale represents a school district where the average student scored one half of a standard deviation higher than the national reference cohort in the same grade [72,78]. The SEDA data provide SDLA for each school district, grade, year, and subject in which there are at least 20 students in each cell or group (i.e., at least 20 Hispanic students in each grade, year, and subject within a school district). Values for observations with less than the required sample size are suppressed for privacy reasons [72]. In addition, SEDA data omit observations in which: the state test participation rate is less than 95 percent; students in the same state, grade, year, and subject took different tests; or the state did not report sufficient data. Overall, SEDA omits 10.9 percent of school district-grade-year-subject observations [72]. The SEDA data also provide information on annual school district-level characteristics including student demographics, which include the proportion of students who: qualify for free or reduced-price lunch; are in Special Education; and are English Language Learners. A socio-economic composite score combines information on median family income, household poverty rate, the proportion of adults who have a bachelor's degree or higher, the proportion of adults who are unemployed, the proportion of households receiving SNAP benefits, and the proportion of households with children that are headed by a single mother. Additional school district characteristics include the proportion of students who are in city/urban, suburban, town, and rural locale schools. For more detailed information, see the Stanford Education Data Archive technical report under https://stacks.stanford.edu/file/druid:db586ns4974/seda_ documentation_4.1.pdf. Third, I link these data at the state level to the CPS for the years 2009-18 [79]. The CPS is collected by the US Census Bureau and has been approved by the Office of Management and Budget (OMB Number 0607-0049). All personally identifiable information is replaced by one or more artificial identifiers, or pseudonyms before the data are made publicly available. For more information, see https://www.census.gov/programs-surveys/cps/about.html. The CPS does not contain information on the documentation status of foreign-born, but following an existing literature I am able to identify probabilistically undocumented migrants among the adult population [80,81]. In particular, I define an individual as probabilistically undocumented if the person is a foreign-born, non-veteran with no post-secondary education. I find that 26.6 percent of the foreign-born population age 20 and above are likely undocumented migrants, which is in line with prior estimates [82]. Among foreign-born Hispanics, about 45 percent are undocumented [83]. --- Sample restrictions I make some sample restrictions that are summarized in Table 1. The SEDA data suppress test scores by race where fewer than 20 students took the test, the participation rate was below 95 percent, students took different tests, or states did not report sufficient data [72]. Missing values are more common among Hispanic and black than white students. I also exclude Alaska and Hawaii, as well as early adopters from the sample, considering that the first counties that implemented SC substantially differ in economic and demographic composition from other counties. This is in line with previous research [14,84]. In a last step, I exclude school districts without recorded scores for Hispanic students. I do this in order to estimate the coefficients for Hispanic, white, and black students using comparable samples. Considering that white students are observed roughly three times as often as Hispanic and black students, comparing estimates across models would otherwise imply comparisons of different samples. Results are robust to alternative sample restrictions, for instance keeping only school districts where we observe all three races, or excluding school districts where only white students' scores are recorded. The analytical sample for English language arts consists of 29,735 school districtyear observations for Hispanic students. The respective numbers are 27,500 for white students, and 15,078 for black students. For Math, it consists of 29,679 school district-year observations for Hispanic students. Again, the respective numbers are 27,394 for white students, and 14,925 for black students. Table 2 provides mean values for school districts in the years before and after the implementation of SC. Hispanic students' SDLA in English language arts is about a third of a standard deviation below the national reference cohort (with a mean of -0.342 in years prior to SC and -0.260 in years after). In math, Hispanic students score about one fourth of a standard deviation lower than the national reference cohort indicated by mean values of -0.281 in years before the implementation of the program and -0.256 in years after. In both subjects, we observe improvements in SDLA over the time period analyzed. White students perform better in English language arts and math than the national reference cohort (0.162 in English language arts and 0.173 in math in years prior to SC). When comparing years before and after the implementation of SC, we also observe an increase in SDLA. Black students perform considerably lower than the national reference cohort (-0.383 in English language arts and -0.424 in math in years prior to SC). Among black students, we observe no improvement in SDLA over the observation period. --- Empirical strategy The first part of the analysis estimates the impact of SC by comparing changes in the SDLA of Hispanic students in adopting school districts with contemporaneous changes in school districts that had not yet implemented SC. I estimate the following difference in differences specification: Y dt 1<unk>4 a <unk> bI post dt <unk> X 0 dt g <unk> m d <unk> r t <unk> dt ;<unk>1<unk> where Y dt stands for the dependent variable (i.e., English language arts or math SDLA among Hispanic students). Equivalent equations are estimated for white and black students. The suffix d stands for school district and t represents the year. The indicator of interest is I post dt. It is equal to one in all school district-years after the activation of SC. X 0 is a vector of covariates that vary between school districts and over time. They include other immigration enforcement policies, a binary indicator of sanctuary jurisdictions, as well as the proportion of students who qualify for free or reduced-price lunch, are in Special Education, and are English Language Learners. Additional controls include a socioeconomic composite score. I control for these characteristics as they are determinants of SDLA [85,86]. The specification also includes school district fixed effects (<unk> d ) to account for any unobserved time-invariant school district-level factors that affect test scores, and year fixed effects (<unk> t ) to account for nation-wide policies or economic shocks that might influence SDLA, such as the enactment of the Deferred Action for Childhood Arrivals program or the No Child Left Behind Act. The models are weighted using precision weighting, which is the inverse of the standard error of SDLA squared. I use two alternative approaches to estimate Eq 1, which are standard two-way fixed effects specifications and the interaction weighted estimation method proposed by Sun and Abraham [87]. This method provides estimates that are robust to treatment effect heterogeneity across cohorts. To estimate parameters following the robust approach, I first define cohorts of school districts that implemented SC in a given year. Second, I estimate cohort-specific average treatment effects using a two-way fixed effects specification that interacts relative period indicators with cohort indicators. Third, I calculate the weights for each relative time period using the sample shares of each cohort in the given period. Fourth, I combine the cohort-specific average treatment effect estimates and the weights to calculate the "interaction-weighted" estimate. The main assumption underlying the empirical identification strategy is that without treatment time trends group specific outcomes would develop in parallel [88]. Considering that SC was implemented in a nation-wide rollout, school districts are classified as treated once SC has been implemented. Under the parallel trend assumption, comparing Hispanic students' achievement in school districts that have implemented SC with that in school districts that have not yet implemented SC provides an estimate of the causal effect of SC on the change in Hispanic students' SDLA [89]. Estimates for white and black students are interpreted in an analogous fashion. Considering that I cannot formally test the identification assumption, I plot pre-and post-trends in test scores to assess whether trends are parallel. Given the staggered rollout of SC, I have more information from years prior to the activation of SC for some counties than for others. Still, the SEDA data provide more limited information from years prior to the implementation. The dynamic difference in differences model underlying the method by Sun and Abraham is given by the following equation: Y dt 1<unk>4 a <unk> P 4 t1<unk>42 d <unk> t I pre d;t<unk> t <unk> P 7 t1<unk>40 d <unk>t I post d;t<unk>t <unk> X 0 dt g <unk> m d <unk> r t <unk> dt ;<unk>2<unk> where Y dt stands for the dependent variable (i.e., English language arts or math SDLA among Hispanic students). Equivalent models are estimated for white and black students. The suffix d stands for school district and t represents the year. The coefficient <unk> -<unk> identifies the impact of SC <unk> years prior to the implementation relative to the reference category of one year prior to the implementation of SC. We observe school districts up to four years prior to the implementation of SC. The coefficient <unk> +<unk> identifies the impact of SC <unk> years after to the implementation and follows school districts up to seven years after the implementation. X 0 is a vector of covariates that varies at the school district level and over time. It includes the same set of controls as provided in Eq 1. The specification also includes school district fixed effects (<unk> d ) and year fixed effects (<unk> t ). Models are weighted using precision weighting. A threat to the difference in differences assumptions is that the groups respond to the announcement of treatment. For instance, Hispanic families may relocate in response to the implementation of SC. Previous research shows that Hispanic families relocate to another US state when local immigration enforcement increases [66,67]. However, considering that the SC program was implemented nation-wide, relocation because of it is not expected. S2 Table also shows that the proportion of Hispanic students in the school district does not change as a result of the program's activation. This suggests that Hispanic families did not necessarily relocate to other parts of the US in response to SC. My second approach is to focus on Hispanic students' SDLA and to estimate stratified models across census regions, rural and urban areas, and the proportion of likely undocumented migrants in the state. As in the approach described above, this analysis builds on difference in differences models and is based on the specification in Eq 1. Results are obtained using the Sun and Abraham method. However, rather than analyzing the entire US, this approach assesses whether there is heterogeneity in the impact of SC across local contexts. First, information on the state of residence allows me to differentiate between four census regions: Northeast, Midwest, South, and West. Second, I use information from the SEDA data on the proportion of students in the school district who are in urban, suburban, town, or rural locale schools. Third, I incorporate data from the CPS and calculate quartiles on the proportion of likely undocumented migrants at the state-level. These analyses also include a placebo test, assessing educational outcomes among Hispanic students in sanctuary jurisdictions, where the activation of SC had a reduced impact. --- Empirical findings --- Difference in differences I begin by assessing the impact of SC on Hispanic, white, and black students' test scores in the entire US. Fig 2 shows results from difference in differences models based on Eq 1. The models are estimated separately for Hispanic, white, and black students, and for test scores in English language arts and math. Each point estimate reports the impact of SC by comparing changes in the SDLA of adopting school districts with contemporaneous changes in school districts that had not yet implemented the program. Five models with different sets of fixed effects are presented. The first model includes year and state fixed effects and the second model adds state-by-year fixed effects. The third model includes year and county fixed effects and the fourth model includes year and school district fixed effects. The abovementioned models are estimated using standard two-way fixed effects specifications. The fifth model is estimated using the Sun and Abraham method and includes year and school district fixed effects. All models control for other policies that were implemented over the time period analyzed. These include E-verify, Omnibus Immigration Laws, state and county-level 287(g) agreements, and sanctuary jurisdictions. The presented models also account for school district level controls, which refer to the proportion of students who have free or reduced-price lunch, are in Special Education, are English Language Learners, and the SES composite score. Corresponding estimates and standard errors are provided in S3 and S4 Tables. The estimate for SC is negative for Hispanic students' SDLA in English language arts. In the first two models, we observe an estimate of -0.030. Introducing county fixed effects in model 3, reduces the estimate to -0.012 likely due to the fact that unobserved characteristics at the county level account for part of the association. The estimate remains similar in size in model 4, which further introduces school district fixed effects. Model 5 also includes school district fixed effects and accounts for potential heterogen

Prior research suggests that disruptive events, such as shocks induced by family instability, neighborhood violence, or relocation, tend to be detrimental for children's educational outcomes, but findings are heterogeneous depending on the type of event. Limited evidence is available on how shocks resulting from immigration enforcement impact educational outcomes among targeted minority groups. This study contributes to the literature by assessing how a policy implementation in the US-Secure Communities-is related to the school district level achievement of Hispanic students. The Secure Communities program is a national level immigration enforcement policy that was rolled out on a county-by-county basis. The program has increased the risk of deportation and led to rising apprehension and insecurity among undocumented migrants and the wider Hispanic community. Using detailed information on the implementation of Secure Communities, data from the Stanford Education Data Archive, and the Current Population Survey, this study estimates dynamic difference in differences exploiting regional variation in the timing of the policy change to assess its impact on educational outcomes. Results show that the activation of Secure Communities is negatively associated with Hispanic students' subsequent English language arts achievement, while white and black students' achievement does not change. Findings further suggest that Hispanic students living in the South, rural areas, and areas with high proportions of likely undocumented migrants are disproportionately impacted by the program's activation. Whereas, Hispanic students in sanctuary jurisdictions, which reduce the likelihood of deportation, are not impacted. These findings indicate that immigration enforcement can have negative consequences for educational and social inequalities in the United States.

effects. The abovementioned models are estimated using standard two-way fixed effects specifications. The fifth model is estimated using the Sun and Abraham method and includes year and school district fixed effects. All models control for other policies that were implemented over the time period analyzed. These include E-verify, Omnibus Immigration Laws, state and county-level 287(g) agreements, and sanctuary jurisdictions. The presented models also account for school district level controls, which refer to the proportion of students who have free or reduced-price lunch, are in Special Education, are English Language Learners, and the SES composite score. Corresponding estimates and standard errors are provided in S3 and S4 Tables. The estimate for SC is negative for Hispanic students' SDLA in English language arts. In the first two models, we observe an estimate of -0.030. Introducing county fixed effects in model 3, reduces the estimate to -0.012 likely due to the fact that unobserved characteristics at the county level account for part of the association. The estimate remains similar in size in model 4, which further introduces school district fixed effects. Model 5 also includes school district fixed effects and accounts for potential heterogeneity in the treatment effect across cohorts using the Sun and Abraham method. The estimate indicates a change of -0.024 in English language arts SDLA. Among white and black students, there is no significant decline in English language arts achievement following the implementation of SC. For math, we observe no relationship between the implementation of SC and Hispanic, white, or black students' achievement. The estimate for Hispanic students is negative but non-significant. Regarding the size of the coefficient, Hispanic students' SDLA in English language arts increased from -0.342 in years prior to the program to -0.260 in years after the implementation (Table 2). This indicates a nation-wide increase of 0.082 points. The negative estimate of SC suggests that the program led to a reduced improvement of Hispanic students' SDLA. In relation to the increase of 0.082, the estimated coefficient of SC at -0.024 within school districts is non-trivial. Fig 3 provides pre-and post-trends for the difference in English language arts and math scores between treated versus control school districts based on Eq 2. The x-axis shows years in relation to the implementation of SC. Four years of lead and seven years of lag are included. T stands for the year of the implementation. In year T, SC has been activated, while it has not yet been activated in year T-1. Non-significant trends in SDLA in years prior to the adoption of SC indicate parallel trends before the policy implementation and are used to assess the parallel trends assumption. The models are estimated separately for Hispanic, white, and black students, and for test scores in English language arts and math. They include school district and year fixed effects and are estimated using the Sun and Abraham method. The models further include the same set of control variables as those presented in Fig 2 In Fig 3, panel A shows that Hispanic students' English language arts scores are similar in all years prior to the implementation of SC. These estimates indicate a stable pre-trend. After the implementation of the program, we observe a decline in SDLA. The impact of SC becomes stronger in the first years after implementation and then stabilizes, which is plausible considering that the policy may progressively expose children to stress as they gain awareness about its consequences before the impact stabilizes. Among white and black students, we observe no sizeable change in SDLA relative to the policy implementation. Panel B shows that math scores decline prior to the activation of SC among Hispanic students. This indicates that the parallel trends assumption does not hold for math. Among white students, we observe relatively high math scores in the first years after the policy implementation. Subsequently, point estimates indicate decreasing scores but confidence intervals become quite large. Among black students, we do not observe a substantial decrease in math scores after the activation of SC. Taken together, the results presented in Fig 3 indicate that the parallel trends assumption does not hold for math, but roughly holds for Hispanic students in English language arts. Still, we need to be cautious when interpreting the coefficients as causal estimates. --- Stratified analyses We assume that Hispanic students are more exposed to SC, because they are more likely to be in close social proximity to undocumented migrants and to experience statistical discrimination than white and black students. However, considering North-South and rural-urban divides in anti-immigrant sentiments and political values, it may be that Hispanic students are more strongly impacted in the South and rural areas. In contrast, sanctuary jurisdictions reduce the likelihood that undocumented migrants are arrested and deported by ICE. Hispanic students in these cities and counties are, thus, expected to be less exposed to SC. Due to variation in the undocumented population, we further expect that Hispanic students are disproportionately impacted by immigration enforcement in states with high proportions of undocumented migrants in the adult population. In the following, I incorporate information from the CPS to calculate the proportion of likely undocumented migrants in the adult population age 20 and above at the state level between the years 2009 and 2018. Fig 4 provides a scatterplot of the share of undocumented migrants in the adult population based on the CPS data on the y-axis. The figure reveals considerable variation in the proportion of undocumented migrants across states. For instance, the proportion undocumented is highest in California where 11 percent of the adult population are likely undocumented migrants, followed by Texas and Arizona with 8 percent. The lowest proportions are observed in states, such as West Virginia and Montana with 0.17 and 0.35 percent, respectively. I calculate quartiles where the top quartile comprises states with the highest proportion undocumented (indicated by dark gray circles) and the bottom quartile comprises states with the lowest proportion undocumented between 2009 and 2018 (indicated by light gray circles). Fig 5 provides estimates from stratified models by census region, rural and urban area, and the proportion likely undocumented. The top panel shows that SC is more strongly related to Hispanic students' English language arts scores in the South than in other regions. The second panel assesses the impact of SC across sanctuary jurisdictions, urban areas, suburbs, towns, and rural areas. We find that the impact is negative and large in rural areas and towns. In sanctuary jurisdictions, SC is not related to Hispanic students' English language arts scores. This goes in line with the expectation that the impact of SC is considerably smaller in sanctuary jurisdictions. The results in the bottom panel show that SC is negatively related to Hispanics students' English language arts scores in states that fall into the top quartile in terms of the proportion of probabilistic undocumented migrants. In the other quartiles, where the proportion of undocumented is lower, we do not observe an impact of SC on Hispanic students' English language arts scores. Estimates and standard errors are provided in S7 Table. Results for math are presented in S2 Fig. About a third of Hispanic students in states with a high proportion of probabilistic undocumented migrants live in the South and the West, respectively. S3 Fig disaggregates census regions into census divisions to gain some insight into how results in the top and bottom panels relate. These estimates suggest that the negative impact observed in states with a high proportion of undocumented migrants is largely due to negative associations observed in the West South Central division (Census region South), which includes Texas, and states that are in the Mountain division (Census region West). The Mountain division includes Arizona, Nevada, and New Mexico. In sum, these results indicate that the impact of SC is especially pronounced in the West South Central and Mountain divisions, towns and rural areas, and in states with a high share of undocumented migrants. --- Sensitivity analyses I have assessed whether the results hold for alternative ways of estimating the proportion of undocumented migrants and alternate sample restrictions. I assess the proportion of likely undocumented migrants among Hispanics (rather than the total population) and the proportion of likely undocumented migrants in the foreign-born population. These lead to less stable estimates, which is expected given the relatively small Hispanic and foreign-born population in some states. Results based on alternate sample restrictions also show that the findings for Hispanic and white students are robust, while the results for black students are relatively noisy. I have further estimated the models with different sets of control variables (including information on the poverty rate and the proportion of Hispanic students in the school district), which leads to very similar results. The main analyses estimated the relationship between SC and test scores for Hispanic, white, and black students separately. S4 Fig instead provides estimates of the Hispanic-white and black-white achievement gaps. The top panel shows that the Hispanic-white achievement gap in English language arts increases with the activation of SC in the first two models. However, once we include county or school district fixed effects, the estimate is attenuated. This indicates that unobserved characteristics at the county and school district level are important for changes in the achievement gap. Estimates for the black-white achievement gap in English language arts indicate no change following the implementation of SC. The bottom panel shows that the Hispanic-white and the black-white achievement gaps in math do not change following the implementation of SC, with the exception of the estimate for the Hispanic-white gap which increases in model 2. Again, the estimate is attenuated when we include county and school district fixed effects. In sum, we find no clear association between SC and racial achievement gaps at the school district level, while the main analysis revealed a negative association between the policy implementation and Hispanic students' achievement. Taken together, these results indicate that SC likely played a notable role for time trends in Hispanic students' educational outcomes, but that other factors were important for achievement gaps. --- Conclusion and discussion This study analyzed whether the shock resulting from the implementation of Secure Communities can be linked to changes in Hispanic students' educational outcomes. There are three main findings. First, SC has a negative impact on Hispanic students' school district level achievement in English language arts. In contrast, white and black students' achievement does not change following the activation of SC. This is in keeping with the view that white and black students are less strongly impacted, as the policy primarily targeted the Hispanic community. The role of SC is substantial in relation to time trends in Hispanic students' achievement, but modest for racial achievement gaps. When compared to other studies, the negative impact is similar in size for test scores [23], though previous studies find somewhat larger estimates for the achievement gap likely due to different sample restrictions [24]. For Hispanic students' math scores, no association is observed. This may be due to the fact that Hispanic students' math skills are less contingent on every-day contact with non-Hispanic neighbors than their English skills, as math can be understood and practiced without firm knowledge of English [20,53]. Still, the finding that the increase in Hispanic students' school district level achievement in English language arts is slowed down by the implementation of SC indicates that the shock resulting from SC has a negative, rather than a positive, impact on Hispanic students' educational outcomes. Second, stratified results reveal that Hispanic students living in the South, rural areas, and in states with high proportions of likely undocumented migrants are disproportionately impacted by SC. The observed patterns provide some insight into potential mechanisms underlying the associations. Anti-immigrant sentiments, which are especially high in the South and rural areas, may hamper integration [61,62,64]. In the rural Midwest, Maya Guatemalan migrants say that they prefer staying indoors as they are scared of being apprehended when driving or walking outside [53]. By reducing interaction with the wider community, immigration enforcement may impact Hispanic students' English progress. Although the majority of Hispanic students were born in the US and speak some English, 30 percent of UScitizen children with undocumented parents are estimated to have limited English proficiency between ages 5-11, as compared to 18 percent among US citizen-children of migrants [20]. Likewise in the South, undocumented families' choice of residence is often not only constrained by financial factors but also by assessments of the level of law enforcement patrols [90]. Given that Hispanic students are more likely to be in social proximity to undocumented migrants in areas with high proportions undocumented, findings also corroborate the assumption that Hispanic students are more exposed to SC due to their higher propensity to have undocumented parents, neighbors, or other community members [15]. Third, findings reveal that there is no association between SC and Hispanics students' test scores in sanctuary jurisdictions. This serves as a placebo test, showing that areas in which the likelihood of deportation was lower were not impacted by SC to the same extent. These cities and counties tend to perceive immigrants as valued members of the community and often provide infrastructure support programs for immigrants aiming to alleviate poverty [64]. The policy implementation analyzed in this paper is specific to the US, but increases in deportations can also be observed in Canada and many European countries [91]. A study from Switzerland indicates that long asylum processes are related to a decrease in the subsequent employment rate of refugees [92]. The authors argue that waiting in limbo fosters psychological discouragement among refugees. In the UK, asylum seekers who receive a negative asylumapplication decision also note that this impacts their quality of life and is linked to feelings of distress, isolation, and statelessness [93]. This evidence suggests that policy-related shocks and apprehension status can have negative consequences for migrants' employment and health outcomes in Europe, which may spill over to their children, as in the US context. This study has several limitations. The SEDA data are published in aggregated form due to privacy concerns. I use the lowest level of aggregation available, the school district, as the unit of analysis. The SEDA data are also disaggregated by race, and I study the performance of Hispanic, white, and black students separately. Incorporating data from the CPS further allows me to stratify the sample by the proportion of probabilistic undocumented migrants in the area. Still, when inferring individual-level processes from aggregate data, results need to be interpreted with care. Furthermore, since SC was eventually implemented nation-wide, there is no truly unexposed area. Sanctuary jurisdictions provide the closest proxy to unexposed areas. The study contributes to the literature by illustrating that shocks induced by SC have a negative impact on Hispanic students' English language arts performance. This is particularly pronounced in the South, rural areas, and areas with high proportions of undocumented migrants. These findings suggest that policy-related shocks can have unintended consequences for educational inequalities in the US, potentially impacting longer term economic trajectories of the children of migrants as well as social inequalities. In future studies, efforts should be directed at observing the mechanisms at the individual level. Educational achievement may decrease among Hispanic students due to increased stress and anxiety experienced among this group and/or result from reduced interaction with non-Hispanic neighbors and lower investment in the community. It will also be worthwhile to further explore differences in the impact of immigration enforcement by subject. --- The data used in this --- Supporting information S1 Standard two-way fixed effects specifications are estimated with clustered standard errors at the county level and precision weights. p <unk> 0.05, p <unk> 0.01, p <unk> 0.001 (two-tailed). (PDF) S3 Table. Estimated associations between the implementation of Secure Communities and school district level English language arts achievement among (A) Hispanic, (B) white, and (C) black students. Data from SEDA 2009-18 and DHS. Precision weighted estimates are based on Eq 1. Models 1-4 show results obtained from standard two-way fixed effect specifications. Model 5 shows results obtained using the method outlined by Sun and Abraham (2021). Clustered standard errors at the county level are in parentheses. a Considering that these are statelevel policies, there is no variation once we include state-by-year fixed effects. p <unk> 0.05, p <unk> 0.01, p <unk> 0.001 (two-tailed). (PDF) S4 Table. Estimated associations between the implementation of Secure Communities and school district level math achievement among (A) Hispanic, (B) white, and (C) black students. Data from SEDA 2009-18 and DHS. Precision weighted estimates are based on Eq 1. Models 1-4 show results obtained from standard two-way fixed effect specifications. Model 5 shows results obtained using the method outlined by Sun and Abraham (2021). Clustered standard errors at the county level are in parentheses. a Considering that these are state-level policies, there is no variation once we include state-by-year fixed effects. p <unk> 0.05, p <unk> 0.01, p <unk> 0.001 (two-tailed). (PDF) S5 Table. Estimated associations between the implementation of Secure Communities and school district level (A) English language arts and (B) math achievement among Hispanic, white, and black students relative to the implementation year. Data from SEDA 2009-18 and DHS. Precision weighted estimates are based on Eq 2. Results are obtained using the method outlined by Sun and Abraham (2021). Clustered standard errors at the county level are in parentheses. p <unk> 0.05, p <unk> 0.01, p <unk> 0.001 (two-tailed). (PDF) S6 Table. Weights for each relative time period obtained in the Sun and Abraham (2021) method. Data from SEDA 2009-18 and DHS. After sample restrictions, there are four cohorts: 2010, 2011, 2012, 2013. The first treated cohort is used as the control group. (PDF) S7 Table. Estimated associations between Secure Communities and school district level English language arts achievement among Hispanic students across (A) census regions, (B) rural-urban areas, and (C) the proportion of likely undocumented migrants. Data from SEDA 2009-18, DHS, and CPS. Precision weighted estimates are based on Eq 1. Results are obtained using the method outlined by Sun and Abraham (2021). Clustered standard errors at the county level are in parentheses. Urban areas exclude sanctuary jurisdictions. a No policy identified in the states. p <unk> 0.05, p <unk> 0.01, p <unk> 0.001 (two-tailed). (PDF) --- Author Contributions Conceptualization: Rosa Weber. Data curation: Rosa Weber. Formal analysis: Rosa Weber. Investigation: Rosa Weber. Methodology: Rosa Weber. Validation: Rosa Weber. Visualization: Rosa Weber. Writing -original draft: Rosa Weber. Writing -review & editing: Rosa Weber.

Prior research suggests that disruptive events, such as shocks induced by family instability, neighborhood violence, or relocation, tend to be detrimental for children's educational outcomes, but findings are heterogeneous depending on the type of event. Limited evidence is available on how shocks resulting from immigration enforcement impact educational outcomes among targeted minority groups. This study contributes to the literature by assessing how a policy implementation in the US-Secure Communities-is related to the school district level achievement of Hispanic students. The Secure Communities program is a national level immigration enforcement policy that was rolled out on a county-by-county basis. The program has increased the risk of deportation and led to rising apprehension and insecurity among undocumented migrants and the wider Hispanic community. Using detailed information on the implementation of Secure Communities, data from the Stanford Education Data Archive, and the Current Population Survey, this study estimates dynamic difference in differences exploiting regional variation in the timing of the policy change to assess its impact on educational outcomes. Results show that the activation of Secure Communities is negatively associated with Hispanic students' subsequent English language arts achievement, while white and black students' achievement does not change. Findings further suggest that Hispanic students living in the South, rural areas, and areas with high proportions of likely undocumented migrants are disproportionately impacted by the program's activation. Whereas, Hispanic students in sanctuary jurisdictions, which reduce the likelihood of deportation, are not impacted. These findings indicate that immigration enforcement can have negative consequences for educational and social inequalities in the United States.

INTRODUCTION Racial and ethnic discrimination has become a phenomenon that is still a problem in the world, especially in the school environment. According to a survey conducted by Komnas HAM entitled "Community Research Survey on Efforts to Eliminate Racial and Ethnic Discrimination in 34 Provinces" found that 81.9% of respondents were more comfortable living in the same family. 82.7% of respondents said they were more comfortable living in the same racial environment. While 83.1% said it was more comfortable living with the same ethnic group. So that the results of the survey above, Komnas HAM researchers said that there would be the potential for racial and ethnic discrimination to have a high probability. (Baskara, 2018) Culturally resonant education is the process of cultivating a respectful, sincere, and tolerant way of life for cultural diversity living in the midst of a plural society. With multicultural education, it is hoped that there will be resilience and mental flexibility of the nation facing the clash of social conflicts. In addition, what is also important is that multicultural education is intended so that all students with all differences get equal education (Raharja, 2010). Cultural responsive education is a theoretical and practical educational model that not only aims to improve student achievement, but also helps students accept and strengthen their cultural identity. According to Ladson-Billing (1995) there are three propositions of culturally responsive education, namely: first, learners achieve academic success; Second, students are able to develop, and have cultural competence and students build critical awareness so that they can participate in overhauling the unjust social order. So that culturally responsive education is also constructivistic learning (Alexon 2010). Strategy is a comprehensive plan in integrating, integrating all capabilities in achieving long-term goals in realizing a success that has been designed. Meanwhile, according to Miller (2007) the strategy will look easy if we have determined the goals we designed. Wheelen and hunger (2003) strategy is a series of structured decisions and actions in determining the performance of a school in the long run. So it can be concluded that strategy becomes a step to the process in determining the direction that a teacher will achieve in learning. In the opinion of Djamarah and Zain (2010) that classroom management is a task that teachers have and must be considered to achieve effective and efficient learning. Suryani and Agung (2012) also said that classroom management is an effort to prepare optimal conditions so that the teaching and learning process or activities can run in an orderly and conducive manner. So that in this case the teacher has the task to create, maintain and manage the system in the classroom and students can take advantage of their abilities both individually and in class groups. Grouping heterogeneous classes can be a way to realize multicultural education, because heterogeneous classes do not separate the sexes in the learning and teaching process (Mael, 1998). This is supported by Ramanda's research (2017) saying that heterogeneous schools have higher social maturity than homogonal schools. Rytivaara (2011) said that using heterogeneous groups against students in the classroom aims to osial or recognize behavior between students. Skills in managing classes can have success in learning, to get an effective class requires skills in an effort to manage the class, so that a teacher can have a positive influence on the learning process. Good learning and teaching methods also require knowledge in the diversity of abilities possessed by students so that a teacher can design his learning regularly and effectively. (Popham, 2011). The potential and diverse characteristics in individual students have diversity that must be considered in managing the classroom, such as motivation, concentration, critical thinking of students to be a benchmark for success in the learning process. (Malawi, 2016). Asmani (2013) also said that there are two indicators that can be used as a benchmark for the success of the learning process, namely absorption of lessons and changes in student behavior. Concentration is one of the factors that affect the low absorption of students. To create an optimal learning environment by providing various facilities for various student learning activities so that effective and efficient learning occurs in accordance with the social, emotional, intellectual environment of students in the classroom so that learning objectives can be achieved. Milner IV & Tenore (2010) state that teachers who have knowledge about themselves and their students have a foundation on which rapport can be built. One of the teachers in the study stated that it is important to understand the power relationship between students, using this understanding to help students who have the power to use it in a positive direction so that they can help create a commitment to learning in the classroom. Teachers and students also have the opportunity to recognize and contradict their prejudices (Banks, 2005;Gay Geneva, 2002). Multicultural education more than just comes through course content and diversity. It contains advancing a comprehensive climate in the classroom and a sense of similarity among learners and facilitating student learning with a variety of assessment and instructional techniques (Gay, 2004). Kraft (2019) suggests that teachers have an impact on standardized tests and competencies such as growth mindset, perseverance, and effort in the classroom. For students from traditionally underserved racial and ethnic groups, assignments to teachers of the same race or ethnicity can further increase students' opportunities for learning (Milner, 2011). With same-race teachers, a student may experience higher expectations, more supportive relationships, culturally relevant instruction, or role models, all of which can improve their academic and nonacademic performance in school (Ladson-Billings, 1994;Milner, 2011). Based on the discussion above, cultural diversity between teachers and students is a complete problem in realizing an effective classroom. This study aims to determine teacher strategies in managing heterogeneous classes and also become a more comprehensive look at the literature on classroom management and diversity and heterogeneous classes. --- METHOD This research uses a type of research in the form of library research. Literature studies are related to theoretical studies and some references that will not be separated from scientific literature. The steps of literature research that will be carried out in this study include; 1) prepare equipment, 2) compile a working bibliography, 3) manage time, 4) read and make research notes. (Sari &;Asmendri, 2020). In this study, data sources obtained from relevant literature such as books, journals or scientific articles related to the selected topic. The data collection technique used in this literature research is looking for data on things or variables in the form of notes, books, papers or articles, journals and others, The research instrument used by researchers is a check-list of research material classification based on the focus of the study, writing schemes / maps, and research record formats. (Sari &;Asmendri, 2020) The data analysis technique used in this literature research is the content analysis method. This aims to maintain the permanence of the assessment process and prevent and overcome misinformation (human misunderstandings that can occur due to lack of knowledge of researchers or lack of literature authors) so that checks are carried out between libraries and re-read the literature. This research report is prepared on the principle of simplicity and ease. This principle was chosen considering the limited ability of researchers who have not been able to conduct an indepth and more detailed literature review. (Sari &;Asmendri, 2020). --- RESULT What we have tried to do in this article is present culturally responsive classroom management practices. Its main objective is as a practice of heterogeneous classroom management strategies and to extend the idea of culturally responsive classroom management taking into account the ethnic backgrounds of teachers and students and the context of the study. (C. S. Weinstein et al., 2004a) outlines several principles of culturally responsive classroom management: recognition of the teacher's own ethnocentrism; knowledge of student culture; an understanding of the broader social, economic, and political systems in education; appropriate management strategy; and the development of caring classrooms. Strategies emerging in this study include the importance and centrality of teachers in the classroom. (a) understanding equality and equality (b) understanding of self in relation to others (c) Teachers integrate into the world of students. --- An understanding of equality and equality The idea of equality and equality has passed through many debates in the social and public policy spheres, but in many contexts there seems to be no clear idea of what justice means and equality. One of the principles in this study is to expand the conception of culturally responsive classroom management is the need for teachers to understand equality and equality in the learning process in the classroom to students. According to Espinoza (2007) similarity is a need that is shared equally, and does not doubt every individual who is different in culture or potential so that it aims for the same achievement. Equality is an opportunity for every individual to obtain his rights and equality depends on the average required by social groups. The concept of equality is associated with fairness or fairness in the provision of education or other benefits and takes into account individual circumstances, whereas equality usually connotes equality in treatment by affirming the fundamental or natural equality of all people (Corson, 2001). The emergence of social inequalities such as gender, culture, potential in society is a relay from one generation to the next through an educational process that is not based on justice and equality. Therefore, there needs to be an effort to open the insight and awareness of teachers and students on the importance of equality and justice as one of the important elements to form a conducive classroom order (Tobrono, 2007). The diversity that exists in students such as ethnicity, gender, religion and the ability to grasp information provided by the teacher must also be considered, because when in class, students have complex individuals such as postence, personality, characteristics and even awareness in learning. Such heterogeneous classes must be solved with an understanding of the equality and equality of teachers in the teaching and learning process (Aziz, 2012). Expectations in the continuity of heterogeneous and multicultural classes are mutual respect and respect for each other regardless of the background and identity inherent in students (Hardy &;Woodcock, 2015). So that differences in cultural diversity and students are one of the things that must be learned and understood by teachers and must be able to manage them to form an effective learning environment in their classrooms (Franklin, 2005). Furthermore, Hardin (2004) also said the role of teacher sensitivity is very important to the diversity of students. Based on the above observations, it shows that culturally responsive class management by paying attention to equality and equality can be a solution of heterogeneous classes in realizing heterogeneous classes that are effective, conducive, friendly to differences. Therefore, culturally responsive classroom management can be a strategy that must be realized and implemented to meet the specific and collective needs of all students. --- Self-understanding in relationships to others Conflict in a heterogeneous classroom cannot be avoided by all teachers and students, especially related to the teaching and learning process (Somantrie, 2011). So that students can be a control when they see their teachers who cannot recognize a difference that exists in the classroom when in a lesson there is an obstacle that has a less beneficial impact between teacher and student, one example is when the teacher uses a teaching method and students find it difficult to grasp what is conveyed by the teacher, and often also the teacher's understanding of problems in the classroom such as bullying must also be understood so that It does not become a conflict between students or psychological students. The relationship between teachers and students in the teaching and learning process is a factor that greatly determines the success of teaching and learning, if the teacher cannot recognize the abilities and potential of students in general, it will be an inhibiting factor. Culturally responsive classroom management represents teachers' understanding of themselves in relation to others. The important thing here is that it is not enough for teachers to understand themselves; Critical reflection of the self is essential, but it is only the first step in developing culturally responsive classroom management. Teachers should seek to understand themselves in relation to their students, their parents, and their student communities. Recognize the similarities and differences between his students and himself. In fact, students make an effort to know themselves as well as others (C. S. Weinstein et al., 2004b). --- Teacher understanding to blend into the student's world The process of interaction between students and educators in a learning environment is called learning (Suardi, 2018). Waty (2017) said the importance of social interaction in the learning process in order to create an active class, one of which underlies social interaction is by building good communication between teachers and students. Communication between individuals and groups can be used by teachers in approaching their students to build good relationships so as to help the teaching and learning process that is easily understood by students. Rosarian &;Dirgantoro 2020). Interaction between students and teachers can be built when seeding learning materials while building communication between students and teachers. The communication will naturally build a relationship between the two parties. Because teachers now not only deliver a material, but teachers also become facilitators during the teaching and learning process. (Yasol) For example, teachers offer learning models that will take place in the next semester to students, such as discussions, presentations in class, or learning outside the classroom. But if students are given the freedom to choose, they must be responsible for what they have chosen. According to research by Rahmawati & Suryadi (2019), teachers as facilitators have an influence on the effectiveness of student learning, so entering the world of students can give them freedom to choose, but teachers must also supervise if students deviate from existing ethics and regulations. C. Weinstein et al. (2003) said that in realizing cross-cultural interaction, teachers must be able to have knowledge to know the cultural background that their shiva has. Teachers should also provide an understanding of the cultural heritage of their group with a view to imparting lessons of good values. Therefore, it is important for teachers to have strategies to always blend in with students with the aim of getting to know their culture, social status and potential so that they can obtain diversity knowledge that students have in heterogeneous classes (Pebriansyah et al., 2020). --- CONCLUSION Based on the results of literature research conducted by researchers, researchers get strategies in managing heterogeneous classes by examining culturally responsive classroom management, so that researchers get the first strategy, teacher understanding of equality and equality. Second, self-understanding in relation to others. Third, teachers blend into the student's world. Of the three strategies, researchers only explain heterogeneous class management strategies in general, the limitations of this study are that they cannot explain the practice of problems directly to heterogeneous classes, the librarian skin data obtained by researchers cannot explain the problems that exist in heterogeneous classes, so there needs to be further research to clarify heterogeneous classes specifically.

The cultural complexity of students in urban areas has influenced schools in the teaching and learning process. Classroom management becomes a concern for educators in managing a heterogeneous class. This becomes one of the important tasks for teachers in creating and maintaining an effective classroom condition. On the other hand, if the heterogeneity of students is not managed properly, it can become a source of various problems in the class. Starting from bullying, uncontrolled learning atmosphere, even attitudes of antipathy towards diversity that trigger other negative attitudes such as discrimination and injustice. This study aims to determine the teacher's strategies in managing a heterogeneous class. This research method uses a qualitative approach and applies the method of library research using sources from Google Scholar, Sage, Garuda, and books. The analysis technique used in this research is content analysis. The results of this study obtained strategies for managing a heterogeneous class, namely 1) the Teacher's understanding of equality and equity, 2) Self-understanding in relation to others, 3) the Teacher blending into the students' world. So when teachers face a heterogeneous class, they can use these three strategies to create an effective and conducive heterogeneous class.

The lives of millions of displaced individuals (including refugees, internally displaced persons, and stateless individuals) are characterized by exclusion, xenophobia, and global apathy. Displaced persons may live in formal or informal camps, or in other settings with few resources and unreliable access to essential commodities and services. Forced displacement, both in the short term and protracted, therefore has complex impacts on health. 1,2 Factors that affect the health of forcibly displaced individuals can be split broadly into two realms: the physical and the social. Displaced communities, both during migration and when they are not moving, are often in locations that have limited access to essential health services. The geographic locations also often increase exposure to toxic and unstable environments, which can have immediate and chronic health impacts. The social attributes of displacement include legal status, nationality, and other socioeconomic factors that can likewise influence access to health services (and health itself), and force communities to live in a state of fear and anxiety. They may also affect the geographic settings in which displaced persons reside. As humanitarian crises grow around the world and new humanitarian crises appear in South Asia, the Americas, Africa, and Europe, so do the camps that are sometimes home to displaced individuals for decades. [3][4][5] In many camps across the globe, there are multiple generations that have been born and spent their entire lives in camps. The populations of these camps are neither fixed nor static, which can strongly influence epidemiological dynamics. [6][7][8][9] Intergenerational factors (arrivals versus those who are born in the camps) create heterogeneity that leads to environmental changes, including those at microbial scales, not excluding changes in the gut microbiome. Changes in demography and environment are rarely studied from an integrated ecological and/or evolutionary lens, and are rarely part of the displaced individuals' health landscape. This absence of scholarship and research allows for the continuation of siloed approaches and creates blind spots in understanding disease dynamics, emergence of new health challenges, provision of long-term care, and inefficient delivery of health services. We propose a more integrated approach to understand health challenges of the forcibly displaced-one that is multidisciplinary and multi-scale, and analyzes the communities and their environment from an ecological framework. We define ecology as the interaction between organisms and their environment at multiple spatial scales. This includes interactions among microbes; among humans, disease vectors, and microbes; and between the physical landscape and humans, disease vectors, and microbes; and does not exclude noncommunicable diseases. Current approaches, even in the context of a One Health framework, can sometimes be siloed or at a single spatial or temporal scale. The common siloed approach through which many services are currently provided leads to inefficiencies in the provision of services, varying temporal coverage based on funding cycles and incomplete pictures of health. Environmental, ecological, and evolutionary considerations are often not even considered, posing further gaps in our understanding of the health of populations within such camps. Though we have information on a higher, general level, we know few specifics about disease dynamics and the relationship of people and their environment in complex humanitarian emergencies. 10 Siloed approaches neglect the reality that all components of humanitarian management have important health considerations when using an ecological lens. We therefore aim to understand the dynamic ecologies that connect vertically (i.e., from the microscale to the macroscale) and horizontally (between individuals, between environments, and between people and their environments) in forced displacement camps. Understanding the ways that upstream factors can influence human health and microbial communities is useful in diagnosing and preventing poor health outcomes. Likewise, this type of big-picture lens is useful from a public health standpoint. A public health system that focuses on all health outcomes-communicable and noncommunicable, including mental health-and likewise considers the multidirectional health implications of human-biophysical environment interactions is useful for preventing disease. For those who are planning or administering camps, having a holistic view of the linkages between humans and their environments may lead to better planned and managed camps. Camp locations could often be better chosen, with health implications for the dwellers, for other communities in the same region, and for the environment. An approach that prioritizes local knowledge will lead to better understanding of the local ecological systems. To achieve this, we argue for considering displaced individual environs as adaptive ecosystems, with interactions at multiple temporal and spatial scales. An ecological lens challenges the siloed approach and allows for benefiting from a new understanding and new tools in the discipline. It helps us better understand the evolutionary and environmental pressures in the camps, and enables us to improve the lives of those who live in them. --- ENVIRONMENTAL AND SOCIAL ECOLOGIES Humans, animals, and the environment influence each other in a codependent manner that varies with time and space, and is affected by both local and global factors. This multiscale dependence, although well recognized by those who are forced to live in the camps-has not been fully understood in research circles. To illustrate the importance of a multiscale dynamic ecological model in understanding the health of displaced individuals in camps, we offer Figure 1, which demonstrates and visualizes some of the ecologies and interactions occurring between the environment and health of displaced populations. Social factors also play a critical role in the camp ecosystem and include the camp's and individuals' legal status, discrimination, mobility, employment, and host community relations. The camp's legal status affects the provision and availability of services throughout the camp, as well as its development and care. The legal status of displaced individuals influences movement out of the camp, including being unable to seek external health care or obtain employment. Employment that is obtained may be in environments where certain diseases are more common. Differences in race, gender, legal status, and other demographic variables can lead to discrimination, inequality, and violence. These have direct and indirect effects on the health of displaced persons, and are compounded further by the individual's vulnerable social and legal position, and lack of legal accountability for perpetrators. Studies have also found an association between gender inequality and environmental degradation. 11 These social factors interact with the broader environment and work in tandem to influence health outcomes, including those related to mental health. Table 1 summarizes a sampling of these factors, though it is by no means an exhaustive discussion of these factors and interactions. --- DISCUSSION The United Nations High Commissioner for Refugees views camps as "temporary solutions of last resort (page 2)," yet the formation of camps is so common that it warrants better planning. 39 Although we fully acknowledge that displaced persons should not be confined to camps and strongly argue for dignified living conditions for those who have been forced to leave their homes, we argue that for those who are in camps, an expanded understanding of the Built environment 12 Infrastructure, including dwellings and water/sewage systems, that are poorly constructed and unsafe can increase exposure to toxic and unstable environments and disease. Physical landscape 13 The physical landscape informs disease ecology, affects resource availability, and increases vulnerability to weather events. Camp location 12,14,15 Camps are often located in suboptimal environments that would otherwise be sparsely inhabited and affect the provision of humanitarian aid. Camp creation leads to landscape changes to accommodate the population and built environment. Fuel demand and supplies 13,16,17 Availability of energy has impacts at the individual level, such as through the ability to cook or have electricity, and at the camp level through provision of health care and waste management. An insufficient energy supply often results in fuel poverty, increased deforestation, and malnutrition. 13 Deforestation 13 Deforestation changes the physical environment of the camps, including vector landscapes, often quite dramatically. It also contributes to climate change, and increased air pollution and irritants for chronic health issues. --- Uncontrolled fires The use of firewood has immediate dangers related to uncontrolled fires, which can spread rapidly in tightly packed camps. Climate change and extreme weather events [18][19][20] Temperature, climate, and extreme weather events can trigger displacement and affect individuals who have already been displaced. 18 These events can also alter the physical landscape, exacerbate the spread of infectious diseases, and affect the provision of humanitarian aid. Water and sewage systems 20,21 Unsafe drinking water and stagnant contaminated water can be sources of infectious diseases and long-term health impacts related to the consumption of naturally derived compounds. [22][23][24][25] Clean water can also affect the provision of health care. Groundwater quality 22,26,27 Insufficient sewage systems may introduce or reintroduce viruses and bacteria into groundwater and surface water, disrupting the natural ecosystem and contributing to disease spread. 22,27 Land clearing and burning, food production, and development or land hardening also release pollutants into these water sources. Livestock [28][29][30] Livestock serve as reservoirs for diseases and can attract disease vectors and multidrug-resistant bacteria, especially if practices are unregulated. Animal waste can also contaminate the water supply. Livelihood production 28 Means of livelihood, especially those that rely on water-intensive agricultural practices or livestock, can affect water resources. 28 Such means are also vulnerable to changes in temperature and climate. Watershed health and viability 26,31 Long-term viability of water sources depends on watersheds around camps, which are affected by the removal of native vegetation, changing hydrology, alteration of the microbial communities, and the addition or increase of pollutants in runoff. 31 Camp legal status 32 The legal status affects the provision and availability of services throughout the camp, as well as the camp's development and care. --- Political and conflict environment Political and/or conflict-related challenges cause variation in camp services over time, especially when camps are in areas of strategic importance. Discrimination, inequality, and violence [33][34][35] Differences in race, gender, legal status, and other demographic variables affect how an individual is treated upon entering and living within the camp. Women and girls are particularly vulnerable to gender-based violence, and lack an established justice system and access to appropriate physical and/or mental care. [33][34][35] Mobility 36,37 Individuals may have their movement restricted or may be completely unable to leave a camp legally, which limits access to health services and their ability to seek employment or education. Occupational opportunity 38 Displaced persons are often limited with regard to occupational opportunities as a result of legal and structural barriers to accessing work. Because of this, many are forced to work in environments where certain diseases are more common or they may work illegally. --- Host community relationships Host community relationships may be strained by actions such as deforestation, unsustainable land use, and employment outside of camps. These relationships can have marked effects on the longevity of the camp. --- Demography Demography has impacts across all sectors. Changes in demography can affect disease burden, required services, livelihood, physical landscapes, fuel demands, and other important components of the camp ecosystem. interacting ecologies will help us improve lives as other policies that provide a dignified existence out of the camps are prioritized. We have thus far focused on situations in which displaced persons aggregate in camps, though many are instead living in urban environments. These environments likewise tend to be unsafe and unhealthy. An ecological lens allows us to understand more fully the living conditions in areas such as these, though other factors (including public health policy) are needed to address the needs of this specific population. 40 We recognize that this lens has its limitations. Individual camps create unique contexts and ecosystems, meaning that ecological interactions may vary greatly among camps. The adoption of this approach may have numerous logistic barriers, though these may be largely a result of the siloed approach this lens attempts to deconstruct. Developing a better understanding of how different components of this ecosystem interact, including at different space-/timescales, could lead to better planning and interventions to improve the health of both populations and environments in these difficult settings. --- RECOMMENDATIONS We need more partnerships that are cross-cutting between silos and that bring together humanitarian aid providers, camp planners, ecologists, public health professionals, health-care providers, environmental practitioners, lawyers, policymakers, and other experts of factors that contribute to the ecology of camps. Integrated policy needs to be developed based on assessing connections among systems so that evidencebased decisions can be made about how interventions may influence outcomes in multiple sectors. Sharing resources and taking an integrated approach will assist in buffering risk and creating more effective and proactive governance. We recommend research and collaboration that increases our understanding of interactions among camps, their environments, and broader ecosystems to incorporate this knowledge into future camps. This includes improving data gathering and analysis to incorporate local knowledge and partners with an emphasis on actively engaging women. 11 We recommend having the data ecosystem be considered more seriously and prioritized in these settings. Making decisions through an ecological lens does not need to preclude immediate aid provision. Through our recommended collaboration, we foresee a pathway to integrate ecological considerations into camp planning, development, management, and sustainment. However, we must first learn how we can adapt this lens in a way that is practical in real-time camp settings. As the first step in this process, we recommend a series of workshops that brings together different actors across silos to make concrete pathways for the inclusion of ecological considerations in camps. This piece is but the first contribution in what we anticipate, and indeed hope, will be much larger conversations around improving the care and services we provide displaced persons.

Health care among forcibly displaced persons is frequently driven by siloed approaches. Aspects of the built environment, social factors, and the bidirectional relationship between the changing ecosystem and residents are often ignored in health policy design and implementation. While recognizing factors that create a preference for siloed approaches and appreciating the work of humanitarian agencies, we argue for a new data-driven and holistic approach to understand the health of the forcibly displaced. It should be rooted in the realities of the emergence of new diseases, dynamic demographics, and degrading environments around the displaced communities. Such an approach envisions refugee and internally displaced camps as dynamic, complex ecosystems that alter, and are altered by, spatial and temporal factors. At the root of this approach is the necessity to work across disciplines, to think holistically, to go beyond treating single ailments, and to develop ethical approaches that provide dignity to those who are forcibly displaced.

Background Son preference has been documented in many low-and middle-income countries (LMICs). Consequences for girls include sex-selective abortions, discriminatory feeding practices, and mortality. Despite consistent evidence of strong son preference in Nepal [1][2][3][4], however, Nepali boys face a survival disadvantage compared to girls [5]. While prenatal discrimination has been well-documented in Nepal [3,6,7], patterns of postnatal discrimination are under-researched. In this paper, we examine gendered nutrition and mortality outcomes for children aged under-five. --- Nutrition and survival Nutrition plays a central role in shaping mortality risks in childhood. Children experiencing severe micronutrient deficiencies and/or severe undernutrition (i.e. stunting, wasting, underweight) face significant risks of infectious illnesses, organ failure, and mortality [8][9][10]. UNICEF [11] estimates undernutrition contributes to more than one-third of under-five mortality globally. To combat early childhood undernutrition, exposure to waterborne pathogens, and diarrheal disease, WHO recommend introduction of breastfeeding within 1 h of birth, exclusive breastfeeding until 6 months, introduction of complementary foods at 6 months, and continued breastfeeding for at least 24 months [12]. Once complementary feeding begins, children should eat a diverse diet to prevent micronutrient deficiencies. Parents may depart from best practices for many reasons, including limited or incorrect information about best practices; insufficient access to a reliable, nutritious supply of food; and/or uneven distribution of food due to personal preferences [13][14][15]. These possibilities are not mutually exclusive-parents may, for example, struggle to access nutritious food, and also not purchase available food because it is undesirable. In some cases, parents prefer packaged, processed foods from markets over fresh local food because packaged foods are perceived as healthier, reflecting both knowledge and preferences [16]. Distal factors impacting undernutrition include household poverty, parental education, geospatial distribution of infectious diseases, food prices, and state policies [13,15,[17][18][19]. Child nutrition is also shaped by intra-household dynamics. One recent study [20] from Dhanusha and Mahottari districts in Nepal considered whether maternal dietary intake and nutritional status and sex differences in growth in the first 20 months of life accounted for the observation that boys grow faster but are more susceptible to undernutrition compared to girls. They found that maternal anthropometry and dietary intake did not differ according to child sex, neither during pregnancy nor in the postpartum period, but that comparable maternal restricted food intakes during pregnancy may negatively impact fetal growth for boys more than for girls. Some literature, mainly from India, suggests girls may be given less food/less preferred foods where competition with brothers for nutritional resources is high [21][22][23]. However, evidence is mixed, with other research finding limited evidence of preferential treatment of boys [24][25][26][27]. Further research is needed to understand when/how/ for what nutritional resources sibling competition intersects with gender to produce nutritional inequalities-particularly in Nepal, where literature on this is scarce. --- Son preference Son preference has wide-ranging negative consequences for girls, including sex-selective abortion [3,7,28,29]; poorer investments in prenatal treatment when fetal sex is known [7,30]; restricted access to healthcare [31]; shorter breastfeeding duration [22][23][24]32]; and lower immunization rates [32]. Where sons are preferred, couples who have daughters may try again for a son even where they have met or exceeded their ideal family size. Son preference thus results in two aggregate demographic trends [33]: a sibling effect, in which girls tend to be born into larger families, and a birth-order effect, in which girls tend to be earlier in the birth-order than boys. Importantly, however, son preference is not universal: As Sapkota et al. [34] show, son preference is greater among women who are less educated, poorer, and/or who live in rural areas. They also document variation in son preference by ethnicity, with women from the Terai Brahman, Terai Dalit, Muslim, and other ethnic minority groups expressing strong son preference compared to women from other ethnicities. Evidence of son preference in Nepal has centered on stopping behavior, contraceptive use, and sex-selective abortion [3,4]. There has been less work on postnatal discrimination. In Nepal, though the reported desired sex ratio at birth has decreased since 1996 [2], demographic manifestations of son preference have increased. Most notably, there is evidence of growing sex-selective abortion, especially in the Kathmandu Valley and amongst wealthier, more educated women [3]. While Leone et al. [35] found the effects of son preference on fertility and contraceptive use in 1996 were moderate, more recent evidence shows the effects have increased over time [2,4]. --- Child health in Nepal Nepal performed very well at meeting Millennium Development Goal 4, which called for a two-thirds reduction in mortality between 1990 and 2015 [36]. Nationally representative data from 2006 to 2016 show that, among Nepali children under the age of five, diarrhea and acute respiratory infections have declined (by 4% and 3% respectively), and treatment-seeking practices have improved over the period [37]. Despite these encouraging trends, sociodemographic and regional disparities in child health and survival have persisted, with children in poor households being particularly disadvantaged. Nonetheless, the under-five mortality rate (U5MR) fell by 72% since 1990, from 142 deaths per 1000 live births in 1990 down to 40 in 2013 [38]. Although this represents remarkable progress, Nepal still ranks 61st globally for highest U5MR, despite high childhood immunization rates and falling rates of low-birthweight (LBW) infants. One explanation may be the very high prevalence of childhood undernutrition. The most recent figures from the Nepal Demographic and Health Surveys (NDHS) in 2016 show that 36% of children under the age of five were stunted, and 27% were underweight [39]. These measures reflect long-term undernutrition, including through non-adherence to WHO recommended feeding practices. Some small-scale studies from Dolakha and Kavre districts and from Dukuchhap village in Lalitpur district focusing on early [40] and mid-childhood [41] suggest girls are more likely to experience sustained undernutrition than boys. Sibling characteristics, birth spacing, and family sociodemographics can influence children's health and nutrition. Using data from a randomized control trial of newborn infections in rural Southern Nepal, Rosenstock et al. [42] found that although there was no gender gap in neonatal mortality overall, within the neonatal period (first 28 days of life) boys faced a greater risk earlier in the period, while girls faced a greater risk of mortality later in this period. For boys, greater risk was explained by biological factors. For girls, sociodemographic factors drove elevated risk-particularly sibling gender and birth order, with girls with no brothers facing the greatest risk of mortality. Girls with no older brother are likely to experience a shorter succeeding birth interval as a result of son preference [43]. The birth intervals before and after the birth of a middle-child are negatively associated with child survival, with much of the association operating through early weaning [44]. Maternal age, education, and economic status all have a positive relationship with breastfeeding duration [45], and maternal socioeconomic status and ethnicity impact initiation of breastfeeding within one hour of birth [46,47]. Mothers in Nepal's mountains are more likely to initiate breastfeeding and complementary feeding following recommended practices, and mothers in urban areas are more likely to use bottle-feeding rather than breastfeeding [48]. --- Study contributions Although son preference is not the only driver of gendered outcomes, it is well-documented as a critical factor impacting child health in South Asia [22][23][24]32]. There has, however, been a relative dearth of studies focused specifically on Nepal. Here, we examine gendered nutritional and mortality outcomes, asking: 1) What are the aggregate patterns of gender differences in mortality across time and place in Nepal? 2) Are girls disadvantaged in infant and early childhood feeding practices? 3) Where disadvantage exists, are feeding practices associated with mortality risks? We question the notion that son preference has a uniform effect on female nutrition and survival, and consider how son preference interacts with birth order and sibling gender to shape health outcomes. --- Methods --- Data We use 5 rounds of the NDHS [49], a publicly available, nationally representative cross-sectional survey conducted every 5 years since 1996. Surveys include women's reports on maternal and household sociodemographics, reproductive histories, and child health [50]. Detailed data about breastfeeding and food consumption were collected only for children born within 36 months of the survey in 1996, and within 59 months of the survey in subsequent years. We therefore restricted the sample to children aged <unk> 37 months in 1996, and <unk> 60 months thereafter. Data from the 2011 Nepal Census were used to construct a map of female U5MR as a percentage of male U5MR across Districts of Nepal. --- Availability of data and materials The datasets analysed for this study are available in the Demographic and Health Surveys repository, (https:// dhspr ogram. com/ data/ avail able-datas ets. cfm). Data were publicly available and fully anonymized, and so ethical approval was not required. --- Dependent variables Because NDHS has undergone revisions over time, some measures are not available in all years. Figure 1 shows availability by survey round. --- Food consumption For living children, mothers reported whether children consumed (1 = yes) 16 different categories of food items within the last 24 hours, including roots and tubers, green leafy vegetables, meat and poultry, vegetarian proteins (e.g. legumes), dairy and grain products, and fats. Interviewers read out: "Now I would like to ask you about liquids or foods that (NAME) had yesterday during the day or at night. I am interested in whether your child had the item I mention even if it was combined with other foods. Did (NAME) drink or eat:" The interviewers then read out each item in the list and recorded whether mothers indicated yes, no, or don't know. Items on the list included, for example "Any meat, such as pork, buff, lamb, goat, chicken, or duck, " "Any dark green, leafy vegetables like spinach, " "Pumpkin, carrots, squash, or sweet potatoes that are yellow or orange inside, " etc. Following WHO guidance for infant and young child feeding [12], we also used these variables to calculate dietary diversity scores for children over 6 months of age (before which exclusive breastfeeding is recommended) based on whether they consumed items in each of seven broad categories within the last 24 hours: grains, roots, and tubers; dairy; legumes and nuts; meat, fish, poultry, and animal organ meat; vitamin A-rich fruits and vegetables; other fruits and vegetables; and eggs. We coded children as having a diverse diet if they consumed foods from at least four of these groups. --- Breastfeeding For living children, mothers provided retrospective reports of whether breastfeeding was initiated within 1 hour of birth (1 = yes). NDHS do not include a question about duration of exclusive breastfeeding; we were unable to examine exclusivity. We examine breastfeeding duration based on mothers' replies to whether they ever breastfed each child; mothers responding 'yes' were then asked, "Are you still breastfeeding (NAME)?" Where women answered 'yes', breastfeeding duration was coded as equal to the child's current age in months. Mothers of weaned children were asked "For how many months did you breastfeed (NAME)?" This was asked for all children <unk> 36 months in 1996, and all children <unk> 60 months in 2001 and 2006. Unfortunately, from 2011 a survey revision means it is only possible to calculate duration for currently breastfeeding children. --- Child survival We utilize a dichotomous indicator of whether the child was living at the survey date (yes = 1). For children who died, we calculated survival duration by subtracting date of birth from date of death, reported retrospectively by mothers. --- Independent variables Our key independent variable is child's gender (female = 1). We also consider sibling gender and birth order because these factors may shape child health outcomes [23,24,32,33]. We measure birth order and sibling gender in 5 categories: (1) firstborn ( 2); second or thirdborn with no older brothers (e.g. a secondborn girl following a firstborn girl) (3); second or thirdborn with an older brother (e.g. a secondborn girl following a firstborn boy) (4); fourth-or-higher birth order with older brother(s) (5); fourth-or-higher birth order with no older brothers. To account for sociodemographic factors which may correlate with feeding practices and son preference, we include maternal age and education (no education; completed primary or less; attended or completed secondary; and attended or completed higher education), urban versus rural residence, household poverty (household wealth in the bottom two quintiles on the NDHS wealth index), household size, religion (Hindu = 1), ethnicity (Brahaman/Chhetri; Tarai/Madhesi other castes; Dalits; Newar/Janajati; and Muslim/Other), and ecological region (Hills, Terai, or Mountain). Hazard models for breastfeeding also include a control for whether the child was alive at the time of the survey. Our models focusing on children's food consumption also include child age (in months). Additionally, given evidence that seasonality is associated with child mortality [51] and dietary intakes [20] in some areas of Nepal (studies from rural Sarlahi and rural areas of Dhanusha and Mahottari districts respectively), we also included a control for interview month in our models of child survival and food consumption. --- Analytic strategy We tested (1) baseline models with sociodemographic controls, (2) models with the five-category birth order/ sibling gender predictor (child birth order excluded), and (3) models including an interaction between child gender and the five-category birth order/sibling gender variable. Our pooled sample across 5 rounds of data included n = 25,648 children born to n = 19,135 mothers within n = 17,560 households. Missing data were handled using listwise deletion. We applied pooled multilevel logit regression models to analyse breastfeeding initiation and food consumption accounting for clustering of children within households. Due to variation in sampling frame and survey questions between years, we ran separate models for breastfeeding duration in 1996 and 2001. However, as the questionnaire and sample design matched in 2001 and 2006, we pooled models in these years. Pooling data where possible allowed us to increase sample size and reduce standard errors, improving the precision of our estimates [52]. In pooled models, we included a control for year to capture potential changes in health and feeding practices over time. Gender differences in initiation and individual food item consumption were estimated using separate multilevel logit regressions for each outcome. We also applied discrete-time hazard models to estimate the hazard of cessation of breastfeeding and child survival. Discrete-time models were selected because these outcomes are measured as discrete monthly data. Hazard models handle censoring well, so the restriction of the sample to children under 36 months in the 1996 survey is unproblematic. There were not enough observations in some months to fit the models using 1-month intervals, so we created 3-month time periods. Truncation occurred after 16 time points for models of breastfeeding duration, and 19 points for child survival. For interpretability, we illustrated the significant interaction between gender and five-category birth order/sibling gender using Kaplan Meier curves. Models were fitted using Stata version 13.1. --- Results --- Descriptive statistics Breastfeeding practices appear to be improving over time (Table 1). In 1996, a little over one-third (37.4%) of the sample was put to breast within 1 h of birth. This figure increased steadily, rising to nearly half (49.1%) Around 50% were female in each round. Mean child age was around 29 months in most years. 1 The percent of children with no brothers was substantially lower than those with brothers. Mean maternal age ranged from 26.5 to 28.0 years across surveys. Education increased substantially: 80.8% of mothers had no formal education in 1996, compared to 33.9% in 2016. Secondary or higher education climbed from 8.6% in 1996 to 46.3% in 2016. There was a noticeable jump in the percent urban across rounds, with 8.5% of the sample living in urban areas in 1996 and 2001 compared to around 20% in 2006 and 2011, with another large jump up to 56.8% in 2016; this latter jump corresponds to governmental reclassification of administrative boundaries rather than largescale migration. Household size declined somewhat over time, from a mean of 7.4 household members in 1996 down to 6.1 in 2016. Across 1996 through 2011, a majority (<unk> 85%) of the sample was Hindu. Caste/ethnicity fluctuated slightly across rounds, but in general around one-third of respondents were Brahaman/Chhetri and another roughly one-third were Newar/Janajati. The next largest group were Dalits, around 15% in most rounds, followed by Tarai/Madhessi Other Castes. The smallest caste/ethnicity group, around 5% across rounds, were Muslim/Other respondents. A minority (from 14.5 to 19.7%) lived in the mountain region, while around 40% each lived in the hill and terai regions. In 2016 the lowest proportion of respondents across rounds lived in the mountain region (9.2%), and the highest lived in the terai region (48.5%). There was some variation in when interviews took place across surveys, but in general on average surveys occurred in spring and summer months. --- Mortality Figure 2 shows female U5MR as a percentage of male U5MR by district (n = 75) according to 2011 population Census data. The 24 red districts have a higher female U5MR-that is, where female U5MR exceeds male U5MR and the sex ratio therefore exceeds 100. As Costa et al. [53] note, boys experience greater biological frailty in early childhood, and higher female mortality ratios run counter to girls' biological survival advantage in early life, indicating gender-based inequalities. From a descriptive overview of Fig. 2, districts with higher female mortality tended to be richer and more urban areas, with lower mortality overall. Table 2 provides results from the NDHS by gender and year for the neonatal (first 28 days), post-neonatal (1-12 months), infant (first 12 months), child (1-4 years), and under-five mortality rates. The female child mortality rate was generally higher than the male child mortality rate in all surveys. Likewise, the female post-neonatal mortality rate was higher than the male post-neonatal mortality rate across time periods. However, this later female disadvantage was masked in the U5MR for 2011 and 2006 by the mortality disadvantage faced by boys in the neonatal period. Girls also had a lower mortality rate than boys in the neonatal period but the 2016 data show a substantial drop in the U5MR for both boys (43% decrease from 2011) and girls (34%), mortality rates for girls again declined at a slower rate-a return to the pattern in previous periods. Girls were particularly disadvantaged in the rate of mortality decline in the post-neonatal period, when boys' mortality dropped by 59%, compared to only 11% for girls. --- Food consumption We found no significant gender differences in food consumption for any of the 16 reported food items in any round, nor for dietary diversity. Results are not shown for brevity, but were tested in separate logit regressions for each food item. --- Breastfeeding We found a marginally significant association between child gender and breastfeeding initiation (Appendix Table S1), with girls having slightly lower odds than boys of being breastfed within 1 h of birth (OR = 0.82; p <unk> 0.01). This association reduced slightly (OR = 0.82; p <unk> 0.05) once birth order and sibling gender were taken into account. All children of higher birth orders had higher odds of being put to breast within an hour of birth as compared to firstborns; sibling gender did not make a substantial difference. We did not find evidence of a significant interaction between child gender and sibling gender and birth order. However, child gender was a strong, significant predictor of breastfeeding duration. Table 3 provides results for discrete-time hazard models predicting breastfeeding duration. Model 1 shows, accounting for all baseline controls, girls faced a higher hazard of breastfeeding cessation compared to boys (HR = 1.16; p <unk> 0.001), i.e. had shorter breastfeeding duration. Higher order births were associated with a longer duration of breastfeeding. The association between gender and breastfeeding cessation was increased (HR = 1.22; p <unk> 0.001) by the inclusion of the five-category birth order/sibling gender variable (Model 2). All groups had a lower hazard of breastfeeding cessation than firstborns, and girls with older brothers had a lower hazard of breastfeeding cessation than girls with only sisters. Girls of birth order four or higher who had older brothers faced the lowest risk. Model 3 showed a significant interaction between child gender and birth order/sibling gender. The main effect of being female (HR = 1.09; p > 0.05) was non-significant; however, this is because of the very strong and significant interaction between child gender and our five-category measure. Female disadvantage in breastfeeding for higher order births was stark. Older brothers had a significant effect on breastfeeding duration, but only for girls. A second-or thirdborn girl with no brothers had increased odds of breastfeeding cessation compared to a firstborn male. Second-or thirdborn girls with older brothers were actually breastfed longer than firstborn males, but not as long as second-or thirdborn males. This is consistent with the idea that at low parity, women plan to have another child regardless of child sex, and so stop breastfeeding early. At higher parity, progression decisions are based on number and gender composition of existing children. To ease interpretation of interaction effects from these models, we produced Kaplan Meier curves, presented in Figs. 345. They show the proportion still breastfeeding on the Y-axis, and time in months along the X-axis. The trend for boys is given by the light grey line, and for girls by the darker line. Figure 3 shows all living children in the pooled sample. A slight gender gap emerged around 12 months, growing with child age. There were three large drops in breastfeeding, around 12, 24, and 36 months, consistent with normative age-based weaning practices. The trend for firstborns closely resembled the overall trend for all children (Fig. 4). There was again an observable female disadvantage in breastfeeding duration. It emerged earlier than in the overall trend, but was still small. A markedly different pattern emerged at higher birth orders. The top left panel of Fig. 5 shows trends for second-and thirdborn children with no older brothers. Female disadvantage in breastfeeding duration appeared around 6 months and grew rapidly, resulting in a very large gap by 24 months. No girls were still breastfeeding past 45 months, while nearly 20% of boys were still breastfeeding at the final observation point. The gender gap between second-and thirdborn boys and girls with 1+ older brothers was small and emerged later (around 12 months), shown in the top right panel of Fig. 5. Patterns for fourth-or-higher order births without (bottom left) and with (bottom right) 1+ older brothers were similar to those for second-and thirdborn children. --- Mortality To assess the link between female disadvantage in breastfeeding and child survival, Table 4 provides results for the pooled discrete-time hazard models predicting the hazard of mortality. Model 1 shows that, controlling for sociodemographic variables, every month of additional breastfeeding conferred an approximately 4% lower hazard of death (HR = 0.96; p <unk> 0.001). This relationship held across Models 2 and 3, with controls for sibling gender and birth order variables and the interactions with gender respectively. Thus, the key gendered nutrition outcome documented in our study was strongly, significantly linked to child survival. However, this did not counterbalance the survival advantage for girls overall-girls had a lower hazard of mortality compared to boys (Model 1: HR = 0.94; p <unk> 0.001; Model 2: HR = 0.91; p <unk> 0.001; Model 3: HR = 0.93; p <unk> 0.05). --- Robustness checks Having an older brother and having a living older brother may differ in terms of discrimination and feeding practices. As a robustness check, we created an additional five-category birth order/sibling gender variable based on surviving siblings. We ran all models with these living male sibling categories. While this resulted in some small fluctuations in effect sizes, results were substantively unchanged. Figure 1A presents an illustrative example of the Kaplan Meier curves for breastfeeding duration disaggregated by this variable. Second, our models of breastfeeding initiation omitted children who had died; it is impossible to know whether the children were weaned and subsequently died or died and therefore were no longer breastfeeding. As a robustness check, we re-estimated the models with children who died included. Again, results were substantively unchanged. Finally, because there are a smaller number of children of fourth or higher birth order with no brothers compared to other categories, we coded a version of the variable comparing firstborns, secondborns with an older brother, secondborns without an older brother, third and higher with and older brother, third and higher without an older brother and fitted all models again. There were some minor fluctuations in effect sizes, but results remained substantively unchanged. --- Discussion There is a large, robust body of evidence documenting negative effects of son preference on girls' health and survival elsewhere in South Asia [1,22,24,32,54], but limited literature in Nepal. Here, we investigated: the patterns of mortality by gender across time and place in Nepal; whether girls are disadvantaged in child feeding practices compared to boys; and, where observed, whether gendered feeding practices are associated with mortality risks. We found geographic variation in mortality differentials: girls face a particular disadvantage in richer, more urban areas. Encouragingly, mortality and nutrition outcomes improved for all children between 1996 and 2016. However, girls were consistently disadvantaged in the post-neonatal period and between ages 1-4 years relative to boys. This is consistent with previous work in Nepal showing boys' greater biological frailty places them at greater risk of mortality than girls in the neonatal period, but girls face a disadvantage in later periods due to postnatal discrimination [42]. These period-specific patterns explain the seemingly contradictory fact that son preference is high in Nepal, but under-five girls do not face an overall mortality disadvantage. There are several possible explanations for period-specific effects. First, postnatal discrimination may operate from birth but be insufficient to counterbalance boys' heightened frailty [55] in the earliest period. Alternatively, parental perceptions/knowledge of infant frailty could encourage strong investments in all children in the first few weeks of life, with a gendered decay over time as infants exit the perceived frail period. This hypothesis is supported by our finding of no gender gap in breastfeeding initiation, which would manifest during this early period, but significant female disadvantage in duration, relevant to later discrimination. Second, parents may be less willing to invest in children perceived as overly frail-it may be that the frailest boys are selected out in the neonatal period due to diminished parental investment. These possibilities are not mutually exclusive. Third, early complementary feeding may be seen as a way of giving favored children (boys) a nutritional advantage. However, such practices would actually place them at greater risk of illness and mortality. We found no evidence of gender differences in complementary feeding practices, as with elsewhere in South Asia [24][25][26][27]. However, we did find a female disadvantage in breastfeeding, with girls with older sisters but with no brothers being most disadvantaged. This finding fits with the theory that son preference manifests not as intentional postnatal discrimination, but rather as asymmetrical preferences for child spacing based on child gender and birth order. Evidence from India [23,24] and Pakistan [22] shows a very similar pattern by child and sibling gender and birth order, attributed to stopping behavior: Mothers at low parity who want additional children may wean early regardless of child gender, while mothers at higher parity may only wean early to try again if they have not already met gender-specific fertility desires. Our study has several limitations. First, it is possible boys and girls receive the same kinds of food but girls are disadvantaged in the amount/quality of food received/are given less nutritious portions, but we could not measure this disparity. For example, if a boy and a girl are usually fed the same curry but the boy systematically receives more pieces of meat and/or leaner, more protein-rich meat while the girl receives less meat and/or fattier, less protein-rich meat, such disparities cannot be measured by the kind of yes/no food category questions available in the NDHS data. Owing to social desirability bias, women may also underreport gender differences. Both factors may introduce a conservative bias in our estimates of postnatal discrimination practices. Second, while extensive literature has measured the impact of son preference, the underlying driver of son preference is arguably the perceived sociocultural and economic value of girls; as with other work in this area, we were unable to directly assess value judgements here. Third, changes in survey design limited our ability to directly compare models across time for some outcomes. Fourth, we were not able to measure exclusive breastfeeding due to data limitations. One cross-sectional study from Bhaktapur (part of the Kathmandu Valley) showed by 1 month of age, one-quarter of infants were no longer exclusively breastfed; by 3 months, only 24% were exclusively breastfed, and by 6 months, the figure dropped to <unk> 10% [56]. It remains unclear whether there are gender disparities in these patterns. To facilitate a critical examination of exclusive breastfeeding practices in future research, nationally representative population data on the duration of exclusive breastfeeding for all children are needed. Fifth, the vast majority of infants are introduced to solid food during a weaning ceremony called pasni [57]. This ceremony occurs earlier for girls, at 5 months, than for boys, at 6 months. We were unable to examine the effect of this cultural practice here as the practice is not directly assessed in the NDHS. In light of continued complementary feeding (in keeping with WHO recommendations), this ceremony would be more relevant to exclusive breastfeeding (not assessed here) than to overall duration, but could also be a factor in gender differences in solid, semi-solid, and soft food consumption. While we did not find evidence of gender differences in food item consumption, future work is needed on whether or how pasni may be associated with gendered nutritional outcomes. Sixth, many factors beyond nutrition may also help to explain gender differentials in mortality, but we were not able to explore these comprehensively here. Nepal is an extremely varied country, with 126 caste/ethnic groups reported in the 2011 census [43]. Topographically it ranges from the terai (or plains), which has a tropical and subtropical climate to the mountains, which include 8 peaks over 8000 m. As of 2015 Nepal consists of seven provinces, which vary dramatically in terms of their economy, education, and ethnic makeup; for example the Total Fertility Rate in Province 3 is just 1.8, while in Province 2 it is 3.0 [39]. Further work is needed to understand regional and sub-regional differences in the relationships we have identified here. Finally, it should be noted that the government classification of urban and rural areas changed substantially between the 2011 and 2016 rounds of the DHS. According to the 2011 census, 17% of the population lived in urban areas, but the widespread reclassification meant that 59% of the population lived in urban areas by 2016 [39,43]. It is important to note that this was due to a reclassification of areas rather than actual urbanization. There were many other administrative changes during this time, including expanding the number of municipalities from 58 to 217; as a result it is not possible to recreate the urban/rural classification from 2011 within the 2016 DHS and direct interpretation of this variable should be treated with caution [39,43]. --- Conclusions The regional patterning of female disadvantage we observed can be used by policymakers to target nutrition interventions to improve girls' health outcomes in regions where they are most at-risk. Health workers providing counseling during antenatal care (ANC) to improve maternal and infant health [58] provide an opportunity for quality counselling to increase adherence to recommended breastfeeding practices for girls. Such embedded community health workers can identify opportunities for targeted provisioning of early childhood healthcare for girls at greatest risk of undernutrition at mortality. This would complement existing governmental aims to improve the quality of maternal health services and reduce health disparities for women and girls in Nepal [59], as well as being consistent with evidence-based WHO recommendations for expansion of ANC coverage [60]. The Government of Nepal is working to increase universal health coverage and reduce health disparities, but there are considerable inequalities in access to and uptake of quality ANC services [58,59,61,62] and even starker inequalities in postnatal care [63][64][65]. Our results show girls' disadvantage is greatest from ages 1-4 years. Community-based health programs, developed in consultation with mothers themselves [60], could be expanded to continue to provide targeted healthcare for children beyond 12 months, with particular focus on nutrition monitoring and health service provision for girls. Nepal's Female Community Health Volunteers (FCHVs) program could be particularly effective in this aim. Founded in 1988 with a primary focus on family planning but expanded to include maternal and child health services in subsequent decades, the program provides health knowledge and connects local families and communities to health facilities [66]. Although some observers have highlighted gaps in the FCHV program [66], particularly focused on program fragmentation and bureaucratic challenges faced by the volunteers themselves, FCHVs and women's groups are effective modes of intervention during pregnancy; interventions utilizing these groups have been found to improve maternal and perinatal health [66][67][68][69]. Similarly, although some supply-side [70] and demand-side [71] barriers to uptake have been identified in India's Janani Suraksha Yojana, support services provided by local resident community health workers have been strongly identified as an enabling factor in improving the uptake of maternal health services. The development of strong interpersonal relationships between the community health workers and pregnant women and mothers appear to be a particularly important factor in facilitating better access to and uptake of public healthcare programs in both Nepal and India [69,71]. The gendered health and nutrition patterns identified here may be reduced through a similar targeting of the nutritional needs of at-risk girls via fostering continuity between ANC and, subsequently, community-based healthcare coverage. Additionally, both to address the need for data on breastfeeding in Nepal, and to inform policy and practice, breastfeeding practices could be recorded by community health workers, anonymized, and made available to researchers to facilitate monitoring of feeding practices and identification of specific at-risk regions and groups. --- Availability of data and materials All data are publicly available secondary data available for download from: https:// www. dhspr ogram. com/ Data/. --- Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40795-022-00543-6. --- Additional file 1. --- Authors' information (optional) Not Applicable. --- Authors' contributions JF and MC jointly conceptualized the study, conducted the analyses, and drafted the manuscript. --- Declarations Ethics approval and consent to participate No primary data involving human subjects were collected or analysed for this study. Ethical approval

Background: This paper examines seemingly contradictory evidence from extant research that son preference is high, but male disadvantage in mortality is increasing in Nepal. To do so, we documented the timing, geographic patterning, and extent of gendered patterns in mortality and feeding practices for children under-five.We applied pooled multilevel regression models and survival analysis to five rounds of data from Nepal's nationally representative Demographic and Health Surveys (1996-2016). We controlled for potential sociodemographic confounders, including child, maternal, household, and regional correlates, and disaggregated findings by birth order and sibling gender.We found evidence of regional variation in mortality, with girls in wealthy urban areas faring the worst in terms of mortality rates. Girls' comparative mortality advantage compared to boys in the neonatal period masks their mortality disadvantage in later periods. Mortality has fallen at a faster rate for boys than girls in most cases, leading to widening of gender inequalities. We also found evidence of female disadvantage in breastfeeding duration, which was linked to higher mortality risks, but no gender disparities in the consumption of other food items. Sibling gender and birth order also mattered for breastfeeding duration: Young girls with older sisters but with no brothers were most disadvantaged.While we did not find evidence of postnatal discrimination in access to solid and semi-solid foods, girls in Nepal face a disadvantage in breastfeeding duration. Girls with older sisters but no older brothers facing the greatest disadvantage, with risks being particularly concentrated for girls aged 1-4 years. This disadvantage is linked to an increased risk of mortality. To address this, community-based health programs could be expanded to continue targeted healthcare for children beyond 12 months of age, with particular focus on nutrition monitoring and health service provision for girls.

,71]. The gendered health and nutrition patterns identified here may be reduced through a similar targeting of the nutritional needs of at-risk girls via fostering continuity between ANC and, subsequently, community-based healthcare coverage. Additionally, both to address the need for data on breastfeeding in Nepal, and to inform policy and practice, breastfeeding practices could be recorded by community health workers, anonymized, and made available to researchers to facilitate monitoring of feeding practices and identification of specific at-risk regions and groups. --- Availability of data and materials All data are publicly available secondary data available for download from: https:// www. dhspr ogram. com/ Data/. --- Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40795-022-00543-6. --- Additional file 1. --- Authors' information (optional) Not Applicable. --- Authors' contributions JF and MC jointly conceptualized the study, conducted the analyses, and drafted the manuscript. --- Declarations Ethics approval and consent to participate No primary data involving human subjects were collected or analysed for this study. Ethical approval of this study was not required because we utilized only publicly available data which were fully anonymized prior to download. --- Consent for publication Not Applicable. --- Competing interests We have no conflicts to declare. • fast, convenient online submission • thorough peer review by experienced researchers in your field --- • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research? Choose BMC and benefit from:? Choose BMC and benefit from: --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background: This paper examines seemingly contradictory evidence from extant research that son preference is high, but male disadvantage in mortality is increasing in Nepal. To do so, we documented the timing, geographic patterning, and extent of gendered patterns in mortality and feeding practices for children under-five.We applied pooled multilevel regression models and survival analysis to five rounds of data from Nepal's nationally representative Demographic and Health Surveys (1996-2016). We controlled for potential sociodemographic confounders, including child, maternal, household, and regional correlates, and disaggregated findings by birth order and sibling gender.We found evidence of regional variation in mortality, with girls in wealthy urban areas faring the worst in terms of mortality rates. Girls' comparative mortality advantage compared to boys in the neonatal period masks their mortality disadvantage in later periods. Mortality has fallen at a faster rate for boys than girls in most cases, leading to widening of gender inequalities. We also found evidence of female disadvantage in breastfeeding duration, which was linked to higher mortality risks, but no gender disparities in the consumption of other food items. Sibling gender and birth order also mattered for breastfeeding duration: Young girls with older sisters but with no brothers were most disadvantaged.While we did not find evidence of postnatal discrimination in access to solid and semi-solid foods, girls in Nepal face a disadvantage in breastfeeding duration. Girls with older sisters but no older brothers facing the greatest disadvantage, with risks being particularly concentrated for girls aged 1-4 years. This disadvantage is linked to an increased risk of mortality. To address this, community-based health programs could be expanded to continue targeted healthcare for children beyond 12 months of age, with particular focus on nutrition monitoring and health service provision for girls.

Introduction The ongoing SARS-CoV-2 (COVID-19) pandemic has triggered national and global restrictive measures in an effort to contain the virus. In the UK, measures such as stay-at-home orders, social distancing, and travel restrictions have been implemented at multiple points over the course of the pandemic, impacting the higher education sector substantially [1]. Despite returning home early in the pandemic, subsequent reopening of university campuses for the new academic year (September/October 2020) led to widescale student movement across the country and coincided with a second surge of COVID-19 within the UK [2]. The introduction of a national tiered system in October 2020 brought rapidly changing social restrictions, impacting not only the delivery of education, but also the mechanics of the wider university workforce. As well as abrupt changes to their method of learning, students have faced changes to their exams, placements, and electives [3]. Both home and international students have contended with prolonged separation from family and friends, with many being required to self-isolate in the confinement of their room or home. In parallel, academic staff have also been faced with sudden and ongoing changes to the delivery of teaching, requiring the adoption of remote working, ensuring in-person teaching is delivered in a COVID-safe way, and the frequent adaptation of materials, all with very often minimal notice and preparation time. This has resulted in an increase in workload related to the ever-changing guidance, restrictions, and support needs of students [4]. In addition to the rapid transition to online teaching, multiple universities developed capacity to conduct mass asymptomatic COVID-19 testing programmes in an effort to combat transmission between students [5,6]. Successful mass testing relies not only on testing uptake, but also on adherence to containment strategies, such as self-isolation, following a positive result [7,8]. ONS data suggest high levels of adherence to social distancing and isolation behaviours in university students [9], with Nixon and colleagues reporting that 99% of their student sample testing positive in the prior fortnight had been self-isolating [10]. It has been suggested that adherence to self-isolation may be more heavily impacted by the availability of resources rather than psychological motivation [11]. Other factors, including a drive to socialise and personal concern about the mental health implications of isolation, may also contribute to students' willingness to isolate [12]. Whilst there has been a deluge of research studying the effects of the COVID-19 pandemic on mental health in the general population, there is less specifically focusing on university students and staff. Further, of the research that does exist, the evidence points to wide-ranging negative impacts on wellbeing, including significantly lower mood [13], reduced mental wellbeing and increased stress [14], and increased depression [15]. There is even less information on the impact of self-isolation on students and the staff supporting them. At present, mental health problems are a primary cause of sickness absence in the UK, accounting for 17.5 million days of lost workforce productivity each year [16]. University students represent a significant proportion of the future workforce and, as a result, understanding the impact of the COVID-19 pandemic and self-isolation measures on students' wellbeing is paramount. This qualitative study was conducted at a large, campus university in central England (University of Nottingham), which implemented mass asymptomatic testing and recorded a large increase in the number of positive COVID-19 cases and, therefore, self-isolation during the Autumn term (2020). The aim of this study was to explore the impact of the COVID-19 pandemic on university students from the perspectives of both students and the staff supporting them, with particular emphasis on the wellbeing and support needs of students who were required to self-isolate. Our intention is to reflect upon the experiences of students and staff, informing future policy and practice in the context of university settings as we continue to navigate the COVID-19 pandemic. --- Materials and Methods --- Study Design This qualitative study included focus groups with students and individual interviews with university staff members from the University of Nottingham campuses in central England. This institution has the ninth largest student body of all British academic institutions [17] with both on and off-campus accommodation and communal living spaces to those in oncampus housing. The institute provides both undergraduate and graduate taught and research courses. A total of 11 online focus groups with 52 student participants and 26 individual interviews with staff members were conducted. The study design aligned with the consolidated criteria for reporting qualitative studies (COREQ) guidelines [18] (Supplementary File S2). The study protocol was approved by the Research Ethics Committee of the University of Nottingham Faculty of Medicine and Health Sciences (Ref: FMHS 76-0920). --- Study Context This study took place between January and March 2021 at the beginning of Spring term, starting shortly after the third national lockdown was announced. During this time, national restrictions were in place, limiting movement outside of the home unless for work that could not be done at home, exercise once per day, and to get essential items such as food. Additionally, household mixing was restricted, and there were restrictions on travel, access to facilities and services. Easing of national restrictions did not begin until after data collection was completed. --- Participants, Recruitment and Sampling Focus group participants were university students who had experienced a period of self-isolation for any reason, including receiving a positive test, being in contact with someone who has tested positive, or pre-cautionary self-isolation. Recruitment occurred through two routes: (1) Students who had taken part in an established cohort study [19], and (2) Students who had received a positive COVID-19 test result through the university medical centre and had, therefore, undergone a period of self-isolation. Both home and international students were recruited. Recruitment took place between January and February 2021. Students received a £20 voucher as compensation for their participation. University staff were purposively sampled for one-to-one interviews, to ensure a variety of roles within the university were represented. Specifically, staff with studentfacing roles, or a role relating to COVID-19 strategies or student support were invited to interview, including staff in teaching and/or research, leadership, support, and pastoral care roles. Individual interviews were offered to ensure staff were able to attend at a time that was convenient to them, bolstering uptake. Recruitment took place between January and March 2021. Students and staff provided both written and verbal informed consent. Sample characteristics are provided in Tables 1 and2. --- Qualitative Interviews and Focus Groups Given the rapidity of the required data collection, we hosted a total of 11 focus groups, including a mix of smaller groups of between two to four participants, larger groups of between five and eight participants and one individual interview with a single participant (due to non-attendance). Focus groups were held online using video-conferencing facilities (Microsoft Teams). Online data collection was necessary due to the national lockdown imposed at the time, as well as social isolation policy. This form of data collection has been commonly used as means of improving access for geographically dispersed or hardto-reach populations [20]. All focus groups took place within a period of three weeks in January and February 2021. Focus groups lasted for 54 to 81 min (mean 70 min). Two psychologists generated the question guide (HB, HK, Supplementary File S3) and focus groups were facilitated and moderated by two PhD students who were part of the research team (SC, MO). All focus groups were conducted in line with NHS England guidance on focus group delivery and followed the same questioning route [21]. All members of the team were trained in qualitative research and interview skills. A total of 26 staff interviews were conducted via online video-conferencing facilities (Microsoft Teams), lasting between 21 and 100 min (mean 65 min). Recruitment for the staff interviews was high to ensure representation of the range of student facing roles and recruitment across all role families continued until data saturation occurred. The individual interview modality was chosen for staff data collection to encourage openness and to enhance uptake by allowing flexible scheduling. Interviews took place between January and March 2021. A separate question guide was used that had been generated by the same two psychologists (HB, HK). Interviews were carried out by two PhD students (MO, LF) and followed the same questioning route. All student focus groups and staff interviews were audio and video recorded and transcribed verbatim. To expedite the process, transcription was conducted by both a professional transcription company and by two members of the research team (MO, SC), with all transcripts checked for accuracy. --- Data-Analysis Data were analysed using inductive thematic analysis [22], which benefits from theoretical flexibility and simplicity, and the ability to identify, organise, and describe the experiences of students and staff through the identification of key themes and subthemes. Focus group and individual interview data were reviewed separately in the first instance to identify unique factors and commonalities in the data, with both datasets subsequently combined to create a nuanced understanding of self-isolation experiences. One researcher (LB) conducted a preliminary scan of the student focus group and staff interview data, allowing generation of initial codes for data extracts. To ensure reliability of the data analysis, a second researcher (HK) independently examined the emerging coding, allowing further refinement. Codes and subcodes were subsequently grouped into themes, ensuring that the data cohered together meaningfully, whilst themes were clear and distinct. Finally, a detailed analysis was conducted for each theme and data excerpts were identified to illustrate the final themes. Combining data sources allowed iterative triangulation of an initial coding scheme using individual and group level input. Staff interviews provided additional contextual information, improving interpretation of the focus group data and providing stronger understanding of the academic landscape during the pandemic. Convergence of focus group and individual interview data enhanced the reliability of the coding scheme [23]. Both researchers (LB, HK) reviewed the final thematic outcomes. NVivo 11 (QSR International Ltd., Melbourne, Australia) was used as a data management tool throughout data analysis. --- Results Four key themes were interpreted from the combined staff and student data: adaptation during the pandemic; the practical, environmental, and emotional challenges of self-isolation; social factors and their impact on COVID-19 testing and self-isolation adherence; and supporting self-isolation. A thematic map illustrating the relationships between the key themes and subthemes is provided in Supplementary File S4. Table 3 shows the list of all the key themes and subthemes and representative quotes. A description of each theme and subtheme, along with representative quotes is presented in more detail in Supplementary File S1. Table 3. Examples of key themes, subthemes, frequency, and their representative quotes. --- Theme 1: Adaptation during the Pandemic A new normal I think that just prolonged isolation in general, the whole thing being in lockdown since March and whatnot since March, but just not really going out and socialising has changed me in a way that made me really comfortable with solitude and a little bit distressed when I'm outside. FG7, S1 I think it's getting back to normal now because, I don't know, I think at first it was a bit weird again to see each other. It almost felt like we were doing something wrong, but we've got over that stage and it's far more natural now to just, you know, chill and chat. FG7, S2 --- Challenges of adapting during COVID-19 We were wondering whether our marks would get negatively affected from not being legally allowed to go to those things. FG1, S8 Most students I know do think that we are definitely at a disadvantage because it's just a completely different environment to be learning in really. FG6, S3 And then we had to do the remote teaching including you know, we're just different because we've got this dual cohort not that started for the first time last year so we had 150 students start in April during lockdown, so we were remote teaching them without them ever having been on campus. And then they came on campus in July to catch up on the practical teaching so they're just completely different to the normal university students. Staff interview 18 I just think there's a huge expectation about you're going to come to university and there's going to be parties and you're going to make loads of friends. And even though they came to university understanding that they weren't going to get that, I guess it was kind of inevitable that to repress and suppress those expectations, was always going to be verging on impossible. Staff interview 14 There's one concern, particularly for the practical people on the ground which is obviously their fear of contracting the virus, you know and the sense that the University is putting them in jeopardy, and is not sufficiently supporting the health and wellbeing, through a health and safety prospective. Staff interview 2 It was quite frightening at first because you just feel like you're forever firefighting, because it was en-masse people were told to, like, you know, pack up and all move down to our hall, I mean, we had no masks, you know, there were people who were frightened. Staff interview 5 Table 3. Cont. --- Theme 2: The Practical, Environmental and Emotional Challenges of Self-Isolating --- Self-isolation environment The thing that I couldn't even cope with for that 24 h I was there, was doing everything in one room, like exercise, uni work, sleeping, relaxing, reading, eating, all of it in one single room. I need separation, I need to be able to step away from things. FG1, S2 I was in a house with 8 so there was 8 of us there and we could all do things like, I don't know, we sat down and watched some TV together or we watched a movie or we played a game. I think just having a lot of people there helped to keep me distracted and not get as worried about the whole situation. FG27, S3 I've had quite a few conversations with international students more recently where I've had a student that recently said "I've not been in the presence of another human being since September" and they've been in Nottingham. They've moved away from their home country to come here and not had virtually any contact with anybody at all and have been self-isolating by their own choice. And again, that is a real issue because they literally have no support network here whatsoever. Staff interview 14 --- Emotional effects of self-isolation To me it didn't really make much of a difference because we were already in lockdown but yeah, I think some of my friends found it a bit harder because they had other friends outside the house that they enjoyed seeing, but personally I was fine with it. FG17, S3 I found it hard to concentrate on my studies because I was stuck in my room all day, I just found the motivation side really tricky. FG1, S7 When I had to self-isolate it was a bit like, for me I was like missing out on the social part of it, which affected my mood a bit and like my motivation to work was really affected at the time. FG17, S2 I was the first person in my household to actually test positive for Covid and have symptoms I felt quite guilty that the rest of my house had to isolate. FG5, S2 How self-isolation affects university service provision Some halls we were at 70% students in isolation. So, the biggest issue was how we serviced these students, how we got food to these students from a catering perspective, dealing with um, you know dietary requirements and getting all of that information through. Um, cause we, we just didn't have a system that could cope with that to be fair. Staff interview 11 I think it would be nice if the staff had something, you know, more than just a letter to say, you know, this week we're going to say thank you to other people, you know, the gardeners for making the area look nice, you know, it should be that the staff can be able to talk to somebody and there's nobody to talk to private and confidential. The staff have got problems at home as well, you know, they're coming into work with families that's ill and then we've got schooling problems. And I don't think the university are looking at that. Staff interview 12 It's been so difficult this year to plan any work, I definitely feel that the majority of my work has been reactive. Staff interview 7 --- Theme 3: Social Factors and Their Impact on COVID-19 Testing, Vaccinations, and Self-Isolation Adherence Testing: social factors, barriers and enablers I'd say that one of the best parts that the university-well, one of the best things the university has done was that big mass testing that they did when everyone was leaving. When I was speaking to some of my friends at other universities, they didn't actually have that, which I was kind of surprised about because I thought it would be every university. It was organised very well, it was carried out super efficiently. FG27, S4 I was definitely happy to have it cause it meant I could come home and not feel guilty for being 'a spreader'. FG3, S1 I think um, that a need to avoid self-isolating has led to people not following the guidance, and has actually led to people harassing other students. Um, because someone goes and has symptoms and gets a test, it then means everyone's got to self-isolate, so I think there's been pressure to either for people not to test, um, and for people to, or not to share that they've tested. Staff interview 20 Compliance with self-isolation I think also people with um like mental illness, or, not necessarily illness but just, um, er struggling with their, with poor mental health. And I think if, if obviously it's better to probably, I mean I know it's technically breaking the law but it's probably better to meet up with someone when you're in self-isolation if you're having for example suicidal thoughts. FG4, S3 I think where other people before they thought they might have to go into isolation they quickly did a food shop, we were able to not do this because we all had friends that weren't in isolation and didn't have any symptoms, so they could bring us food to last us until we could get a delivery slot. FG5, S1 I think not all students take it equally seriously, and I think that's the main thing is that a lot of them just kind of don't see the point, and there's also a lot of misleading information that young people are less likely to get covid badly. FG3, S1 They absolutely weren't happy and I don't think they understood why they should be self-isolating. But on the other hand we have students who always follow the rules and do as they're told. Staff interview 15 Almost 100% of those students who were found to repeatedly breach Covid rules, and their letters of appeals to the registrar, cited severe mental health problems as a reason to not adhere to the rules... and actually we found that almost 100% of the students didn't have significant mental health problems. Staff interview 17 Theme 4: Supporting Self-Isolation: Factors with the Potential to Help Self-isolation mind-set It was quite nice and it very much lifted the mood of kind of-because before isolation it's, kind of, you don't really-you kind of just take it for granted because, as I said, I wasn't going to the university really at all, so actually being able to isolate and then come out made me feel a lot better about being out. FG5, S4 I think when they first got there, whole blocks were having to isolate. It was almost excitement, 'ohh, we'll do it together', do you know what I mean? Staff interview 22 I think after a couple of days I started coping a bit better because I had ways to distract myself, occupy myself with work or other things. So it got a lot easier throughout it. FG27, S1 I think that what the university wants to say is that it's important to self isolate, so, and it is also an incentive for them to self isolate for all the period, not only for the few days and maybe you've got your fever is just disappeared and you're young and so you want to go out. Giving them something to do for seven days, ten days makes them maybe feel that it's really important to do that and it's also an incentive. Staff interview 16 University support I was made aware of the options that were given because I remember when the first lockdown started, we received loads of emails from the university telling us about options if we wanted to get in contact with anyone about mental wellbeing, or any kind of support from either the university or your own course or department, which I found very helpful but personally I didn't use any of those. FG14, S2 I really liked the university doing was they had a lot of sessions about kind of mental health, but one of the problems is you had to search for them, or they were hidden in the, halfway through an email, so it wasn't that the university hadn't provided it, it was that kind of, it was there but you have to go and find the necessary links and kind of do it. FG2, S7 I mean for the students in Halls it's better because they should get their food deliveries every day and they've got the duty hall managers that will be checking in and checking they're all right. But for students who are living in their own accommodation, they're going to become very isolated and I mean there's no guarantee that they've got anybody than can get food to them. Staff interview 9 I mean generally there would have been face-to-face contact. This year we've tried to do that via Teams, but students haven't particularly engaged in that, to be fair, it's been, it has been quite difficult in that respect. Staff interview 11 --- Social support I cope with that a lot of the time by socialising and seeing people a lot, um so I was lucky my girlfriend came and spoke to me through my window. FG1, S7 What made it a lot easier for me was like a lot of my friends knew I was in isolation so they'd call me very often. But I guess if it was a day where I didn't get like a phone call or one of them was busy or something like that, because I wasn't living with my family or housemates or anything, I'd just spend like the whole day by myself. FG27, S2 I think being part of the community where they kind of make it easier and more enjoyable to stay at home, I think that definitely, definitely made it easier to stay at home and adhere to it. FG4, S3 I think that's been especially hard on the first year students, especially first year international students where they don't have a support network. Staff interview 14 I think the most vulnerable were also the first year students so they were here for the first time. They didn't know anybody and it's tougher for them so probably they also tend to approach the university more because they don't know how to ask for help and support. Staff interview 16 I think someone said earlier, but it was a bit of a, like being in first year and with like um, seven people you'd sort of just met, um there was a bit of a like blame game going on where it was like, oh who got the test, who meant we had to isolate, like that caused a bit of tension, and so yeah that was sort of a challenge like to get through like, when you've only known these people y'know a few weeks. FG2, S6 --- Adaptation during the Pandemic --- A New Normal Many students reported initial difficulties adapting to periods of self-isolation, particularly because they felt the lines between self-isolation and lockdown became blurred due to national restriction measures. This sentiment was exacerbated for students living in halls, who were often exposed to extended periods of isolation with other students due to high case rates. For many, these lockdown measures elicited changes to their sense of normal. --- Challenges to Adaptation The challenges of adapting to the pandemic were multifactorial for students and staff, but many reported that these challenges reduced as the pandemic progressed and they adjusted to meet the new normal. For students, challenges centred around disruptions to their studies (particularly as a result of self-isolation), which required increased flexibility from the university and variable teaching methods. Some staff felt that the online teaching modality had created a unique cohort of students with unique needs, with multiple students reporting that they felt disadvantaged by the new learning environment. A broader challenge was the discrepancy between students' expectations of a normal university experience and the infringements placed on this by national measures, particularly for first-year students. For staff, there was a similar adjustment period to the variable teaching methods and different ways of working, alongside trying to meet the expectations of students (which caused students to drop out in some cases). For staff in student-facing roles, there was a general concern about contracting COVID-19, despite being provided with protective equipment. Although the transition was mostly well received, multiple staff members noted that ground staff did not feel their own health and wellbeing was sufficiently supported, creating anxiety and the sense that the university was putting their staff "in jeopardy". --- The Practical, Environmental and Emotional Challenges of Self-Isolating --- Self-Isolation Environment The environment in which students self-isolated was a significant challenge for many, particularly for those confined to small rooms with no outdoor spaces. Feelings of boredom and entrapment were common. The restrictions presented by being required to isolate within a household also had broader impacts, affecting students' relationships with other housemates (both positively and negatively). Despite this, both students and staff generally felt it was more positive to self-isolate with others rather than alone. From a staff perspective, most appreciated the role that environment played in coping with self-isolation, particularly for students living in university accommodation. For staff working with international students, there was recognition that self-isolation presented nuanced challenges, particularly for new international students without existing social networks or supports. --- Emotional Effects of Self-Isolation The emotional effects of self-isolation were variable; some students reported feeling unaffected by self-isolation, whilst others reported experiencing low mood, loneliness, frustration, and a lack of purpose. Many students reported notable decreases to their motivation due to the disruption that self-isolation had on their daily routines and social connectedness. Some students also described feeling guilty if they had received a positive COVID-19 test prompting isolation, given that this would impact others in their accommodation. --- Self-Isolation Impacts on University Service Provision Staff reported being significantly impacted by students' self-isolation. This included dealing with frustrated or non-compliant students, ensuring those who required greater mental health input were cared for, and outreach to students who were self-isolating regularly. Staff noted this added significant burden to their workloads and for some, affected their mood. These experiences were further exacerbated by external pressures, such as adherence to government regulations and balancing this with recognition of the mental health impacts that isolation had on students. In some cases, staff reported "firefighting" throughout the pandemic, being unable to do their usual work, or make developments in other areas of their role due to these additional pressures. It was also noted that additional support for staff would be welcomed, with one staff member feeling that there had been limited support provided. --- Social Factors and Their Impact on COVID-19 Testing and Self-Isolation Adherence --- Testing: Social Factors, Barriers and Enablers Many students felt that frequent asymptomatic testing was useful. However, some staff reported groups of students who refused to participate in testing because they feared repercussions (such as having to self-isolate, being blamed for making others isolate, or being unable to go home for Christmas). Students and staff felt that greater flexibility in testing delivery and incentives might improve adherence. --- Compliance with Self-Isolation Almost all students in the study described feeling that self-isolation was important but acknowledged that some students did not comply with regulations. Many factors appeared to affect students' self-isolation adherence behaviours, and this variability was recognised by staff. These factors included: moral views about adherence (the 'right' thing to do), lack of understanding of the rules, protecting others, rules not being enforced by the university, peer pressure, fear of missing out on social activities, inconsistent or confusing communications, and repeated self-isolation for long periods of time. Students also suggested this resulted from the perception of being at low risk or through general misinformation about health risks. However, there were also many situations where students considered compliance in balance with other needs, such as the availability of external support (to support practicalities such as shopping), professionalism (if they felt non-adherence would put their future career at risk), risk to others (if they felt their overall risk of having COVID-19 at that time was reasonably low), or for mental health support. Some staff also felt the reasons given for non-compliance were not genuine in some cases, for example blaming non-compliance on mental health difficulties that did not exist. 3.4. Supporting Self-Isolation: Factors with the Potential to Help 3.4.1. Self-Isolation Mindset In several cases, students adopted a positive outlook about self-isolation. These students felt appreciative of their normal, everyday lives, and for some, isolation fostered a sense of community with others in the same situation. Students described using distraction as a primary tool to help cope with being in isolation. Staff noted that it was important for the university to incentivise compliance with self-isolation by providing activities to support students for the full period of isolation. --- University Support For those self-isolating, the university provided a variety of support measures, including: the logistical aspects of self-isolation (such as meal delivery, care packages, and rubbish collection), emotional support (such as general phone or email communication to offer support, or online self-help tools), activities (such as online social events, tiered isolation), and teaching variations (offering flexible options for teaching). In a small number of cases, students felt that this support was not enough or had not been communicated well. A small number of staff and students also felt the support was inequitable across faculties and different accommodation locations, with some students receiving greater support than others. This was also visible in how students described their knowledge and experiences of the support available, with some receiving additional support and others saying that receiving this would have been useful. Staff also described the increased workload and pressures associated with trying to sufficiently provide this range of additional support services to students. Despite support being available, some staff reported low student uptake of the supports offered, in particular the services designed to support student mental health. --- Social Support As a final consideration, peer support was seen by students and staff as being extremely valuable during the pandemic, for both emotional wellbeing and practical support through self-isolation. Both staff and students recognised the need for increased social support for first year and international students, on the basis that their social support networks were likely impeded because restriction measures limited opportunities to build friendships. There was a predominantly positive view towards self-isolating in a group compared to isolating alone, as many students and staff felt that this social support made a significant difference to wellbeing. However, this also came with its own challenges, particularly if there were disagreements, or extended periods of self-isolation due to large numbers of students in a household who may potentially test positive. --- Discussion The current study provides a qualitative exploration of university students' and staffs' experiences of the COVID-19 pandemic, with a focus on the impact of self-isolation. Although the UK government's roadmap [24] provides a positive route out of national restrictive measures, rising rates of new COVID-19 variants, particularly amongst university students who remain largely unvaccinated, may continue to warrant containment behaviours. We present learnings and potential long-term implications that can be gleaned from the experiences of both students and staff, representing a sector of the current and future workforce. Students and staff reported initial difficulties adapting to the imposed restrictions and shift to online education. Students, particularly those living in halls, described feeling as though they were trapped in prolonged periods of isolation, given spikes in infection rates within the student body, in tandem with strict national lockdown measures. This posed a significant challenge to students' expectations of university life, with both students and staff noting that the diminished opportunity to socialise meant many new students were missing out on vital formative experiences. The long-term impact of such experiences should not be overlooked; studies demonstrate that relationships with peers and a sense of university belonging significantly impact students' social, psychological, and academic outcomes [25,26]. Our findings demonstrate that first year students and international students were particularly impacted by these periods of isolation and lockdown, struggling to develop friendships and social networks as would normally be expected. Although it is unclear whether we will see further periods of enforced isolation following testing, universities must consider the impact that over a year of engaging in stringent protective behaviours has had on student wellbeing, offering opportunities to foster connectedness for students who might have missed them these during the pandemic. Rapid changes to the modality of teaching provision, i.e., moving from face-to-face to online teaching, were challenging for both students and staff alike. For students, selfisolation was particularly disruptive to their studies and led to concern about the impact of isolation on their learning outcomes. Concern was also noted in relation to general educational outcomes, with some students feeling they had been disadvantaged by this shift. To date, evidence on learning outcomes from online teaching has been mixed, with some studies reporting comparable outcomes to face-to-face teaching [27] and others concluding that online learning leads to worsened outcomes [28]. Both students and staff noted concerns about the impact of online learning on educational outcomes. Monitoring the impact that this year of flexible learning has had on students' attainment, whilst considering mechanisms to enhance learning outcomes in those affected, will be invaluable. Assessing for professional or educational disadvantages to this model of learning will be particularly important as students transition into the workforce. Similarly, as universities continue to navigate new formats for education delivery, the impact of this delivery on staff who might already face high workloads should be assessed, with appropriate supports put in place to support staff wellbeing. The emotional impact of self-isolation was multifaceted and diverse, with some students coping well and others experiencing deleterious effects on their mood. For some, self-isolation also fostered a sense of positivity and appreciation of their daily lives post-isolation. In line with these findings, early qualitative work suggests the pandemic has fostered opportunities for appreciation and gratitude by considering one's personal experience in comparison to others who have been adversely affected (i.e., relative fortune) and has increased time spent with family and engaging in new hobbies or activities [29]. However, burgeoning research has reported the negative impacts of the pandemic and national lockdowns on students' ability to regulate their attention and motivation levels [30] and to develop studying networks [31]. This was particularly true within our sample, with students reporting reduced motivation to engage with their academic work during isolation. Van Der Feltz-Cornelis and colleagues found that social-isolation was a significant predictor of vulnerability to psychological distress in both students and staff members [15], with additional research reporting increases in self-reported stress, anxiety, loneliness, and depressive symptoms [32][33][34] during the pandemic more broadly. Our findings suggest that environment during isolation is a significant contributor to wellbeing, with those isolating with others seeming to fare better. Similarly, Worsley et al. found that student accommodation offers the opportunity to cultivate a sense of community, with accommodation-based pastoral staff providing an important source of support for general student wellbeing and mental health concerns [35]. This echoes our finding that for some, isolation fostered a sense of community with others who were also isolating. However, for others, feelings of boredom, loneliness, and a lack of separation between work and personal space likely contributed to downturns

This qualitative study explored the impact of COVID-19 self-isolation and social restriction measures on university students, through the perspectives of both students and the staff supporting them. The study comprised 11 focus groups (students) and 26 individual interviews (staff) at a higher education institution in England during a period of national lockdown (January-March 2021). Participants were university students (n = 52) with self-isolation experiences and university staff (n = 26) with student-facing support roles. Focus group and interview data were combined and analysed using an inductive thematic approach. Four themes emerged: 'Adaptation during the pandemic', 'Practical, environmental, and emotional challenges of self-isolating', 'Social factors and their impact on COVID-19 testing and self-isolation adherence', and 'Supporting self-isolation'. Students and staff struggled with the imposed restrictions and shift to online education. Students found it difficult to adapt to new expectations for university life and reported missing out on professional and social experiences. Students and staff noted concerns about the impact of online teaching on educational outcomes. Students endorsed varied emotional responses to self-isolation; some felt unaffected whilst others experienced lowered mood and loneliness. Students were motivated by pro-social attitudes; campaigns targeting these factors may encourage continued engagement in protective behaviours. Staff struggled to manage their increased workloads delivering support for self-isolating students. Universities must consider the support needs of students during self-isolation and prepare for the long-term impacts of the pandemic on student wellbeing and educational attainment. Greater support should be provided for staff during transitional periods, with ongoing monitoring of workforce stress levels warranted.

' ability to regulate their attention and motivation levels [30] and to develop studying networks [31]. This was particularly true within our sample, with students reporting reduced motivation to engage with their academic work during isolation. Van Der Feltz-Cornelis and colleagues found that social-isolation was a significant predictor of vulnerability to psychological distress in both students and staff members [15], with additional research reporting increases in self-reported stress, anxiety, loneliness, and depressive symptoms [32][33][34] during the pandemic more broadly. Our findings suggest that environment during isolation is a significant contributor to wellbeing, with those isolating with others seeming to fare better. Similarly, Worsley et al. found that student accommodation offers the opportunity to cultivate a sense of community, with accommodation-based pastoral staff providing an important source of support for general student wellbeing and mental health concerns [35]. This echoes our finding that for some, isolation fostered a sense of community with others who were also isolating. However, for others, feelings of boredom, loneliness, and a lack of separation between work and personal space likely contributed to downturns in their mood. Importantly, both access to greenery or green space and the ability to be physically active offer protective mechanisms against the negative mental and physical impacts of isolation [36,37]. Although we may see a return to pre-pandemic levels of anxiety and depression, this pandemic has highlighted the impact that personal environment and accommodation can have on mental health. Looking forward, as many workplaces and academic environments begin to embrace a work from home culture, consideration for the potential negative impacts on social network building, a lack of division of personal and professional space, and an increased risk of loneliness should be considered. Despite negative media attention, university students demonstrate high rates of adherence to self-isolation following testing [9]. Students in this study described being mostly adherent, although multiple factors appeared to influence adherence to self-isolation. Students frequently reported pro-social factors as positive influencers of adherence (e.g., moral obligation). Recent research suggests a drive to socialise also impacts adherence to COVID-19 protective behaviours [12]. Indeed, both staff and students viewed peer support as being instrumental during the pandemic, bolstering emotional wellbeing and providing practical support through periods of self-isolation. Students and staff noted that communication factors, such as confusing messaging or general misinformation about health risks may negatively impact compliance. Younger adults frequently derive health information from social media platforms [38], however, there is ongoing concern about the spread of misinformation on through these channels [39]. Importantly, government-supported messaging accounted for only 11% of COVID-19 information on YouTube [40]. Ultimately, students appeared to consider adherence to isolation in balance with practical, professional, and mental health factors. These factors indicate that a multi-pronged approach to facilitating adherence in young adults is required. First, staff noted that adherence to isolation could benefit from incentivisation, aligning with government recommendation [41]. Universities should focus on promoting clear and targeted messaging through relevant channels, such as social media, to help combat misinformation and non-compliant behaviour. Novel campaigns targeting pro-social behaviour may reinforce key drivers of student adherence. Consideration should also be given to the future impacts of isolation on students' professional goals, for example working with students to develop learning plans for practical placements impacted by isolation. Finally, some students felt that support was not clearly sign-posted and differed by department, whilst staff felt that the offered supports were underutilised at times. The discrepancy between perceived availability and uptake of services warrants further exploration to ensure barriers to access are minimized. Beyond self-isolation, the competing demands placed on young adults to manage practical and professional needs with protective behaviours should be recognised, particularly as rates of COVID-19 infection continue to rise nationally. To date, there has been limited exploration into the impact of the pandemic on university staff. Stadtlander and Sickel conducted interviews with university faculty and found that whilst staff were generally satisfied with their home workplace, they felt more negative about online work and a loss of freedom and independence. The authors noted that some staff experienced an increase in their working hours as a result of the transition to online working, with some feeling overwhelmed by this transition [42]. The current study purposively sampled staff in student-facing roles from a variety of role families to represent the make-up of the university workforce. Our work suggests staff across all role families struggled to adapt. Many staff reported continued contact with students throughout the pandemic, leading to personal concern about contracting COVID-19, particularly when thinking about their own families and caring for vulnerable others. This anxiety occurred despite recognition that the university provided sufficient protective equipment. The need to support students who were self-isolating also presented a significant challenge to staff, who were expected to deal with a range of resulting issues, from mental health difficulties to non-compliance. Some staff noted changes to their mood as a result, with others finding this additional work added substantially to their workload. For many, this new and reactive way of working induced significant stress, meaning that they were unable to direct attention to other aspects of their role during this time. This led some to feel that greater support for staff was required from the university. Although the initial stressors of adaptation have subsided, we may begin to find that there are mental health sequelae of the prolonged state of reactivity and stress induced by the pandemic. We have seen the impact of this reactivity on the mental health of health care workers [43]. In the current study, it was noted that whilst staff felt services were widely available to students, university support programmes were lacking for staff. As we continue to move through the pandemic, care should be taken to monitor staff stress levels and mental wellbeing, ensuring staff specific support services are available and accessible. --- Strengths and Limitations This study has several strengths and limitations. The interest and uptake of the focus groups and interviews amongst students and staff demonstrates that remotely conducted focus groups are an acceptable and suitable approach for exploring experiences of selfisolation in both university students and staff during the pandemic. Study rigor was strengthened as all focus group data were collected by two researchers. Additionally, whilst analysis was conducted by one researcher, a second was involved in checking of codes to ensure reliability and rigour of the data. The student sample included both home and international students, although the proportion of home students in our sample is slightly higher than the proportion of home student registrations at the university during the 2019-2020 academic year (82.7% versus 67.9%) [44]. Although measures were taken to improve representative sampling, many students had undertaken asymptomatic testing and were actively involved in COVID-19 related research. Most students also reported being compliant with guidelines, thus, the views of those who are disengaged, ambivalent, or largely non-compliant are likely not represented. We also attempted to recruit a minimum of four participants per focus group and to maximise attendance by offering gift vouchers and scheduling focus groups around the academic schedule. However, given the rapidity of data collection and student non-attendance, we hosted a mixture of larger and smaller student focus groups and one individual interview. Despite the discrepancy in group sizes, the authors felt that the data collected from the smaller groups and individual interview were meaningful, informative, and warranted inclusion within the study. Although our recruitment strategies aimed to achieve demographic representativeness, there were more female than male participants in the sample, with this proportion being higher than the estimated proportion of females across higher education in the UK (80.7% versus 57%) [45]. Studies demonstrate that females find coping with social isolation more difficult than their male counterparts [46] and are more likely to engage in protective behaviours [47] and comply with public COVID-19 measures [48]. Given the gender discrepancy of the sample, our findings are likely skewed towards female viewpoints and experiences and should, therefore, be considered within this context. Future research should endeavour to over-recruit male participants to ensure a reflective gender balance is obtained. During the recruitment period, students received a high volume of COVID-19 related university email correspondence. As our recruitment predominantly occurred through email, students may have missed recruitment emails and focus group reminders. To improve attendance, future studies might consider offering greater incentivization and alternative communication strategies, perhaps through social media or text messaging. Additionally, the proportion of undergraduate compared to postgraduate students was slightly higher than that of student registrations (88.56% versus 72.7% undergraduates) [44]. The majority of students lived in off-campus accommodation and, therefore, the experiences of students living in on-campus university halls of residence may be under-represented. Future research should attempt to explore the views of those who are non-compliant with isolation measures, perhaps through recruitment from university disciplinary channels. This study has important implications; it is one of the first studies to explore the impact of self-isolation on students and staff in British higher education. However, the data should be considered in terms of the context in which they were collected. Focus groups and interviews were conducted during the third national lockdown, during which many students were, in effect, experiencing self-isolation to some degree. Thus, some of the findings may not be specific to containment behaviours, but more reflective of the experience of lockdown measures in general. Groups took place shortly after the winter break, and so attitudes towards testing and uptake of symptomatic testing may be influenced by the fact that these were encouraged before travelling home, and so may not be representative of students' attitudes during other time-points. Finally, it is worth noting that all students and staff were from a single institution, which experienced a large COVID-19 outbreak at the beginning on the Autumn term, resulting in a large number of students self-isolating and the imposition of stringent local restrictions. However, data were collected during a national lockdown and, thus, likely represent the student and staff experiences of large campus-based universities in England. --- Conclusions The COVID-19 pandemic has significantly impacted students and staff, who were required to make rapid adjustments to novel learning environments. Although adherence to containment behaviours appears high amongst students, current and future mental health implications of isolation should be considered. Ensuring adequate practical, social, and emotional support for both students and staff will be paramount moving forward, given that containment behaviours may be required until high rates of vaccination are achieved in this population. Greater support should also be provided for staff during transitional periods, with ongoing monitoring of workforce stress levels warranted. --- Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to risk of participant identification. --- Supplementary Materials: The following are available online at https://www.mdpi.com/article/10.3390/ijerph182010675/s1, File S1: Study themes and subthemes with supporting quotations, File S2: Consolidated criteria for reporting qualitative studies (COREQ), File S3: Student and staff interview Guides. Supplementary File S4. Thematic map illustrating the relationships between key themes and subthemes. Informed Consent Statement: Informed consent was obtained from all participants involved in the study. --- Conflicts of Interest: All authors were employees of University of Nottingham (UoN), the institution at which data were collected. JC sits on the Executive Board for UoN. JB and CD were involved in the delivery of the university's asymptomatic testing service but were not involved in data collection or analysis for this research. No other conflicts of interest were declared.

This qualitative study explored the impact of COVID-19 self-isolation and social restriction measures on university students, through the perspectives of both students and the staff supporting them. The study comprised 11 focus groups (students) and 26 individual interviews (staff) at a higher education institution in England during a period of national lockdown (January-March 2021). Participants were university students (n = 52) with self-isolation experiences and university staff (n = 26) with student-facing support roles. Focus group and interview data were combined and analysed using an inductive thematic approach. Four themes emerged: 'Adaptation during the pandemic', 'Practical, environmental, and emotional challenges of self-isolating', 'Social factors and their impact on COVID-19 testing and self-isolation adherence', and 'Supporting self-isolation'. Students and staff struggled with the imposed restrictions and shift to online education. Students found it difficult to adapt to new expectations for university life and reported missing out on professional and social experiences. Students and staff noted concerns about the impact of online teaching on educational outcomes. Students endorsed varied emotional responses to self-isolation; some felt unaffected whilst others experienced lowered mood and loneliness. Students were motivated by pro-social attitudes; campaigns targeting these factors may encourage continued engagement in protective behaviours. Staff struggled to manage their increased workloads delivering support for self-isolating students. Universities must consider the support needs of students during self-isolation and prepare for the long-term impacts of the pandemic on student wellbeing and educational attainment. Greater support should be provided for staff during transitional periods, with ongoing monitoring of workforce stress levels warranted.

INTRODUCTION Historically, both cities and villages have been characterized by the existence of common places, where the residents can engage in trade, social relations, culture demonstrations, politics, and other forms of interactions. From the Greek AGORA to the Roman Forum, the fairs and long trade routes of Medieval and early Modern historical frameworks, until today's fairs, these places had a meaning to the people living in their historical times that nowadays can only be reconstructed selectively. They were more than a simple marketplace to buy goods for a living, they were a location of public interaction, cultural effervescence, and socio-economic development. No institution was more important than fairs in the early commercialization of the European economy, following what Robert Lopez called 'the Birth of Europe', from the late tenth and eleventh centuries. This was the very era, according to the well-known if now unpopular Pirenne thesis, in which Europe experienced the revival of long-distance trade and the consequent growth of towns, as the chief motors of economic growth. Lopez also viewed the ensuing three centuries as the 'Commercial Revolution' era, during which Europe acquired the veritable foundations of modern capitalism in new commercial and financial institutions. Even those unsympathetic to the Pirenne or Lopez theses might still agree that international fairs based on long-distance trade did play a crucial role in European economic development during these three centuries. (MUNRO, 2001). The history of common places outside of Europe partly resembles and contrasts these descriptions. The common places of indigenous populations before colonization had their distinct features, even though they shared many of the functions of common places across the Atlantic. More importantly, the colonization set in motion a process in which both enclosure of commons and of (often violent) "commoning" played a crucial role (Greer, 2012). During these processes, new colonial forms of common places were established. With the advance of industrialization, constitution of bigger cities and automatization of daily activities, common places grew in number and size, but they were more like exhibitions and big festivals. More recently, they have been replaced by big stores, shopping malls and the "social networks" in the virtuality of the internet. The open, free, and public places of the old times became owned by big tech companies who operate on the internet, and big corporations and store chains in the "physical world". In this historical process, food played a crucial role, and exemplifies many of the crucial features of the transformation of common places. Nowadays, farmers markets and rural fairs are becoming unlikely and unusual ways for people to buy their "daily bread" and engage in social exchange. In the current system (McMichael, 2005) the food conglomerates do not vinculate the knowledge and reputation of the food to producers, but instead to brands and chains. The traditional ancient territories codify product specifications that guarantee product-specific qualities (Hofmeier, 2023) and farmers are being used as a "selling factor". The new nutritional trends (McMichael 2000) promote "superfoods" or "food 4.0" that explicitly favor enclosed forms of food production and marketing. There are movements, sometimes but not always tied to the academic sphere, who advocate for critical reconsideration of "privatization" spaces, and a return to more communal ways of life (Alkon and Guthman, 2017). Often, these movements act based on a worldview in which the role of networks between human and non-human "actors" is acknowledged, at least implicitly. Still, a frequent and crucial element of these approaches is the establishment of a more direct connection between people (Malak-Rawlikowska et al, 2019). The first step for many of them (Zoller, 2021) is a return to "slow" food culture and tradition. Others emphasize a need to adapt to a new digital, cosmopolitan, and urbanized reality, while not losing the more traditional principles of decentralized food production and marketing. To achieve this ambitious model of a food system combining digital and traditional as well as urban and rural elements, many movements advocate a partial return to "the olds ways"' with more urban and periurban agriculture. Concepts such as the one of retro-innovation call for "the purposeful revival of historic practices, ideas and/or technologies (...) with mechanisms by which old (retro) elements are becoming drivers for new (innovative) ideas" (Zagata, Sutherland, Hrabák, Lostak, 2020). They attempt to involve people with nature, binding humans to their environment. A quick look in a medium or big cities on the other hand convey the inability of traditional farmers or direct purchase in farms to fill the void, drive to food insecurity, shortage of agricultural resources, energy crisis, increase of population and urbanization (Oh and Lu, 2023). Nevertheless, much of the new generation are too guided by a "virtual behavior" and even seem to lack interest or social skills to engage in the traditional communities, fairs, and events. In this situation a new concept is gaining shape, so called "food hubs" or "regional food hubs". The food hub concept is about establishing new ways to organize distribution chains, ways that ensure a more open and integrated system, a rapprochement of producers and consumers, the integration of community into the food system, and a non-"reductionist" approach to human cooperation involving markets (Berti, G.; Mulligan, C. 2016). One of the tasks is to bring the farmers market culture, basket delivery and country fair to a new digital world and to approach these new generations. In this system, producers can be integrated in a virtual platform capable of reorganizing the distribution chain (including functions of management, logistics, sales, and cooperatives) (Palmer, 2013), and filling the gap between farmers and consumers to develop more resilience and local food system (Felicetti, 2014). This is supposed to play a critical role in generating positive economic, social, and environmental impact Barham (2012). So, with these new societies and groups formed we question if the food hubs can play a role in a more democratic and open system of community integration? Which system of governance and cooperation between farmers and consumers can be brought to discussion? We address these questions applying methods from social practical theory (see section 2). Section 3 conceptualizes (regional) food hubs, section 4. --- THE SOCIAL PRACTICAL THEORY In the pursuit of observing social episodes, focusing on the behavior and ethical practice, the Social Practical Theory (SPT) gives a theoretical background to provoke a better understanding of the relationship between attitudes and behaviors, holding an important place in the ethical consumption literature (Carolan, 2021). More than that, this theory is also able to address the agency-structure debate in a broader way, which is also a core subject of analysis of structured multidimensional institutions and collective bodies. The use of SPT can bring a non-reductionist way of perceiving the action of organizations and social networks (Owens, 2000). It challenges a more "neoclassical" and linear approach of rational action by individuals, the interaction of which was a "between individuals" in some sort of vacuum place (Burgess, Harrison, & Filius, 1998). Its scope is more holistic, and to some extent transcends the classical perspectives of "homo economicus" and the "homo sociologicus" (Reckwitz 2002). The SPT tends to focus more on social and symbolic behaviors and practices to address questions. It asks which types of action are regarded as desirable and which norms are legitimate. It also acknowledges the importance of cognitive-symbolic structures (of which language is a prominent example) in the reproduction of social order, even in cases in which a normative consensus does not exist (Reckwitz, 2002). This proves to be helpful in analyzing new and dynamic organization in a "digital and postmodern world". To understand social phenomena through the lens of this theory you must not see the act by itself (praxis) but also the motivational and emotional state, the premeditated act (background in terms of time and knowledge) which many authors refer to as "Pratik". Those elements can't be analyzed or split without losing coherence. The act must be seen as a practice in a holistic way to do so, easy measures must be combined with more abstract information, establishing correlation and evidence who appoint to some direction. This model still works with an understanding of the often-linear nature of a mostly logical behavior done by a mostly rational actor (e.g., Burgess, Harrison, & Filius, 1998;Hunter, 1927;Owens, 2000) but not limited by it. In this way, it turns out to be a suitable theory to study a phenomenon that is not controlled by the environment and the object of study. To Warde (2005), "the principal implication of a theory of practice is that the sources of changed behavior lie in the development of practices themselves." For this present paper, the factual practices and the theory are combined to gain a better understanding of ethical consumption (e.g., Fonte, 2013;Huddart Kennedy, Cohen, & Krogman, 2015) where it "happens not through conscious decision making, but rather through routines that are tacit and unthinking" (Schoolman, 2016). This plays an important role in choosing this theory. This approach will help us to navigate between information collected from papers and work on data and conclusions from authors exposed later. --- CONCEPTUALIZING (REGIONAL) FOOD HUBS When taking into account the existence of a good number of farmers who embrace the organic way to production, focusing on a sustainable production of high-quality health products, and residents of big cities involved in a struggle to get a more natural diet and the anguish of being a part of unequal, corporatized and cold society we can find an easy match. The words "natural, organic, small farmers, with no additives, handmade" sounds a melody for the ears of most consumers and a reality for most organic farmers. A problem for many opportunities of trade is the need of the party to meet each other, be aware of others and what they pose, and offer. The need for this place, or marketplace, is something that appears to be frugal, but is fundamental. The boiling of trade, culture, science, and politics don't take place in the "marketplaces" in the city center for mere fortuity, but because of the huge transit of people and goods in one public place. Awareness is a fundamental part of short chains analyzed by Marsden (2000) and poses the biggest challenge to make bounds between farmers and consumers. The vintage ways to do it, by farmers markets and festivals alone aren't sufficient to fill the necessity of interaction and contact, or still to compete with the facilities of marketplaces, like Amazon fresh or supermarkets in the city. For that, a new approach can be suggested, like food hubs. The term food hubs sometimes can be found in the literature as Regional Food Hubs, and presents conceptualizations with some variation, however, it usually ends up being a delimiter and "centralizer of products, services and information" including "social aggregator", linked to the distribution of food locally produced products in short chains. According to the report "California Network of regional food hubs" (2010), the term is an integrator of distribution systems together among themselves which coordinates produces, stores, processes, markets and distributes regionally produced products, and may also cover additional activities related to marketing, tourism rural areas, production, gastronomy, among others. To show the wideness of the term, we can observe just by looking at the table constructed by (Berti, G.;Mulligan, C., 2016), where they appoint some dichotomic approaches of Food Hubs in literature. Management andTourism, Curitiba, v.10, n.1, p. 426-449, 2024 Table 1. Dichotomic approaches to Food Hubs. Source: Berti, G.; Mulligan, C., 2016 In one way, food hubs have characteristics connected to the supply chain, more specifically to the short supply chain, and to the distribution channels, constituting a "hub" that provides a relevant apparatus for thinking about commercial transactions, effecting, and centralizing many activities. In this field, it can act more like a facilitating "hub" for other stakeholders, carrying out loss-making activities and organizing the chain, or even centralizing the process carried out within the property to the consumer's door in the most varied stages. --- International Journal of Scientific On the other hand, they are involved with NGOs and cooperatives with the intent to challenge the concept of "neo classical maximization of profit" for a system which considers environmental and social issues. They hope to build a more sustainable food system, with an approach on social engagement which provides good access for small farmers to the market and healthy and regional food for consumers. They have this approach because of community cohesion and collaboration rather than a simple trade focused only on the product "per see". For Palmer (2013), the term carries "different meanings for different people". Using his studies in Canada as an example to contextualize food hubs as a fluid entity, with many gray areas in its definition, but having as a key point the integration of the chain and coordination of the distribution of products to the consumer. While Campbell and MacRae (2013), state that the regional food hub is a company or organization that actively manages the aggregation, distribution and commercialization of food products identified by origin mainly from local products and regional producers to strengthen their ability to satisfy wholesale, retail, and institutional demand. Attached to this thought, Felicetti (2014) considers the issue of "regional" relevant in this analysis, the search for sustainability, engaged regional actors, the generation and participation in integrated food systems, which generates a differentiation in relation to global marketplaces and other chain aggregators that eventually use the same nomenclature. According to Rogers (2013), they contribute to expanding access to food, especially fresh food, taking advantage of modern tools to make the distribution process more efficient, with online orders, collection points and delivery routes, while being convenient, with the focus on efficiency. One of the relevant characteristics of food hubs is their integration, which allows producers to place their products in a collaborative way and sell them in expressive volumes, in addition to that, the possibility of accessing markets that were inaccessible by themselves (Felicetti, 2014). Another interesting survey shows that food hubs in California are engaged with each other with common goals to bring together small and medium-sized producers, promoting the growth of local food production systems focused on bringing together cooperatives, NGOs, and sustainable enterprises (California Network of regional food hubs, 2010). Regarding sustainability issues, Barham (2012) points out that food hubs must be financially viable, while also having positive economic, social, and environmental impacts within their communities, as demonstrated by the achievement of a certain production, community, or environmental services and activities. Having, therefore, a role of social aggregator, economic development, and environmental positive role. The food hub, therefore, has characteristics for this thesis aligned with the thought of being and acting as a physical or virtual distribution center that aggregates different skills and has an impact on various dimensions of society. Food hubs are very flexible in the way they can organize themselves and one of the main concerns is the virtual integration of producers and consumers. Within this method they can be molded in a way to emulate some of the new channels of trade who are becoming dominant in the market, but still bring the cooperative, social activism and democratization of supply chains. These are not easy tasks, because a food hub in the best of intentions can be not so different from any food company integrating food chains. The other display to present themselves, like virtual platforms, the food hubs can have a more social background and pursue a direct involvement of social activism combined with a strong voice for the farmers which tends to have an approach to be more focused in the interaction and the change of food systems than just thinking of a theoretical "classical firm". This combination reflects the way food hubs behave and results in interesting aspects. One of the characteristics is the projection of the farmers on the internet in a multiple sense. The farmer has the opportunity not only to show his products, and to make a business relation, but use this open channel to express his opinion, daily life, and expectations. This mix of social media network, market and associationism brings to the internet something very common in country fairs and even farmers markets, in that way they can not only target more people but can engage the new digital generation. On the consumer side, they enact the possibility to directly affect some communities and farmers at the same time strategies such as CSA (Community Supported Agriculture) and other production models. They open a line of contact with farmers who possibilities the awareness process through the interaction with the farmers themselves or produced content on the food hubs portals and through this interaction possibilities not only clarification, but to generate a bound. A great aspect of the viability of these enterprises is the fact that they have a possibility of an online buying process, a recurrent form by the subscribe options, and facilitate the process of acquiring food at the same time bringing the needs of fresh vegetables delivered to the door in more than weekly timing. This surpasses the convenience of buying in supermarkets and at the same time helps engage with the other aspects of the food hub. --- THE SEARCH FOR MORE SUSTAINABLE AND DIGITAL SUPPLY CHAINS IN AN OPEN, COMMONS-BASED FOOD SYSTEM From the preceding section, it should have become clear that food hubs situate themselves at the nexus between the two transformations that many observers believe to be the context that will define socio-economic developments in the next decades, the sustainability issues, and the digital revolution. In this section we first describe how these two transformations are related to food hubs. We then highlight some of the tensions and contradictions that emerge due to the food hubs aspirations to play a role in combining sustainable and digital futures. --- FOOD HUBS AND THE SUSTAINABILITY TRANSFORMATION The increase in production and the interest in sustainable and healthy food is on the agenda of many initiatives and debates. Often decentralized, with discussions revolving around the search for a rapprochement between producers and consumers, in a kind of rediscovery of the countryside and food, focusing on the search for "local" consumption (Seyfang, 2008). This debate becomes fundamental due to the crisis that the current system is facing in social and environmental terms, against the backdrop of the need to reorganize the global food market. These new markets, such as "food hubs", are also organized with the help of movements that go beyond criticism of the current system of production, distribution, and marketing. They try to think of solutions to many of the challenges of the 21st century, such as food security, degradation of the environment and production systems, helping to produce less harmful impact in terms of Co2 and other emissions. In addition to security and the climate issue, other topics such as the loss of people's food culture, biological and food diversity, food waste and health problems arising from lack, excess or inadequate food, are addressed by their correlation with some of the challenges that society must face. Today, there are initiatives that propose and, sometimes, aim to think about the food systems themselves in favor of reducing the environmental impact and improving the quality of food, connected to movements of traditional/local cuisine and organic/biodynamic agriculture. With regard to practical issues, in order to fulfill the initiatives implemented in the issue of urban/peri urban agricultural production, formation of production cooperatives, consumption and more efficient commercialization modalities in terms of environmental impacts (Blake, 2010), strategies range from new forms chain and commercialization to the productive model inside the rural properties. A natural path for these movements is to look at the past, a renaissance in old fashion, but in a new light, adapting many of the characteristics which had been lost in the new way of life of the 21st century. In this scope, the farmers' fairs and traditional basket delivery have a problem of scope and scale to fulfill the complex and demanding food system involved in big cities around the globe. --- FOOD HUBS AND THE DIGITAL TRANSFORMATION The aspect of consumer behavior and the "internet age" also play a bigger part in the need to adapt old practices to accommodate a new, more digital generation of consumers. From work to pleasure, computers, smartphones and the broad use of the internet take a main role in everyday life. In the advent of COVID-19 the "delivery culture" took an even stronger place, putting regular shoppings at risk and street commerce and markets fairs at the brink of collapse. Some authors (like Perea et al, 2004) already preview the expanding interest to consumers to migrate to online shopping, appointing some limitations of that time on platforms and consumers behaviors and knowledge. Late works (Hand et al, 2009) show there is still a conditional online groceries shopping, but in a growing trend, post covid papers like Jensen et al (2021), not only show a doming ascension of online shopping, but an intent to the consumers continue to do that after COVID-19 even in the end of pandemic. The reasons pointed out are many: from the growing percentage of population already in a digital society, the traffic and lack of time in big cities, the price of products and the lack of desire/option to have social interaction in shopping runs. The shopping experience also changes from a rudimentary sea of pages to a few omnichannel where you can find almost anything: products, social interaction, and media like amazon on you have the delivery of products of thousands of brands, channels for debate and "amazon prime videos" with streaming media content. Although, especially the "iGen generation" was branded as isolated with lack of social skills, some movements, and trends for them are trying to engage in many social endeavors from justice, global warming solutions and identitarians respect. This behavior is because the time adolescents spend on digital media may displace time that might have been spent on in-person social interaction (Twenge, Spitzberg and Campbell, 2019), producing the lack of past values and community sense in a corporate owned digital world are bringing a dystopian sense of future for much of the youth. The way to do it, and the place of community and social relation are something still much debated without a clear deduction. In the beginning of the 2000's the World Wide Web entered a new step, going from a more static structure to a more flexible one, with the possibility of interaction and collaboration. Much of this change emerged by the introduction of social networks like Facebook and twitter, and the user generated content that emerges from the interaction and collaboration of the users. Instead of the content of the web being a passive experience by the user, they became the center of the internet (Schrape, 2021). The prospect of a web center in dynamic platforms instead of deposit of content makes a possibility of bringing to the internet some practices which have had good results, but sometimes, limited scope in the society like shared economy, crowdfunding, solidary markets, and other initiatives. The collaboration and new possibilities of this new way to use the www (world wide web) food hubs can emulate mostly 3 aspects: Hybrid Economy, Platform Economy and sharing economy. The hybrid economy according to Rifkin (2014), can have at the same time characteristics and participants of capitalist private markets and collaborative commons. In that social capital and means of production both have relevant play, with competition and cooperation mixing in a "coopetition", not only focusing on "exchange value" but also in a "sharable value". About this issue, the sharing economy has a central role, the collaboration and couse of resources are a fundamental part of this new system. In food hubs this can be in terms of activities like trucks, platforms or know-how and experiences. This reduces the capital and maintenance of the structures and incentives, collaborative investment, and innovation. And to juxtapose all of that, the platform is the "avatar" of a food hub, centralizing and organizing most of the activities, reducing translation cost, facilitating the transition in terms of payment, logistics and social interaction. In this way, a digital food hub can present some flexibility and be part of private ownership, but still have a high level of "common participation" with public and collective money and funding. Also, there are visible elements of sharing of resources like trucks, deposits, and websites with the possibility of collective buying from suppliers and collective selling by the platform. This platform enables the use of the internet to organize and make almost all aspects of commercialization, payment system and logistics at the same time as a channel for marketing and to enable communication and initiation between members of the hubs and the community. --- CAUGHT BETWEEN TWO TRANSFORMATIONS? Food hubs can thus be described as not only taking part in the two big transformations characterizing current socio-economic developments. They appear to be among the few types of entities that seek to actively position themselves at the nexus between those two. In contrast to many other attempts to reconcile sustainability and digitalization, their structures are intrinsically connected to both transformations. At the same time, being positioned at the nexus of these two issues can also create the risk of being worn down between the two. Usually, this risk is described as conventionalization. More concretely, food hubs seem to face a dilemma: If they wish to play an active role in the sustainability transformation, they must engage in the emulation of a public forum. But to compete with other business models in the digital sphere, they must invest in structures that make them appear to be just another virtual marketplace. --- THE FOOD HUBS DILEMMA In this paper so far, we discuss a lot of ways on how these "food hubs" can help to stimulate connections between consumers and producers and share content with social awareness or even generate a more socially responsible and environmentally conscious food chain and market. All the sea of possibilities and characteristics brought by some authors to this paper can be exemplified in these food hubs, who sometimes are mistakenly guided to become a "just another marketplace" by a "trend" called "conventionalization". The conventionalization thesis refers to the process by which organic farming increasingly shares the same basic features as conventional production (Rover, Pugas, De Gennaro, Vittori and Roselli, 2020), when an alternative system has a differential proposed driving to the center, instead of a common standard. One good example is organic agriculture, which carries an approach very different from the "criticized industrial agriculture". However, modern organic agriculture is showing signs of increasing intensification and specialization, like trends in conventional agriculture (Wit and Verhoog, 2007). But nowadays, we have some producers presently and close to the farmers, this movement could be described "beyond organic" like agroecology, permaculture and other aspects incorporated in a productive system. Not just the conventionalization poses a risk of taking off the intrinsic characteristics of a "food hub" into becoming a "regular all brand store". The element of Greenwashing is also a threat to this business model. This subject could disrupt the food hub model, gathering into a kind of the basics and natural aspects of it, turning into a "regular all brand store". So, in this process, the regular consumer will be unable to feel and observe significant differences between them. Greenwashing is known to be used by many companies to "show" "progressive, green, and sustainable practices", who are not doing in extent or profundity being more like a "marketing" or "just for show" purpose, when firms exhibit positive green communication or pretend to be environmentally friendly (Santos, Coelho, e Marques, 2023). Looking backwards to the concept of food hubs, they appear to have a relationship with many aspects and practices which are not "taken for granted". They supposedly need a certain number of characteristics to be so-called food hubs in the sense of this literature, but much of that depends on the way they are constituted, managed, and operated. Like many other concepts, the food hub itself can be unfit to be called that sometimes. There isn't any net of farmers or sellers theoretically established. There are many gray areas in the way the hubs operate themselves. For that, it seems that the food hubs can play a role in a more democratic and open system, in relation to the community in the matters of integration. But how that can be achieved depends on the system of governance and values held by the hub. For those assumptions, we need to observe within practical examples how this may be relevant to a better model of food hubs fulfilling the objectives and at the same time avoiding conventionalization to have resilience. Observing the complexity of characteristics in this system which operate as multiple tasks, it is common to be in the middle of contradictions and more holistic characteristics, such as a multi-platform and hub system. These "food hubs" generate an environment where they act like a tool, model, product seller and more. These "labels" don't exclude themselves; they are complementary in play in the role capable of enabling the operational capabilities and potential of these "food hubs". The platform is itself technical and capable of showing products, making payment, and managing logistics power, as tools to provide competition, drive and get a whole sum of customers who now shop online and want to have the products delivered to their doors. This new way of commercialization now turns out to be able to put traditional regional organic farmers into direct contact with these customers. But the simple gain in efficiency and being a tool capable of organizing the production chain is not enough to make deep changes in the food system, but a need to modify the mentality. This could be illustrated in the article "sustainable agrifood systems for a post-growth world", where the authors defend these changes of perspectives: We can perceive much of those values are in terms of awareness (care) in a way people are bound with others voluntarily and not by control or hierarchy. Also be part of a collective thinking as commons, apart from being in a private closed system help in the process. In addition to that, a vision of social and productive systems focused on harmony with nature is an aspect very present in agroecology, not focusing on a "marginalist maximization of profit" minded action. Authors like Cohen (2016), observe the connection between producers, consumers and the option for a more democratic way of organization and shared production like cooproduction was a way to achieve a more sustainable system. For that the cooperatives in hybrid model, from consumers and producers can be a desired solution. To bring another example, Vivero-Pol (2019) appointed common governance as a key to better manage the natural resources and food in an integrated way with natural systems. The products themselves are an important organism of the system, alas, producing and delivering healthy and sustainable food is one of the main concerns expressed in this paper. Not trying to be reductionist, delivering food from farmers to table in urban areas with competitive prices and few food waste is a great deal and for that they already have many merits, and doing so is possible to make a "bridge" between consumers and producers, but this maybe is just one step. In this scenario governance plays an important role as well the "social capital" of the food hubs. A company who fails to have a good set of rules or a "sustainabilityoriented members" with a good set of skills and awareness can diverge from the path of a "shared economy" local organic food with social engagement to a standard business or just dissolve themselves when the "trending and early interest" vanishes. For that one the first aspects are the horizontalization of organization in a food hub, with shared decisions and democracy. For that, the presence of cooperatives or even capital from some of them can play an important role. This is relevant to generate the sense of commons and the sense of belonging in guaranteeing a better environment to collaborate and avoid "privatization" and "opportunistic behavior" by some members. The awareness between members is fundamental to strengthening the connection in the food chain and generate a sentiment of solidarity in the hub. A short chain approach, with a lot of direct contact and face-to-face interaction is important to improve the social dimension of the food hub. This normally is linked also with a degree of decentralization and autonomy in a way to form in an easy way small and very bond groups how in other "levels" interact with other small groups. For that, consumers must feel as "co-productors", where their role matters to guide and help the impact of the rural reality and productive system, using schemes like subscriptions or even CSA (community-supported agriculture). The farmers in this context also felt a responsibility to the consumers (co-productors) more deeply than a simple way to sell some products, making them more aware of the impact of prices, necessity or more healthy food and good quality products. In the same direction, being close to NGOs and social movements helps to get close to potential new collaborators and to organizations who spread awareness about the necessity of a more sustainable food system. This collaboration can be achieved by sharing platforms, content, organizing events together or other joint projects. Many of these organizations are involved in intellectual and awareness work and are very demanding on "practical action" and "real world" experiences. --- POLICY OPTIONS AND PUBLIC SUPPORT Most of the world's farmers have a government incentive to practice agriculture, in some cases related to some "environmental services" like preservation of the land or protection of some species or traditional "brand" of animals or plants. Many of those come for direct payments (European Union and USA) or tax exemptions (most of the BRICS). In the constitution of food hubs, the role of government is still to be defined, but as are in many commons, states can make a huge contribution or let this model decay. Perceiving "Brazil and USA" the extensionists of rural development of the state have an important role in the constitution of the food hubs. The impact can come in terms of knowledge, in how to "create" a food hub, with technical/agricultural or financial (financing food hubs is a big deal for traction of the system). This action can be noted in some studies when most food hubs relate to public initiatives to fund and finance this business with alliance between government, NGO, and cooperatives. But the organization of food hubs in country zones is not the only role of the government. Logistics play a huge role in this most door-to-door delivery, initiatives like delivering baskets and pick up centers in the UK who can help these hubs to be more efficient, special zoning for urban agriculture to reduce the food miles (Bowyer, Caraher, Eilber and Carr-Hill, 2009). Lending microcredit to help with the constitution of new farmers in existing or needed hubs (Sagbo and Kusunose, 2020), like the example of Gana which pose a positive relationship between farmers' decisions (Martey, Etwire, Wiredu and Dogbe, 2014). Governments can play a crucial role in terms of infrastructure, finance and legislation of the cities and rural areas. To achieve that turn to be relevant included in urban and rural developing plans, the necessity to adapt cities to be more "welcoming" to a local door-to-door delivery, increasing "interpost" and spaces for farmers markets can work as "pick up points" and areas of events and interaction. There is no "neutral city design" and for that, the government should prepare new models of sharing economies, hybrid and virtual systems like food hubs and many others. --- CONCLUSION Digital food hubs are platforms which can enable a flexible system of cooperation and organization with a shared and hybrid economy, with support for a CSA (community supported agriculture), cooperatives and other forms of participative organization. This allows food hubs to play a role in constituting a more democratic and open system with increased community integration. Nevertheless, being a food hub is not enough to play this role. In fact, the preconditions that need to be fulfilled are manifold, and not easy to implement. First, food hubs themselves need to find governance structures that

This article covers the concept of food hubs and their relation with the common places, and other possible interactions within the civil society, in the eyes of the theoretical point of view of the "social practical theory". Resembling the conceptualization of the food hubs more in a general way considering the problems and particularities of these systems, recognizing the different business models in the market. Additionally, we will explore some of the values and prerogatives spread by those hubs in connection with the environmental and climate change assumptions and scientific research, using a short supply chain to distribution channels in their commercial transactions. Understanding the role of the hubs in building a sustainable transformation together with those movements, NGOs, private and government initiatives additionally with the digitalization of information and relations between societies.

Dogbe, 2014). Governments can play a crucial role in terms of infrastructure, finance and legislation of the cities and rural areas. To achieve that turn to be relevant included in urban and rural developing plans, the necessity to adapt cities to be more "welcoming" to a local door-to-door delivery, increasing "interpost" and spaces for farmers markets can work as "pick up points" and areas of events and interaction. There is no "neutral city design" and for that, the government should prepare new models of sharing economies, hybrid and virtual systems like food hubs and many others. --- CONCLUSION Digital food hubs are platforms which can enable a flexible system of cooperation and organization with a shared and hybrid economy, with support for a CSA (community supported agriculture), cooperatives and other forms of participative organization. This allows food hubs to play a role in constituting a more democratic and open system with increased community integration. Nevertheless, being a food hub is not enough to play this role. In fact, the preconditions that need to be fulfilled are manifold, and not easy to implement. First, food hubs themselves need to find governance structures that prioritize horizontal organization, with democratic process of decision making. is a step necessary to explore the potential of this way to organize farmers and consumers. Second, food hubs need to invest in human capital, third, structures of collaboration with NGOs, cooperatives and social movements must be established and developed, to finally, issues of ownership and the involvement of the public sector must be addressed, including with respect to the role of hybrid and public forms of ownership to ensure a more common and open function and avoiding the risk of conventionalization. In the context of the emergence of social media and a step further in digitalization of social relations and commerce, models like food hubs can be an alternative to traditional food system, preserving some of the social aspect of farmers markets and state fair, and at the same time having a digital approach suitable to new generation and convenience of contemporary life. The challenge to make a model that can merge the short chain farmers to consume a close approach like short supply chain to a digital order door-to-door delivery. Governments can have a big role in the formation and resilience of the hubs. Contribution can be in terms of infrastructure and rural support (agriculture experts, business, and financing), but also with city planning and rules, giving spaces for the "common areas" like parks, farmers market able to interact and make events, organize the supply chain, and escalate the existing system. Many governments already have specific programs to enhance food hubs, having already done some impact in this model.

This article covers the concept of food hubs and their relation with the common places, and other possible interactions within the civil society, in the eyes of the theoretical point of view of the "social practical theory". Resembling the conceptualization of the food hubs more in a general way considering the problems and particularities of these systems, recognizing the different business models in the market. Additionally, we will explore some of the values and prerogatives spread by those hubs in connection with the environmental and climate change assumptions and scientific research, using a short supply chain to distribution channels in their commercial transactions. Understanding the role of the hubs in building a sustainable transformation together with those movements, NGOs, private and government initiatives additionally with the digitalization of information and relations between societies.

Introduction Obesity is a chronic disease and an established risk factor for many noncommunicable diseases, including ischemic heart disease, stroke, diabetes, chronic kidney disease, and cancers [1]. It is associated with an increased risk of premature death, with the impact being larger with multiple years of the condition [2]. Over the past few decades, global obesity rates have shown a constant increase, and by 2016 thirty-nine percent of the world's adult population was overweight and 13% was obese [3]. Iran has experienced a similar trend in recent decades [4]. In 2011, the prevalence of overweight and obesity in Iranian adults was reported to be 34.5 and 21.5%, respectively [5]. However, rates vary substantially across the country, ranging from 27.0 to 38.5% for overweight and from 12.6 to 25.9% for obesity [6]. Regional variations in obesity prevalence have also been found in other parts of the world [7][8][9]. Much of these variations have been explained by the interactions between environmental and individual factors, including sex, age, and socioeconomic status. Hence, a predictable pattern exists for obesity in subpopulations. In low-and middle-income countries middle-aged adults (mostly women with a high socioeconomic status) are more likely to be overweight or obese, whereas in high-income countries both sexes across all age groups are affected; however, the prevalence tends to be higher in lower socioeconomic strata [10]. --- Obesity in Iranian Adults Existing studies in Iran have also demonstrated the environmental and individual factors that contribute to obesity; middle age, a lower educational status, and living in urban settings were found to be positively associated with overweight and obesity [5,11]. However, further information on regional variations in obesity rates is needed since these studies often do not indicate geographic differences, which tend to be as important as national data for public health strategies. Additionally, to reduce health disparities across socioeconomic groups, determination of the obesity-related inequality is helpful for understanding the need to refocus health policies aimed at decreasing the obesity prevalence at the population level. Providing evidence on the social disparities of the disease could also assist decision makers in effectively choosing the related health service expenses that need to be covered under basic insurance plans [12]. Consequently, this study sought to assess the prevalence and sociodemographic factors associated with obesity using data from the most recent nationally representative survey, i.e., the National Survey of the Risk Factors of Noncommunicable Diseases (STEPS), conducted in 2016. Moreover, the geographic distribution and socioeconomic inequality of overweight and obesity in Iran are determined with the intention of informing health policy decisions with the currently most at-risk groups for overweight or obesity. --- Methods --- Study Population This descriptive study was performed using data from the seventh-round STEPS survey. This nationwide household survey was implemented in Iran in 2016. To create a nationally representative sample, adults aged 18 years or older were selected based on a multistage clustered probability design, and weighting adjustments were applied. A detailed review of the survey methodology has been published previously [13]. In this survey, trained in-home interviewers collected a wide range of information on sociodemographic characteristics, health behaviors, history of metabolic risk factors, household assets, healthcare utilization, and anthropometric variables using a standardized protocol. Ultimately 30,541 participants aged 18 years or older were interviewed. However, the analytic sample for the current study consisted of 28,321 adults aged 18 years or older who had complete data on sociodemographic characteristics and physical measurements. The sample population was drawn from both rural and urban settings across 29 out of 31 provinces. The provinces of Qom and North Khorasan opted out of the survey research. --- Dependent Variable Anthropometric measurements were performed by trained healthcare workers during the physical examinations using standardized tools and measurement processes. The body mass index (BMI) was calculated from height and weight data as weight (in kg) divided by height (in m) squared. The continuous BMI variable was categorized based on the World Health Organization (WHO) recommended cut-off values for ease of interpretation [3]. Therefore, individuals were classified into the following 4 groups; underweight (BMI <unk>18.5), normal (BMI 18.5-24.9), preobese (BMI 25.0-29.9), and obese (BMI >30.0). --- Independent Variables Sociodemographic determinants of adult obesity including age, sex, marital status, literacy, employment, health insurance coverage, and residing area were considered as independent variables. Participants were categorized by age into the following groups: 18-27, 28-37, 38-47, 48-57, 58-67, and <unk>68 years. Sex was defined as either male or female. Literacy was divided into 3 categories based on the number of years attained (0-5, 6-12, or >12). Additionally, individuals were allocated to 1 of the following employment status classifications: unemployed or retired, student, or employed. Insurance coverage status was categorized as covered or not, and residing area as rural or urban setting. --- Wealth Index A living standard variable is needed to calculate inequality in obesity. Income and consumption expenditure are the direct economic indicators of socioeconomic status. However, no data was collected about household expenditures or income in the 2016 STEPS survey. Therefore, a proxy measure was created by applying the principal component analysis statistical method to the data on household durable assets, housing characteristics, and access to public services. Variables included in the computation of the wealth index were: possession of a house, car, phone line, LCD TV, cooler, vacuum cleaner, dishwasher, washing machine, oven, freezer, side-by-side refrigerator, mobile phone, or personal computer; existence of a separate kitchen area, bathroom, and central heating system; and access to electricity, a water pipeline, a gas pipeline, and internet. The Kaiser-Mayer-Olkin measure of sampling adequacy and the Bartlett test of sphericity were used prior to PCA to determine whether the asset dataset was suitable for data reduction. The Kaiser-Mayer-Olkin value was above 0. Bartlett test of sphericity was significant; therefore PCA was appropriate for this dataset [14]. The analysis resulted in the extraction of 6 components with eigenvalues >1, which could explain 53.7% of the variation in the dataset. The first principal component was used to construct the wealth index and it was divided into quintiles for further analysis. The first quintile represented the lowest socioeconomic level. The principal component analysis was conducted using IBM SPSS Statistics for Mac (version 26; IBM Corp., Armonk, NY, USA). --- Statistical Analysis The prevalence of the 4 BMI categories was described for the whole country and for each province separately. To assess the association between the sociodemographic factors and the prevalence of preobesity and obesity, a <unk> 2 test was applied. For this purpose, individuals were categorized as being either preobese/obese (BMI >24.99) or not preobese/obese (BMI <unk>24.99). A multiple logistic regression model was fitted to determine the association of each sociodemographic characteristic with preobesity and obesity, while it was adjusted for the others. The dichotomous variable for BMI was used as the dependent variable in the model and sociodemographic variables were specified as categorical covariates. The estimates were performed using IBM SPSS Statistics for Mac (version 26; IBM Corp.). The threshold for statistical significance was set at <unk> <unk>0.05 for all analyses. To assess the degree of inequality in the distribution of preobesity and obesity across socioeconomic groups, the concentration index was calculated for the overall population and for each province. However, since the pattern of inequality may also be correlated with sex, the population was further stratified accordingly to estimate the sex-specific inequality. The authors used Stata statistical software (release 15; StataCorp LLC, College Station, TX, USA) for inequality analyses. To gain more insight into the spatial distribution of the mean BMI, provincial mean BMI were shown on a map of Iran using geographic information system analysis (ArcGIS online). A separate map was created depicting the distribution of concentration indices across provinces. --- Concentration Index The concentration index is the most appropriate measurement for assessing socioeconomic inequalities in health among the few summary indices commonly used for health inequity [15]. The concentration index is measured based on the concentration curve. The curve displays the cumulative percent of the healthrelated outcome on the y-axis against the cumulative percent of the study population, which is ranked by socioeconomic status on the x-axis (from the lowest to the highest category). The index value ranges between +1 and -1, where 0 reflects complete equality, meaning that the health outcome is equally distributed across socioeconomic levels. The index has a positive value if the health variable is concentrated more among those who are better off. A negative value emerges when the lower socioeconomic levels bear a disproportionate share of the health outcome. The concentration index is determined as twice the area between the obtained concentration curve and the 45° line of equity. The greater the distance, the larger the absolute value of the index and the greater the degree of health inequality [16]. However, the bounds of the concentration index for dichotomized outcome variables are affected by the mean of the health variable. In order to rectify this issue, the concentration index can be normalized by using correction methods [17]. In this study, inequality analyses were undertaken considering the Wagstaff correction method. --- Result Data on 28,321 individuals (51.6% women and 48.4% men) were analyzed. Nearly half of the study population was aged between 28 and 48 years. The majority were married and resided in urban settings. More than a third had a high educational status. However, employment rates varied considerably between the sexes and were significantly lower for women than for men (i.e.,10.1 vs. 72.1%, respectively). The distribution of the study participants by sociodemographic characteristics per sex is provided in Table 1. From the study sample, 10,115 (35.7%) had a normal weight, 1,140 (4%) were classified as underweight, 10,492 (37%) were preobese, and 6,574 (23.2%) were obese. In general, more men than women were within the normal range for BMI (i.e., 41 vs. 30.7%, respectively). Preobesity was also more common in men than in 1 shows a map of Iran showing the geographic pattern of mean BMI portioned by province. Higher mean BMI ranges were more common among the northwestern and central territories, yet lower ranges were seen among the southeastern provinces of the country. Table 2 demonstrates the preobesity/obesity prevalence by sociodemographic characteristics. As indicated, the proportion of preobese/obese individuals increased with age until the age of 68 years, after which the prevalence fell slightly. Preobesity/obesity was less common in single adults, while almost two thirds of the married and divorced or widowed individuals had a BMI >25. A negative educational gradient, i.e., a rate decrease with higher education levels, was observed in the study sample. As per occupational status, the prevalence of preobesity/obesity was significantly higher in homemakers or retirees compared with their employed and student counterparts. In general, high BMI were more prevalent in those who were insured and in those residing in urban settings. All sociodemographic characteristics were significantly related to the prevalence of preobesity/obesity in the bivariate analysis. Results from the multivariate logistic regression analysis of the preobesity/obesity determinants are presented in Table 3. Participants in the age group 48-57 years who were married and lived in urban settings were more likely to be preobese or obese. Male sex and being employed or a student were negatively associated with high BMI. After accounting for covariates, educational attainment was no longer a statistically significant predictor. Analysis of the socioeconomic inequality in the prevalence of preobesity/obesity presented a prorich direction. The concentration index for the whole country was 0.15 and it was statistically significant. Sex-stratified concentration index analyses also resulted in positive values and were significant for both sexes. However, women had a larger concentration index than men (i.e., 0.19 vs. 0.11), suggesting a stronger disparity in preobesity/obesity prevalence across socioeconomic levels in women. The provincial concentration indices are summarized in online supplementary Table 2. The values obtained ranged from 0.00 to 0.29. The most unequal provinces in terms of preobesity/obesity distribution across socioeconomic levels were: Kohgiluyeh and Boyer-Ahmad, Kerman, and Hormozgan. The provinces of Markazi, Ardabil, and Gilan had the most equal distributions. The geographic distribution of concentration indices is shown in online supplementary Figure 2. --- Discussion The main findings of this study showed that 60.3% of the participants were affected by preobesity or obesity. The preobesity prevalence was 39% in men and 35.2% in women. The obesity prevalence was 15.6% in men and 30.4% in women. The importance of sociodemographic predictors of preobesity or obesity were also outlined. Individuals in the age group 48-57 years who were married and lived in an urban setting had an increased risk of being preobese or obese. Male sex and being employed or a student were inversely associated with a high BMI. Socioeconomic inequality in preobesity/obesity distribution was quantitatively measured using the concentration index approach. The index revealed a prorich direction which was more severe in magnitude for women compared to men. Overall, the preobesity/obesity rate in Iran is in line with the rates in Egypt (61.1%) and Saudi Arabia (65.9%) in the Middle East region [18,19]. It is higher than the prevalence in Europe (53.1%), and it is approaching the rate in the USA (68.5%) [20,21]. At the national level, a rising trend was observed compared to the prevalence rates reported in similar nationwide surveys performed in 2005 (49.9%) and 2011 (58.1%) [5,11]. A growing prevalence of obesity had also been reported in previous studies performed in Iran and could be attributed to the urban development and lifestyle transformation marked by a sedentary lifestyle and non-healthy eating habits in recent decades [22,23]. Findings showed subnational variations in the prevalence of obesity in Iran which were consistent with results from a systematic review conducted earlier in the country [23]. Preobesity/obesity rates in our study ranged from 40.8 to 70.1% in adults across the provinces. Population differences in genetic susceptibility, cultural characteristics, environmental opportunities for physical activity, and access to healthy diet, as well as regional variations in weather conditions affecting outdoor recreation time, may all have contributed to the observed differences. Age was positively associated with excess body weight until the age of 58-67 years, where the reverse relation began. Similar findings have been reported in other studies [24][25][26]. Aging is accompanied with increased adipose tissue in the abdomen and deposition of fat in skeletal muscles. In addition, as people age, sedentary behaviors become more common, resulting in a chronic positive energy balance which leads to excessive fat tissue accumulation [27]. However, the negative effect of age on body weight in the elderly observed in this study and those in others might be explained by several physiological changes in neuroendocrine function which could lead to early satiety and a low appetite in older ages [28]. Aligned with other studies, this study showed that abnormal BMI was more prevalent among women compared to men [5,[28][29][30]. Sociocultural beliefs which limit outdoor activities as well as biological factors such as hormonal signaling influencing energy expenditure during menstruation, pregnancy, and menopausal transition may influence the predisposition of women to excess weight gain [31,32]. Being single was negatively related to preobesity/obesity prevalence, suggesting that marriage is associated with lifestyle changes that could adversely affect physical activity and eating habits. This supports previous findings in the literature and has important implications for intervention programs targeting families [30,33]. An inverse association between educational attainment and preobesity/obesity prevalence was determined in the present study, which confirms previous findings in Iran and most countries in the Persian Gulf region [11,33]. However, inconsistent findings have also been observed in the literature [20,24]. It is believed that education has a protective effect by providing greater access to health information, increasing the health risk perception, and improving self-control [34]. With regard to occupational status, being employed or a student was found to be a negative predictor of preobesity/obesity presence, which corroborates previous results [5,20]. High BMI were more prevalent among homemakers and retirees, who are believed to have low physical activity levels. Results showed that individuals with health insurance were more likely to be preobese or obese. This implies that insured individuals take on more health risks and are less health conscious than those who are noninsured. It is believed that the success of health care and financial coverage plans in promoting the health of individuals is associated with multiple factors including health education [35]. However, further studies are needed to shed light on this finding. In parallel with previous studies conducted in Iran, the Persian Gulf region, India, and China, residence in urban areas, where a sedentary lifestyle and high-calorie food intake are common, was positively associated with preobesity or obesity [5,30,33,36]. However, it is worth noting that a reverse pattern has also been reported in some regions of the world such as Europe and Russia [20,24,37]. This study also highlighted the existing disparities in preobesity/obesity prevalence across socio-economic groups in Iran. A positive association was observed between being preobese or obese and socioeconomic status as measured by wealth. This finding supports previous studies undertaken in Iran and other developing countries like India and Mexico [38][39][40]. Overall, in low-and middle-income countries, obesity and its related comorbidities are more prevalent in people of a higher socioeconomic status (referred to as prosperity diseases in previous literature) [41]. However, as economies grow, lower socioeconomic groups become more exposed to unhealthy environments and the obesity burden shifts to lower socioeconomic strata [29]. In Iran, despite recording positive economic growth and rapid urbanization in the past decades, a prorich inequality pattern was obtained for preobesity and obesity prevalence in this study [42]. It is assumed that economic development and urban development negatively influenced the lifestyle of people of a higher socioeconomic status by increasing access to ultraprocessed food products, decreasing walkable environments, and promoting sedentary behavior. However, individuals with a lower socioeconomic status were still more likely to have physically demanding occupations and maintain their traditional diet which could reduce the risk of obesity. Portioning the concentration index across provinces can lead to a better understanding of the regional variations in equity in preobesity/obesity distribution. Provinces with a lower overall social health index ratio showed greater inequality in the preobesity/obesity distribution [43]. The major strength of this study is the fact that a large nationally representative sample of adults from rural and urban communities of Iran, which contained both sexes and all socioeconomic levels, was used for the analyses. This could lead to more reliable inferences about the population and provides important information for future public health interventions. Another strength is that the data on height and weight were collected objectively by trained health workers according to a standardized protocol, which results in more accurate estimates of BMI than those obtained from self-reported measures. However, this study has potential limitations as well; first, since secondary data sources were examined, the assessment of all factors associated with obesity prevalence was not possible, and second, considering the cross-sectional nature of the study, drawing conclusions about causal relationships was not possible. --- Conclusion The findings of this study suggest that policies aimed at reducing preobesity and obesity should remain a public health priority in Iran. Given the provincial variation in the prevalence of obesity, there are key geographic areas such as the northwestern and central territories that require additional consideration. Moreover, there are specific determinants of obesity common to all regions of the country, including sex, age, marital status, and education, which can be targeted in national obesity prevention and health promotion strategies. To alleviate the observed socioeconomic inequality in obesity distribution, community-based efforts including nutrition education and physical activity interventions need to have a stronger emphasis on higher socioeconomic groups which tend to be left uncared for in most public health programs. --- Statement of Ethics This study was approved by the Shahid Beheshti University of Medical Sciences Ethics Committee (reference No. IR.SBMU. RETECH.REC.1399.482; approval date: August 18, 2020). All of the participants signed an informed consent form before enrolling into the 2016 STEPS survey. Data was deidentified prior to analysis, and this study was performed according to the Declaration of Helsinki. Participation was voluntary and informed consent was obtained from all of the participants prior to enrolment. The survey protocol was approved by the Ethical Committee of the National Institute for Medical Research Development. --- Conflict of Interest Statement The authors have no conflict of interests to declare. --- Author Contributions Each author substantially contributed to conduction of this research and drafting of this paper. R.A. was the main researcher and involved in study design, literature search, data analysis, data interpretation, article drafting, and finalization of this paper. A.-A.K. was involved in data cleaning, study design, data interpretation, and article drafting. M.-R.S. was the head of the team and involved in study design, literature search, data analysis, data interpretation, article drafting, and finalization of this paper.

Introduction: This paper outlines the prevalence, disparities, and social determinants of preobesity and obesity in Iranian adults. Methods: Data on 28,321 adults who participated in the 2016 National Survey of the Risk Factors of Noncommunicable Diseases (STEPS) survey were analyzed. The body mass index (BMI) was calculated from physically measured height and weight. To assess the association between sociodemographic factors and the prevalence of preobesity and obesity, a χ 2 test and a logistic regression model were used. Socioeconomic inequality was quantified by a concentration index. Disparities in provincial mean BMI and concentration indices were shown on the map of Iran using geographic information system analysis. Results: Overall, 60.3% of the participants were affected by preobesity or obesity. The preobesity prevalence was 39% in men and 35.2% in women. The obesity prevalence was 15.6% in men and 30.4% in women. The mean BMI for the country was 26.5. Higher ranges were observed across the northwestern and central territories. Female individuals in the age group 48-57 years who were married and lived in urban settings had an increased risk of being preobese or obese. The concentration index revealed a prorich inequality, with a greater magnitude among women. Conclusion: The findings suggest that policies aimed at reducing preobesity and obesity should remain a public health priority in Iran. However, a greater emphasis should be placed on the northwestern and central territories and on higher socioeconomic groups.

Introduction Conspiracy theories (CTs) refer to false beliefs in which the ultimate cause of significant social and political events are attributed to secret collusions manoeuvred by multiple actors who work together for their personal gains (Douglas et al., 2019). CTs tend to gain momentum during social and political crises such as pandemics (Bierwiaczonek et al., 2020). Since December 2019, the coronavirus disease (COVID-19) pandemic has put pressure on hospitals in many countries with increasing number of hospitalizations and deaths (World Health Organization, 2021). In the interim, due to the lack of solid scientific and governmental consensus on the causes of spreading and solutions to contain the virus, narratives and accounts about the origins of the virus, its cures, and the probable social and political powerful actants involved are ripping through social media (Ahmed et al., 2020). Therefore, social media has become a hosting platform for promulgating CTs worldwide expeditiously. Misinformation, disinformation and conspiratorial fantasies shared on the social media are overshadowing fact-based information (Pulido et al., 2020), implying the implicit endorsement of CTs that tags with sharing. Results also demonstrated that a remarkable percentage of people explicitly reported believing in COVID-19 CTs (Miller, 2020a;Uscinski et al., 2020). Belief in CTs about COVID-19 has impinged on several aspects greatly. These comprise of yielding poor well-being (Chen et al., 2020), resistance to preventive actions (such as wearing a face mask, washing hands and practicing social distancing) (Motta et al., 2020), racial hostility and exclusionary narratives (Depoux et al., 2020), violent behaviors (Jolley & Paterson, 2020), fears of vaccination and unwillingness to be vaccinated (Romer & Jamieson, 2020). In view of the detrimental effects associated, a surge of scholarly interest in possible factors contributing to such belief emerged. Drawing on the existing literature, the factors can be divided into situational, dispositional and political dimensions. Situational factors are characterised by epistemic motives (e.g. understanding desire for certainty and desire for control) whereas dispositional factors refer to individual characteristics that influence behaviors and actions (Cassese et al., 2020;Miller, 2020aMiller,, 2020b;;Uscinski et al., 2020). From the situational perspective, Miller (2020aMiller (, 2020b) found that the pandemic situation induces feeling of uncertainty which in turn amplifies belief in COVID-19 CTs. Similarly, learned helplessness (a perception of failing to control or change the aversive situation) gives rise to conspiratorial thinking (Farhart et al., 2020 as cited in Cassese et al., 2020). Also, individuals psychologically predisposed to conspiratorial thinking tend to believe in COVID-19 CTs (Miller, 2020b;Uscinski et al., 2020). Conversely, resilience as a predisposition force serves as a buffer against COVID-19 CTs (Miller, 2020b). From the political aspect, Republicans in the United States were found to believe in COVID-19 CTs more than Democrats or Independents (Uscinski et al., 2020) due to extensive criticisms targeting the Republican president in handling the pandemic. Typically, CTs are not supported by evidence that withstands scrutiny but this does not stop them from growing (Lewandowsky & Cook, 2020). A CT could turn out to be true if substantial evidence on the causes and procedures of the theory are put forward in support of the theory (Dentith, 2014), supporting that conspiracies may not necessarily be "a figment of anyone's imagination" (Goertzel, 2020, p. 493). However, some individuals especially ideologues are reluctant to accept the evidence when it opposes their prior beliefs and tend to "defend their prior attitudes and actively challenge attitudinally incongruent arguments" (Kraft et al., 2015, p.121). This can be attributed to motivated reasoning where individuals tend to maintain their pre-existing attitudes through biased reasoning processes, producing defences that mainly serve their personal desires rather than accurately reflect the evidence (Kunda, 1990). In view of this, a CT is likely to remain a CT, even in the emergence of supporting evidence, for individuals with motivated reasoning as they have the tendency to accept what they choose to readily believe with less scrutiny than that which they desire to believe. In attempting to understand the influence of CTs on individuals, researchers turn to examine the factors that render people susceptible to CTs. In this regard, media skepticism which is the core to media literacy has received considerable attention due to the proliferation of falsehood on social media (Fletcher & Nielsen, 2019;Vraga & Tully, 2019). Media skepticism enables users to combat disinformation and misinformation (Ashley et al., 2010;Maksl et al., 2015). The nuances of COVID-19 CTs aplenty in the absence of more informed knowledge create fear and insecurity that affect society negatively. Therefore, the present study aims to provide a better understanding of individuals' behavior with regard to believing in CTs during the pandemic by examining the association of skepticism towards social media content and COVID-19 conspiracy ideation which is relatively a new research direction. While the association between skepticism and belief in CTs seems to suggest a direct relationship, very often, individuals' behaviors are navigated by a complex interplay of a variety of components, let alone strive through difficult periods such as the COVID-19 pandemic. Therefore, a simple correlation between social media skepticism and belief in CTs is insufficient to provide an understanding of the complex behavior of the CT-oriented people. It is claimed that differences in dispositional and personality characteristics may affect the extent to which individuals believe in CTs (Swami et al., 2016). Conspiracists' belief is also linked to the dark triad which is made up of three personality traits (i.e., Machiavellianism, narcissism and psychopathy) (Douglas et al., 2017;March & Springer, 2019;Nowak et al., 2020). Association between skepticism and belief in COVID-19 CTs can be influenced by dispositions of social media users. Underlying this background, it is timely and useful for researchers to investigate these personality traits as third variables that may alter individuals' disposition toward the belief in CTs. In the present study, these personality traits will be introduced as moderating variables in explaining the relationship between social media skepticism and belief in COVID-19 CTs. The present study sets out to achieve two main objectives: (i) to examine the relationship between social media skepticism and belief in COVID-19 CTs and (ii) to test the moderating role of dark triad personality traits in the relationship between social media skepticism and belief in COVID-19 CTs. --- Background of the Study --- Media Skepticism and Belief in Conspiracy Theories Skepticism is a "tendency towards a kind of judicious doubt and promotes recognizing any information or rationale offered as limited by the perspective and/or motivations of its narrator" (Feuerstein, 1999, p.45). It invokes reflective thought and careful consideration of any beliefs (Feuerstein, 1999). Media skepticism refers to mistrust toward media messages and the way information source functions (Tsfati, 2003). It is conceptualised as questioning about fairness, factuality, accuracy and completeness of media messages (Maksl et al., 2015) with emphasis on critical thinking (Feuerstein, 1999). The mechanism of skepticism aids to reduce susceptibility to misinformation effects as it encourages interrogation of origins and accuracy of information on media (Lewandowsky et al., 2012;Tsfati, 2003). Media skepticism was found to be the predictor of assessment of news source credibility (Carr et al., 2014). Aside from that, skepticism can also be a solution to various problems entrenched in contemporary news consumption where rumours, misinformation and conspiratorial thoughts spread much faster than truth on social media. However, results showed that social media users possess a sense of 'generalised skepticism' which serves them as a deterrent for not scrutinizing information critically (Fletcher & Nielsen, 2019). Media and news literacy orientations can promote skepticism toward political CTs (Kahne & Bowyer, 2017;Maksl et al., 2015;Vraga & Tully, 2015). An emerging body of literature has been conducted to investigate the potential of media literacy that aids to enhance individuals' skepticism in combating misinformation on social media. For example, Kahne and Bowyer (2017) found that greater media literacy enables individuals to assess evidence-based posts more accurate than misinformation posts. Besides, warning about misleading and false information on Facebook also influenced users' judgment toward the accuracy of such information (Clayton et al., 2020). Likewise, news literacy on social media sites like Twitter would reinforce perceptions of high-quality information and increase doubt about low-quality or false information (Tully et al., 2020). In other words, skeptics demonstrate better adjudication of information accuracy. --- Moderating Role of the Dark Triad The dark triad is a set of personality traits namely Machiavellianism, narcissism, and psychopathy. These three traits "often show differential correlates but share a common core of callous-manipulation" (Furnham et al., 2013, p. 199). Machiavellianism portrays a tendency to exploit, deceive, and manipulate others (Christie & Geis, 1970). Machiavellians tend to be cynical and distrustful (Christie & Geis, 1970;Furnham et al., 2013;Jones & Paulhus, 2011). Machiavellianism is also characterised by a strategic-calculating orientation (Jones & Paulhus, 2011), which fits well with the conception of the Machiavellian as a conservative and utilitarian (Christie & Geis, 1970;Jones & Paulhus, 2011). Driven by self-gain, individuals with Machiavellianism are more likely to appear low on honesty-humility (Ogunfowora et al., 2013). The emergent literature has shown that Machiavellianism is positively associated with conspiracist ideation (March & Springer, 2019;Nowak et al., 2020). Often, Machiavellians perceive that those with power in the government would engage in similar tactics and they are inclined to conspire too if they were given the power (Douglas & Sutton, 2011). Research showed that suspiciousness, political cynicism and lack of trust in the government can contribute to belief in CTs (Imhoff & Lamberty, 2018;Swami et al., 2016). All these results strengthen the assertion that tendency to believe in CTs can be associated with Machiavellianism characteristics. Machiavellianism could attenuate the negative relationship between social media skepticism and belief in COVID-19 CTs. Skepticism is "a constructive response to political blunders and public affairs news media, representing a critical but open posture toward news media and politicians" (Pinkleton et al., 2012, p.26). It can foster a heightened attention to media and media content by scrutinizing facts and motives with an open mind and free of any resentment (Tsfati, 2003). Skeptics may not manifest Machiavellian attributes such as cynicism and manipulation. Past studies found that skeptics are oriented to know the truth by critical thinking and enhanced information seeking (Vraga & Tully, 2019;Yamamoto & Kushin, 2014) while Machiavellians tend to build cynical distrust (Dahling et al., 2009) that reflects their resentment and hostility towards media (Pinkleton et al., 2012;Tsfati, 2003). In fact, healthy skepticism can turn to cynicism if individuals feel that "critically consuming news and information has no value to them or is a waste of their time" (Vraga & Tully, 2019, p. 5). Narcissism is typically characterized by dominance, increased sense of entitlement and superiority (Paulhus & Williams, 2002). Narcissists are engrossed with how others perceive them (Horvath & Morf, 2009). Narcissism was found to be positively related to belief in CTs (Cichocka, Marchlewska, & de Zavala, 2016a;Cichocka, Marchlewska, de Zavala, & Olechowski, 2016b;March & Springer, 2019). The propensity of them believing in CTs can be due to their desire to retain a favorable impression of others. The relationship between narcissism and belief in CTs was attributed to the tendency of the individuals with high narcissistic behavior to be the centre of attention (March & Springer, 2019), and hence, any sharing of CTs on social media allows them to receive the extent of attention they desired for. Narcissism promotes paranoia, a pattern of thinking that triggers an irrational suspicion or mistrust of others and perceive their actions as intentionally malevolent (Cicero & Kerns, 2011;Cichocka, Marchlewska, & de Zavala, 2016a). Research reported that paranoia is positively linked with belief in CTs about political and social events caused by high authoritarians (Darwin et al., 2011;Wilson & Rose, 2014). The negative relationship between skepticism and belief in COVID-19 CTs on social media could be contingent on users' narcissism. There exists evidence showing that narcissists with excessive self-focused attention has less propensity to engage in reflective thinking processes (Littrell et al., 2019). Previous work also demonstrated a significant association between narcissism and overconfidence (Macenczak et al., 2016) while overconfidence has a negative relationship with cognitive reflection (Noori, 2016). Narcissists tend to act without undertaking deliberate information processing (Lee & Seidle, 2012). In this view, narcissism characteristics may not be agreeable with skepticism which entails critical thinking and careful information processing. Given this, it could be assumed that social media users with high scores on narcissism characteristics may not judge social media content critically and hence they are susceptible to agreeing with CTs. Psychopathy is typically perceived as thrill seeking, lack of empathy and low in interpersonal emotions (Paulhus & Williams, 2002). It is characterised by social dominance, self-confidence, and callous manipulation (Hare & Neumann, 2008). Psychopaths were found to have the tendency to believe CTs (Hughes & Machan, 2021;March & Springer, 2019). Previous research showed that strange beliefs/ magical thinking has a significant positive association with ideation of CTs (Barron et al., 2018) because higher odd beliefs/magical thinking reinforces a sense of control that belief in CTs could provide (Swami et al., 2011). Psychopathy may affect the relationship between skepticism and belief in CTs. Individuals with psychopathy often show deficits in learning (Oba et al., 2019) and demonstrate a reduced capacity to process multicomponent perceptual information simultaneously (Hamilton & Newman, 2018). In addition, psychopaths have difficulties in switching their attention and often fail to integrate additional information into their thinking (Burley et al., 2020). According to Jones and Paulhus (2011), psychopathy was positively associated with impulsivity. Impulsivity was a potential obstacle to the development of adaptive social problem solving skills and also the learning of effective social problem solving strategies (McMurran et al., 2002;McMurran et al., 2005). All these psychopathic characteristics may impair social media users' skepticism. --- The Present Study Given the spread of COVID-19 CTs on social media, it is timely to know what factors have an association with tendency to believe in such theories. The existing literature reviewed above found that skepticism on media information negatively predicts belief in CTs whereas the dark triad personality traits are positively associated with the acceptance of CTs. Therefore, the present study develops the following hypotheses to test the relationship between social media information skepticism and belief in COVID-19 CTs as well as to provide a more in-depth understanding on the moderating influence of the personality traits on this relationship. H1: There is a negative relationship between social media information skepticism and COVID-19 conspiracy theory beliefs. H2: Machiavellianism moderates the negative relationship between social media information skepticism and COVID-19 conspiracy theory beliefs such that the relationship is weakened for individuals who score high on Machiavellianism. H3: Narcissism moderates the negative relationship between social media information skepticism and COVID-19 conspiracy theory beliefs such that the relationship is weakened for individuals who score high on narcissism. H4: Psychopathy moderates the negative relationship between social media information skepticism and COVID-19 conspiracy theory beliefs such that the relationship is weakened for individuals who score high on psychopathy. --- Method Procedure A cross-sectional design was established to collect data from young adult social media users from Kuala Lumpur, Malaysia, using an online survey. The rationale of choosing this segment as research target population was supported by the fact that about 81% of the Malaysian population (as of January 2020) were identified as active social media users (Müller, 2020) of which the majority of them were in their 20s and 30s and mainly residing in urban areas (Malaysian Communications and Multimedia Commission (MCMC), 2018). Results also showed that Malaysian social media users aged between 18 to 24 years old were the biggest age group that trust social media (Hirschmann, 2021a). An online survey was created using Google Form. The questionnaire started with a brief introduction to the research. Participants were assured that their responses would be kept strictly confidential and used only for research purpose. Their participation is completely voluntary and as such they were free to withdraw from the survey at any time. In the first section of the questionnaire, a filter question was designed to ensure that the survey is completed by participants who was a user of at least one of the social media sites such as Facebook, Instagram, Twitter etc. The other sections covered the sociodemographic characteristics of participants and items measuring the predictor and outcome constructs. Questions on the constructs were asked in no particular order to reduce common method bias (MacKenzie & Podsakoff, 2012). Participants were required to complete all items in the survey to ensure no missing data. We used G*power to calculate the required sample size for the study. We performed the most common recommended setting for social science research to determine the sample size for a moderation model (Hair et al., 2017). Setting G*power at "F tests analysis", statistical test: "Linear multiple regression: fixed model, R2 deviation from zero", the type of power analysis: "A-priori: Compute required sample sizegiven <unk>, power and effect size", effect size f 2 = 0.15 (medium effect), <unk> err prob. = 0.05, and power (1-<unk> err prob) = 0.80, number of predictors = 7, the total sample size obtained was 103 with the actual power of 0.80. A research assistant was recruited to distribute the link of the survey among potential respondents. Using convenience sampling method, a total of 439 responses were collected which met the required sample determined by G*power. Data collection took less than one month from 15 July to 10 August 2020. The protocol of the study (including the research procedure, the rights and safety of the participants, and the method of data collection) was approved by the Review Board of XXX University Malaysia to comply with research ethics policy. --- Participants Out of the 439 participants, 117 were male (26.7%) and 322 were female (73.3%), respectively. Majority of them ranging from the age of 18 to 22 (81.3%), were single (97%) and Chinese (90.6%) and pursued tertiary education (62.9%). Less than half of the participants (44.6%) reported spending 4-6 h a day on social media sites, followed by 36.2% who spent 1-3 h on social media. Social media was the second most reliable source of news for 25.3% of the participants after online newspapers with 41.5%. Table 1 shows the demographic profile of respondents. --- Measures Social Media Skepticism A 7-item scale was adopted from Maksl et al.'s (2015) study where respondents' perception on whether news media are "fair, tell the whole story, are accurate, can be trusted, get in the way of society, solving its problems, and whether news media prioritize being the first to report a story" were sought (Maksl et al., 2015, p. 34). Participants were asked to think of information on social media when reading and choosing the option which best indicates their agreement or disagreement with each statement. Some items include "I think the news media are fair," "I think the news media are accurate," "I don't think the news media can be trusted." Responses were rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). Some items were reverse coded so that a higher score indicated a higher level of skepticism towards the information on social media. Cronbach's alpha of the scale reported in previous research was acceptable (Maksl et al., 2015;Vraga & Tully, 2015). The present study found an internal consistency of.91 for the scale. --- Belief in Conspiracy Theories about COVID -19 A 5-item COVID-19 conspiracy scale was developed for the present study. The scale included the most popular CTs in the world about COVID-19 as of July 2020 documented in the literature (Cassese et al., 2020;Georgiou et al., 2020;Miller, 2020aMiller,, 2020b)). The five items include: "Coronavirus has escaped from a lab in one of the cities of China, Wuhan," "Coronavirus is originated in the US," "Many dead bodies affected by Coronavirus were secretly burned in China," "The Covid-19 is a man-made virus," and "Coronavirus was a part of a Chinese covert biological weapon programme." The questionnaire was validated by a pilot study. Participants rated each statement on a 5-point scale ranging from 1 = strongly disagree to 5 = strongly degree. The Cronbach's alpha for the scale was.75. --- The Dark Triad The dark triad comprises Machiavellianism, narcissism, and psychopathy. The present study used the Short Dark Triad (SD3) scale developed by Jones and Paulhus (2014). The scale comprises of 27 items, with 9 items for each trait. Some items for the section of Machiavellianism include "It's not wise to tell my secrets," "I like to use clever manipulation to get my way," "Whatever it takes, I must get the important people on my side." Some items for narcissism include "People see me as a natural leader," "I hate being the center of attention," and "Many group activities tend to be dull without me." Some items of psychopathy are "I like to get revenge on authorities," "I avoid dangerous situations," and "Payback needs to be quick and nasty." All responses were rated on a 5-point Likert scale ranging from 1 = strongly disagree to 5 = strongly degree. Previous research showed an acceptable reliability for each sub-scale (Nowak et al., 2020). In the present study, internal consistency for Machiavellianism was <unk> =.75, for narcissism was <unk> =.61 and for psychopathy was <unk> =.72. Although the alpha value of narcissism is less than 0.7, there is a general rule which recognises internal consistency value between 0.6-0.7 as acceptable (Ursachi et al., 2015). --- Analysis The data was analysed using IBM SPSS Statistic version 25. Pearson Correlation analysis was computed to determine the relationship between all study variables. SPSS PROCESS macro by Hayes (2018) was used to test the moderating effect of Machiavellianism, narcissism and psychopathy on the negative relationship between skepticism towards information on social media and belief in CTs about COVID-19 while controlling for gender, age and education level. SPSS PROCESS macro uses ordinary least squares (OLS) regression to estimate the interaction in the moderation models. The predictors (skepticism, Machiavellianism, narcissism and psychopathy) were mean-centered when conducting the moderation analysis. The moderation analysis was conducted separately for each dark triad traits. --- Results Before testing the hypotheses in the present study, common method bias was tested using the most widely used technique -Herman's single factor test (Podsakoff et al., 2003). From the factor analysis, the total variance explained by one factor is 16.59% which indicated the absence of common method bias. Pearson Correlation analysis was used to determine the negative relationship between skepticism towards information on social media and belief in CTs about. As shown in Table 2, it was found that skepticism towards information on social media was significantly and negatively correlated with belief in CTs about COVID-19 (r = -.10, p =.029), supporting H1. Moreover, belief in CTs about COVID-19 was found to be significantly and positively correlated with Machiavellianism (r =.11, p =.019), narcissism (r =.11, p =.024) and psychopathy (r =.30, p <unk>.001). The moderating effect of Machiavellianism on the negative relationship between skepticism towards information on social media and belief in CTs about COVID-19 while controlling for gender, age and education level (H2) was examined using PROCESS macro. Skepticism did not significantly predict beliefs in CTs about COVID-19 (B = -.04, SE =.04, t = -0.95, p =.345) while Machiavellianism significantly predicted belief in CTs about COVID-19 (B =.09, SE =.04, t = 2.54, p =.012). The overall model was significant, R 2 =.047, F(6, 432) = 3.57, p =.002, which explained 4.7% of the variance in belief in CTs about COVID-19. The interaction effect between skepticism towards information on social media and Machiavellianism was statistically significant (B = -.02, SE =.01, t = -3.12, p =.002) as displayed in Fig. 1. The interaction model was <unk>R 2 =.021, F(1, 432) = 9.72, p =.002, which explained 2.1% of the variance in belief in CTs about COVID-19. Based on the conditional effects, significant relationship was found between skepticism towards information on social media and belief in CTs about COVID-19 for high level of Machiavellianism, B = -.13, t(432) = -3.14, p =.002, 95% CI [-.22, -.05]. Meanwhile, no significant relationship was found between skepticism and belief in CTs about COVID-19 for low and moderate level of Machiavellianism (see Table 3). Thus, when Machiavellianism is high, the negative relationship between skepticism towards information on social media and belief in CTs about COVID-19 is diminished, thus supporting H2. When investigating the moderation effect of narcissism (H3), it was found that skepticism towards information on social media did not significantly predict belief in CTs about COVID-19 (B = -.06, SE =.04, t = -1.49, p =.138) while narcissism significantly predicted belief in CTs about COVID-19 (B =.12, SE =.04, t = 2.77, p =.006). The overall model was significant, R 2 =.049, F(6, 432) = 3.71, p =.001, which explained 4.9% of the variance in belief in CTs about COVID-19. The interaction effect between skepticism towards information on social media and narcissism is statistically significant (B = -.02, SE =.01, t = -3.24, p =.001) as shown in Fig. 2. The interaction model was <unk>R 2 =.023, F(1, 432) = 10.52, p =.001, which explained 2.3% of the variance in belief in CTs about COVID-19. Based on the conditional effects, significant relationship was found between skepticism towards information on social media on belief in CTs about COVID-19 for moderate, B = -.08, t(432) = -2.01, p =.046, 95% CI [-.15, -.002] and high level of narcissism, B = -.15, t(432) = -3.37, p <unk>.001, 95% CI [-.23, -.06], whereas there is no significant relationship between skepticism and beliefs in CTs about COVID-19 for low level of narcissism (see Table 4). As a result, the negative relationship between skepticism towards information on social media and belief in COVID-19 conspiracy theories is weaker for individuals with moderate and high levels of narcissism, thus supporting the moderating effect of narcissism (H3). As shown in Fig. 3, psychopathy did not significantly moderate the relationship between skepticism towards information on social media and belief in CTs about COVID-19 (H4). Skepticism towards information on social media did not significantly predict belief in CTs about COVID-19 (B = -.04, SE =.04, t = -1.12, p =.265) while psychopathy significantly predicted belief in CTs about COVID-19 (B =.24, SE =.04, t = 6.32, p <unk>.001). The overall model was significant, R 2 =.10, F(6, 432) = 8.41, p <unk>.001, which explained 10% of the variance in belief in CTs about COVID-19. The interaction effect between skepticism towards information on social media and psychopathy is not significant (B = -.01, SE =.01, t = -1.21, p =.227) with <unk>R 2 =.003, F(1, 432) = 1.47, p =.227, which explained 0.3% of the variance in belief in CTs about COVID-19. Thus, H4 is not supported. --- Discussion The present study aimed to examine the association between social media information skepticism and belief in CTs about COVID-19 and the moderating role of the dark triad personality traits on this association among Malaysian young adult social media users. As postulated in H1, skepticism towards information on social media was negatively associated with the ideation to CTs about COVID-19, suggesting that the attitude of doubt towards content on social media aids to buffer susceptibility to COVID-19 CT effects. The result substantiates the assertion that skepticism negatively predicts the acceptance of misinformation and disinformation because it motivates people "to question the origins of information that may later turn out to be false" (Lewandowsky et al., 2012, p. 120). Skepticism involves analytical thinking (Feuerstein, 1999) and analytic thinking reduces belief in CTs (Swami et al., 2014). The present study further corroborates the result that individuals who scored high on media skepticism may assess the news program as less credible than those who scored low on media skepticism (Carr et al., 2014). The result also provides support for the negative association between news media literacy and the likelihood of endorsing CTs (Craft et al., 2017). Skeptical news media consumers are less likely to be exposed to nonmainstream media due to their belief in the lack of reliability of online news sources (Tsfati, 2010). Also, they are more likely to engage in situations and activities that require a deep elaboration (Tsfati & Cappella, 2005). Their skepticism towards information on social media could be associated with their media and news literacy (Vraga & Tully, 2019). Skepticism is consequential in the context of social media because it is receptive to misinformation (Craft et al., 2017;Kahne & Bowyer, 2017). Those with high score on media skepticism tend to repudiate the overwhelming reports in media (Tsfati, 2003). If given a choice to adjudicate the social media content while maintaining the interest for information accuracy, they should be less agreeable to the content on such outlets (Tsfati, 2003). This tendency will be advantageous for society as it then reduces the negative impact of false or misleading information (Lewandowsky et al., 2012). Therefore, cultivating skepticism towards media messages, including online information, should remain as the fundamental of media and news literacy programs (Ashley et al., 2010;Maksl et al., 2015) because it promotes accurate judgment of truth claims (Kahne & Bowyer, 2017) and may mitigate the tendency to endorse CTs. Machiavellianism was found to be the moderator on the relationship between skepticism and belief in CTs about COVID-19 (H

The prevalence of conspiracy theories about COVID-19 in the wide-open social media environment has raised considerable concerns about the adverse effects of believing such theories. The previous work showed that skepticism is negatively associated with acceptance of media content. However, this association has yet to be investigated between skepticism towards content on social media and conviction of COVID-19 conspiracy theories. Furthermore, this association can be a function of social media users' dark triad traits comprising Machiavellianism, narcissism and psychopathy. A convenient sample of 439 Malaysian young adult social media users was recruited to complete the survey. Results showed that skepticism has a negative relationship with belief in conspiracy theories about COVID-19. However, the strength of the relationship is varied depending on respondents' Machiavellianism and narcissism levels. The negative association between skepticism and belief in COVID-19 conspiracy theories is weaker for individuals with high scores on Machiavellianism. High and moderate scores on narcissism were also found to diminish the negative relationship of skepticism with COVID-19 conspiracy ideation. The results yield implications for enhancing healthy skepticism to mitigate belief in conspiracy theories promulgated on social media. Nonetheless, Machiavellianism and narcissism appeared to weaken the power of skepticism on conspiracy theories.

social media because it is receptive to misinformation (Craft et al., 2017;Kahne & Bowyer, 2017). Those with high score on media skepticism tend to repudiate the overwhelming reports in media (Tsfati, 2003). If given a choice to adjudicate the social media content while maintaining the interest for information accuracy, they should be less agreeable to the content on such outlets (Tsfati, 2003). This tendency will be advantageous for society as it then reduces the negative impact of false or misleading information (Lewandowsky et al., 2012). Therefore, cultivating skepticism towards media messages, including online information, should remain as the fundamental of media and news literacy programs (Ashley et al., 2010;Maksl et al., 2015) because it promotes accurate judgment of truth claims (Kahne & Bowyer, 2017) and may mitigate the tendency to endorse CTs. Machiavellianism was found to be the moderator on the relationship between skepticism and belief in CTs about COVID-19 (H2), suggesting that high scores on Machiavellianism weaken the association between skepticism towards social media content and conspiring propositions. Participants characterized by Machiavellianism are more likely to have a stronger conviction about COVID-19 CTs and are less likely to evaluate news from a healthy and critical perspective. Machiavellians are cynical and distrustful in nature as well as strategic in manipulation (Furnham et al., 2013;Jones & Paulhus, 2011). Therefore, they may not engage in information evaluation process with a healthy skepticism. As such, Machiavellianism may diminish the negative relationship between skepticism and belief in CTs. The finding of the present study lends credibility to past research that showed Machiavellianism positively predicts the generic conspiracist beliefs (March & Springer, 2019), personal willingness to conspire (Douglas & Sutton, 2011), endorsement of conspiracy theories about the COVID-19 outbreak (Malesza, 2021) and COVID-19 vaccine intentions (Hughes & Machan, 2021). Findings also supported the moderating role of narcissism on the relationship between skepticism and belief in. This reflects that high and moderate levels of narcissism can diminish the strength of the relationship between skepticism and the ideation of COVID-19 CTs because they are less likely to engage in a thoughtful message processing and reflective thinking (Lee & Seidle, 2012;Littrell et al., 2019). The results of the present study substantiate past research that showed narcissism is a robust predictor of conspiracy ideation (Cichocka, Marchlewska, & de Zavala, 2016a). This can be attributed to the narcissists' paranoia behavior which was found to be associated with conspiracy beliefs (Wilson & Rose, 2014). Furthermore, narcissists are attention-seeking (Raskin & Terry, 1988) and their desire to be the center of attention can also be linked with their tendency to have conspiracy mentalities (March & Springer, 2019). The present study failed to support the hypothesised moderating role of psychopathy on the relationship between skepticism and belief in COVID-19 CTs (H4). This result suggests that psychopathy characteristics are less likely to impede skepticism in adjudicating information critically. One possible reason to explain the absence of the moderating influence of psychopathy lies in the scale items that mainly capture impulsivity feature of psychopathy (Jones & Paulhus, 2014). Past research also demonstrated that psychopathy, conceptualised as impulsivity, emotional reactivity, destructive and volatile nature, is not associated with the ideation of CTs (March & Springer, 2019). However, other characteristics of psychopathy which cover social dominance, exploitation, and manipulation were found to be the predictors of belief in CTs (March & Springer, 2019). --- Implications The present study yields some theoretical implications. Firstly, it investigated skepticism as one of the core components of media literacy along with the three dark personality traits, thus providing a better understanding on the complexity of individuals' behavior towards the acceptance of information about COVID-19 on social media. Secondly, the significant negative association between social media skepticism and belief in COVID-19 advanced the literature of media literacy and CTs. Prior studies did not examine the impact of skepticism on social media pertaining to belief in CTs associated to the unprecedented crisis like the pandemic, thus the present study provides a ground for further postulates in the future. Thirdly, the moderating role of Machiavellianism and narcissism on the relationship between social media skepticism and the ideation of COVID-19 CTs provides theoretical evidence supporting the dominant role of individuals' dark personality characteristics. The present study also produces several practical implications. Results showed that skepticism predicted COVID-19 conspiracy beliefs, implying that skeptical social media users are less susceptible to conspiratorial content and more likely to show more responsible online behaviors. Emphasis on skepticism towards social media content may possibly help cultivate the good practice of fact-checking and critical thinking. Skeptics are more likely to refute the influence of the COVID-19 CTs. In light of this, media literacy advocates should emphasize skepticism as an essential disposition to empower users' ability to distinguish conspiratorial content and resist falsehood as a result of inaccurate reports about the pandemic that effectuate uncertainty and stress among public. The descriptive results of the present study showed that the participants of the present study considered social media news more reliable than printed newspaper or television, supporting the Statista survey that showed majority of the Malaysian respondents (86%) get their news online, including through social media, and 30% of them get their news through print media (Hirschmann, 2021b). Nonetheless, Malaysian young adults appeared to have low awareness of potential media effects and critical thinking abilities of information encountered on digital media (Chin & Zanuddin, 2019). They were also found to report some difficulties in evaluating the relevance and usefulness of information on digital media (Pandian et al., 2020). All these results call for an attention of media literacy education to empower individuals to become more critical social media users with a focus on healthy skepticism. In addition, fostering skepticism toward information on social media while avoiding Machiavellianism characteristics (such as cynicism, manipulation and exploitation) should be the driven forces in media literacy programmes. Healthy skepticism on social media content should be inculcated while the elements of accuracy, trustworthiness, fairness, and completeness of information are valued. Having a misanthropic view of the world may intensify tendency to believe CTs. While skepticism is negatively linked with CTs belief, Machiavellianism can be detrimental as reflective thinking is absent, instead, cynical hostility and rejection of media driven by antagonism are dominant. Results of the present study demonstrated that users high on Machiavellianism are more likely to accept COVID-19 CTs because they may not be armed with skepticism. Given this, indirect intervention programs through creation of awareness (such as # Say "Yes" to Responsible Sharing of News campaign) are recommended to prevent social media users from communicating inaccurate information to promote anti-malevolent behaviors on social media. The result showed that high and moderate levels of narcissism can weaken the negative relationship of skepticism on COVID-19 CTs, implying that social media users may be defenceless to conspiratorial content if their skepticism is diminished by narcissism. The negative impact of rampant narcissism impairs users' concern about the quality of information on social media and thus intensifies their propensity to endorse CTs. Narcissism is manifested as a stable trait (Vater et al., 2014) and can be harnessed for good intentions. Research showed that some characteristics of narcissism such as inflated self-image and sense of importance appeared to be associated with higher self-esteem, greater resilience and less depression (Brailovskaia et al., 2019;Mehdizadeh, 2010). In view of this, those concerned of the inimical effects of CTs on social media can promote positive narcissism for selfenhancement and discourage users from engaging in deconstructive dimensions of narcissism that lead to vindictive motivations such as creating and promulgating CTs on social media. --- Limitations and Future Research Although the findings from the present study have made several contributions to the literature, it was limited by its reliance on a correlational design. Therefore, the present study failed to provide results for causality of COVID-19 CT belief (Creswell & Creswell, 2014). The sample of the current study was recruited conveniently and was predominantly constituted by female young adult social media users who were mainly Chinese, thus imperilling the generalizability of the results (Babbie, 2017). A replication of the study can be conducted to improve on the inclusiveness of the sample which shows an adequate and accurate representation of the population. The present study used self-report measurements which are subject to participants' social desirability response bias, thus compromising the validation of scoring in the research measures (Arnold & Feldman, 1981). In addition to skepticism as one of the core components of media literacy, media locus of control and automatic versus mindful thought processing have the potential to protect media audience from being susceptible to the effects of conspiratorial information on social media (Maksl et al., 2015) which can be the focus of future research. While the present study focused on dark personality traits that may affect media consumption, future studies are recommended to investigate the moderating role of big five personality factors on the relationship between social media skepticism and conspiracy beliefs. In addition, the scale used to measure COVID-19 CTs was not inclusive due to the dynamic nature of such theories during the pandemic. Therefore, future research is suggested to expand the scale comprising the recent CTs. --- Data Availability The raw data of the present study is available at https:// osf.io/k4vab/ --- Declarations Ethical Approval The protocol of the study (including the research procedure, the rights and safety of the participants, and the method of data collection) was approved by the Review Board of Xiamen University Malaysia to ensure the principles of research ethics. Informed Consent Informed consent was obtained from all individual participants of the present study. --- Conflict of Interest The authors declare that they have no conflict of interest. CRediT Authorship Contribution Statement Ashraf Sadat Ahadzadeh: Conceptualization, Investigation, Methodology, Writing-original draft, Funding acquisition. Shin Ling Wu: Formal analysis. Fon Sim Ong: Writing -review & editing.

The prevalence of conspiracy theories about COVID-19 in the wide-open social media environment has raised considerable concerns about the adverse effects of believing such theories. The previous work showed that skepticism is negatively associated with acceptance of media content. However, this association has yet to be investigated between skepticism towards content on social media and conviction of COVID-19 conspiracy theories. Furthermore, this association can be a function of social media users' dark triad traits comprising Machiavellianism, narcissism and psychopathy. A convenient sample of 439 Malaysian young adult social media users was recruited to complete the survey. Results showed that skepticism has a negative relationship with belief in conspiracy theories about COVID-19. However, the strength of the relationship is varied depending on respondents' Machiavellianism and narcissism levels. The negative association between skepticism and belief in COVID-19 conspiracy theories is weaker for individuals with high scores on Machiavellianism. High and moderate scores on narcissism were also found to diminish the negative relationship of skepticism with COVID-19 conspiracy ideation. The results yield implications for enhancing healthy skepticism to mitigate belief in conspiracy theories promulgated on social media. Nonetheless, Machiavellianism and narcissism appeared to weaken the power of skepticism on conspiracy theories.

Introduction Sexually transmitted infections (STI) have been on the rise in the United States, growing to epidemic proportions among adolescents and young adults (AYA) aged 15 to 24 years [1]. As a group, AYA account for the highest rates of STI, contributing more than 50% to the $16 billion in lifetime cost of the infection [2,3]. In addition to the heightened risk of STIs, AYA residing in low-income, densely populated urban communities experience racial and ethnic disparities and often have limited access to STI screening, treatment, and other sexual and reproductive health needs [4,5]. AYA have also been characterized as a vulnerable population due to cognitive factors such as inaccurate risk perceptions, increased biologic susceptibility (eg, cervical ectopy), engagement in unprotected intercourse, and sexual partnership patterns such as serial monogamy and concurrency [6][7][8][9]. National surveillance of the high school youth in the United States show that more than half of high school students are sexually active [9], and without access to services, they may suffer STI-associated sequelae. Despite the emerging public health threats and incidence of STIs in this age group, many AYA are not tested for STIs [10]. While the Centers for Disease Control and Prevention (CDC) guidance on the approach to STI prevention in youth is clear about the need for developmentally appropriate approaches to care, youth continue to face barriers to accessing sexual and reproductive health clinical services and education. STI prevention efforts have primarily relied on individual approaches and heavily target girls and young women in heterosexual relationships, leaving young men with significant unmet sexual and reproductive health needs [11][12][13]. Young women are also diagnosed and treated for STIs more often compared to young men, who are often asymptomatic. As such, partner notification and treatment is a key strategy for disease control. Individuals with bacterial STIs often notify and refer their partners for treatment, but the relationship context influences the timing, manner of communication, and outcome of notification. Even so, less than half of male partners in a heterosexual relationship notified for STI treatment are successfully treated [12]. In response to this public health failure, many states are establishing regulations to implement expedited partner therapy [14][15][16]. In addition, the US Preventive Task Force recently recommended the integration of behavioral counseling as part of STI management, especially for adolescents at higher risk for new or recurrent disease [17]. This clinical change is estimated to have a moderate benefit and will reduce the probability of acquiring new STIs. The context for STI treatment services has also shifted from public health departments to primary care settings. Unfortunately, national data suggest that less than half of eligible patients are screened for STIs and many primary care providers serving youth fail to provide needed services [10,11,18]. Given the personal and societal costs associated with STIs, new models of care that integrate effective public health interventions into primary care settings are warranted. Public health departments have effectively used individual health educator models for STI/HIV risk reduction counseling but with limited effectiveness due to limited partner engagement. Efforts for engaging sexual partners of AYA using an integrated approach that combines behavioral interventions and health education and other methods are needed as individual level approaches fail to harness the power of joint intervention to improve sexual decision making and behavior. Sexual relationships among AYA have long been believed to be delicate, short-lived, and lacking in commitment and mutual trust as compared to adult sexual partnerships. Evidence to the contrary proves that although youth, particularly males, tend to engage in extra-dyadic sex (sex with someone other than their partners) [6] and have higher tendencies toward concurrency and serial monogamy [19], STI in AYA is more nuanced and complex, often occurring in the context of committed relationships with or without concurrency. AYA relationships possess levels of trust and engagement, qualities that are relevant for and amenable to modifications for improving communication and promoting emotional and sexual satisfaction and negotiation skills. Such interventions can culminate in better self-management and a reduced risk of acquiring or transmitting STIs. Among adults, engagement of couples (or sexual dyads) for STI/HIV prevention have proven effective [20,21]. Research with AYA suggests that understanding early relationships may help prevent STIs but that we have failed to use our knowledge about the structure and function of romantic relationships occurring during this critical developmental learning period to engage youth for individual and dyadic behavior change. Formation of romantic relationships is a key task of adolescence with significant implications for young adult adjustment and health outcomes [22][23][24]. Male partners can be engaged through partnerships with female sexual partners [13]. Adolescent sexual dyads are formative and sufficiently stable for dyadic intervention, with relationships lasting 8 months on average [22]. Recent research with adolescent dyads also demonstrates that male and female partners report shared decision making and strong feelings toward their partners [24,25]. A dyadic approach to STI prevention allows both parties to see mutual responsibility for protecting each other from STIs, facilitates them working together, highlights the context and connection to disease acquisition, and provides a safe space for communication about difficult issues such as concurrency, while giving room for the dyad to learn key communication skills and receive support and guidance from a health professional. Although several studies demonstrate the effectiveness of adult-focused couple interventions for STI/HIV prevention, no published studies address the AYA dyad as the unit of transformational change or have contextualized the interpersonal dynamics influencing individual sexual health behavior change within young romantic partnerships. To address this issue, the Sexperience study was developed using the tenets of transformational motivation theory to bolster the integrative model of behavior change as a conceptual framework [26,27]. The aim of this study is to examine the feasibility, acceptability, and effectiveness of a partner intervention in a community facing significant STI/HIV health disparities. --- Methods --- Study Design The Sexperience study is an exploratory randomized controlled trial designed to examine the feasibility, acceptability, and differences in short-term behavioral outcomes between youth who receive both individualized counseling and joint debrief with sexual partner and those who receive sexual risk reduction counseling alone. The study question operationalizes the tenets of transformational motivation theory [27] as a part of the conceptual framework (Figure 1) based on the integrative model of behavior which asserts that being in a relationship can change the member's behavior from one of self-centeredness to one that is driven to improve the relationship and is health enhancing. The integrative model of behavior serves as a backdrop for integrating the transformational motivation theory for dyadic research with AYA as a strategy to reach both female and male patients for treatment and preventive services [26]. The experience gained and lessons learned from previous research are invaluable tools for achieving the desired goals of this study [28]. A total of 50 AYA dyads aged 16 to 25 years, engaged in heterosexual intercourse, who reside in the greater Baltimore (Maryland) metropolitan area (BMA) and are willing to recruit their main sexual partner for the study will be recruited and randomized. Participants are being recruited from an academic adolescent/young adult medicine practice and the BMA using social media (Facebook and Instagram). The index participant and their recruited partner complete an individual behavioral intervention session separately (Sister to Sister or Focus on the Future) with a gender-matched health educator. Dyads are then randomized to completion or to receive an additional dyadic debriefing session. Participants separately complete an exit survey at the end of the intervention and a telephone interview 6 weeks postintervention to determine the feasibility, acceptability and perceived impact of the intervention on mutual sexual and condom negotiation, consistency of condom use, and communal coping skills (Figure 2). --- Inclusion Criteria Index participants are eligible if they are aged 16 to 25 years, live in the BMA, agree to recruit their main partner to the study, engage in male-female (heterosexual) intercourse, are willing to participate in a single individualized session with a health educator followed by a joint intervention, are willing to be contacted in 6 weeks for a telephone interview, have no reports of intimate partner violence with their current sexual partner, and have a current sexual partner within 5 years of their age. The partner must be aged 16 to 30 years based on adhering to prespecified age differentials to coincide with state law and developmental expectations, engage in male-female (heterosexual) intercourse, permanently reside in the BMA, and be willing to participate in a single individual session with health educator with or without a dyadic debriefing session with their partner and both health educators depending on randomization. --- Exclusion Criteria Index participants and partners who are unable to communicate with staff or participate in study procedures due to cognitive, mental, or language difficulties will not be eligible for recruitment into the study. Dyads will also be excluded if in same-sex main partnership or a member of the dyad is currently enrolled in another sexual behavior study, the patient/partner is currently pregnant, one or both partners has a known concurrent HIV infection, one or more partners has a pending incarceration, if there is more than 5 years age difference between the two partners, or there is evidence of intimate partner violence in the relationship. --- Setting Baltimore, Maryland, is the primary site of this study. Baltimore is a large city whose citizens face significant STI and HIV disparities. Baltimore was ranked the city with the highest STI rates by the CDC in 2020 [29]. The Maryland Department of Health and Mental Hygiene has determined that every county is a chlamydia hot spot, and county maps demonstrate that among hot spots, the city of Baltimore is the most densely affected. Although there are extremely low rates of youth noninsurance, the use of primary care providers for routine health maintenance drops off considerably. The Adolescent and Young Adult Clinic at Johns Hopkins established a Title X reproductive health clinic to serve the youth of Baltimore to aid in resolving the reproductive health access issues facing urban youth. The clinic sees over 200 unduplicated positive STI cases and partner referrals annually. Patients may also be referred for treatment through the Johns Hopkins Wellness Center, the Johns Hopkins Pediatrics Emergency Department, or the Baltimore City Health Department clinical sites. --- Recruitment, Informed Consent, and Randomization Recruiters will be available via phone during office and some evening hours in the clinic. Patients will be referred for study contact by their providers. Patients who agree to be contacted will be approached by a recruiter in person or via phone and if eligible, the dyad will be scheduled for written informed consent and intervention procedures. After enrollment, participants will be assigned to the control group or intervention group using a computer-generated randomization assignment. Envelopes based on dyadic study group number will be available to the research staff at the time of enrollment to adhere to the randomization sequence. --- Sexperience Intervention Following contact of prospective participants via social media or through the AYA practice by a member of the research team or the research coordinator, willing participants who agree to invite their sexual partners will be enrolled into the study when they bring their partners along to the clinic on a specified date (usually within 2 weeks of agreeing to be in the study). Researchers will provide information to participants regarding the study and its processes individually and then seek informed consent. All participants who sign the informed consent with their partners will be randomized to either the intervention arm or control arm. Integral to the Sexperience intervention is the use of effective behavioral interventions designed to reduce STI/HIV risk behaviors, Focus on the Future and Sister-to-Sister, which will be used for male and female participants, respectively. The Focus on the Future intervention [30] is an individual-level single session intervention designed to reduce the spread of heterosexually transmitted STI and HIV in African American males. This is achieved through education on correct and consistent condom use and motivation for sexual behavioral change. Sister to Sister [31] is a similar intervention designed for heterosexual females and aims to reduce the risk of HIV and other STI acquisition in AYA women. Sister to Sister, like Focus on the Future, is a 1-time brief intervention that encompasses training for skills building in sexual negotiation, demonstrations on the right way to insert a condom, and role playing. Both of these interventions have been shown to be effective in reducing STI risk and improving healthy sexual behavior. The novel component of this intervention is that sexual partners will come together at the end of the individual sessions for a joint debriefing session that includes viewing of the condom negotiation videos and role-playing communication for condom use. At the end of the session with the Sexperience health educators, the patient will be able to state/demonstrate (1) understanding of the definition of STIs, (2) how to prevent future STI episodes, (3) proper use of male and female condoms, and (4) use of effective communication and condom negotiation strategies with their partner. Low-resource patients have been receptive to health educator support, and our preliminary studies indicate that improved communication between an adolescent and her partner occurs with the support of a professional who can educate and motivate adolescents with an STI and provide expertise on STI management. Video/audio recording of dyadic intervention sessions will be used to ensure intervention delivery is consistent with the protocol and to identify content themes to bolster our understanding of relationship dynamics and the potential of communal coping. Text message boosters related to care-seeking behavior, condom use, and condom access will be provided over the 6 weeks of the study. Messages will be standardized and delivered via an automated 1-way system. No protected health information is disclosed as a part of this communication, but youth will be encouraged to provide an extra layer of protection on existing cell phone security to allow for an additional degree of privacy related to message content. Welcome messages will be sent right after enrollment, "Welcome to the Sexperience study," followed by 1 text every week, including: "Don't forget to try out your new communication skills," "Did you practice the Sexperience role play? Text the # of times you practiced together," "Condoms prevent STDs. Stop by the clinic if you need some," "Condoms can improve pleasure, if you need some stop by the clinic," and "Want to up your game? Practice your condom techniques." In week 6, we will send "Thanks for trying Sexperience, we'll call you soon for feedback." SMS messages will be sent through TextNow for communicating with participants in both arms of the study. TextNow is one of the fastest growing technology companies with strict policies to protect the user by preventing unauthorized disclosure of information and ensuring physical, electronic, and procedural safeguards are in place according to industry standard procedures and security procedures. Finally, all participants will be encouraged to use standard safety mechanisms such as a pin or password to lock access to their cell phones by others. Participant dyads in the control group will also receive the 20-minute Focus on the Future for male and Sister-to-Sister for female participants and one-on-one sessions with a gender-matched health educator to guide the patient through skills-based risk reduction counseling. As noted above, these interventions have been evaluated (with African American males and females residing in communities with high STI rates) and were found to increase condom use, reduce unprotected intercourse, and reduce new STIs among participants and are considered effective interventions for clinical implementation by the CDC. Participants in the control arm will also receive positive health-related attention-control text messages focused on general health (healthy eating, exercise, spending time with family and friends, and healthy sleep). No text messages related to relationships or condom use will be sent. As with the intervention group, the message "Thanks for trying Sexperience, we'll call you soon for feedback" will be sent in 6 weeks. None of these texts will request a response. --- COVID-19 Procedure Adjustments Due to the pandemic, the remainder of participants will be recruited and enrolled virtually. Social media will be used for recruitment, and Zoom videotelephony software will be used to complete the enrollment visit. Recruiters will screen potential recruits over the phone and if eligible, set up a Zoom meeting for enrollment. Participants will be consented with the oral consent script. Recruiters will read the baseline Audio Computer-Assisted Self-Interviewing to the participants and record their answers. There will be separate Zoom meetings for the individual gender-matched intervention, and if the participants are randomized to the intervention group, they will join after for the joint debrief session. The exit survey through Qualtrics will be read to the participants afterward by the recruiters, and their answers will be recorded. Gift cards will be mailed to participants after the enrollment visit has been completed. The 6-to 8-week follow-up survey will remain the same and be completed over the phone with participants. --- Outcomes The primary outcome variable for this pilot study is feasibility of recruitment. Recruitment feasibility will be defined as the number of dyads completing the intervention out of the number of eligible dyads. The feasibility of short-term retention will be calculated as the proportion of enrolled participants who complete a follow-up interview at 6 weeks. The secondary outcome variable for this pilot study is acceptability, measured as participant satisfaction with their intervention experience. Participant satisfaction will be measured using a brief survey asking participants to rate their experience with qualitative feedback documented. --- Sample Size Determination and Statistical Analysis A total of 50 adolescent dyads will be enrolled. The choice of 50 dyads is inherently arbitrary and was the number considered realistically feasible to enroll within the limited time of this pilot study. While the study will lack statistical power to detect significant differences in outcomes between the two study arms, point estimates derived from this work will be invaluable for future research proposals to evaluate efficacy. Feasibility and acceptability will be evaluated against benchmarks from previous studies. Observed recruitment rates from the proposed pilot study will be compared to expected recruitment rates (determined from previous studies that recruited adolescents with an STI diagnosis). Prior studies recruiting from the AYA clinic have achieved recruitment rates ranging from 48% to 91%. A similar approach will be used to compare satisfaction scores to an a priori benchmark of patient satisfaction. Based on previous work, average patient satisfaction greater than 80% would be necessary to conclude that the intervention is acceptable. Descriptive statistics will be used to make a comparison of the baseline characteristics between the intervention and control group. Statistical analysis will be based on the standard for randomized controlled trials, and both primary and secondary outcomes for each group will be assessed using an intention-to-treat analysis. Data distribution will be assessed, and comparisons of the 2 groups will be conducted using generalized estimating equations. P<unk>.50 will be considered statistically significant. The multiple imputations method will be used to address missing data. --- Risks and Ethical Considerations There are no significant medical risks involved in this study. While the study will attempt to preserve patient confidentiality, there is potential for breaches of confidentiality, but it is miniscule given the limited number of personnel with access to participant information. Finally, while the Sexperience intervention focuses on positive communication and condom negotiation, dyads may experience conflict or disagreement during the context of the joint session. While the health educator will be trained in basic conflict resolution, there are staff onsite to provide both mental health (social work, mental health counselors, psychologists) and seasoned security staff available to intervene as needed. There are no additional risks associated with these telephone calls. Human subject approval has been granted for beta testing of this research protocol by the Johns Hopkins institutional review board (IRB00083609), and the study was registered at ClinicalTrials.gov [NCT03275168]. All patient data are kept confidential and secure. Individual patient data shared during health education sessions and clinical visits (sexual history data, lab results, clinical findings) will not be disclosed to the partner in accordance with Health Insurance Portability and Accountability Act regulations. However, the patient will be encouraged to notify partners regarding behavioral risks that impact their partner per standard of care and public health practice. The research team on this project has training and experience in the protection of human research participants and the treatment of protected health information. All staff have completed human subjects training through the institution's research compliance training program as a part of orientation training and as a condition of their employment. --- Data Safety and Monitoring The study will use a data safety monitoring plan consistent with National Institutes of Health guidelines. The research team will meet regularly to communicate about ethical issues related to the study. A data safety and monitoring committee was created; members have no formal association with the study and their selection is based on their expertise in AYA clinical interventions, sexual and reproductive health service delivery, and biostatistics. --- Payment, Remuneration, and Benefits Study incentives are US $50 for the enrollment visit and US $50 for the 6-week follow-up interview for each participant, both patient and partner. Those participants who elect to complete the in-depth interview will receive an additional US $50. There are no anticipated costs to participants that are unaccounted for in the support and remuneration plan for the study or usual care. Transportation support may be provided for participants with transportation issues. There may be no direct benefits to participants in this study. --- Results Study recruitment began in January 2018 and was initially estimated to be completed in February 2021. As of March 30, 2021, 485 persons were assessed for eligibility, of which 312 (64.3%) were ineligible. Of the 173 eligible participants, 87 (50.3%) declined to participate and 42 (24.3%) were not randomized due to certain factors (Figure 3). So far, 22 dyads (n=44) have been recruited. Of the index patients randomized, most were females (20/22, 91%), with at least a high school diploma (19/22, 86%), and an average of 9 lifetime sexual partners. Index patients and their partners did not differ in age, with mean ages of 21.7 (SD 2.6) years and 22.6 (SD 2.5) years, respectively, and age at first sexual encounter 15.4 (SD 2.5) years and 15.1 (SD 1.8) years, respectively. --- Discussion --- Summary Preliminary findings from the study show that a dyadic intervention for heterosexual AYA is both feasible and acceptable. To our knowledge, this study will be the first of its kind to engage AYA dyads in a joint STI risk counseling intervention. While many current research studies have used couples engagement for prevention of sexually transmitted diseases, most have focused primarily on HIV to the neglect of other STIs (Neisseria gonorrhoeae, Chlamydia trachomatis, Trichomonas vaginalis), which account for the preponderance of STIs [20,32,33]. Similar studies have been limited to adults or male-male partnerships to the exclusion of younger age groups and heterosexual dyads [9,12,13]. The results from this critical inquiry will further clarify the potential role for a joint health educator-administered sexual education and counseling intervention in enhancing AYA access to sexual and reproductive health services in primary care settings and will determine the feasibility, acceptability, and effectiveness of using the sexually involved dyad as a vehicle for behavioral change. It will also give us an insight into the role of actively engaging sexual partners in joint communications for promoting healthier sexual behaviors on STI prevention compared to the general practice of individualized risk reduction counseling, partner notification, and expedited therapy without mutual engagement [14,17,34]. Dyadic interventions have been used in many medical conditions including parent-adolescent dyads for improving HIV prevention and self-efficacy [35], obesity prevention, and smoking and sedentary lifestyle cessation and were influential for sparking behavior changes [36] and preventing HIV transmission between serodiscordant same-sex male couples. A dyadic intervention for heterosexual couples in Uganda was equally effective for HIV prevention [32]. Consistent condom use and condom negotiation skills reduce STI transmission from an infected partner. Given the high impact of STIs among AYA across the United States, this study will be a first step toward the use of previously untested strategies and a remarkable addition to the current armamentarium in the fight against high STI rates in minority groups and associated sequelae. The study findings will provide new information on more effective ways to engage sexually active AYA in a discussion on sexuality and skills building for a healthier and safer life. --- Limitations The study procedures target African American youth living in an urban community that is currently ranked number 1 in STI rates by the CDC. While we acknowledge that this population may not reflect the distribution of AYA across the United States, the distribution of STI is mirrored in the demographics of study participants. This initial study, although not generalizable to the general US population, is a first step toward larger and more heterogeneous studies. Since its inception, the study has been modified to accommodate challenges that were not anticipated prior to the launch of the study. One such challenge was the problem with the clause that restricts study recruitment to only persons with a previous or active STI diagnosis. This was met with difficulty in recruiting persons who were otherwise eligible for the study but expressed reluctance in volunteering STI information. Based on the this, a decision was reached to remove previous (or current) STI as a prerequisite for study enrollment. This modification to the study design was approved by the institutional review board and is consistent with our approach to this exploratory study, and its effect on the speed and ease of recruitment will be evaluated at the end of the study. Additionally, the current COVID-19 pandemic has been a major hindrance to the recruitment of participants into the study, as recruitment has been conducted through the clinic and via social media advertisements and interventions were previously delivered at in-person clinic visits. In response, the remainder of participants will be recruited and enrolled virtually. Social media will be used for recruitment and Zoom will be used to complete the enrollment visit. Study visits have also been moved to virtual meetings. There will be separate Zoom meetings or rooms for the individual gender-matched intervention and, if the participants are randomized to the intervention group, they will join their partners in the main Zoom room for the joint debrief session. Both participants recruited prior to the onset of the pandemic and during the pandemic will be followed up using phone interviews as designed. The effect of the COVID-19 pandemic on study recruitment or outcomes will be evaluated at the end of the study. --- Conclusion In conclusion, engaging heterosexual urban AYA dyads in a joint intervention for STI prevention and improving self-efficacy for sexual risk reduction is both feasible and acceptable. If successful, the outcome of this study will add to the current literature on the approaches to STI prevention and its success will inform its application in risk reduction counseling for youth who are most at risk for STI in the United States. --- Conflicts of Interest MT receives research supplies from SpeeDx LLC through Johns Hopkins University and serves on the Trojan Sexual Health Advisory Council (Church and Dwight Inc). 25.05.2022. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on https://www.researchprotocols.org, as well as this copyright and license information must be included. --- Abbreviations

Background: Adolescents and young adults (AYA) aged younger than 25 years have the highest rates of sexually transmitted infections (STIs) in the United States. Current STI prevention strategies for AYA rely primarily on individual approaches, leaving sexual partners with significant unmet sexual and reproductive health care and health education needs. Dyadic interventions may hold promise for harnessing the power of communal coping within relationship dynamics to enhance sexual decision making, communication, and behavior changes that reduce the future risk of STIs. Objective: This paper describes the protocol and research methods of a dyad-based behavioral intervention that augments individual evidence-based interventions with joint health education counseling for heterosexual AYA dyads within a primary care setting. The trial aims to improve partner communication and collaborative sexual decision making and promote the adoption of sexual behaviors such as consistent condom use. The primary objective of this study is to assess the feasibility, acceptability, and effectiveness of a dyadic intervention targeted at preventing STIs in heterosexual couples in an urban setting. Methods: A total of 100 AYA (50 dyads) aged 16 to 25 years, engaged in heterosexual intercourse, who reside in the city and are willing to recruit their main sexual partner for the study will be recruited and randomized into 2 groups, an intervention arm and a control arm. Participants will be recruited from an AYA medicine clinic and by using social media (Facebook and Instagram). The index participant and partner will complete a single individual session separately (Sister to Sister or Focus on the Future) with a gender-matched health educator. Dyads will then be randomized to receive an additional joint debriefing session together to discuss relationship dynamics, condom negotiation, etc. Participants will separately complete a telephone interview 6 weeks postintervention to determine the feasibility, acceptability, and impact of the intervention on mutual sexual negotiation, consistency of condom use, and communal coping skills, etc. Results: So far, 25.4% (44/173) of eligible participants have been enrolled and randomized. Participants are mostly female (20/22, 91%), with at least a high school diploma (19/22, 86%), and 9 average lifetime sexual partners. Acceptability is high, with 98% (43/44) of participants expressing satisfaction with their study experience; 100% of dyads recruited were still together at 6-week follow-up. Conclusions: Findings from this study will add to the current literature on the approaches to STI prevention, and its success will inform its application in risk reduction counseling for youth who are most at risk.

mine membership in populations of interest must be obtained routinely and reliably, and be stable over time or updated periodically. Identifying race/ethnicity for patients may be relatively straightforward; many systems capture information on racial/ethnic identity with a single, self-report item that does not change over the course of a patient's association with the health care system. However, the situation is more complicated for socioeconomic status (SES). SES is a multidimensional construct with many possible indicators (Liberatos, Link, and Kelsey 1988;Braveman et al. 2005). Many systems do not routinely collect SES information. Further, some indicators, such as income, are sensitive topics and may be uncomfortable for patients to report or for health care systems to collect (Liberatos, Link, and Kelsey 1988), and they may change over time. For these reasons, the use of area-based SES indicators obtained from address data linked to geocoded census information is sometimes used as an alternative to self-report indicators (Bonito et al. 2012). Area-based SES indicators are convenient, as patient home addresses are routinely collected and updated by health care systems for administrative and clinical purposes. Geocode linkage can often be accomplished expeditiously and at low cost (Krieger et al. 2002a(Krieger et al., 2003b)), and census data are in the public domain. Previous studies have examined the use of area-based SES measures in monitoring disparities as part of public health surveillance (Krieger et al. 2002a(Krieger et al., 2003a(Krieger et al.,b, 2005)). Despite this, there is little evidence identifying which area-based SES indicators may be best for monitoring inequalities within other health care delivery systems. Unanswered questions include the following: How well do areabased indicators correlate with self-report indicators such as educational attainment? Which geographic level is optimal? From which domain of SES, such as income, education, or occupation (Braveman et al. 2005), should an indicator be chosen? Do single-item indicators perform as well as multidimensional indices? Our goal was to identify the best area-based SES measures from the perspective of a primary care delivery system in terms of accuracy and ease of obtaining and maintaining data for ongoing surveillance. To accomplish this, we compared commonly used area-based indicators to each other and to self-reported educational attainment. Based on previous work in the public health setting (Krieger et al. 2005), we hypothesized that area-based SES measures would correlate well with self-reported educational attainment, that the census tract would provide data for the most patients with the greatest specificity, that area-based indicators related to income would best predict health care outcomes, and that single-item indicators would perform as well as multidimensional indices. --- METHODS --- Study Setting and Population We examined a cohort of adult (age >18 years) patients who were known to receive most of their care in one of 18 practices in our primary care network from January 1, 2009, to December 31, 2011, using a previously validated algorithm (Atlas et al. 2009). These practice settings included an academic, hospital-based clinic, community health centers, and private practices. The Partners HealthCare institutional review board approved the use of patient data for this study. --- Assigning Geocodes Patient address information was obtained from electronic registration data. We attempted to link each address to a specific geographic area ("geocode") at each of three levels: census block group (BG), census tract (CT), and ZIP code (ZIP). Census block groups are small, relatively homogenous areas of approximately 1,000 persons (U.S. Bureau of the Census 1994). Census tracts aggregate several block groups, and represent approximately 4,000 persons (U.S. Bureau of the Census 1994). As their name implies, both BGs and CTs are constructed by the U.S. Bureau of the Census for the purpose of assessing population demographics. By contrast, ZIP codes are constructed by the U.S. Post Office, for purposes of mail delivery, and may include more socioeconomic heterogeneity than BG or CT defined areas (Krieger et al. 2005). Previous studies have reported that ZIP code-based SES indicators detect weaker associations between SES and health outcomes, and sometimes even reverse the direction of association, compared to BG and CT defined areas (Krieger et al. 2002b(Krieger et al., 2005)) always, correspond to the ZIP code of the mailing address. Assigning a ZIP code tabulation area requires the use of geocoding software to obtain a Federally Information Processing Standard code for an address, rather than simply using the ZIP code in an address field from patient records. This requires extra time and expense, but it may yield greater accuracy. We wanted to know how indicators constructed using the simpler process of taking ZIP codes in the address field to define the geographic region (and using ZIP Code Tabulation Area data provided by the census) would compare to the more complicated process of formally linking an address to a specific BG or CT. If address information cannot be linked to a BG, CT, or ZIP, then an area-based SES indicator cannot be constructed. This type of missingness is generally greater for smaller (e.g., BG) geographic areas, as opposed to larger ones (e.g., ZIP). Furthermore, due to privacy concerns and nonresponse for various census items, the U.S. Census does not release information for every geographical area (U.S. Bureau of the Census 2003). This is also more likely to occur in smaller geographical areas, and thus ZIP code defined areas may be more likely to have information for a given indicator, than BG or CT areas (U.S. Bureau of the Census 2003). However, if a ZIP code does not overlap a ZIP code tabulation area, there will be no data for that ZIP code. We used Census 2000 data (US Bureau of the Census 2003), because Census 2010 data was not yet available for all variables of interest. We used ArcGIS software to geocode addresses from our cohort (Esri, Redlands, CA, USA). --- Socioeconomic Status Indicators We assessed several socioeconomic status indictors in our study. Self-reported educational attainment was obtained from patient registration records. Educational attainment was categorized as less than high school diploma (<unk>HS) or high school diploma or higher (<unk>HS). Self-reported educational attainment is a widely used indicator of SES and has been associated with many disparities in health (Braveman et al. 2005). We also examined several established areabased SES indicators which reflect different dimensions of SES, including income, education, and occupation (Krieger et al. 2002a(Krieger et al., 2003a(Krieger et al.,b, 2005;;Braveman et al. 2005). We constructed these indicators at all three geographic levels (BG, CT, and ZIP) based on patient address information. Area-based SES indictors can be constructed as "quantiles," where geographic areas are ranked within the study sample and grouped. However, a quantile approach may produce categorizations that vary across space and time. For example, a BG median household income which is in the lowest quantile in a sample that largely contains wealthier areas might be in the highest quantile within a more impoverished sample. Therefore, we also employed an a priori categorical cut point approach, which used previously defined thresholds to assign groups. However, some area-based indicators, as described below, did not have well-defined cut points, and thus only the quantile approach could be used. After examining a range of area-based SES indicators (data available upon request), we report on those that represented the best performance within each domain. For income-based indicators, we used (1) median household income, categorized into four groups using a priori cut points based on a government definition of low-income areas (Department of Housing and Urban Development 2013) (<unk>60 percent state-wide median household income [the worst-off group], 60-100 percent state-wide median household income, >100-140 percent state-wide median household income, and >140 percent state-wide median household income[the best-off group]); (2) median household income, as quartiles (with the lowest median household income quartile representing the worst-off group and the highest representing the best-off group); (3) percent of area persons living in poverty, categorized into four groups based on previously established cut points (Krieger et al. 2005; Department of Housing and Urban Development 2013) (>20 percent [the worst-off group], >10-20, 5-10, and <unk>5 percent [the best-off group]); and (4) percent of area persons living in poverty, as quartiles (with the quartile with the highest percent living in poverty representing the worst-off group and the lowest percent living in poverty representing the best-off group). For education-based indicators, we used percent of area individuals with college degree or higher educational attainment by quartiles (with the lowest quartile representing the worst-off group), as there was no established cut point. With regards to occupation-based indicators, we constructed one based on percent of area individuals unemployed, using a governmental definition of high unemployment areas (Employment Training Panel 2011) to categorize using the following cut points: >125 percent the state-wide unemployment rate (the worst-off group), >100-125 percent the state unemployment rate, 75-100 percent the state-wide unemployment rate, and <unk>75 percent the state-wide unemployment rate (the best-off group). Finally, we used two previously validated multidimensional indices that were constructed from several different census indicators (Figure S1): the Neighborhood Deprivation Index (Messer et al. 2006;O'Campo et al. 2008;Schempf et al. 2011) --- Health Outcomes We selected outcomes that encompassed multiple domains of health care quality and in which SES gradients were expected (Agency for Healthcare Research and Quality 2012a,b). Our dimensions of health and health care quality (Agency for Healthcare Research and Quality 2012a,b) included disease prevalence, chronic disease management and effectiveness of care, preventive service provision, resource utilization, and patient centeredness of care. For disease prevalence, we used the prevalence of diabetes, as defined by a previously validated algorithm (Grant et al. 2012). While disease prevalence may not be a valid indicator of health care quality in general, several studies demonstrating effective diabetes prevention efforts in the health care setting (Tuomilehto et al. 2001;Knowler et al. 2002) support examining this condition. For chronic disease management and effectiveness of care, we assessed the proportion of patients with coronary heart disease and/or diabetes mellitus whose most recent low density lipoprotein cholesterol in the prior year was <unk>100 mg/dL (Agency for Healthcare Research and Quality 2012a,b). For preventive service provision, we assessed the proportion of eligible adults (those aged 52-75 years without prior total colectomy) who had completed colorectal cancer screening within the recommended interval (Centers for Disease Control and Prevention 2011). For health service resource utilization, we assessed the proportion of patients considered "frequent utilizers" of the emergency department, with greater than three visits in 12 months (Liu et al. 2013). For patient centeredness of care, we assessed the proportion of patients who reported poor communication with their health care provider (defined as reporting that their provider sometimes or never explained things in a way that was easy to understand, listened carefully, showed respect for what they had to say, or spent enough time with them) in a subset of sample population who received the Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey (Agency for Healthcare Research and Quality 2012a,b). --- Data Analysis We first performed descriptive statistics of our study cohort. We then examined the missing data for each SES indicator and further characterized the Evaluating Area-Based Socioeconomic Status Indicators area-based indicators' missing data into two groups: missing because an address could not be assigned to a given geographical level, and missing because census data for that indicator was not reported for the geographic area. We compared group missingness using logistic regression with generalized estimating equations to account for repeated subjects (SAS PROC GENMOD). Next, to understand how closely associated our self-report SES indicator, educational attainment, was with our area-based SES indicators, we calculated a regression based metric, the Relative Index of Inequality (RII), for each of our comparisons, similar to prior studies (Krieger et al. 2003a;Khang et al. 2004). An alternative to the risk ratio (RR) between the worst-off and best-off groups, the RII has the advantage of incorporating all the available data in its calculation and may be less affected by extremes in high and low categories with low numbers of patients (Wagstaff, Paci, and van Doorslaer 1991;Krieger et al. 2003a;Regidor 2004). Following the standard calculation method (The Public Health Disparities Geocoding Project 2004; Khang, Yun, and Lynch 2008), each area-based SES group was assigned a ridit score, between 0 and 1, which represented the midpoint of the group's position in the cumulative distribution from worst-off (1) to bestoff (0), using the entire network sample. We calculated the RII and 95 percent Confidence Interval (95 percent CI) using log-Poisson regression with robust error variance (SAS PROC GENMOD) (Zou 2004). We selected this method because the RIIs presented can be interpreted as prevalence ratios (Kunst and Mackenbach 1994;Khang, Yun, and Lynch 2008), which is more interpretable than odds ratios given that many of our outcomes occur with high frequency, and because Poisson regression is preferred for this purpose (Khang, Yun, and Lynch 2008). In our study, the RII reported represents the prevalence ratio of a given outcome between a group having the worst-off and best-off status. We also conducted sensitivity analyses using traditional RRs and RIIs calculated with log-binomial regression, and the results were not substantially different (data available upon request). To determine if a particular geographic level, BG, CT, or ZIP consistently detected the larger gradients, we performed all analyses at each of these geographic levels. While we report nominal 95 percent confidence intervals for ease of interpretation, we were concerned about multiple comparisons. Thus, for significance testing, we compared RIIs derived from the regression equations using an alpha =.00025 level of significance, which preserves a type 1 error rate of 5 percent across 200 comparisons, using the Bonferroni correction (Pocock, Geller, and Tsiatis 1987) --- RESULTS The study cohort included 142,659 adult patients seen in our network between January 1, 2009, andDecember 31, 2011, and linked to a specific primary care provider or practice. The median age was 50.2 years, 57.5 percent were women, 69.0 percent had commercial insurance, and 7.6 percent had less than a high school diploma (Table 1). --- Completeness of SES Indicators With regards to SES indicator data, missing data were low overall ( data than educational attainment data (p <unk>.0001). Similarly, ZIP-level data had significantly lower missingness than CT, BG, and educational attainment for all other SES indicators studied (data not shown). With regards to the reason for missing data, ZIP-level data were more likely to be missing due to missing census data, as opposed to inability to assign a patient to that level. For example, using median household income data, 86 percent of patients could not be assigned a ZIP-level indicator due to missing census data, compared to 52 percent for CT and 54 percent for BG (p <unk>.0001 for both comparisons). Results were similarly significant for all other census data (available upon request). Comparing Patient-Reported Educational Attainment and Area-Based SES Indicators Overall, area-based SES indicators were strongly associated with educational attainment (Table 3). For example, using the quantile-based median household income indicator, the prevalence of having <unk>HS educational attainment was 46, 55, and 46 times higher in the worst-off, compared to the bestoff, group at the BG, CT, and ZIP level, respectively. Results were similar for the other area-based indicators (data not shown), and no geographic level was consistently more strongly associated with educational attainment. --- Association among Area-Based SES Indicators and Health Care Outcomes Area-based SES indicators consistently detected associations between low SES and poor clinical outcomes that were in the same direction as and similar in magnitude to associations detected by self-reported educational attainment (Table 4). For example, with regard to patients reporting poor provider Evaluating Area-Based Socioeconomic Status Indicators communication, the RII for the worst-off, compared to best-off, self-reported educational attainment groups was 2.66 (95 percent CI: 1.92-3.69), meaning that the prevalence of poor communication would be 266 percent higher in the worst-off compared to the best-off group. By comparison, the RII in the worst-off, compared to the best-off, group using median household income cut points at the ZIP level was 2.79 (95 percent CI: 1.72-4.50), 2.29 (95 percent CI: 1.36-3.87) for the groups defined by percent with college degree at the BG level, and 2.51 (95 percent CI: 1.51-4.18) for the group defined by the AHRQ index at the CT level. Overall, RIIs were smaller for area-based indicators than self-reported educational attainment for most outcomes, but the differences were small: the RII of at least one area-based SES indicator was not statistically significantly different from the RII of self-reported educational attainment for every outcome except diabetes prevalence, and for both the chronic disease management and patient centeredness outcomes, several area-based SES indicators detected larger disparities than self-reported educational attainment. With regard to geographic level, comparisons of BG-, CT-, and ZIPbased indicators revealed very similar RIIs. Comparing ZIP to BG/CT (Table 4), there was no statistically significant difference in 69 of 80 (86 percent) comparisons. The BG-or CT-level indicator detected a greater RII in 10 percent, and the ZIP indicator detected a larger RII in 4 percent. Compared to other ZIP-level indicators, ZIP median household income with a priori cut points was not statistically significantly different in 27/35 comparisons. ZIP median household income with a priori cut points detected a statistically significantly greater RII than other ZIP-level indicators in 5/35 comparisons, and a significantly smaller RII in 3/35 comparisons. Examining differences in cut point versus quantile approaches to indicator construction, the RIIs were again similar (Table 4). When compared to the cut point-based indicator at the same geographic level, there was no statistically significant difference in the RII detected by quantile-defined indicators. For example, ZIP-level median household income by cut point detected an RII of 3.14 (95 percent CI: 2.95-3.35) for diabetes prevalence, compared to an RII of 3.27 (95 percent CI: 3.07-3.48) for the quantile-based indicator. Finally, we examined differences in the use of single-item indicators versus indices (Table 4). Compared to NDI and AHRQ indices constructed at the same geographic level, there was no statistically significant difference for the RIIs detected by cut point-based median household income for all outcomes except diabetes prevalence, where the AHRQ group-based indicator detected a significantly larger RII at the BG, CT, and ZIP levels, and the --- DISCUSSION We sought to identify which area-based SES indicator should be used to monitor health disparities from the perspective of a primary care delivery system in terms of accuracy, ease of determination, completeness, and ongoing maintenance. First, area-based SES indicators demonstrated little missing data, especially using ZIP codes, with no consistent difference in magnitude of the RII by geographic level. Second, area-based SES indicators were strongly correlated with self-reported educational attainment, a widely used and well-validated SES indicator (Braveman et al. 2005). Third, patient-reported educational attainment and a variety of area-based SES indicators detected similar differences across a range of clinical outcome categories. Finally, cut point-based indicators performed as well as quantile-based indicators, and single-item area-based SES indicators performed as well as indicators constructed from multidimensional indices. These results suggest that area-based SES indicators such as median household income or percent living in poverty from ZIP code tabulation areas may represent a simple, easy-to-obtain way to track the impact of SES disparities on health outcomes in primary care practice networks. Area-based SES indicators constructed using data from small geographic areas, such as BG and CT, were associated with greater missing data. In addition, we did not find that the BG or CT consistently detected substantially greater health care disparities than ZIP-level indicators. For some outcomes, such as frequent ED utilization, BG and CT did detect greater disparities, but for others the opposite was true. Despite the statistically significant difference in some cases, it is not clear that the magnitudes would be importantly different from the perspective of a health care system user-either a CT-or ZIP-based indicator in these cases would still highlight an important inequality and suggest further investigation regarding whether it could and should be resolved. Thus, for interpretation, the qualitative similarities across indicators are of greater importance than quantitative differences. Supporting this view, we found no cases of a qualitative change where a ZIP-level measure Evaluating Area-Based Socioeconomic Status Indicators detected a gradient in the opposite direction of that detected at the BG or CT level. This is opposite of what we expected based on prior studies, which used public health data (Krieger et al. 2002b), and did find several of these qualitative changes. There are several possible explanations for this finding. First, BG-or CT-defined areas are thought to contain more demographically homogenous groups. While more homogenous demographics may be useful if trying to impute area values for individuals, BG and CT areas are also smaller, and a person may spend more time away from this exposure. ZIP-defined areas, which are larger, may represent greater exposure to deprivation. The relative contribution of individual characteristics (to the extent this is estimated by an area-based SES indicator) and neighborhood context may vary by outcome category, reflecting both compositional and contextual effects of SES on health outcomes. For example, the associations among area-based SES indicators and frequent ED utilization were generally larger for BG-and CT-level indicators, though not always statistically significantly, than ZIP-level indicators, but the reverse was true for patients reporting poor communication (Table 4). Because BG and CT levels are more demographically homogenous than ZIP, this may indicate that composition, that is, who are the people in that area, is more important than context, that is, what is that area like, with regard to this particular outcome. Second, BG and CTareas had more missing data. If this missingness was not random, particularly if less-well-off areas were underrepresented, then this may have introduced bias which "cancelled out" some of the beneficial effect of greater demographic homogeneity. Prior studies have not been able to compare area-based indicators to patient self-report of educational attainment, a commonly used SES indicator. Area-based SES indicators were strongly associated with self-reported educational attainment, which suggests that they may identify similar patient groups. Perhaps because of this close association, both area-based indicators and educational attainment detected expected disparities across a wide range of health outcomes. Finally, our results support the use of cut point-defined, single-item indicators, as opposed to quantile-defined or multidimensional indices. Factors that support their use include consistency over place and time compared to quantile approaches, and lower missing data compared to indices. In addition, indices rely on specific weighting among components, and so may not be applicable in settings or time-periods that are different from those in which they were derived (Braveman et al. 2005). One concern with area-level SES indicators is the "Ecologic Fallacy" described by Robinson (Subramanian et al. 2009). This occurs when inference is made by imputing an aggregate-level value for an individual parameter and associating it with an individually measured outcome. For example, this could occur if we used the ZIP median household income to impute the income for a patient living in that ZIP, and then claimed an association between individual income and an outcome, say diabetes. Because we would not, in this case, know whether the individual with the outcome actually had the imputed income, the conclusion drawn, for example "low income is associated with increased risk of diabetes" could be erroneous. However, in this manuscript, we have not used area-level data to impute individual values, but rather used individual-level data, that is, the patient's address, to assign an exposure, namely residing in an area with particular features, which likely reflects a mix of both neighborhood composition, such as the characteristics of people in the neighborhood, and neighborhood context, such as available resources. Since both the exposure and the outcome were measured at an individual level, there is no danger of the ecologic fallacy (Krieger et al. 2005;Subramanian et al. 2009). This study has several important limitations. Our results are representative of a health care system that serves a single metropolitan geographic area in the northeastern United States. Whether these results are applicable to other areas is not known. However, with regard to detecting health inequalities, our results are consistent with those derived from national samples (Agency for Healthcare Research and Quality 2012a,b), and there is a diversity of practice types within this system. Second, we have evaluated this tool for use in health care system monitoring, not for etiologic research. If attempting to "adjust" for the effect of SES on particular clinical outcomes, or establish the outcome's association with SES, self-report educational attainment, or income information, may be more useful. Finally, there is no "gold standard" indicator of SES to compare area-based SES indicators against. However, our technique of comparing area-based indicators to self-reported educational attainment and to area-based indicators from other domains allows a "ballpark" estimate of the "true" effect of low SES. While each indicator measures a different exposure, we found that the magnitude of the RIIs were generally similar. This simultaneous assessment of self-reported and area-based SES measures is an important strength compared to prior studies (Krieger et al. 2002b(Krieger et al., 2003b(Krieger et al., 2005)). These limitations are balanced by several strengths. Previous studies addressing the use of area-based SES indicators for monitoring disparities have used data from public health surveillance efforts, and while important, they may not translate to the health care delivery setting with different sources Evaluating Area-Based Socioeconomic Status Indicators of data and different outcomes. In addition, the comparison of ZIP code-based indicators to ones defined using census geography (BG and CT) has practical implications for primary care networks. The ability to assign SES indicators to a high proportion of patients using ZIP codes, and have them perform similarly to BG-or CT-level SES indicators, increases the feasibility of monitoring SES inequalities within health care systems. Being able to construct indicators without having to use specific geocoding software significantly reduces the expense and complexity of indicator construction. Additionally, while we have focused on using area-based SES indicators for disparity surveillance, they may still be useful for research studies within practice-based research networks. For systems where researchers have access to educational attainment, or other patient-reported SES indicators, the addition of area-based SES indicators can allow for the use of multilevel frameworks when seeking to disentangle compositional and contextual effects of SES (Subramanian et al. 2009). One important area of development in measuring SES using publicly available data is the construction of the HOUSES index (Butterfield et al. 2011). While not currently in widespread use, this method uses publicly available housing data for index construction. As this method develops, it will be useful to compare its performance in detecting SES disparities to other existing methods. In conclusion, area-based SES indicators can be used to assess disparities in clinical outcomes in a large health care network. Almost all indicators detected similar disparities, so based on considerations of completeness, ease of use, and consistency over time and place, we recommend ZIP-level median household income or percent living in poverty, based on a priori cut points, for monitoring SES differences in health outcomes within a health care system. --- SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article: Appendix SA1: Author Matrix. Figure S1. Formulae Used to Calculate Area-Based Socioeconomic Status Indices. Evaluating Area-Based Socioeconomic Status Indicators

Objective. To determine which area-based socioeconomic status (SES) indicator is best suited to monitor health care disparities from a delivery system perspective.

Introduction Menstruation is an issue that impacts many facets of life such as social participation, mental and physical health, education, and employment [1][2][3]. Globally, women and girls face numerous challenges in managing their menstruation. These challenges arise from cultural taboos, lack of knowledge, limited access to safe and secure water, sanitation, and hygiene (WASH) services, affordable menstrual products, appropriate disposal structures, and pain management [4][5][6]. In 2016, the Joint Monitoring Program of the World Health Organization (WHO) and UNICEF defined adequate menstrual hygiene management as access to clean absorbents including sufficient washing, drying, storage and wrapping of reusable absorbents; adequate frequency of absorbent change; washing the body with soap and water; adequate disposal facilities; privacy for managing menstruation; and basic understanding of menstruation and how to manage it with dignity and without fear or embarrassment [7,8]. More recently, this has been expanded to menstrual health and hygiene (MHH) to include the sociocultural and economic factors that inform menstrual management and impact women's lives [3,8]. Improving MHH will contribute to achieving the Sustainable Development Goals (SGDs) on gender equality (SDG5), good health (SDG3), education (SDG4), and clean water and sanitation (SDG6) [9]. Yet, most girls and women in low and middle-income countries (LMICs) continue to have inadequate MHH [10]. Current options for menstrual products include disposable pads and tampons and cost-effective and environmentally friendly reusable pads, period pants, and the menstrual cup. However, many women and girls from LMICs use ineffective and unhygienic menstrual materials such as cloth, toilet paper, or old underwear [11]. Discomfort during menstruation due to ineffective menstrual products and pain relief may result in heightened anxiety, lack of confidence, and absenteeism from school, work, sports, and other socially or economically beneficial activities [12,13]. Moreover, lack of accurate menstrual knowledge and the proliferation of harmful sociocultural norms can lead to myths and taboos that also contribute to the negative menstrual experiences of young women in LMICs [3]. A systematic review in 2016 highlighted the potential for interventions to improve MHH-related outcomes among women and girls but noted a lack of evidence on optimum models for delivery of MHH interventions [10]. Since then, several intervention studies have been conducted in LMICs but vary in quality and focus almost exclusively on girls in school [3]. Little is known about how girls and young women respond to or experience nonschool, community-based MHH interventions, and which factors inform intervention or product uptake [14]. The aim of this pilot study was to assess and refine the design and implementation of a comprehensive MHH intervention incorporated into a cluster randomized trial of a package of integrated HIV and sexual and reproductive health (SRH) services for youth delivered in community-based settings across Zimbabwe (the CHIE DZA trial) (registered in clinical trials.gov:NCT03719521). Pilot findings informed two studies: the implementation and scale-up of the MHH intervention within the CHIE DZA trial; and a nested longitudinal MHH sub-study investigating the acceptability, uptake, and effectiveness of an MHH intervention on MHH knowledge, practices, and perceptions among young women aged 16-24 years in Zimbabwe (Fig. 1). The objectives were to investigate (1) the uptake of a novel comprehensive MHH intervention, (2) menstrual product preference, and (3) the external sociocultural, economic, physical, and environmental factors that informed intervention uptake and menstrual product choice among young women in a community-based SRH programme. --- Methods --- Study setting and participants This study was nested within the ongoing CHIEDZA trial that seeks to determine the impact of an integrated community-based package of SRH and HIV services for 16-24 year olds on population-level HIV prevalence and other health outcomes. The two-arm trial is conducted in 24 clusters (a geographically demarcated area that includes a community centre and a primary health care clinic) in three provinces in Zimbabwe (Harare, Mashonaland East, and Bulawayo), with eight clusters per province. Each province is stratified 1:1 to either existing, routine health services (control arm), or to receive a package of SRH and HIV services (intervention arm). The intervention arm services are delivered in community centres over a 2-year period. All residents aged 16-24 years in the intervention clusters are eligible to access CHIEDZA services. In the current paper, we describe an MHH pilot study which included female participants accessing CHIEDZA services at the four intervention clusters in the poor and urban settings of the Harare province. --- The MHH intervention Formative work that included a combination literature review, stakeholder engagement, and qualitative focus group discussions (FGDs), in-depth interviews, and participatory workshops were conducted with young women (aged 16-24 years old), community health workers (CHWs), and other community stakeholders such as relevant community-based organizations and the Ministry of Health and Child Care to inform and develop a Theory of Change (ToC) for the MHH intervention. Details of the formative work will be published elsewhere. The ToC recognizes that successful and sustainable MHH interventions need to address stigma and harmful myths and taboos around menstruation, access to MHH education and products, and access to pain medication. The aim of the ToC was to encapsulate how different components of the intervention would contribute to the intended outcomes (Fig. 2). Based on this, the MHH intervention, to be piloted within the ongoing CHIEDZA trial, was designed to include analgesics (paracetamol or ibuprofen), two pairs of underwear, a bar of soap, a simple period-tracking sheet, comprehensive MHH education including an MHH educational pamphlet, and a choice of either a menstrual cup or reusable pads (Fig. 3). The MHH package was provided freeof-charge and delivered to young women within CHIEDZA trial. The MHH intervention was delivered by an intervention team that included two nurses, three community health workers (CHWs), two youth workers, and one counsellor. The entire intervention team went through a 2-week training that addressed (1) youth-friendly service delivery, (2) logistics management, (3) principles of counselling, (4) research ethics and Good Clinical Practice, and (5) engaging young people and other community members. --- Study procedures The pilot study was a prospective mixed-methods study conducted from April to July 2019 in all four intervention sites in Harare. In evaluating the implementation of the pilot study, we routinely monitored uptake and coverage data from the CHIEDZA trial and attended weekly CHIEDZA intervention team meetings to more fully understand the context in which the MHH intervention was being conducted [15]. --- Quantitative methods Each attendance at the CHIEDZA service is tracked in real-time using a biometric system [16]. Fingerprints, Fig. 1 Framework for how the pilot informs the CHIEDZA trial and the nested MHH longitudinal sub-study age, and sex for each participant is recorded and delinked from participant's name, birthdate, address, or other identifying information. Additionally, the services taken up by participants at each attendance are recorded on electronic tablets. Data for the female participants who accessed CHIEDZA services in the four Harare sites were used to assess intervention uptake and product choice. MHH intervention uptake was calculated as the proportion of women who took up MHH services of those who attended CHIEDZA. Z tests were conducted to compare the proportions reporting MHH intervention uptake and MHH product choice disaggregated by age group. --- Qualitative methods We conducted one focus group discussion (FGD) with the intervention team and two FGDs and four in-depth interviews (IDIs) with female CHIEDZA participants to explore the factors that influenced uptake of the intervention and choice of menstrual product. The intervention team's FGD included all eight team members: there were three males (aged 24-32 years old) and five females (aged 24-44 years old). The FGD explored their understanding of, and attitude towards, the MHH intervention, their experiences of delivering the intervention, and their own experience of using the available products. For the FGDs with participants, we used purposive sampling to select 12-15 women to participate in two FGDs disaggregated by age (16-19 and 20-24 years old). Participants were selected to include those who did, and did not, take up the MHH intervention, and those who chose the menstrual cup or reusable pads respectively. At midline (two months into the pilot study), participants were screened (over the period of a week) by a female research assistant (RA), informed about the study, and asked if they were willing to be contacted via telephone to participate in the FGDs and subsequent IDIs. Two FGD participants from each age group were purposively selected to represent different product choices (two for the menstrual cup and two for the reusable pads), for follow-up IDIs to explore the factors that informed menstrual product choices. The FGDs and IDIs were conducted face-to-face by an experienced female RA in either Shona or English (as agreed by the participants), used semi-structured topic guides and were audio-recorded. FGDs took 45-60 min and IDIs took 30-45 min. FGDs took place at the CHIE DZA site outside of usual opening hours to ensure confidentiality. IDIs took place at a time and place most convenient to the participant. Written informed consent was provided before the FGDs or IDIs were initiated. Audio recordings of the FGDs and IDIs were then directly transcribed into English for analysis. Data was analysed using thematic analysis based on the following broad themes: current MHH knowledge, perceptions, and practices, facilitators and barriers to product choice, and product user experience [17]. Codes were generated based on these themes and sub-themes emerging from the transcripts. All transcripts were coded manually by MT and reviewed by a senior social scientist (JR). Themes and coding were continually reviewed and refined to capture emerging new codes. Verbatim quotes from interview participants were captured to highlight thematic areas and to increase our understanding of the context. Quantitative and qualitative data were analysed independently and then the findings triangulated to deepen our understanding of how the intervention was working, how it was being received, and how it could be improved. These findings were then further interrogated during the weekly CHIEDZA intervention team meeting. Collectively, the group would reach a consensus on any changes or actions needed based on the evidence from the data and their experiences. --- Results --- MHH intervention uptake Of the 1732 eligible participants who sought services at the CHIEDZA centres between April and July 2019, 1414 (81.6%) took up the MHH intervention at their first visit. There was no evidence of a difference of uptake between the four sites. There was strong evidence for a difference in uptake by age group with 690/816 (84.6%) of 16-19 year olds compared to 724/916 (79.0%) of 20-24 year olds accessing the MHH package (p = 0.003). From the qualitative data, key themes related to factors that influenced uptake of the MHH intervention were access to free menstrual products and analgesics, youthfriendly intervention delivery, and access to MHH information. --- Access to free menstrual products and analgesics Almost all participants, particularly younger women, cited the MHH intervention as the reason behind their initial CHIEDZA visit: "I personally came here with an intention to get pads and when I entered into the CHIEDZA booth, I saw a very friendly service provider and felt comfortable and free to talk" (FGD, 16-19 years old). A key motivating factor for the observed high uptake was the provision of free reusable menstrual products, particularly reusable pads. Most participants reported having to use old socks or cotton wool in the absence of the menstrual product they would prefer to use or were able to afford before the economic downturn in Zimbabwe: "sometimes I would also use cloths when I didn't have enough pads to last my period" (FGD, 20-24 years old). Almost all participants reported that they were "grateful" and "happy" that the intervention provided them with a choice of menstrual products which they could not afford and did not have access to before. MHH intervention team members also reported that the free products were "the most important hook" for young women that accessed the CHIEDZA services: "...those who are coming in for menstrual hygiene, they are going out and inviting others for menstrual hygiene. They are only telling them that if you go to the community you can get pads, you can get a cup or something. They are not really raving about other services but it's all about menstrual hygiene... the communities love the products we have" (FGD, Nurse). Some participants reported experiencing pain during their menstrual periods and cited access to monthly analgesics as an additional reason for taking up the MHH intervention. Almost all participants chose to take up the MHH intervention. Of those that did not, most only declined uptake because their menstrual product of choice was unavailable on the day of their visit. --- Youth-friendly intervention delivery and access to MHH information Participants had learned about the MHH intervention in CHIEDZA through community mobilization efforts facilitated by the intervention team: "...we were on our way from clinic and we were told to go to the community hall to get some pads" (FGD, 20-24 years old). Once at CHIEDZA, participants reported being "treated well" and many found the service providers "friendly" and "helpful". MHH intervention uptake was both a function of the provision of needed menstrual products and youth-friendly service provision facilitated by the delivery team. Participants highlighted how the intervention staff, unlike their parents or teachers, provided much needed access to MHH information in a safe, nonjudgemental environment and in a way in which resonated with them: "I think this is a good programme because it helps us. Some children might have questions but they are not able to ask their parent, they might not be open to their parents but CHIEDZA, you are free to ask and say things you want" (FGD, 16-19 years old). Many participants described how their only conversations about menstruation began and ended at menarche and were limited to menstrual product use and basic hygiene guidance. The MHH intervention provided young women an avenue to learn more about menstrual health from trained staff and to talk through their MHH-related concerns and anxieties in a safe space: "When I entered into the booth, I saw a very friendly service provider and felt comfortable and free to talk and I was able to express my feelings and to openly seek the help I needed, and I was assisted there. Getting pads was now an extra benefit" (FGD, 16-19 years old). Participants highlighted the MHH-related education sessions with the intervention staff. These sessions gave participants an opportunity to feel the menstrual products, to observe menstrual product use demonstrations, and to openly discuss myths and taboos around menstruation in private consultation. All the intervention team members reported that the MHH component of CHIEDZA was received with "gratefulness" and was the "most popular" service. Overall, both the provision of free menstrual products and youth-friendly service provision were highlighted as key facilitators to intervention uptake. However, older participants seemed to be more motivated by the former rather than the latter as most of them had children with them or household responsibilities to get back to and thus did not have the time to engage in the MHH-related education sessions or other youth-friendly activities within CHIEDZA. --- Menstrual product choice among participants Of the 1414 participants who took up the MHH intervention, 1244 (88.0%) participants chose to receive the reusable pads and the remaining 170 (12.0%) chose a menstrual cup on their first visit. There was strong evidence of a difference of product choice by age, with 50/ 690 (7.2%) of 16-19 year olds to take up the MHH intervention choosing the menstrual cup versus 120/724 (16.6%) of 20-24 year olds (p <unk> 0.001) (Fig. 4). The qualitative data with clients and the intervention team highlighted key themes related to factors that influenced product choice: barriers and facilitators to the uptake of menstrual cups; and barriers and facilitators to the uptake of reusable pads. --- Barriers to the uptake of menstrual cups Sociocultural norms negatively influenced both the delivery and the uptake of the menstrual cup among participants. Participants reported that they and their caregivers in the community were hesitant about inserting the cup into their vagina. Most feared that the cup would "take their virginity". Some feared that the "big" cup would be too difficult or painful to insert. Others thought the cups were stiff and hard and feared that the cup would stretch out their vagina making them undesirable for men to have sex with. The FGD with the intervention team members also highlighted these fears, and this affected their ability to promote cup uptake, with many explaining their struggles with delivering clear messaging to allay these concerns: "I think it's also an issue of needing to look at our cultural values. It's about what are we told from childhood about virginity and inserting such big things. So that issue is a concern and how would you tackle it if you want to introduce a cup" (FGD, Community health worker). Participants mentioned that the intervention team appeared reluctant to talk about and distribute the cup, and some expressed that they would have been willing to trying the cup if it had been talked about during service delivery: "I just heard someone saying cup, but I did not know which cup she was referring to... I got pads instead of cup, and then I was like a cup? I did not get it, I thought she is talking about a cup of tea. I honestly thought that the CHIEDZA people had been unfair, and I wanted to go back and to get a cup" (FGD, 20-24 years old). --- Facilitators promoting the uptake of menstrual cups The main factors that facilitated menstrual cup uptake included anecdotal evidence of menstrual pain relief and the prevention of leakage. Some participants chose the cup over the reusable pad as it could be cleaned and dried discreetly. Even though most participants chose the reusable pads, many raised concerns over their need to "keep [their menstruation] a secret", and challenges in washing and drying their pads outside where other family members and neighbours could see. --- Fig. 4 Menstrual product choice disaggregated by age group The majority of participants gave negative accounts of their menstrual experiences and often described menstrual blood as "dirty" or "impure". Many spoke about being restricted from housework, social or religious gatherings or sports while others expressed fears of leaking or "spoiling" their clothes leading to being "teased" by boys or chastised by other females during their menstrual periods. Informed by these negative perceptions and experiences of menstruation, some participants chose the menstrual cup over the reusable pads as they felt it was less likely to leak: "The time for menstruation is so annoying because you will be anxious and you will be afraid of spoiling your clothes, you may not feel comfortable with a pad, so I decided to try a cup" (IDI, 20-24 years old). Others that chose to take up the menstrual cup also cited "pain relief" as reason for uptake: "[Other participants] told me that period cup heals period pains slowly. If you continue to use it you end up not having period pains" (FGD, 20-24 years old). Limited access to water in the communities and seasonal rains also informed product choice. Some of the participants who opted for the cup cited the fact that it did not require much water to wash or time to dry as reusable pads as the reason for their choice: "the cup is very smart, and you don't need to do much washing, but you just remove it and empty it" (IDI, 20-24 years old). --- Barriers and facilitators promoting the uptake of reusable pads The main facilitators included peer influence (particularly for participant's aged 16-19 year olds) and the similarity between reusable pads and disposable pads. Participants noted that the reusable pads were most similar to disposable pads they had used before and therefore the less "scary" than the cup. Many of the participants, particularly those aged 16-19 years old, were encouraged to visit CHIEDZA by a classmate that had previously visited the intervention site: "We were in class and Rashna brought her pads and started to show us,'see what they look like, and see what they look like!' So I said, 'Where did you get them?' and she said 'I got them from the community centre, on Tuesdays.' Today is my second time coming here, when I came for the first time, I was given pads" (FGD, 16-19 years old). Overall, the few barriers to reusable pads uptake seemed to be linked to the environmental factors that facilitated menstrual cup uptake. --- Discussion In this pilot study of a novel comprehensive MHH intervention, the uptake was high. Key factors to intervention uptake, and the SRH program more broadly, were the availability of free menstrual products and analgesics, youth-friendly intervention delivery, and access to tailored MHH education. Most young women preferred reusable pads to menstrual cups. Barriers to reusable pads uptake were limited to environmental factors such as limited access to the amount of water needed to wash the pads, concerns about appropriately drying pads in the wet season and discomfort around openly drying the pads. Highly influential facilitators to uptake of reusable pads included peer influence amongst younger women, the appearance of the pads, and the comfort and familiarity derived from the similarity of the reusable pads to the more well-known disposable pads. Despite anecdotal evidence of menstrual cups reducing leakage and menstrual pain and being easier to clean than reusable pads, uptake was negatively informed by strong sociocultural beliefs around the preservation of virginity, lack of promotion of the cup from the intervention team, fear around the size of the cup which was perceived to cause vaginal stretching or pain, and/or anxiety around an inability to insert it properly. This study highlights the importance of context-specific interventions and informed choice when offering MHH services to young women. Our study findings support similar findings from other LMICs that note that most young women face challenges in accessing menstrual products and only learn about menstruation at menarche and even then, the information is limited and skewed by local myths and taboos [18,19]. These barriers then lead to feelings of isolation, shame, and fear that negatively inform how young women experience menstruation over time [3]. Data on community-based MHH interventions in LMICs is limited [3,20]. This study provides evidence of the need for MHH interventions in community-based settings. Young women want access to MHH information, analgesics, and menstrual product choice and, if possible, young women positively respond to these services being delivered in youth-friendly spaces by supportive and friendly staff. The principles of Behaviour Centred Design (BCD) theory posit that behaviour change interventions must disrupt the external environment with a "surprise" that causes a shift in the target individual that results in the desired change or "performance" [21,22]. Nested within a larger SRH intervention, the qualitative data from the pilot study suggests that the MHH component may have been the most attractive service of the SRH package-effectively acting as the "surprise" that disrupted the community environment causing young women to come to CHIEDZA with the prospect of being rewarded with much needed free menstrual products, information, and support. Our hypothesis is that MHH services can facilitate access to broader SRH and HIV services within CHIEDZA. While the pilot data does not provide quantitative evidence that other SRH services in CHIEDZA were subsequently taken up, we can posit that the MHH intervention increased female engagement with CHIEDZA services. We will be able to investigate this further in the larger study. Robust evidence for the potential "pull factor" of MHH interventions in SRH programming could present a strong case for the integration of MHH in SRH and HIV programming for young women. Reusable menstrual products are both cost-effective and environmentally friendly. There is also anecdotal evidence from study participants that menstrual cups reduce instances of leakage and period pain, particularly from those participants that previously used tampons to manage their periods. Despite these benefits, barriers to menstrual cup uptake centred around sociocultural norms that discourage insertable products due to (1) fears of "losing virginity" by rupturing the hymen, or (2) fears of hurting or stretching the vagina due to the size of the cup. Young women, particularly unmarried women, are often discouraged from inserting products into their vagina to preserve the hymen as a sign of one's purity and virginity before marriage [23]. Our results highlight that these sociocultural norms influenced, not only, the participants but also the intervention team members as well. Despite extensive training on menstrual cup use, most of the intervention team members reported to have found it difficult to promote the menstrual cup because of their sociocultural beliefs. The tenets of BCD state that sustainable and effective interventions need to consider factors outside of the behaviour setting as these external factors inform participant behaviour [21]. Our findings suggest that product choice goes beyond provision-external environmental factors such as access to water and sociocultural factors in the community also play roles in the decision-making process. With a growing body of work to scale-up use of the menstrual cup in LMICs, understanding social and contextual factors will prove critical to improving acceptability [24]. Importantly, as reflected in our findings, intervention acceptability has to be considered and addressed from the perspective of both the participants and the service providers [25]. MHH is now recognised as an important public health issue worldwide [26], with an increase in MHH advocacy and research globally [14,27]. Collaborative networks such as the Menstrual Health Hub and the African Coalition for Menstrual Health Management and advocacy efforts such as the "MHM in Ten" initiative, globally recognized "Menstrual Hygiene Day", and the annual co-hosting of the MHM in Water, Sanitation and Hygiene (WASH) in Schools virtual conference have all played important roles in placing MHH at the centre of international research and development dialogue and in mobilizing efforts to address the MHH needs of girls and women in LMICs [27][28][29][30]. However, apart from a small feasibility study investigating menstrual practices and perceptions around the use of the Duet (an insertable menstrual product) and a cup acceptability study with 54 young women, little scientific data on MHH needs or experiences among young women in Zimbabwe is available [23,31]. A systematic review of menstrual health interventions conducted in 2020 also highlights limited research into the lived experiences of young women that engage with MHH interventions in LMICs with most of the existing data focusing on school-based interventions and education outcomes [14]. This pilot provides crucial MHH programming information that has contributed to the development and implementation of improved a multi-component MHH intervention that aims to improve MHH knowledge, practices, and perceptions among young women in Zimbabwe. Findings from the pilot informed the following changes to the MHH intervention within CHIEDZA (see Fig. 5): 1. Implementation of trained menstrual cup ambassadors to increase cup promotion and sensitization and to provide ongoing support for new cup users 2. Procurement of smaller and softer menstrual cups that are less intimidating to new users and are less likely to cause discomfort or pain during insertion 3. Implementation of group-based MHH education sessions aimed at demystifying menstruation and facilitating MHH dialogue for participants 4. Inclusion of community members such as mothers, fathers, partners, and caregivers, in MHH discussions Strengths of the study were the availability of real-time quantitative data and the use of a mixed-methods approach to gain an in-depth understanding of the various contextual and individual factors that influenced both the engagement of young women with MHH intervention and menstrual product choice. Limitations of the study are that data were only collected on menstrual product uptake at single point in time and uptake does not necessarily translate to product use. Additionally, pilot data was collected from a fairly small sample in poor urban settings in Harare, Zimbabwe therefore the data may not be generalizable to all young women, such as women in rural or high-income settings. Due to the sensitive nature of the discussions, qualitative data collection may have been informed by social desirability bias leading to inaccurate reporting of factors informing menstrual product choice. --- Conclusions To our knowledge, this is the first study to investigate MHH intervention uptake and product choice in a community-based setting in an LMIC. Overall, the pilot study results showed an unmet need for comprehensive MHH interventions in the community. Results also highlighted the strong influence of sociocultural and environmental factors on menstrual product choice and economic factors in informing participant engagement overall. Community-based interventions should be context-specific and multicomponent focused to fully address the MHH needs of young women. Importantly, access to MHH education, pain management medication, and a choice of MHH products in youth-friendly SRH programming may act as facilitating factors to increase female engagement in SRH services and improve young women's SRH outcomes over time. --- Availability of data and materials All data generated or analysed during this study are included in this published article (and its supplementary information file). --- Supplementary Information The online version contains supplementary material available at https://doi. org/10.1186/s40814-020-00728-5. Additional file 1. MHH Intervention Uptake and Product Choice at First Visit. Quantitative data extracted from biometric system highlighting MHH intervention uptake and product choice during pilot period. --- Abbreviations --- MHH intervention and participant flow for main MHH intervention Authors' contributions MT designed the MHH intervention, analysed and interpreted the data regarding intervention uptake and menstrual product choice, and drafted the manuscript. JR analysed the qualitative data and was a major contributor in writing the manuscript. HW and SF assisted analysed the qualitative and quantitative data collected and contributed to the writing of the manuscript. TL collected qualitative data from participants. RF, CDC, CVM, and ED contributed to the coordination of the study and the critical revision of the manuscript. NR and TB assisted with quantitative data capturing and analysis. All authors read and approved the final manuscript. --- Ethics approval and consent to participate This study was approved by the Medical Research Council of Zimbabwe [MRCZ/A/2387], the London School of Hygiene and Tropical Medicine ethics committee [16124/RR/11602] and the Biomedical research and Training Institute Institutional Review Board (MRCZ/A/2387). Written informed consent was provided before the FGDs or IDIs were initiated. Pseudonyms were used during the FGDs and in subsequent use of the quotations to facilitate confidentiality and maintain anonymity. All recordings were downloaded on to a password-secured laptop for processing purposes and deleted once the data analyses were completed --- Consent for publication Not applicable --- Competing interests The authors declare that they have no competing interests. --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background: Menstrual health and hygiene (MHH) is a human rights issue; yet, it remains a challenge for many, especially in low-and middle-income countries (LMICs). MHH includes the socio-political, psychosocial, and environmental factors that impact women's menstrual experiences. High proportions of girls and women in LMICs have inadequate MHH due to limited access to menstrual knowledge, products, and stigma reinforcing harmful myths and taboos. The aim of this pilot was to inform the design of an MHH sub-study and the implementation and scale-up of an MHH intervention incorporated into a community-based cluster-randomized trial of integrated sexual and reproductive health (SRH) services for youth in Zimbabwe. The objectives were to investigate (1) uptake of a novel MHH intervention, (2) menstrual product preference, and (3) the factors that informed uptake and product choice among young women. Methods: Female participants aged 16-24 years old attending the community-based SRH services between April and July 2019 were offered the MHH intervention, which included either a menstrual cup or reusable pads, analgesia, and MHH education. Descriptive statistics were used to quantitatively assess uptake and product choice. Focus group discussions and in-depth interviews with participants and the intervention team were used to investigate the factors that influenced uptake and product choice. Results: Of the 1732 eligible participants, 1414 (81.6%) took up the MHH intervention at first visit. Uptake differed by age group with 84.6% of younger women (16-19 years old) compared to 79.0% of older women (20-24 years old) taking up the intervention. There was higher uptake of reusable pads (88.0%) than menstrual cups (12.0%). Qualitative data highlighted that internal factors, such as intervention delivery, influenced uptake. Participants noted the importance of access to free menstrual products, analgesics, and MHH education in a youth-friendly environment. External factors such as sociocultural factors informed product choice. Barriers to cup uptake included fears that the cup would compromise young women's virginity.

INTRODUCTION Sexually transmitted infections (STIs), are infections that are passed from one person to another through sexual contact. STIs are a major public health concern both in developed as well as developing countries. More than 1 million STIs are acquired every day globally and the Human Papillomavirus (HPV) is the most common type of STI around the world. In 2020, WHO estimated 374 STIs affect individuals of all ages, particularly young people as during this stage various physical, mental and social developments take place in an individual. College students are most active due to their freedom of living away from home, alcohol use and influence by peers which leads to curiosity, experimentation resulting in the involvement of risk-taking behaviour and casual sex activities. Moreover due to limited knowledge regarding STIs, their symptoms and causes among college students they get more involved in premarital sex and risky sexual practices. 5 It is important to understand the region specific factors influencing the knowledge, attitude and practices with respect to the sexual knowledge and practices in youths to raise awareness among them and to plan interventions to overcome the barriers, if any. The current study assessed the knowledge, attitude and practices of students of Panjab University, Chandigarh, regarding sexually transmitted infections. --- METHODS A cross sectional study was conducted among the students of Panjab University, Chandigarh (North India). The study was conducted from May 2022 to July 2022. The sampling frame consisted of both male and female students of Panjab University located in Chandigarh. Sample Size was calculated by using the following formula with approximation for large population: N = Z 2 1 -<unk>/2 P (1 -P)/d 2 Where, P = Anticipated population proportion, 1 -<unk>= Confidence Coefficient, d = Absolute precision, and Z is the value of standard normal variate. Taking awareness of STI among adolescents as 89%, assuming 95% confidence interval and 5% absolute precision sample size comes out to be 157. 6 The unit of analysis was the individual students, who were administered the survey instruments. The participants were selected by simplified random sampling technique. Hence a sample of 252 students was assessed. These students was categorized into two groups according to their sex, i.e., males and females. A sample was then drawn from each category having 127 females and 125 males. The students were enrolled in the study after verifying their identity card. Confirmation of enrolment was done through physical proof of the library card. Exclusion criteria: 53 Students from the Centre for Public Health were excluded as participants for the study, 7 incomplete questionnaires were also excluded from study. A self-administered, semi structured and validated questionnaire, by experts from Centre for Public Health, Panjab University and Government Medical College and Hospital sector 32, Chandigarh was used for collecting data about knowledge, attitudes and practices regarding Sexually Transmitted Infections from students after explaining them the objectives of the study and taking their written consent. The students were assured and not ensured regarding confidentiality of their identity so that they can be comfortable in giving correct answers even to the sensitive questions. Statistical Package for Social Sciences (SPSS) version 28 computer software was used for coding and data entry for descriptive and inferential analysis of data. Appropriate descriptive analysis was used to generate frequency distributions tables for knowledge, attitude and practices regarding STIs. ANOVA and Pearson correlation was used to find correlation between knowledge, attitude and practices related to STIs. --- RESULTS --- Demographics of respondents The study was conducted amongst 252 students. There were 125 (49.5%) males and 127 (50.4%) females. The age group of participants was classified into 17-19 (8.7%), 20-24 (66.7%), 25-29 (22.6%) and >30-50 (2.0%). Total 77.8% of participants belonged to urban areas while 22.2% were from rural areas. Regarding religion of participants 60.3% were Hindu, 25.4% Sikh. --- Assessment of total knowledge about STI 90.5% had the correct knowledge that multiple sexual partners can increase risk of STI. 82.5% of participants answered correctly that it is irresponsible to have sex with someone without a condom and 69.8% answered that STI can be transferred from mother to child during pregnancy. Only 68.3% participants answered correctly that a person can have STI without any visible signs and symptoms. Out of 252 participants, 67.9% responded that it is possible to get STI when you first have sex and 59.9% responded that STI can cause long term health problems. 26.2% of participants did not know that the early stage of STI is not contagious and 25% did not know that STI can lead also to infertility. 92.50% of respondents had heard about AIDS, 68.70% of syphilis, 65.90% of Gonorrhoea and 60.30% of Hepatitis B, while only 2.40% responded to Genital warts as a STI. Participants were aware of symptoms of STI, 77.70% responded itching/burning sensation on genital areas 70.90% sores on sexual organs, 66.90% discharge from genital area. Virus causes STI was responded by 80.40% followed by bacteria (65.20%) and fungi (33.60%). 92.1% responded vaginal sex, 73.00% anal intercourse, 65.5% oral sex while 22.60% responded kissing as the main act for STI transmission. The Internet was chosen as a source for providing information on STI by 77.80% followed by TV/radio (68.30%). --- Assessment of attitude about STI Around 86.1% agreed with the statement that "one should go for an STI test". When participants were asked if sex before marriage is becoming a normal practice, 62.7% agreed. 46% of respondents disagreed that having multiple sexual partners was not a problem. While 38.1% agreed that banning prostitution can control spread of STIs? 64.7% of respondents stated that one should not have extramarital affairs, 24.6% supported isolating people living with STIs for the sake of public safety. Majority of respondents opined on introducing sex education at school level. --- Assessment of practices about STI Only 10.7% of respondents had tested for STI at some point of time whereas the majority of them (78.2%) never tested. 79.8% had no sign and symptom of STI. 19.4% agreed with the statement about having multiple sexual partners while 65.9% were reluctant to disclose on having such partners. 64.7% of the respondents denied experiencing injecting intravenous drugs. 71% of respondents felt comfortable discussing topics such as sex and STIs with family members, teachers, friends, coworkers.10.3% experienced violence during sexual encounters with a spouse or other sexual partners. Only 16.7% agreed to drinking alcohol in the past. On statement about premarital sex, 54% refused to disclose such affairs. However, 26.6% opined they did know how to use a condom correctly. --- Correlation of knowledge, attitude and practices with STI The Pearson correlation was used in this study to find a correlation between the total scores on knowledge, attitude, and practices. The study found a statistically significant weak correlation was found between knowledge and practice (r=0.357). --- DISCUSSION The current study assessed the knowledge, attitude and practice regarding sexual transmitted infections among university students of Chandigarh. The present study found that the majority of the participants were aware about diseases like HIV/AIDS (92.5%), 60.3% of hepatitis and 68.7% of syphilis, 65.9% of gonorrhoea. Majority of participants agreed one should get a STI test and sex before marriage is becoming common practice. Around one fifth of the participants agree with having multiple sexual partners. A similar proportion stated that biological and socio-cultural settings make women more vulnerable to STI. Majority of respondents have never tested for STI, 19.4% had multiple sexual partners. This gap in stressing upon the severity of diseases led to differential awareness regarding these diseases among college/ University students. Furthermore gonorrhoea and syphilis are a part of STIs which are covered under NACP as a subunit of it and there is no separate dedicated programme for these diseases further leading to less awareness among students. Educating children about these diseases in their school curriculum can be a promising step in raising the awareness and further capping the morbidity due to these diseases in India. The present study found sexual intercourse (94%) and blood transfusion (73.4%) were the main causes of STI transmission. The study conducted among Turkish adolescents found that 92.6% respondents reported sexual intercourse ever in their life where as a study among Nigerian adolescents found that 61.9% had received blood transfusion in their life. 8.9 High prevalence of sexual intercourse in current study can be attributed to the easy availability of condoms, casual attitude towards sexual intercourse these days and residing away from home with no one watching them. 10 More than three fourth of the students reported the internet as source of information for STIs while two third of the total participants responded TV/Radio as an information source. A study in Nigeria reported that the main source of information about STI was TV and radio (68.7%). 11 It is worth noting that the internet and visual media have largely replaced many traditional sources of information in this age group. The wide coverage and easy accessibility to the Internet in India has made people have access to information via mobile phones. Further the awareness about the STIs by the Government of India on television and radio has also helped in spreading the awareness about these infections. 12 Our study found that 21.8% of the participants experienced premarital sexual intercourse. A study conducted in Malaysia reported that around 20% of students had experienced sexual intercourse. Premarital sexual intercourse is no longer a taboo these days. Many people are considering it normal and thus indulging. Sexual activity during young adult period is a normal physiological phenomenon and instead of making it a thing of taboo, people should be taught about the safe sexual practices by imparting them with sex education and providing various contraceptive methods otherwise it will lead to spread of STIs. Due to the fact that the data was self-reported by the students, the figure in the current study may be an underestimate of their true promiscuous behaviors. 13 In the current study 82.1% of participants agreed that in every sexual encounter, a condom should be used consistently and correctly. A study conducted in a semi urban area of Ethiopia showed that only 36.3% used condoms as a contraceptive. Regarding the use of contraceptives, participants in the current study expressed a liberal attitude. 14 The finding may be understood in light of recent major AIDS prevention campaigns that promote the use of contraceptives. The campaign is primarily conducted through mass media, to which everyone is frequently exposed. 15 In this current study only 20% participants were involved in premarital sex. A study among college students of Mumbai found that 66% respondents had premarital sex. The reasons for involvement in premarital sex included relying on postsexual encounter emergency contraceptive pills, having no fear of pregnancy because "we had sex only once," "only people with HIV should use condoms," lack of access to contraceptives during spur-of-the-moment acts, and embarrassment over buying them from the store. 16 Having multiple sexual partners was favoured by only 19.4% of respondents which are less as compared to a study done in Maharashtra, where 44% respondents agreed to have multiple sexual partner. 17 Even though college students are well aware of the risk of having sexual activity resulting in STIs, they continue to engage in it due to influence of their peer groups of having multiple sexual partners, further this can also be attributed to background prevalence of sexual perversions in the society mainly influenced by the social media which is accepting having multiple sexual partners of either sex. 18 Only 25.4% of respondents in the current study agreed that culture does not restrict them from sexual activity. 19 In present study 81.8% favoured imparting sex education to students whereas in a similar study in Ambala (Haryana), 93.5% were in favour of sex education. 20 For youths, culture could have a double-edged effect by inhibiting their ability to discuss STDs openly with their parents and teachers. It is generally accepted that culture protects youth from STDs, but this is not true. The message for protecting against STDs can be introduced in the context of our religion and culture, particularly in light of our concern over the desire to teach STDs in schools and colleges. 21 But the study has few limitations. Because this was a sensitive issue study, some respondents might have answered some of the sensitive questions wrong leading to social desirability bias. Some participants only responded partially to the questionnaire. --- CONCLUSION The study findings show mixed responses in assessing youth knowledge level on various facts related to STIs. Majority of the students had correct knowledge and awareness on STI transmission routes, STI symptoms, and STI complications; correct and consistent use of condoms as a preventive measure. A favourable attitude toward STI on some of the statements, such as one should go for a STI test; condom should be consistently and correctly used in all sexual intercourse; and sex education should be introduced at school level' were found in the current study. On the contrary, some respondents expressed oppositional attitudes toward STI in response to claims like banning prostitution can control the spread of STI and women are more responsible for prostitution than men. Furthermore, a sizable proportion of respondents had a free and relaxed attitude toward premarital, multiple-partnership, and extramarital sexual affairs. Many young people appeared to have engaged in risky sexual behaviours, including having their first sexual experience at a young age, having multiple partners and premarital affairs, having affairs with the same sex, using violence during sex, and referring to alcohol consumption before having sex. --- Recommendations Government should provide free services and encourage younger generation for sexual health screenings by setting up regular on-campus or reachable STI screening clinics. Higher education institutions should strengthen their current health education and awareness programmes by including sex education in their curriculum. In view of current widespread availability of the Internet, this may be used as a tool to increase coverage on STIs and related topics in the form of text messages as well as transparency in the content available on websites. Funding: No funding sources --- Conflict of interest: None declared Ethical approval: The study was approved by the Institutional Ethics Committee

million new infections with one of four STIs: chlamydia (129 million), gonorrhoea (82 million), syphilis and trichomonas (156 million). 1 In India according to the National Family Health Survey (NFHS-5), it was also noted that the STIs among the age group 15-24 years was 5.6% in males and 2.0% in females. 2 STIs are caused by microrganisms like bacteria, viruses and parasites that pass from one person to another by sexual contact (vaginal, oral, and anal), infected needles, and skin to skin contact and ABSTRACT Background: Sexually Transmitted Infections are a major public health concern. Young adults have risk-taking behavior and casual attitude, engaging themselves in risky sexual activities thus contracting STIs. This study aimed at assessing knowledge, attitude, and practices regarding sexually transmitted infections among students of Panjab University, Chandigarh. Methods: A cross-sectional study was conducted among 252 students of Panjab University, Chandigarh, selected here through simple random sampling. A self-administered, semi structured and validated questionnaire was used for collecting data. The data was analysed using SPSS and presented as percentages, mean, standard deviation and ANOVA.The current study had 125 (49.5%) males and 127 (50.4%) female students. 228 (90.5%) had correct knowledge that multiple partners can increase risk of STI. Around 171 (67.9%) responded that it is possible to get STI after having sex first time while 151 (59.9%) responded that STI can cause long term health problems. 217 (86.1%) of respondents agreed with statement "one should go for STI test". While 96 (38.1%) respondents believed that prohibiting prostitution would control spread of STI. Majority of respondents (81.8%) opined for introducing sex education at schools. 65.9% of respondents were reluctant to disclose about multiple sexual partners. 179(71%) respondents felt comfortable discussing STIs with family members, friends, and co-workers. Conclusions: The current study found the need of introducing sexual health education among university students to fill gaps in existing knowledge which can further reduce the risk of STIs among them and ultimately help in achieving the targets of Sustainable Development Goals.

Introduction Adolescents and ethnic subgroups have been identified as having high risks of overweight and its associated complications [1][2][3], and overweight at young ages has strongly predicted overweight and its associated morbidity and mortality during adulthood [4,5]. Dramatic changes occur in health-related behaviors during adolescence [6], including changes in eating habits, and decisions about, and the length of time, spent in sedentary or physical activities [7]. Such unhealthy habits developed during adolescence have been found to persist into later life stages [8]. Long periods of sedentary behaviors and short periods of physical activity combined with consumption of high-calorie foods play a major part in weight gain during adolescence [9]. Among the various sedentary activities with limited energy expenditure, such as computer use, reading, playing video games, and driving a car, extensive television watching has been widely reported as a risk factor for overweight and obesity [10,11]. Usage of computers and mobile phones has further increased these days as most of the schools have adopted online teaching methods amid the COVID-19 pandemic [12]. Prolonged use of computers and mobile phones have been shown to cause neck and back pain and altered body posture including forward head posture across all age groups [13,14]. Socioeconomic status and ethnicity have been associated with each other [15]. Obesity and its associated complications might be relatively prevalent in some ethnic groups because of their cultural characteristics expressed in everyday lifestyles and habits [16][17][18]. Various previous studies from the UK, US, and Europe have indicated that behavioral differences related to ethnicity might lead to overweight during adolescence [6,19,20] and studies of adolescents from various ethnic backgrounds have found that some sedentary and physical activities associated with unhealthy diet patterns differently influenced body mass index (BMI) [21][22][23]. The results of these studies have been used to devise preventive strategies to prevent adolescent obesity [24,25]. Saudi Arabia is a rapidly developing country recently undergoing important demographic and public health changes [26,27]. Lifestyle modernization is decreasing the average amount of the average Saudi's physical activity [28,29]. Some research has found that, on average, about 34% of Saudi university students are overweight [30,31]. Although things are changing now, Saudi culture was considered to be relatively closed till recently. Females were not allowed to drive a car and all the girls had to wear a special dress before going out of their homes. In some places, they were not even allowed to go out of their homes without an accompanying male guardian. Such rules are not prevalent in the cultures in other countries, especially from where expatriates come from. These rules can have both negative and positive effects, specifically among adolescents. The kingdom's development and growth is attracting large number of skilled and unskilled workers from around the world, most of whom immigrate with their families. According to the Central Department of Statistics and Information, in 2014, about 33% (about 10.1 million people) of the population was immigrants [32]. Although some studies have investigated overweight, obesity, nutritional status, physical activity, and associated factors among Saudi students, to the best of our knowledge no studies have examined these characteristics among non-Saudi students or compared non-Saudi to Saudi adolescent students. Consequently, this study compared a sample of Saudi to non-Saudi adolescent students regarding the amount of time spent on physical activities, watching television, computer use, and their food preferences. The study's main objective was to compare differences between Saudi and non-Saudi adolescent students regarding time spent watching television, using computers, engaging in physical activity, and their food preferences. The relationships between these lifestyle behaviors and body mass index by Saudi nativity and gender were tested. Our results are intended as a foundation for devising preventive strategies for adolescent obesity. --- Materials and Methods --- Sampling and Data Collection Children aged 12 through 18 years old attending various schools in Riyadh were invited to participate in this study. A self-administered questionnaire was used to collect the data. The study's purposes and goals were explained to the adolescents and their parents and their written consent was obtained before the survey was administered. They were assured of the confidentiality of all the information they provided for the study. Ques-tionnaires with missing data and respondents with any type of disability were dropped from the analysis. The institutional review board approved all procedures performed in accordance with the Helsinki Declaration for ethical standards of human research. --- Variables and Measurement The questionnaire included the questions about demographic and anthropometric characteristics, daily after-school routine, physical activity, watching television, using computers, and food preferences [21,26,[33][34][35]. Their weight (kg) and height (cm) were measured to the nearest 0.1 kg using a digital metric scale, and to the nearest 5 mm using a wall-mounted height board, respectively, in the school medical room. BMI was subsequently calculated as weight divided by height square (kg/m 2 ) [36][37][38][39]. Waist circumference (cm) was measured as the minimum circumference between the hip bone and the rib cage [40]. Respondents were asked to assess the average time per day they spent while watching television or using computers. Television viewing (h/day) time included time spent watching television, videos, or DVDs. Computer use (h/day) time was defined as the time spent while using a computer, laptop, or playing video games. They were also asked to report the number of times per week (times/week) during which they participated in any physical activity enough to cause sweating. Section on food preferences included questions about breakfast, daily milk consumption, taking home made lunch to school, and frequency (times/week) of consuming high-fat fast foods and high-sugar soft drinks. --- Analysis Graph-Pad Instat 3.0 (GraphPad Software, San Diego, CA, USA) was used to analyze the data. Non-parametric (Mann-Whitney U) tests were used to assess the statistical differences in BMI between Saudi and non-Saudi respondents in relation to various parameters including physical activity, watching television, computer use, and food preferences. Differences were considered statistically significant at p <unk> 0.05. --- Results Of the 300 students who agreed to participate in the study, 250 (83%) returned the questionnaires. Of them, 36 respondents did not meet the inclusion criteria, and the final sample comprised 214 (71%). Table 1 shows the respondent's personal characteristics by gender and nativity. The mean age of respondents was 14.41 years (SD = 2.36). About 46% of the sample was Saudi. The non-Saudi respondents were from Australia, India, Pakistan, Afghanistan, South Africa, or Egypt. At the time of the survey, 35% of the sample was in secondary school and the rest were at the pre-university level. About 89% (n = 190) of the respondents reported that at least one parent suffered from diabetes, heart disease, or another health problem; 88% (n = 189) reported that they never smoked cigarettes, and about 83% (n = 177) considered themselves fit and healthy compared to their friends. --- BMI and Lifestyle Differences The mean BMIs of the Saudi and non-Saudi respondents were 24.02 kg/m 2 (SD = 1.34) and 24.21 (SD = 4.34), respectively, which was not significantly different. However, the difference between Saudi (24.51) and non-Saudi (23.74) boys was statistically significant p <unk> 0.05 (Table 2). Overall 8% of the 214 respondents reported that their favorite after-school activity was playing outdoor games, such as football. The other respondents reported that their favorite activities were using the internet or playing computer games. For their transportation to and from school, 67, 26, and 7% used a car or bus, walked, or cycled, respectively. Tables 3 and4 report the lifestyle differences and their relationships to mean BMI between the Saudi and non-Saudi boys and girls, respectively. --- Watching Television About 87% of the Saudi boys and about 69% of the non-Saudis boys reported watching television more than two hours per day. The difference in mean BMI between the Saudi and non-Saudi boys who watched the most television per day was statistically significant (Saudi = 24.57 vs. non-Saudi = 23.67, p <unk> 0.05). High proportions of the girls also reported watching television more than two hours per day, (Saudi = 76% and non-Saudi = 63%), but the association between watching television and mean BMI was not significantly different for the girls. --- Computer Use Amongst all of the respondents, 94% reported to ownership of a personal laptop or a desktop computer at home, and 85% of the overall sample reported computer use of more than two hours per day. Among the boys, 98%of the Saudis and 80% of the non-Saudis and, among the girls, 89% of the Saudis and 74% of the non-Saudis reported computer use of more than two hours per day. Among the boys with more than two hours per day of computer use, the non-Saudis had significantly lower mean BMI (Saudi = 24.52 vs. non-Saudi = 23.61, p <unk> 0.05). --- Physical Activity Among the boys, almost all of the Saudis (98%) and most (80%) of the non-Saudis reported physical activity (enough to cause sweating) two to five times per week. Similarly, among the girls, 98% of the Saudis and 73% of the non-Saudis reported two to five days of physical activity per week. There was a significant difference in BMI among the boys, but not among the girls, at this activity level (Saudi = 24.52 vs. non-Saudi = 23.70, p <unk> 0.05). --- Food Preferences Overall, the majority of the respondents 81% reported eating breakfast daily, 36% reported not drinking milk every day, and about 45%of the respondents did not take a lunch to school and they either skipped lunch or ate purchased food. --- High-Fat Fast Foods and High-Sugar Soft Drinks Just 3% (n = 7) of the respondents reported that they did not like to eat high-fat fast foods and preferred to eat at home. Similarly, just 6% (n = 12) of the respondents reported not liking high-sugar soft drinks. Among the Saudi boys, those who consumed high-fat fast foods were overwhelmingly likely to do so more than three times per week (97%). On the other hand, 28% of the non-Saudi boys ate these foods once, twice, or three times per week and 67% of them ate them more than three times per week. Comparing the boys who ate high-fat fast foods the most often, the non-Saudi boys had a significantly lower mean BMI (Saudi = 24.47 vs. non-Saudi = 23.73, p <unk> 0.05). The consumption patterns of the Saudi and non-Saudi girls were similar, but the Saudi girls were less likely than their male counterparts to consume these foods more than three times per week (Saudi girls = 79% vs. Saudi boys = 97%), and the difference in mean BMI between the Saudi and non-Saudi girls was not statistically significant at any consumption level. Regarding high-sugar soft drinks, 83% of the Saudi and 67% of the non-Saudi boys consumed them most frequently (more than three times per week), and their mean BMIs were different (Saudi = 24.49 vs. non-Saudi = 23.95), p <unk> 0.05). The girls also were most likely to consume soft drinks at the most frequent level (Saudi = 83% and non-Saudi = 62%) but the difference in mean BMI among those girls (Saudi = 23.88 vs. non-Saudi = 25.34) was not significantly different. --- Discussion This study compared Saudi to non-Saudi adolescent students regarding watching television, computer use, physical activity, and food preferences. To the best of our knowledge, only a few studies have also examined cross-cultural differences between Saudi and non-Saudi adolescent students. Saudi boys who reported the highest levels of television and computer use, moderate physical activity, and the highest frequency of fast food and soft drink consumption had significantly higher mean BMIs than the non-Saudi boys at those levels. A further comparison of the Saudi and non-Saudi respondents regarding sedentary and physical activities and food preferences found that mean BMI was partly influenced by ethnic differences. Ethnic differences in body composition mediated by energy-balance related behaviors (intake and expenditure) were found in studies set in other countries, such as Germany and the Netherlands [8,41]. Increased time spent on computers and mobile phones due to online teaching in schools or work from home, decreased physical activity due to precautionary lockdowns, and unchecked eating patterns while spending more time in sedentary activities in homes has become the COVID-19 pandemic lifestyle in all the age groups. Although this study was conducted before the beginning of theCOVID-19 pandemic, its objective has strong relevance in the current situation. A similar study should be replicated on a large scale to study the effect of the COVID-19 pandemic lifestyle on our lives in all the age groups. One of the previous studies has shown that body composition indices and sitting time associated with media use were higher among Saudi boys and that expatriate boys and girls were physically more active compared to Saudis of the same age group [36]. Our results also show that Saudi boys had a higher mean BMI than non-Saudi boys that might be explained by the longer time period spent watching television, using computers, and their less frequent physical activity. For example, none of the Saudi boys watched television for less than one hour per week as compared to 14% of the non-Saudi boys, and none of them were physically active more than five days per week compared to 7% of the non-Saudi boys. The Saudi boys also reported more frequent consumption of high-fat fast foods and high-sugar soft drinks than the non-Saudi boys. Previous studies reported that the amount of time spent watching television was related to overweight and obesity [26,42,43], and the amount of time spent in sedentary activities negatively influenced body composition more than medium or high intensity physical activity positively influenced it [44]. This is more important in the current situation when physical activities of students have decreased due to precautionary lockdowns while schools have started online teaching that has considerably increased sedentary activities, including time spent on computers and mobile phones. Cycling or walking to or from school was previously found to significantly influence BMI [9,45,46]. The majority of the respondents in the current study reported traveling to school either by car or school bus. Longer travel time to and from school using these means necessitates sitting for periods found to be an important influence on student's nutritional status [47] and this might be a reason for the high prevalence of overweight Saudi students. Differences in BMI between ethnic groups with similar daily activities and food preferences might be due to interactions between environmental and social factors that affect individuals' decisions about physical activity and food preferences [11,48]. Saudi Arabia's extreme dry climate is another factor that might reduce students' physical activities. It has been reported that students should be encouraged to increase their physical activity and eat a healthy balanced diet that includes Vitamin E to improve academic performance and executive function [49]. Optimal daily physical activity has also been regarded as a determining factor for academic performance [37]. The vast majority of our respondents (94%) reported owning a personal laptop or desktop computer at home. Personal television sets, computers, laptops, tablets, and so on in children's bedrooms are key reasons for their long periods of screen time. Watching television might indirectly increase food consumption while reducing physical activity and energy expenditures [33,50] and watching television while eating has been identified as a risk factor for obesity [51]. Television advertisements for fast foods also have been found to influence eating behaviors, particularly among children [22,52]. Some studies have indicated that the amount of time spent engaged in sedentary activities, such as watching television, during childhood and adolescence predicts overweight and obesity in adulthood [44,53,54]. This link was identified in a review prepared for the World Health Organization [55,56], and, based on those studies, decreasing children's television time was recommended [23]. It is important to note that the influences of television time on health might differ across cultures and by the nature of the viewed content [22]. High consumption levels of sweetened drinks and snack foods have been associated with weight gain among adults and children [57,58]. However, because genetic metabolic rates might be ethnicity specific, individuals with similar physical activities and diet preferences might have different risks of weight gain. The timing and frequency of meals also might influence that risk [59]. Imbalances between proportions of artificially sweetened drinks and fatty fast foods on the one hand and fresh fruits, vegetables, and dairy products on the other hand influence children's health [60]. Individuals' eating habits also are influenced by social and lifestyle factors [22]. Our results supported the findings of a previous study conducted in Saudi Arabia that found 30% of the boys and 21% of the girls ate fast foods [30] and another study reported a high rate of energy drink consumption (56% of the boys and 35% of the girls) [61]. A 260 mL soft drink provides about 188 kcal of energy [62], which, if not counteracted by physical activity, is stored as fat and results in weight gain [63]. Although no direct links have been established between the composition of consumed food and weight gain among adolescents, eating late at night, omitting breakfast, and frequent snacking between meals have been found to variously influence individuals [64]. In one study, the majority of the respondents (81%) reported not omitting breakfast because breakfast is the most important meal of the day [65]. Although eating breakfast increases the total number of daily calories consumed, the likelihood of weight gain is less among those who eat breakfast [66,67] and their cognitive and memory functions are better [68,69]. In our study, about 36% of the respondents reported not drinking milk daily, and 45% of the respondents did not take a lunch to school. The importance of milk, fresh fruit, and home-cooked food has been widely reported [63,70]. Thus, parents should insist that their children at least one meal each day with the family and without television. When parents eat with their children, the children tend to have healthier diets [71]. People have adopted various bad habits related to eating patterns such as eating late at night, omitting breakfast, frequent snacking between meals, etc. during the current pandemic situation. This, associated with decreased physical activities and increased sedentary activities, would affect our bodies badly. This needs to be studied further in ordered to prevent associated consequences. Interestingly, there were no significant differences in mean BMI between the Saudi and non-Saudi girls within the lifestyle categories of the variables. The pattern of differences between the Saudi and non-Saudi girls was very similar to the pattern of differences between the two groups of boys. Compared to the non-Saudi girls, the Saudi girls were relatively likely to watch the most number of hours of television and spend to most number of hours on a computer per day (>2), to have low levels of physical activity (<unk>2 times per week), and to consume fast foods and soft drinks more than three times per week. The finding is not consistent with previous studies that found direct relationships between preferences for high-fat and high-sugar content foods and overweight among girls [72]. In our study, the majority of the non-Saudi respondents were from Egypt. Some studies have reported similar lifestyles and food preferences among Gulf Cooperation Council and their neighboring countries [73,74], which might partly explain our non-significant results for the girls. Since the studied age group is more active on various social media platforms, promotion of healthy lifestyle including food habits and benefits of physical activity can help prevent associated risk and complications. We also recommended that the political authorities should implement various measures and protocols to alleviate the potential risk related to unhealthy lifestyles. Age difference between 12 to 18 years is too large. We propose that future studies should divide the age by groups and be conducted on a larger sample size to get better clarification about the variables. Various studies have associated the social, economical, and educational levels of parents, especially mothers, in the health status among children. We propose that this should be considered in similar future studies. --- Limitations The current study did not include equal representation of Saudi and non-Saudi children. It used a self-report questionnaire and the respondents might have exaggerated or under-reported their activities. Girls in adolescence often use various weight control diets, which could have affected the results of this study. --- Conclusions High levels of sedentary and low levels of physical activities as well as high consumption of high-fat fast foods and high-sugar drinks threaten the health of Saudi adolescents. Cultural differences in lifestyle between Saudi and non-Saudi families should be considered when developing programs to improve knowledge, attitudes, and behaviors regarding diet quality and physical activity. The objective of this study is more important in the current situation where increased time spent on computers and mobile phones due to online teaching in schools or working from home, decreased physical activity due to precautionary lockdowns, and unchecked eating patterns while spending more time in sedentary activities in homes has become our COVID-19 pandemic lifestyle in all the age groups. A similar study should be replicated on a large scale to study the effect of this lifestyle on our lives in all the age groups. --- Data Availability Statement: The datasets used in this study are available from the corresponding author on request. --- Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Sub-committee of Rehabilitation Research Chair (file ID RRC-2017-013, date of approval 12 December 2017). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study as well as their parents. --- Conflicts of Interest: The authors declare no conflict of interest.

among Adolescents in Saudi Arabia.

Introduction Israel is not isolated from the global migration process, and it is required to provide a medical, educational, and socio-cultural response to the integration of tens of thousands of African asylum seekers. A very significant proportion of them reside illegally in the country, in difficult living conditions on the social margins, while suffering from general, occupational, and family instability and experiencing a sense of alienation and social nonbelonging. Asylum seekers experience hostility and racism in the public space from an early age, including comments on skin color, cursing, spitting, insults, and abusive graffiti on walls in residential neighborhoods and schools [1][2][3]. According to the Population and Immigration Authority, there are currently about 50,000 African asylum seekers living in Israel, most of them from Eritrea and Sudan. About 8000 children (most of them born in Israel) study throughout the country, of whom about 2300 study in separate schools in south Tel Aviv. The Israeli government pursues a policy of exclusion and does not take responsibility for asylum seekers. Moreover, in recent years the Israeli government has made various attempts to reduce the number of asylum seekers and make their lives difficult: relocating them to open detention facilities, providing financial assistance for voluntary return, resettlement in third countries, and more [4,5]. In such a reality, municipal entities have had to take on the direct care of the new residents, allocating funds for this purpose from the municipal budget. They receive most of the services from third sector organizations and community organizations. In addition, the presence of asylum seekers in the southern neighborhoods of Tel Aviv has changed their social fabric and even created conflicts with the local population [6]. In Israel, according to the 'Compulsory Education Law', education is compulsory and free for all children from the age of 3 to 17, regardless of their citizenship. Refugee children have access to the primary and secondary education system under the same conditions as Israeli citizens. However, for most of them the situation is quite different: They are perceived as strangers and are located at the bottom of the social ladder [7,8]. As such, they are one of the weakest groups in Israel, excluded from many services and living in economic distress. Many children suffer from significant gaps and delays in learning, and the schools' ability to provide a comprehensive and effective response is limited. The result is that children do not receive academic support and are unable to complete high school successfully. This fact is a blow to the young refugees' dreams of a better future [9][10][11][12]. It is agreed that the subjects of this study-educators in schools that integrate children from families of asylum seekers-are among the significant factors for the children. These educators develop on their own initiative culturally adapted teaching practices for integration, adjustment, inclusion, and reduction in gaps. They act out of respect and sensitivity to the culture and tradition of the country of origin and play the role of intercultural mediators who promote socialization processes in order to facilitate coping with a lack of information, bureaucratic barriers, and educational conflicts [13][14][15][16]. Studies [17][18][19] indicate that educators' mediation actions may establish health and educational knowledge among parents and children and help foster a continuous and supportive trust-based communication system. Moreover, in order to foster effective educational work, to understand that all the teaching and interactions with students take place in a cultural context that is neither neutral or accidental, and to maintain open, positive, and lasting communication, educators are required to possess intercultural competences that include knowledge, skills, and understanding of diverse cultural frameworks [20,21]. From the global aspect, in view of the demographic changes and globalization processes, and out of a desire to enrich the theoretical and practical knowledge regarding the integration of African asylum seekers in Israeli sociocultural spaces, I chose to focus on health and education mediation processes occurring between African asylum seeker parents and their children's educators-educators who have a responsibility to develop a climate of integration between new and old groups and majority and minority groups and to encourage their involvement in the public space. This article examines the perceptions and actions of educators in Tel Aviv's elementary schools and high schools, who act as mediators for asylum seekers in Israel. The research question is as follows: What are the health and educational mediation actions of educators among asylum-seeking families? At the beginning of the article, I review the community of asylum seekers in Israel from two aspects: medical and educational and the role of educators as mediators. I then present the research method and analysis of the findings regarding the mediation actions relating to health and educational issues. At the end of the article, a summary and conclusions are presented. --- The Medical Aspect According to the law regulating the provision of health services in Israel, most asylum seekers are not entitled to health services because they do not have a civil status that entitles them to resident rights. They are only eligible to receive emergency medical care in hospitals. In Tel Aviv, there are the following: (1) a designated clinic for the statusless, (2) a Levinsky clinic for sexually transmitted diseases and AIDS, and (3) a refugee clinic operated by volunteers [22]. In addition, minors from families of asylum seekers and refugees born in Israel and whose parents stayed in Israel for six months are entitled to subsidized medical insurance from the Ministry of Health. However, due to financial difficulties, some parents cannot insure their children, and they are left without health insurance. Data from the Ministry of Health show that more than 20% of refugee minors lack health insurance [23]. The coronavirus pandemic has exacerbated the situation of asylum seekers in Israel, a weakened and excluded population without any social and health support network, which is at greater risk of infection. The isolation and elimination of jobs meant that 80% of the asylum seekers found themselves in a severe crisis, without work, a livelihood, health insurance, entitlement to unemployment benefits, and severance pay. They find it very difficult to recover from the effects of the coronavirus crisis and return to their daily routine, which was difficult and complex to begin with [24]. --- The Educational Aspect The Government of Israel has determined that following the Compulsory Education Law, children from families of asylum seekers and refugees without status residing in Israel will study in the public education system from age three [4,25,26]. However, children are not entitled to pedagogical and social support such as allocating school hours to improve the acquisition of the language of instruction, didactic diagnoses, or participation in programs to treat complex learning disabilities, language difficulties, or emotional problems [27]. In this complex reality, the challenge facing educators working in the intercultural arena with children from African asylum-seeking families is to understand the sociocultural differences and find effective and appropriate educational solutions for parents and children. This requires developing intercultural competence among educators as a central tool for creating cooperation between them and the parents while establishing ways of communication and trust [20,28]. Educators with intercultural competence are able to develop a climate of integration between cultures within the educational framework, fulfill the role of mediators to promote new medical, educational, and cultural knowledge, find appropriate solutions to their diverse problems, and prevent discrimination of asylum seekers. They can also make it easier for parents and children to deal with barriers and conflicts [29][30][31]. Studies in education indicate [7,10,32] that educators play an essential and significant role concerning children from families of asylum seekers. The children describe a positive relationship of trust between them and the teachers. According to them, teachers are a significant factor influencing their motivation, self-confidence, and academic success. Educators in multicultural and multilingual settings who teach children from migrant, refugee, and asylum-seeker families should have intercultural competence and plan classroom lessons based on principles of culture-relevant pedagogy. Educators also play the role of language and culture mediators and find themselves explaining educational and health issues to children and parents. Their main role is to cultivate ways of communicating and mediating messages that are not clear to asylum seekers [28,29].. Researchers in education and migration [33][34][35] show that personal relationships, empathy, personal exposure, and listening to personal issues help children and parents deal with social tensions and promote a positive attitude toward the general society. According to Berhanu [36] and Sisneros [37], intercultural mediation encourages members of different cultural groups to strengthen their "belonging" and identification with the dominant group. It does not mean abandoning the culture of the country of origin in favor of the dominant culture; rather, it means facilitating the process of forming a multicultural society that gives a sense of space and allows the various groups within it to adapt to the system of values, norms, and behaviors of the majority. The purpose of this study is to examine and describe the actions of educators in an elementary school to mediate health and educational issues for African asylum seekers in Israel. --- Methods This qualitative-phenomenological study emphasizes the subjective interpretation that people give to the socio-cultural reality they find themselves in and draws its data from the natural array. The qualitative paradigm opens up new possibilities for observing human behavior, understanding it, and creating knowledge. The role of the researcher is to investigate a phenomenon, find its meaning, and interpret it [38]. The phenomenological approach deals with the study of the essence of human experience. The phenomenon is the object of inquiry, and the researcher strives to explore the meaning of the experience for the subjects in their daily lives. Phenomenological research deals with individuals' experiences and how they perceive them. Researchers in this approach do not take reality for granted but examine the nature of things according to the contexts and circumstances in which they observe. A phenomenon perceived from one point of view will not be the same as that perceived from another point of view. Therefore, the essence of phenomenological research is in creating theoretical generalizations [39,40]. Moreover, this approach makes it possible to collect data from educators as a primary source and learn about their perceptions, actions, feelings, unique experiences, and the interactions in which they are involved. As a result, research findings may assist in constructing educational programs that promote the integration of marginalized minority populations in a multilingual and multicultural society [41]. Fifteen educators participated in the study: three men and twelve women. They all teach in multicultural and multilingual primary schools, where children from immigrant families, refugees, and asylum seekers comprise the majority of the student body. The institutional vision of these schools incorporates a pluralistic conception rooted in the values of equal rights and tolerance, respect for other languages, and acceptance of difference without preconditions. All educators are aged 25-48, with five years or more of teaching experience. They all revealed an educational concept that emphasizes the importance of personal contact with children and their parents, despite language difficulties and cultural gaps. One of the educators taught for three years in the United States in a culturally, religiously, and nationally diverse school. He gained experience teaching a foreign language and contacting parents from cultural and national minority groups. The study was conducted in the southern neighborhoods of Tel Aviv, where most asylum seekers live, alongside the migrant population. The presence of African asylum seekers has changed the urban space. As part of the change, for example, they are seen in various places in the city, riding bicycles, cleaning streets, walking by the sea, gardening, and being employed in restaurants. Their presence is also sometimes revealed in protest measures such as demonstrations, handing out leaflets, or strikes. Surveys [42][43][44] show that veteran Israeli residents of Tel Aviv feel an economic, social, and national threat from asylum seekers, even from those perceived as "real refugees." Moreover, the majority of the Israeli public fears that asylum seekers will increase the level of crime in society and become a burden on welfare systems, and among many there is growing support for tightening immigration policy and reducing social rights [45,46]. Due to fears of the development of social exclusion, which could negatively affect the conduct of the city of Tel Aviv on various levels, the municipality established in 1999 an assistance and information center for the community of foreigners called "Mesila", serving migrant workers, asylum seekers, and refugees. All activities are funded by the municipality and donations from private individuals. The assistance center provides information services in all areas of life as well as a response to primary needs such as food, clothing, etc. in situations of stress [47].. The study data were collected from personal interviews with 15 participants. Each interview lasted about an hour and was conducted at the participants' workplace. The topics discussed in these interviews revolved around the actions of educators to mediate health and educational issues for African asylum seekers. The data were analyzed through content analysis that focused on what participants said rather than how things were said. According to Shkedi [48], content analysis is a window into the inner experience of the interviewee. The analysis focuses on the words and descriptions of the educators as reflecting their actions, perceptions, feelings, beliefs, and knowledge. Moreover, Krippen-dorff [49] claims that content analysis allows the description of data and the drawing of valid conclusions for their broad context. I strictly followed the necessary ethical rules: maintaining the anonymity and confidentiality of the respondents and the data, avoiding abusive questions, and giving educators a choice whether to participate in the study or not. --- Results Examining the actions of primary school educators in mediating health and educational issues for African asylum seekers revealed four content categories of mediation: (1) a healthy lifestyle: nutrition and hygiene; (2) emotional-behavioral; (3) learning disabilities and special needs; (4) diseases, vaccines, and medical treatments. --- A Healthy Lifestyle: Nutrition and Hygiene Here is an example that illustrates the educators' actions as mediators of healthy lifestyle issues through a dialogue with parents that includes explanations of the importance of a daily shower, brushing teeth, and trimming nails: "On the subject of cleanliness there is neglect: they do not wash their hair, nor do they trim their nails. I think we need to comment, explain to children and parents. If you explain, you can see the change. I try to explain to parents in simple words, in pictures, in body gestures, as they do not speak Hebrew, how important it is for the children to shower and brush their teeth. We even did a "body cleaning" day and gave them toothbrushes and toothpaste. Sometimes the kids come a few days in a row with the same clothes. They have a shower at home, but the parents arrive late, the children are already asleep, without showering, and get up in the morning for school. Parents do not really have time for children; they work crazy hours, so I also explain to children that it is important to shower even if mom and dad are not home (M)." M. seems to be aware of the gaps in norms of behavior between what is acceptable in the child's home and what is acceptable and desirable in Israeli society, the receiving society. It is important for her that the parents understand the message, so she tries to be creative and mediates the message through pictures, simple words, and body movements. Her words "if you explain you can see the change" prove that she believes that if she makes the parents listen, understand, and believe her, change can occur. M. also tries to explain the fact that the children do not shower every day, claiming that the parents work until late, are busy with existential troubles, and do not always have the time, strength, or patience to take care of the children's hygiene as well. B. describes an ongoing dialogue with the parents on the issue of eating sweets. Out of concern for children's nutrition, she tries to encourage children and parents to change their eating habits: "From morning and throughout the day they eat sweets, gummy candies, lollipops. The parents buy them sweets because they think it calms the children. When the children cry, get angry, argue, the parents think that candy is a solution. I talked to the parents, I explained to them, but nothing helps, they continue to buy sweets. I'm not sure their eating habits can be changed. It can be difficult to convince them to eat more fruits and vegetables because their price is very high (B)." B. tries to influence the issue of sweets but is unsuccessful. Parents seem to be aware of the harm to their children's health but are unable to reduce their consumption of sweets, perhaps because of their cheap price compared to the high price of fruits and vegetables, or perhaps because they have difficulty setting boundaries for children, explaining, and educating out of parental authority, thus choosing sweets as an easy solution. Perhaps the parents know how difficult their children's lives have been, and they are allowing them to eat sweets to try to make them happy. The healthy eating issue also came up in interviews with other educators. Here is another example, where A. shows concern about eating habits at home: "The school has delicious and varied sandwiches for breakfast. I at least know that here they sit at the table eating a sandwich and vegetables, because I do not know what is going on in the house, in terms of setting boundaries, hygiene, or how many sweets they eat (A)." This example also illustrates the teachers' concern about what is happening in the children's homes from various aspects: education, health, and nutrition. The children from the refugee families seem to benefit from a government-funded feeding program, which is intended, among other things, to improve the nutrition of children from vulnerable groups for whom healthy food is not available. From the point of view of the educator, there may be a gap in eating habits between what is acceptable at school and what is familiar at home. --- Emotional-Behavioral These two educators describe actions aimed at helping children and parents deal with social, emotional, and behavioral difficulties, with frustration and social processes accompanied by expressions of violence and racism: "I have been teaching at the school for seven years, and I do not believe there will be a change. I'm not optimistic. It should be understood that there is a very big difference between "real life" in the neighborhood and "protected life" in school. There is chaos, confrontation, and violence in the neighborhood. The Israeli residents curse the Africans, shout at them to return to where they came from, call them names such as "monkeys" "garbage", "blacks". There is a lot of frustration, verbal violence, and sometimes even beatings. The children live in daily risk. At school we make great efforts to enable them to live a normal life. I want to allow children to dream in a safe environment, to teach them to take responsibility, to strengthen their self-esteem, their motivation. The best way is through personal conversations. I make time for these meetings, and sometimes stay in the afternoon as well. Through the personal relationship I manage to touch them personally, to really listen to them. The kids know I'm here for them, they can trust me (L)." "I have an open communication channel with the parents, and yet their starting point is so unequal, and they live in such difficult conditions, the parents work 12-14 h a day, for them they try to do the maximum, but they cannot support the children. To empower parents emotionally and help them deal with frustration and helplessness, we set up a forum that meets once a month. The forum is attended by parents, representatives from the municipality, and teachers. We sit together, the parents raise issues that exist in the neighborhood, at school, and think together about a solution. The forum gives parents a sense that there is someone who cares, who listens to them, who tries to help (T)." Both educators, L. and T., act as mediators in a multicultural environment, caring for the children's mental health and trying to alleviate emotional frustration through personal dialogue with the children and with African asylum-seeking parents. The mediating discourse takes place not from a paternalistic position, superiority, and power of a dominant culture, but as a dialogue at "eye level", while establishing trust, closeness, and openness [29].. Researchers [33][34][35]50] have found that close relationships based on listening, respect for diverse beliefs, empathy, personal exposure, and concern can help address challenges and tensions while reducing the gap between the dominant society and the minority group. --- Learning Disabilities and Special Needs In the educators' reports, I found evidence of educators' actions aimed at finding appropriate solutions to learning disabilities, communication problems, developmental delays, and special needs of children, while mediating the knowledge to the parents. Pedagogical activities included individual learning, increasing parental involvement in the learning process, and informal diagnoses. "Children and parents do not receive support for learning from an official office such as the municipality or government-only volunteers. As far as the state is concerned, they are not recognized, they do not have an identity card, they are just a number. It all depends on the initiative and goodwill of the teachers. I take the children to individual meetings, teach them the new words, practice until they start reading in Hebrew. In case I find more complex difficulties, I try to diagnose the problem. Luckily, I also took special education courses. I have to work differentially in class (R)." The Israeli education system expects educators in classes with children from African asylum-seeker families to bear educational responsibility and to lead all students to meet the required achievements and standards. That is, educators are required to have a broadbased pedagogical knowledge base, good will, and the ability to adapt to the linguistic teaching-learning methods of the children in the classroom without official municipal or political support. Studies [51][52][53][54] show that classroom educators respond on a practical level to differences and take a concrete pedagogical approach while addressing teaching methods, cognitive abilities, and language adaptation for cultural minorities. One of the ways in which educators address learning difficulties and special needs is through personal-differential teaching that allows each child to learn regardless of their abilities. There is research evidence for the effectiveness of differential teaching in achieving learning goals, using methods such as individual work, peer teaching, and providing children with a choice of topics and teaching styles [55]. Moreover, the personal relationship between educator and child may increase children's self-confidence and selfefficacy and help deal with stress [55][56][57]. Gieras [58] also believes that personal connection, an emotionally supportive discourse, and positive reinforcements encourage the children and increase the understanding that their success depends on them and their investment. "Without parental involvement we will not be able to advance the children in schooland this is an almost impossible task. To increase the partnership, we have a group on WhatsApp, I upload pictures of the kids, write what we learned, what the tasks are. Some parents understand Hebrew a little better, and they translate and pass on the information to the other parents in Tigrinya (F)." F. reveals a pedagogical perception according to which cooperation with parents is of great importance, especially when it comes to parents from a minority group on the social margins, whose socio-behavioral norms are different from those accepted in the receiving educational setting. It seems that asylum-seeking parents lack the knowledge, skills, and social support to deal with the difficulties posed by the receiving society, and educators understand that in order to fulfill their role as mediators, they must strengthen cooperation with parents. For example, the encounter with the African asylum-seeker mothers was an opportunity for direct mediation that enriched the mothers with new knowledge about engaging in learning processes without using violence toward the children, even in situations of tension and argument. Studies examining the relationship between parental involvement and student achievement from minority groups [17,19,[59][60][61] show that there is a positive relationship between the two. It is therefore important that the educational setting develop and expand the range of activities that encourage parental involvement. --- Diseases, Vaccines, and Medical Treatments Educators reported personal conversations with parents when the children were sick, not feeling well, or injured. They explained to parents where to go and what to do. They mediate information about local clinics where they can receive treatment, even though they do not have health insurance: "I check the children if they have a fever, mycosis, skin sores, or other medical problems. If there is a problem, I call the parents via WhatsApp. They are in no hurry to come and pick up the child from school. The parents take the child to the doctor and call me because they do not understand what the doctor wrote. Sometimes I write a letter to a doctor, and sometimes I have to explain where the clinic is for someone who does not have health insurance. The next day the child comes to class because they do not understand the concept of "staying at home" (B)." The educator helps parents deal with medical issues while drawing their attention to such issues. She mediates information written by the doctor and accepted norms of behavior such as that a child stays home when he is sick, and also helps them to overcome language difficulties in the face of medical factors. The mediation reveals to the asylumseeker parents the way in which the Israeli health system is run and encourages them to take responsibility for the children's health, to take time to take care of the children, and to be with them at home until they recover. Mediation seems to reduce the gaps and help the asylum seekers to learn and understand that they can receive medical care even under living conditions in which they are not considered equal citizens in the receiving country. Reality forces H. to play a new role as a "health-educational mediator." Health information is not accessible to helpless parents, many of whom were left without health insurance during the coronavirus pandemic, after losing their jobs. After all, health insurance is activated only when asylum seekers work for an employer. Moreover, they are not entitled to state support benefits, and many of them suffered from food insecurity during the pandemic: "No one explains to them how the health care system works in Israel. What should they do? When? Who do you turn to in an emergency? Where is there a volunteer clinic? Do you have to buy medical insurance? I feel I am the source of knowledge, the "address" for them. When they have a problem-not necessarily related to education or school-they turn to me. During the corona period, it intensified. The educators were the source of information during the lockdowns, vaccinations, and the issuance of a green pass. Many asylum seekers have difficulty receiving green passes or immunization certificates because the government system does not have all of their details. The educators were a listening ear to them and helped them even when the schools were closed. The school administration also approached several organizations to provide them with food, baby products and more. Each educator examined the needs of her families (H)." --- Discussion The findings of the study revealed four content categories of mediation: (1) a healthy lifestyle; (2) emotional-behavioral; (3) learning disabilities and special needs; (4) diseases, vaccines, and medical treatments. The educators reported personal relationships and individual conversations with the parents in all four categories. They provided them with relevant health-educational information about options for receiving medical care even for those without insurance and explanations of accepted social norms and behaviors in the receiving society. The parents saw the educators as an "address" for every question and problem in all areas of life. It also shows that in the reality of the Israeli government's exclusion policy regarding the integration of African asylum seekers into society [4], educators are forced to take on roles traditionally entrusted to the state and become health and education mediators for African asylum seekers. Thus, educators have become agents of socialization, and their role takes on a new meaning as a personal-social factor mediating between parents and the Israeli health and education system, at their formal and informal levels; in doing so, the educators go beyond their formal roles. Educators find themselves in a space that deepens educational inequality while facing many difficulties resulting from children's lack of background knowledge and community characteristics and lack of professional guidance. Despite this, they act as mediators and express a conception of human pedagogy that follows cultural responsiveness [31,[62][63][64], i.e., responding out of respect for the other culture [65,66] and maintaining personal communication based on trust with the parents-listening, explaining, translating, advising, encouraging, and supporting. The personal relationships led parents to perceive the educators as "honest" mediators with intercultural competence that includes knowledge of the other culture, communication skills, the ability to resolve conflicts between groups, empowering social justice, reducing gaps, and dealing with stereotypes and prejudices [20,67]. Moreover, in the local Israeli context, the findings indicate that the mediation process helps to reduce tensions between the home system and the general social system outside it. In examining the global context, Gratton, Gutmann, and Skop [68] studied families and children in migration processes around the world and found that these tensions have an impact on family functioning, child integration in the education system, and family integration in the receiving society [69]. The findings also reveal the difficult situation of asylum seekers throughout the coronavirus pandemic, during which the medical mediation of the educators to the parents was more significant and important than ever. With the loss of livelihood, asylum seekers also lost their only way to receive medical care in the community. Additionally, during the pandemic, by virtue of the Patient Rights Act, they could only receive treatment in emergency rooms. The problems were many and complex: (1) the medical information was not accessible to asylum seekers; (2) there was a fear that asylum seekers would arrive in the emergency rooms contrary to the guidelines of the Ministry of Health; (3) asylum seekers suffering from serious or chronic illnesses-some of which were life-threatening-did not receive appropriate treatment; (4) the lack of medical insurance did not allow asylum seekers to be tested for the coronavirus [56]. In a reality where the employment crisis led to an economic and humanitarian crisis in all areas of life of the asylum-seeking community, and they were left without a state support network, educators played an important role as a source of knowledge in everything related to vaccines, the rules of conduct in lockdowns, what is allowed and what is forbidden and the possibility of receiving a "green pass". Moreover, the educators also acted to prevent a humanitarian catastrophe. They collaborated with volunteers and non-profit organizations, kept in touch with parents, found out what they were lacking, and took care to provide food, basic products, diapers, medicines, and more. These findings may have implications for the international arena, as the mediation activities described can be a working model for promoting and integrating other immigrant populations elsewhere in the world. The mediation activities position educators as a significant personal and ethical anchor for asylum-seeking parents and their children. These activities are important because they place the African asylum seekers at the center of the educational endeavor and as a top priority of the educational setting; they strive to address the challenges of integration between groups and give asylum seekers an opportunity to integrate, reduce gaps, and possibly even live independent lives. --- Conclusions The term "multiculturalism" carries sociological significance and refers to a situation in which different cultures exist within a single political framework. A multicultural approach is a model for educational action that includes recognition of cultural diversity, the right of individuals to preserve their cultural uniqueness while ensuring their full access to rights, resources, and opportunities in society, all with the aim of addressing social (in)equality and promoting integration and inclusion [70,71]. This study reveals the authentic voice of primary school educators: perceptions, experiences, and mediating actions in multicultural educational frameworks that incorporate children from African asylum-seeking families, who are different from the general society in race, religion, skin color, culture, ethnicity, and culture. This study reveals a dual reality: on the one hand, African asylum seekers face bureaucratic, social, economic, and social challenges. They experience alienation, exclusion, and violence that amplify the difficulties in the process of social integration. Additionally, it seems that many Israeli citizens do not show openness and tolerance towards them. Moreover, the coronavirus crisis has exposed government policy and the injustice it causes to asylum seekers, as the closure of the economy and the cessation of work propelled the population of asylum seekers to the brink of an economic and mental crisis as they face economic insecurity and inability to pay rent [24]. On the other hand, a positive point of view is revealed when parents see the educators as loyal partners, as a source of knowledge, as figures who can be consulted to receive help in routine and times of crisis. Educators revealed an educational perception in which they understand the other and his or her point of view, identify difficulties, and work to reduce information and knowledge gaps. They do not seek to change asylum seekers' lifestyles and do not show a preference for Israeli culture over the culture of their country of origin; rather, they enable the preservation of asylum seekers' tradition while deepening their familiarity with the education system, health system, and accepted norms of behavior in the receiving society. The most prominent pedagogical principles adopted by the educators in the mediation process were the establishment of personal relationships and the strengthening of cooperation. The personal connection between the educators and the parents grew through continuous communication and mutual trust that enabled effective functioning and cooperation, finding solutions in a good spirit, and even organizing properly for emergencies [72]. Mediation through personal relationships helped parents correctly interpret the Israeli educational, health, and cultural context, to receive health care even in an environment that often-revealed hostility, resistance, or lack of acceptance [25]. In addition, the mediation process was accompanied by mediation of language gaps and the understanding that each of the parties had different norms of behavior and values. According to Arzubiaga, Noguerón, and Sullivan [17], in order to overcome the difficulties, prior recognition of the worldview of each of the parties is required. Such prior acquaintance may prevent frustrations and disappointments. The findings of this study are inconsistent with other studies that examined the relationship between educators and refugee parents and asylum seekers [73][74][75], which found that most parents do not trust the educational staff and therefore avoid contact, personal or family exposure, and prefer to maintain limited contact with the educational establishment. It is possible that in the Israeli context, African asylum-seeker parents did not find a sympathetic ear in other urban or political sources, so that educators who expressed empathy, good will, and acceptance were perceived as trustworthy and reliable in the eyes of the parents. From a practical aspect, the findings of the study highlight the importance of expanding the direct and personal dialogue between educators and parents from asylum-seeking families or with parents from other minority groups. It is important and necessary that in the process of teacher training, students cultivate empathy, sensitivity to others, intercultural competence, and develop skills of socio-cultural mediation that deepens the acquaintance between the parties and avoids expressions of racism, exclusion, and power. Such training will raise educators who can deal with complex situations while addressing academic, emotional, and cultural differences related to the learners' abilities. --- Data Availability Statement: Not applicable. --- Conflicts of Interest: The author declares no conflict of interest.

Israel is not isolated from the global migration process. It is required to provide a medical, educational, and socio-cultural response to the integration of tens of thousands of African asylum seekers. This qualitative-phenomenological study collected data from 15 educators as a primary source and learned about their actions to mediate health and educational issues for African asylum seekers. The findings reveal four categories: (1) a healthy lifestyle; (2) emotional-behavioral; (3) learning disabilities and special needs; (4) diseases, vaccines, and medical treatments. It seems that educators are forced to take on roles traditionally entrusted to the state, and they have become agents of socialization who mediate between parents and the Israeli health and education system through personal relationships and individual conversations. This study reveals a dual reality: on the one hand, African asylum seekers experience alienation, exclusion, and violence; on the other hand, they gain a positive point of view when parents see the educators as loyal partners and sources of knowledge who can be consulted to receive help in routine times and during the coronavirus pandemic, a time in which they lost their livelihood, health insurance, and ability to understand the new rules of the lockdowns.

Introduction A software development company flashes the following text on the front page of their web site: "We help you with user-friendly solutions..." It was contradictory then to learn later in an investigation of the company that the company has little interest in usability design and evaluation, let alone in their users' experience of usability. It is on that backdrop that we went further in our investigation to try to figure out why that is. That is the crux of the matter of this article: Why? Over the last decade, there has been marked improvement in the software development industry's perception of usability work (Chilana, Wobbrock et al. 2010). Previously thought of as a vague and immeasurable system attribute, usability is now being acknowledged within the industry as a means through which companies can maintain their competitive edge and customer base (Mao, Vredenburg et al. 2005). There is also now greater acknowledgement of the usability designer role as a valid profession (Gulliksen, Boivie et al. 2006), and more individuals with this skill are involved in development projects (Venturi and Troost 2004). For a field that has struggled for legitimacy since its establishment (Rajanen and Iivari 2007), these positive developments would seem to suggest that usability work has achieved considerable integration, such that it is now a common consideration in software development projects. Findings from practice paint a more complex picture; while there has been some progress, it is not as widespread as is to be expected (Gulliksen, Boivie et al. 2006;Ji and Yun 2006;Venturi, Troost et al. 2006;Abrah<unk>o, Juristo et al. 2010). Indeed, it seems to have had no discernible impact as most organizations still lack a strategic approach to usability work (Seffah and Metzker 2004;Zhou, Huang et al. 2008;Ardito, Buono et al. 2013). Practices such as a focus on the user and end-to-end user involvement are yet to become the norm (Mao, Vredenburg et al. 2005). Usability designers still do not have the clout they need to function, and they often end up being passive bystanders during software development projects (Gulliksen et al. 2006). Generally, there a gap between intention and reality in the software industry remains (Bygstad, Ghinea et al. 2008). In this paper we seek to understand the reasons behind this and therefore ask, "How can we explain the reluctance in practice to integrate usability work into software development?" We conducted an interpretive case study of an agile software development company to examine the attitudes towards the integration of usability work. In accordance with previous studies of the integration of usability work into software development (Bygstad et al. 2008;Iivari 2006;Gulliksen et al. 2003;Boivie et al. 2006) we have taken a process view rather than a product view. To provide a rich picture of participants' rationalisations on this matter, we applied the theoretical lens provided by system justification theory. The theory is one which builds on individuals' natural tendency to rationalize and how they use such rationalisations to justify the social systems they exist in -even at the expense of personal and group interest (Jost et al. 2004). In the next section we provide an overview of existing research on the integration of usability work within the software development industry. In section 3 we describe system justification theory as our theoretical lens. Section 4 argues for the research design, the data collection and analysis, and it provides an overview of the selected case organisation. Section 5 presents our findings where we show the system justification mechanisms being engaged in the case organisation. We discuss the findings in section 6 and relate these to the existing literature. --- Related research In this section we provide an overview of existing research on the integration of usability work. --- Usability work integration Usability work consists of activities such as usability design and usability evaluation carried out to ensure a desired level of usability in the final product. There is an interplay between these activities evident in how products from the design phase form the basis for the evaluation phase; and how findings from the evaluation phase are fed back into the usability design phase for subsequent development of the product (Hornbaek and Stage 2006). Usability work is said to be integrated when there is a strategic decision to fully incorporate it into an organization's business processes (Venturi, Troost et al. 2006). When this occurs, the following six attributes can be expected to characterise such an organization (Venturi, Troost et al. 2006). These are: • Usability work would be included in a timely way into the product lifecycle • The usability team would have access to proper skills and experience • There would be a proper usability infrastructure in place to support the usability team --- • Management would be fully committed to usability work and to ensuring it is done • Usability work awareness and culture would be properly disseminated inside and outside of the organization • Results from usability work activities would inform design decisions. Another approach is the usability maturity model (Nielsen 2006) that can be used to assess the integration of usability work in an organization. The model describes an eight-level evolutionary path which shows the stages an organization progresses through as its approach towards usability matures, see Table 1. --- <unk>Insert table 1. Caption: Usability maturity model (Nielsen 2006)>> --- The integration of usability work in practice Several studies (Venturi and Troost 2004;Mao, Vredenburg et al. 2005;Venturi, Troost et al. 2006;Bygstad, Ghinea et al. 2008;Ardito, Buono et al. 2013) have examined the integration of usability work within the software industry. Their main finding is that even though there have been marked improvements in organizations' awareness of usability work and its importance, organizations where usability work is fully integrated are still rare (Venturi, Troost et al. 2006;Bygstad, Ghinea et al. 2008;Ardito, Buono et al. 2013). More specific observations include: little or no user involvement in software development projects (Vredenburg, Mao et al. 2002;Vukelja, Müller et al. 2007); usability issues being handled by software developers despite their limited understanding of usability work (Gulliksen, Boivie et al. 2004;Vukelja, Müller et al. 2007); and the disparity between their mental models and that of users (H<unk>egh and Jensen 2008). Hence, usability work is ad hoc and unsystematic and there is a wide gap between what organizations acknowledge usability work to be and what they actually do (Venturi and Troost 2004;Ji and Yun 2006;Vukelja, Müller et al. 2007;Bygstad, Ghinea et al. 2008;Zhou, Huang et al. 2008). Some researchers have attributed the non-integration of usability work to a lack of education among software developers and management (Vukelja, Müller et al. 2007). It has also been observed that organizations are unwilling to expend resources on usability work (Bygstad, Ghinea et al. 2008) because of ambivalence about benefits for the organizations (Mao, Vredenburg et al. 2005;Bygstad, Ghinea et al. 2008;Zhou, Huang et al. 2008). Cost benefit trade-offs have been highlighted as a reason why organizations prefer to use heuristic evaluations rather than the more expensive field study techniques (Vredenburg, Mao et al. 2002). Organizational cultures have been studied and particular cultural characteristics are more suited to certain types of usability work (Iivari 2006). Other reasons are that usability is a non-issue among clients as these individuals do not specifically request for it when initiating a project (Ardito, Buono et al. 2013). Time pressures in software development projects have been noted (Boivie, <unk>borg et al. 2003). There is also the issue of the type of development method, which is a criticism that has been levelled against agile methods in particular (Blomkvist 2005;Chamberlain, Sharp et al. 2006). Solutions to usability work integration are dominated by methods and prescriptive approaches in the literature and these have been criticised for failing to acknowledge the reality of practice (Gulliksen, Boivie et al. 2006). In some cases these solutions have even been observed to work to the detriment of usability work. A case in point is the usability cost benefit model which has been highlighted as being inherently capitalist in orientation and encouraging focus on the paying customer and not the end user (Rajanen and Iivari 2007). One approach which is increasingly gaining ground in the usability work integration debate concerns the organizational character (Cajander, Gulliksen et al. 2006). Obstacles to the integration of usability work within software development are a consequence of how organizations perceive usability work and if not conducive, integration becomes impossible (Gulliksen, Boivie et al. 2006). In this regard it has been shown that managers have a structural perspective where the emphasis is on efficiency and economy and that this explains their viewing of usability work as a non-rational investment, even if they acknowledge its relevance (Cajander, Gulliksen et al. 2006). Agile software development has received much interest among practitioners over the last years, and has led many software development companies to adopt to agile principles, e.g., a focus on flexibility and amenability to changes in business requirements through iterative processes (Dyb<unk> and Dings<unk>yr 2008). This in turn has led to much research interest into how usability work is achieved within agile settings; and it has been observed that agile processes present non-trivial constraints for usability work (Ferreira, Sharp et al. 2011). Research into the integration of usability work with agile software development has just started and most results are normative so far with little empirical investigation or detail (Wale-Kolade, Nielsen et al. 2013). Both agile development and usability work are similar in many respects, e.g., iterative approach, user involvement, and prototyping (Blomkvist 2005), and several attempts at combining the approaches have however led to little integration (Lárusdóttir, Cajander et al. 2012). The integration seems to be problematic due to several factors, e.g., power struggles between software developers and usability designers (Chamberlain, Sharp et al. 2006), and difficulties with separating customers from users (Blomkvist 2005). These difficulties can be overcome (Kautz 2011), but there are very few reported successes (Ardito, Buono et al. 2013). --- System Justification Theory System justification theory was developed to provide insight into how and why individuals legitimize and support their social systems even when these systems embody values that conflict with their own self-interest (Jost, Kivetz et al. 2005), and even when made aware that a different social system might better meet their needs (Napier, Mandisodza et al. 2006). It asks why individuals reinforce prejudicial attitudes that exist within their social systems. Here social systems refer to the structured network of social relations which may be tangible such as the families, institutions and organizations within which people live their daily lives or abstract and intangible, such as the unwritten but clearly recognizable rules and norms that prescribe appropriate social behaviour (Thorisdottir, Jost et al. 2009). With regards to the theory, it has been asserted that if individuals' blind adherence to the status quo is to be overcome and if interventions are to be effective, then there is a need to understand the rationale for their actions (Haines and Jost 2000). This is as ignoring these self-justifying tendencies would only lead to errors in the prediction of individuals' attitudes and inadvertently the solutions proposed to tackle them (Blasi and Jost 2006). The disconnect between the existing recommendations on the integration of usability work and practice and their inability to resolve the usability problem in software development, cf. (Ardito, Buono et al. 2013), may be seen as in accordance with this argument put forward by the theory. Further the priority the theory places on understanding individuals' attitudes and motivations is a practice which in the usability literature has been described as necessary to change organizations' approach to usability work (Cajander, Gulliksen et al. 2006;Gulliksen, Boivie et al. 2006). This shows the theory's relevance to our study. The theory asserts that individuals in engaging in system justification are driven by their social and psychological needs to legitimize the status quo and see it as good, fair and inevitable (Jost, Banaji et al. 2004). It is this emphasis on the motivational component of individuals' defence of the status quo that distinguishes system justification theory from status-quo-biased theory. The latter being a theory that considers such bias a consequence of a purely cognitive process involving the consideration of factors like transaction costs during choice making (Kay, Gaucher et al. 2009). Conversely in system justification theory this bias presents itself as a paradox, manifesting in individuals' legitimation of their social systems even when aware of the shortcomings of their social systems (Gaucher and Jost 2011). System justification theory describes the mechanisms that might be employed by individuals when justifying their social systems (Jost, Banaji et al. 2004;Jost, Kivetz et al. 2005;Blasi and Jost 2006). They include in no particular order, stereotyping, internalization of structural inequality, and rationalization of the status quo: • Stereotyping: Describes how individuals use mental representations of social categories within a system to defend disparities in how individuals within a system are treated (Kay and Jost 2003) and system passivity towards these disparities (Napier, Mandisodza et al. 2006). An example is the 'poor and lazy' vs. 'rich and hardworking' stereotype which leads individuals to justify why the poor should be neglected by the system and the rich esteemed (Jost, Kivetz et al. 2005;Singh, Mathiassen et al. 2010). Another form of stereotyping is complementary stereotyping where compensating favourable virtues are ascribed to a disadvantaged category (e.g., 'poor but happy') and corresponding unfavourable vices to the advantaged category (e.g., 'rich but sad') (Kay and Jost 2003).This latter form of stereotyping has been observed to allow individuals to maintain a psychological sense of equality ('no one has it all'), and justify inequality ('people get what they deserve and deserve what they get') (Jost, Kivetz et al. 2005;Kay, Jost et al. 2005). • Internalization of structural inequalities: Describes how disadvantaged individuals within a system legitimate the manner in which they are treated by the system in a bid to maintain their view of the system as fair and just (Blasi and Jost, 2006;Napier et al. 2006). It may in some cases involve some bias in information processing such that these individuals construe the situation in a manner that portrays them as deserving of whatever unfairness they experience (Napier et al., 2006). It may also be present as implicit or explicit group favouritism whereby members of disadvantaged groups evaluate advantaged groups more favourably than they evaluate themselves (Gaucher and Jost 2011). While this occurrence lessens whatever emotional distress the individual might experience, it leads to a 'depressed-entitlement' effect (Blasi and Jost, 2006). This is a condition where such individuals out of the continued persistence of the issue become completely resigned to the current situation and feel they do not deserve better from the system (Napier et al., 2006). Thus they become less inclined to advocate for change. • Rationalisation of the status quo: Describes how individuals justify not only their own decisions and behaviours but also the decisions and behaviour of others as well as outcomes where no decision maker was involved (Blasi and Jost, 2006). These justifications may involve denying or ignoring system failures in a bid to avoid the negative emotions that would result from admitting to these system failures (Napier et al. 2006). Additionally they might present as'sour grapes' rationalizations whereby objects perceived as unavailable or unattainable are disparaged; and as'sweet lemons' rationalizations where emphasis is placed on the desirable features of present unimpressive attainments (Kay, Jimenez et al. 2002;Sinha 2013). System justification theory further highlights that varying situational and dispositional conditions may cause the specific justifications and how they are triggered. This enables explanation of why in some contexts or among certain individuals, the tendency to engage in system justification is stronger in some situations and weaker in others (Kay and Friesen 2011). Criticisms by outsiders have been identified as the most potent form of the system threat antecedent (Kay and Friesen 2011). --- Research Method The research is based on the case study approach (Benbasat, Goldstein et al. 1987;Lazar, Feng et al. 2010;Yin 2013) to investigate a contemporary phenomenon in depth and within its real-life context. Both within research on software development and on usability work, the use of the case study approach has been advocated because it is more suited to the dynamic and complex nature of the field (Zhang, Li et al. 2009). The case study approach is particularly appropriate as the research question addresses the 'how' and 'why' of the integration of usability work into software development. In designing the case study we adopted the single case strategy in line with the common case rationale (Yin 2013). The aim is to capture the circumstances and conditions of an everyday situation to gain insight into the social processes surrounding the phenomena being examined. In addition we have assumed an interpretive stance (Walsham 1995) with an added inclination to probing beyond surface level explanations (Miles and Huberman 1994;Myers and Newman 2007). The case company was selected because the company at one level portrays itself as delivering software products and services with a high level of usability and at the same time performing agile software development. The software company is small with approx. 30 employees and as such typical for software companies in Scandinavia. Six informants were selected including the key product manager, see Table 2. In order to avoid elite bias (Myers and Newman 2007), the informants were selected based on their involvement in product development, rather than their position in the organization. The interviews were semi-structured following general advice on qualitative interviewing (Patton 2005;Myers and Newman 2007) and utilizing an interview guide. --- <unk>Insert table 2. Caption: Profile of informants>> The interview guide framed the interviews in a semi-structured format, asking the participants questions and an hour and a half for each interviewee. The interviews were audio recorded. In the post processing the interviewer also contacted the interviewee to clarify some of their answers. The analysis was done using system justification theory as the theoretical lens (cf. section 3) and applying contents analysis (Lazar, Feng et al. 2010). The interviews were first read to gain an overview and form an initial structure and interpretation. Subsequently, the interviews were analysed in detail by identifying the quotations interpreted as relevant for the topic and then coding these quotations using the categories from the system justification theory. Finally, the quotations and their coding were reviewed and reconsidered to ensure that the coding was consistent and in accordance with the theory. The interviewees' expressions and tone of voice were also taken into account during the analysis. We have in the analysis applied a process view rather than a product view (Floyd 1987). This means we have a particular interest in how they work, how they develop software, and how well they integrate usability work. This view is in line with previous studies of the integration of usability work into development (Bygstad et al. 2008;Iivari 2006;Gulliksen et al. 2003;Boivie et al. 2006), and this is also the focus chosen in research on integrative methods, e.g., (Göransson et al. 2003;Gould & Lewis 1985). In a similar manner we have analysed our case data with a view to the process questions such as 'in which ways?', 'how well?', and 'why not?' --- Analysis and Findings In this section we first introduce the case company and provide an overview of how they develop software; we then present the findings. --- The Case Company The case company is located in Scandinavia and started out as a software development company more than ten years ago. A few years ago the case company was purchased by a leading international company within enterprise systems. The primary task of the case company is to develop software. It operates as an independent subsidiary of the parent company, and it has a high level of autonomy and control over its operations. The product suite consists of web-based applications for processing projects, time, travel and expenses. The company has long-term relationships with a few large customers and its products are not aimed at a mass market. The case company has a forum of super users that is established prior to customers purchasing the product and is active after deploying the product. The super users' forum acts as a liaison between the software company and its customers. These super users are typically the IT personnel in the customers' organizations and they are also responsible for selecting the configurations of the product suite. On their web site it says: "We help you with user-friendly solutions..." Usability of the product suite is not only a sales tactic, it is also a business goal. The company is small and staffed with 29 employees. The software company's product development team consists of nine people. Six of these are developers, one is a functional tester, one is the product manager, who acts both as a Scrum master and a product owner, and one is a lead consultant. The work environment is open and encourages collaboration, sharing of information, and a large degree of autonomy for developers. The company has a customer centre that provides consultancy, support, training, and implementation services directly to customers. The customer centre also provides help desk service to the users. --- Agile development in the case company The software company has for the past five years been using their own, adapted version of Scrum (Schwaber and Beedle 2002). Their version of Scrum is modified with respect to how they organise their daily stand-up meetings and sprint planning. For instance they have two daily stand-up meetings unlike the one that is suggested by Scrum. The first meeting is between the product manager and the lead consultant and it lasts no longer than 15 minutes. According to the informants, this arrangement was introduced to further the communication efficiency between the developers and the consultants by: (1) creating a single gateway through the product manager, and (2) preventing consultants from overloading developers with requests. Later in the day there is a standard stand-up meeting for the product development team lasting no longer than 15 minutes and aiming at coordinating the work of team members. At the start of a new iteration sprint the team spends half a day on sprint planning. Sprint planning includes which tasks should be completed, but it excludes effort estimation based on hours. Rather they plan based on story points. The interview with the product manager reveals: "Earlier we use [d] to plan each task with hours on each developer during the sprint planning and that means if one developer has 250 hours we will break that down and he will take on the tasks that he wants to have from the list and suggest some hours on that. What we have done recently since April this year has been a shift toward that we have a pool of tasks and we have not planning on hours." The product manager who is also acting as Scrum master monitors the time spent by developers on tasks in order to minimize delays in subsequent, dependent tasks. The team has 14-days-sprints and maintains production line setting where the scope of each sprint is not fixed. Sometimes the outcome of a sprint is bug fixes on a version of products or product upgrades. Sometimes the outcome is new functionality specifically requested by key customers or newly acquired customers. Sprint reviews occur sometimes at the end of the sprints. --- Usability work in the company The informants' responses show a reasonable level of knowledge of usability work. Themes emerging concern users' interaction with the system, learnability, and usability. This is captured in several statements, e.g., the product manager states: "I would describe it as the user interface and paying attention to it would be making the software as good as possible for the user to use. So it is not only about the functional part. One thing is that it has to be functional but it also should be a good experience to use it and that is the layer that I would put into usability." One of the developers states that usability is about having the "user perspective" and highlights the importance, as a developer, of having an outside perspective as "sometimes if you are working on a product, you work every day with the same details, you get kind of blind in your ideas." The informants are also able to describe various ideal usability work practices and their perceived benefits, see Table 3. --- <unk>Insert table 3. Caption: Evidence of the informants' knowledge of integration of usability work into their agile software development>> The informants' understanding of usability work is not limited to their ability to provide articulate descriptions. The functional tester has previously been employed as a usability designer in another company. The product manager has also taken courses in usability while at university and is generally familiar with usability design principles. This background knowledge on usability work is a prerequisite for the business goal to pursue a high level of usability. It is however worth noticing that on the one hand they acknowledge that the integration of usability work could be valuable to them, but on the other hand they do not integrate usability work into their development process. The current development practices in the case company are not integrating usability work. One of the developers said: "We don't actually really speak to the end users." On the question of whether they do usability testing with users the product manager replies: "... no I wouldn't say so. The only way we have touched that would be when we did the expense overview, but that was also much based on functionality." One of the developers describes how a usability evaluation session is carried out: "... we hired a graphic designer, we did not do much testing just some testing to verify with our biggest customers that the new icons and colours were okay." Apart from the functional tester, the other members of the product development team do not feel that the integration of usability work into their development processes is something that needs to be approached more systematically. These team members describe themselves as user-driven, asserting that their products "should be self-describing." According to the lead consultant, it is "very important to us to have a system which is good for the users." This again confirms the business-oriented usability goal of the case company. Yet the team never conducts an evaluation of the usability of the products. The development team collects users' feedback; but that concerns solely perceptions of the company and the functionality offered in the product suite. The developers, the lead consultant and the product manager are aware of the inherent and potential danger in this approach, but they have chosen to accept this. The product manager asserts that it is possible that their product "could be really a [poor] system and we are just surfing around happy here, thinking everyone loves our system and they are really like [not]." This shows a discrepancy between the desire to develop for high usability and how systematic the manager pursues it. The functional tester, who is also a trained usability designer, further states: "It is funny because at this user conference, we showed our users a really small change which was only one line of code and it was that when you click on expense print out, you get a view of what it would look like. Before you had to use the [image] of the expense and then you had to click print once more and then you had to click on the printing interface once more. The company got applause for this and everyone was shocked. It took the developer maybe a minute to do that and they got applause. It was a good example of usability, that only small things can get our customers to go 'wow.'" The quote shows that the developers have no practical awareness of usability work since they have not tested its usability with their users and they have little sense of the ease with which they could potentially increase usability. The company has no infrastructure in place for usability work. There is no usability designer role in the product development team. It is likely that the company is too small to have a full-time usability designer, but there is none assigned to a part-time role either. The functional tester with the usability designer skills and who might have functioned in a part-time role as usability designer has been assigned a different role and to different tasks. Management is not committed to a process that includes usability work as captured in the following comment by the product manager: "So I would say that we consider [usability work], but we don't do it, we don't have like this is the usability way... this is the design team, this is the development team, we don't do that, but it is always there. It is more implicit than explicit, we would never release anything without having checked that it works with the client, but how we do it is not set and sometimes we do it too late." The reluctance to integrate usability work in an explicit way is not a consequence of their agile development approach or specifically their use of Scrum. As emphatically stated by one of the developers, "No it nothing to do with Scrum.... it is nothing to do with which project methodology you use, it is the decision you as a company take." This view is also shared by the functional tester who asserts, "No. Actually I think it [Scrum] is really good for usability, because we can do some rounds and think about it.... We have the possibility to do it, if everybody agrees that now this is necessary." According to the developers, Scrum is a flexible approach as it allows for more "change than in waterfall" and that one can "test things in an earlier stage using Scrum than in waterfall." The company's approach to usability design seems independent of their agile development approach, but there is no deliberate integration of usability work into the development process. The company as summed up by the functional tester is "missing a routine" to usability work even though it "has the tools", and she and others have the knowledge needed. This resembles the level of developer-centric usability in the Nielsen model, cf. Section 2. The product development team realises the value of usability, but responsibility for ensuring the usability of their products has been left to the developers, who on their part are based on intuition and an understanding of what is efficient. The analysis shows that the company can best be described as paying some kind of lip service towards usability work. They claim usability work to be important, but they do not practice it in a systematic or explicit way. They can talk about it and they can explain the different aspects in detail including its potential benefits. They have however deliberately chosen not to pay more attention to it. --- A system justification analysis The analysis of the development team's rationalizations in defence of their approach to usability work can be extended by applying the theoretical concepts of system justification theory. The following analysis shows the three mechanisms we found in the interviews: stereotyping, internalization of inequality and the rationalization of the status quo. Stereotyping captures the descriptive terms participants used when talking about the usability designer's role and their users. The internalization of inequality captures the viewpoint of the usability designer and how this individual now in the functional tester role rationalised the approach of the organization. The rationalization of the status quo includes two different aspects both leading to the preservation of status quo: justifications that highlight perceived impracticality of usability work, and justifications that highlight the company's present achievements. --- Stereotyping The stereotyping addresses both usability designers and users. --- Usability designers as creative, but orthodox Stereotyping is observed in the development team's descriptions of usability designers and the approach the company has adopted towards this particular role. For instance the developers recognize the benefits a usability designer would bring as highlighted in Table 2. Further terms such as "creative," "fun to work with," and "particular" were used by the developers to describe some of their impressions of usability designers. However, the product manager argued at the same time that usability designers were often out of touch with reality and users' needs and that they had failed to recognize that the "rules had changed within usability." Further that: "designers think in more general terms, you know less is more, sometimes they talk about the audience, but they forget that it is a bit different... Of course it is fun to have an application that looks cool, but the most important thing is that it runs fast." This line of rationalization came as a response to a question of why the company do not employ a usability designer (perhaps part-time) in the light of usability being one of the business goals. From a business perspective it may well be reasonable not to employ a usability designer and then in stead rely in the software developers to take responsibility for usability design. From the product manager's defensive attitude we suggest an alternative interpretation that is part of a stereotyping of usability designers. The above statements also came in a discussion concerning an aspect of their products' interface which apparently has been criticized by the usability designers they have met over the years, on the basis of it violating usability design principles. The product manager responded rhetorically, "should we listen to the professional guys or should we listen to our clients?" As we shall see below they have a process where they listen to super users, but not to all their users. The developers also mentioned that usability designers ignore that the "demands for usability were different when it comes to administrative systems" and how the users of such systems are more focused on speed rather than whether the application is "nice and colourful." The developers mentioned that their clients had affirmed that that is the main reason they choose the company's products. The developers view usability designers as individuals who are out of touch with reality and are too focused on the mundane or even orthodox. For example, one of the developers when asked about the impact a usability designer has had on an application that was redesigned, replies that it is more "glossy and colourful." This is effectively a very limited view on usability and usability work. The developers generally rationalised that having a usability designer would lead to unnecessary interface changes. We are not suggesting that this stereotyping of usability designers is the main issue when the product manager and the developers justify a discrepancy between the business goals and the process through which they work with usability; but we are suggesting that it is a part of it. --- Super users as central gatekeepers; end users as futile The product development

In this article we report from a case study of a software development organization and we study in particular the developers' and product managers' attitudes towards integrating usability work into software development. We offer explanations based on system justification theory illuminating what would-be integrators might be up against. The analysis shows how the developers only pay lip service to usability work and how they treat users superficially. It further shows how that leads to stereotyping of usability designers and users in order to preserve status quo, and how internalization of inequality between the developers and usability designers rationalizes the preservation of status quo. These findings will have significant implications for managers of software development.

all their users. The developers also mentioned that usability designers ignore that the "demands for usability were different when it comes to administrative systems" and how the users of such systems are more focused on speed rather than whether the application is "nice and colourful." The developers mentioned that their clients had affirmed that that is the main reason they choose the company's products. The developers view usability designers as individuals who are out of touch with reality and are too focused on the mundane or even orthodox. For example, one of the developers when asked about the impact a usability designer has had on an application that was redesigned, replies that it is more "glossy and colourful." This is effectively a very limited view on usability and usability work. The developers generally rationalised that having a usability designer would lead to unnecessary interface changes. We are not suggesting that this stereotyping of usability designers is the main issue when the product manager and the developers justify a discrepancy between the business goals and the process through which they work with usability; but we are suggesting that it is a part of it. --- Super users as central gatekeepers; end users as futile The product development team distinguishes between'super users' and 'end users.' The super users perform the dual role of acting as a gateway and gatekeeper between the customer organization and the software company. These super users act as primary internal support persons in the customers' organisations and are typically employed in the information technology department or financial department of the client organization. According to the developers the super users are the "power users who use all of the system and not the normal users" whose use of the system is limited to the occasional "delivery of hours and recording of expenses." The company place a high priority on the super users. According to the product manager, this is because they are the "contact point" and therefore it is not "natural" to relate with the end users. According to one developer, the super users are "experienced so small glitches or something that could be done easier does not get their focus, the focus is more on the functionality and not 'can we do this in this way?'" The product manager mentioned that if the company decides to make any changes to the user interface, with a focus on making it easier to use, "then we would [need to have] good arguments for doing that and we would need to train them [super users] in using the interface." In other words, they would need to justify why they would want to improve the usability of their products, as in this case for the super users, it would be a "hassle." The developers also argue that it is not necessary for them to focus on the end users since their products are "for companies anyway, for enterprises; [and] not for end users." To explain the incongruence in this statement, one of the developers stated, "it is for end users, but not primarily for end users and in that case, it is more functionality that is important than if it is green or red." In other words, the developers feel the end users are incapable of knowing what attributes a good product should possess or making judgements related to it. Additionally, the developers acknowledged that there are differences in the tasks these two sets of users carried out on the system, but they do not perceive that it is necessary to examine the use experience of the end users. The head consultant asserts that the registration of hours on the system that the end users carried out is typically a "pain... for most people" and their aim is that "it should go very easy and that is why we have to focus on user friendliness." This consultant talking for the product development team said they do not want a situation where the end users will have the "thought after they have used or [entered in] expense, this is awful and I don't want to do it again." In spite of these affirmations, the development team takes the perspective of the super users and even when the focus is on functionality. Asking the head consultant about this, highlighting the incongruence, he again justifies their focus on the super users, stating that it is important "they [super users] should know all the functionality and they can select what is the best for [their] company." The developers rationalise that these super users should be given high priority in the development process. This priority is in the form of courses arranged for these users, occasional testing of new functionality with the super users, and eliciting of company-related feedback from these users. For the end users, it should be enough that they follow the help guidelines the company has installed in the products. --- Internalization of inequality In examining the internalization of inequality we focus on the functional tester who is a trained usability designer and who is most disadvantaged by the current arrangements within the company. The frustration this individual feels can be gathered from: "I work as a functional tester, this is my job and they knew that I had a background for usability, but I have no percentage, 10% or 20% of my time I can focus on usability" The functional tester admits to the failures of the company in relation to integrating usability work into the development process. She mentions how the user perspective has been neglected; and she refers to how the focus is on the super users and not on the end users the few times usability testing has been performed. This, she notes, has negative implications as the super users are using other parts of the system than the end users. She also elaborates more on some of the occurrences that has been used by the developers as a basis for their justifications, e.g., user feedback was positive only because "it is typical Scandinavian to give good feedback" because "the salesmen had done a good job of advertising the product as one with good usability," and because the company mainly obtained user feedback from the highly prioritized super users. She further states that the "general understanding" among the developers is that having decided on the right colours solves most usability issues. For her usability is that: "a specific user can solve his problem. The user interface can have something to do with usability, but not necessarily I think. I think [the user interface] is just the tool. It could also be work flows or something like this... But it is... not only the user interface, but I understand the developers think it is." She also described how she has once presented at a seminar to argue for more focus on usability in the development process, and she also suggested that they try usability testing during sprints for a month. However, such considerations were not acknowledged. This failure to make any changes, it seems, has led to her becoming gradually more accepting of the way things are. This is captured in the following: "I am really interested in this [usability work] and I would like to do it more. But I know it is not my main job and I have a lot of things to do in my functional testing and so on. That is how it is." As a sign of this gradual shift towards acceptance, we observed that the functional tester sometimes legitimates current practices in the organization. For instance when asked why no usability testing is being done, she mentioned that it is "because we are a small organization, and often we do not have time." She understood the position of the product manager since there are "a lot of tasks to do and customers want new releases all the time and bugs fixed really fast." She also talks about how she thinks the "developers have a focus on usability," and "that they have it in their head," thereby contradicting her earlier assertion on the developers' lack of interest in usability work as part of the process. The functional tester also lacks experience with usability work in agile development processes. She has emphasized the need for usability testing to be conducted in between sprints. However, when asked how easy this was to do in her previous place of employment, she admitted these suggestions were mostly in her "head" and that she had never seen it done in practice. However, she do not see her "disadvantaged state" as being a consequence of the company's use of agile processes. This is summed up in the following: "I really like the agile way of working and I am part of the team and I don't have to spend as much time discussing or arguing." The balance of power is clearly not in functional tester's favour, yet through the internalization of inequality she also in part rationalizes the organization's approach to usability work. --- Rationalisation of status quo The rationalization of status quo is twofold that usability work is not worth integrating into the development processes ('sour grapes'), and that the developers are already doing everything right ('sweet lemons'). --- 'Sour grapes' rationalizations of usability work Usability work is described by the developers as "more theoretical than practical." Though the developers have provided descriptions of how usability work should ideally be done (cf. Table 3), one of the developers argues that "in our project it is not a realistic scenario" to perform usability work like user trials in their projects. Usability testing is considered to be something necessary only when it would give a direct effect and not something to be done systematically. One developer describes testing as an impossible feat as it is "difficult to know who your next user would be," even though he admits that this could easily be resolved by taking "anyone from the street and testing the user interface." The developers also talk about the cost of testing, in terms of the manpower and analysis involved. Other concerns include time in relation to how actions like prototyping during sprints and having these tested with users would lead to project delay. The product manager remarks, "I think in a perfect world where we did not have to think so much about how much time we use when we are doing things, we could focus on usability." Employing a usability designer (even part-time) as part of the development team is not seriously considered, even though they acknowledge that it might be beneficial. One of the developers talked about how such an individual as part of the team "could get caught up in the same mind-set, the same ideas, [and] the same pattern of doing things" and that it is preferable for such an individual to be outside the organization so as "to have an unbiased view of the product." This developer also argues that the company is not "big enough a company or organization to be able to have someone with that sole responsibility." Thus, the potential benefits of usability work are diminished -what could potentially be sweet grapes are turned sour. --- 'Sweet lemons' rationalizations of self One of the developers argues that, "we try to do a reasonably good job, so we try to make it workable... we do compromise on some things sometimes, we don't go by the book but till now it seems as if the people who use our product, we get good feedback". This feedback they describe as including clients' favourable comparisons of the company's products against those made by their competitors, high number of business referrals, and low customer attrition rates. This positive feedback is used to extol the benefits of the company's current development process. For example, the product manager states: "Our users are giving us high credit even though our applications defy usability design guidelines and it is because we focus on functionality." The developers also emphasize the altruistic aspects of their development process. One of the developers highlights how they have included the functional layer into the graphical user interface to make the applications more responsive for the users. Another developer in relation to this admits this has limited their ability to make changes to the interface. However, this is again justified on the basis that it allows them to ensure that customers' familiarity with the products is maintained. But a developer states that "we have solved this reasonably. Most of our functionality is user friendly." Thus, the developers contend that they have good development practices and that this has been proven by the lack of customer complaints. That helps inflating the belief in what they are currently doing and in preserving status quo. --- Discussion At a distance the case data may lead us to consider it along the same lines as previous research which has examined the gap between the research and practice and concluded how prescribed or formalized development methods are rarely adopted in their entirety and never used as prescribed (Stolterman 1991;Fitzgerald 1998;Nandhakumar and Avison 1999;Fitzgerald, Russo et al. 2002). Previous research has also highlighted how such adaptation of methods and processes are often based on pragmatic considerations (Stolterman 1991;Fitzgerald, Russo et al. 2002). At one level of analysis our findings concur with this as we have also found discrepancies between research and practice, e.g., the development team has made modifications in their practice of Scrum as an agile method. The system justification theory as an analytical lens has led to a closer look at the case data, and at a qualitative explanation of the case company's contradictory stances towards the integration of usability work into their software development processes. The analysis has shown that not only do contradictory stances exist, but the analysis also suggests the source of this. There are several contributions in the analysis and the findings from the case data. Firstly, the analysis based on system justification theory tells us that sometimes such adaptations of methods and processes with respect to usability work may well arise from irrational considerations such as a preference for the status quo and apathy towards change. The theory also enabled us highlight the rationalizations which could be used to legitimate such bias. In this study, this referred to the development team's stereotypical representations of users and usability designers, the designer's internalization of inequality. Lastly the'sour grapes' rationalizations where the developers saw usability work as impractical; and their'sweet lemon' rationalizations of self which bolstered their confidence in their approach. There was in general a self-reinforcing cycle which ensured a continued sustenance of the apathy towards usability work in the organization and which ensured there would be no integration occurring in the organization. Secondly, the findings from the case highlight unanticipated attitudes -in this case the tendency to engage in system justification and the rationalizations which accompany this. Such unanticipated attitudes have not been explicitly considered in existing research. The usability designer in the case will typically be described as an "advocate of usability" (Boivie, Gulliksen et al. 2006), and she had gradually as a result of the internalization of inequality become less an advocate of usability and begun legitimatizing the status quo. Further personal skills such as having a great deal of stamina and a strong belief in what one is doing have been listed in (Boivie, Gulliksen et al. 2006) as prerequisites for usability designers in organizations where issues of usability are not prioritised. Our findings suggest that even a usability designer possessing these skills may be overshadowed by internalization of inequality. We suggest that the possibility of its occurring should not be overlooked, hence adding to the results in (Boivie, Gulliksen et al. 2006). Thirdly, similarly on the part of the developers, previous research has observed that developers see users as similar to themselves (Bader and Nyce 1998), and that developers have an inadequate sense of usability of their own software (H<unk>egh and Jensen 2008); and further that usability evaluations can debunk developers' erroneous assumptions of usability (H<unk>egh and Jensen 2008). Our findings suggest that the effect of usability evaluation may not be realized. The developers in this study were well aware that there was a disparity in their perception of usability and that of their users. They also acknowledged that they did not have the necessary expertise to ensure usability, yet in spite of this they chose to justify the company's approach to usability work. Therefore it is necessary to be aware of the existence of such system justification tendencies in order to recognize their influences and to be able to discern when they are being used. With this finding we thus add to (H<unk>egh and Jensen 2008) the awareness of potential system justification taking place. Along similar lines (Rajanen and Iivari 2007) argued for using the cost benefits of usability to justify changing the development process to include usability work; and our findings suggest that might be overtaken by a cover-up to preserve the status quo. Fourthly, existing research development-centric companies are more receptive to the logic of usability and more inclined to move to a higher level of maturity if pushed (Nielsen 2006), and that usability designers need to be able to tailor their message and focus on selling usability (Iivari 2006). Our case and the analysis of it suggests that such an intervention or mere allusions to a company's problematic usability processes can be construed as criticisms of the organization and trigger system justifying tendencies hence with the preservation of status quo as a consequence. The implication of our study is however that we concur with (iivari 2006) that it is necessary to understand an organisation's culture fit usability processes to the organizational culture. Based on our finding we further suggest that these organisational cultures are not necessarily easy to understand and that sometimes a deeper assessment of underlying justifications may be necessary. Fifthly, the stereotypical attitudes used in our case company to describe usability designers and users are similar to what has been reported in previous research. Usability designers have been described as retrogressive by developers due to their focus on user studies prior to the start of the development (Norman 2006), usability designers being viewed as technically incompetent due to their lack of understanding of the development process (Kollmann, Sharp et al. 2009), that end users are being seen as peripherals (Shackel 2009), and that they should be kept at arm's length (Gulliksen 2007). Our findings show that such views were indeed also found in our case, but we suggest that these views are not inherent or intrinsic to software development, but may in stead be caused by a gradual rationalisation and justification for status quo and against change. Sixthly, it is not uncommon to report from practice that software development companies do not find that integration of development processes and usability processes can be reconciled due to different cultures (Bygstad, Ghinea et al. 2008). Our findings corroborate this, but we suggest that we have added to a more detailed understanding of why that is -in some cases. This also correlates with the reports that the integration of usability work in development contexts is less a consequence of the method used but more dependent on how usability work is perceived within these contexts (Ferreira, Sharp et al. 2011;Wale-Kolade, Nielsen et al. 2013). Based on the developers' views in our case and their support of agile development processes we will question whether the agile culture have also triggered their system justification tendencies. There is nothing to suggest that in the case data. --- Conclusion In a case study of the integration of usability work into software development processes it has previously been reported that there is a gap between intention and reality. Our study confirms the presence of this gap in the case company, but it also points out that the gap can -as in our case -be based on fundamental justifications, not only presenting itself as a gap between what is intended and what is actually done. The case shows an example of a company where the perceptions of usability work were far from favourable. We have been seeking answers to the question of how they perceive usability work and we have provided evidence to show how they at one level praise usability, but in their practice they are not adhering to their own praise. In order to understand these contradictory attitudes we have used systems justification theory as a theoretical lens through which we have been able to show how their justifications have led to: • Stereotyping of super users, end users and usability designers. • The usability designer's internalisation of this stereotyped inequality. • Rationalisation of status quo by the developers. We have shown with quotations from interviews with the informants how the case organisation expresses this kind of justification. This is hugely incongruent. The findings can be used in at least two ways: (1) it highlights what a software manager, a usability designer, or a would-be integrator may well be up against; and (2) it points to how to potentially restore some congruence simply by demonstrating the severity of the implicit justifications leading to the explicit incongruence. --- 2016 --- Document Version Early version, also known as pre-print Link to publication from Aalborg University --- The limitation of the case study lies primarily in the use of system justification theory is that it does not tell when and how individuals' tendency to engage in system justification might be reversed (Gaucher and Jost 2011). We also see that the system justification theory does not address in any significant way a business perspective on the case company. Future research would benefit from studying the same phenomena in a business perspective and try to see to what extent and in which ways business rationality may lead the development team to refrain from going deeper into usability processes and practice. Another limitation of the presented case study is also that we have almost exclusively taken a process view; this has downplayed the issue of whether the developed system has a sufficient degree of usability and if that has been achieved implicitly without any explicit or systematic usability design processes. Hence, further research should be conducted to show how justification relates to perceived as well as measured usability.

In this article we report from a case study of a software development organization and we study in particular the developers' and product managers' attitudes towards integrating usability work into software development. We offer explanations based on system justification theory illuminating what would-be integrators might be up against. The analysis shows how the developers only pay lip service to usability work and how they treat users superficially. It further shows how that leads to stereotyping of usability designers and users in order to preserve status quo, and how internalization of inequality between the developers and usability designers rationalizes the preservation of status quo. These findings will have significant implications for managers of software development.

INTRODUCTION During public health emergencies like the 2003 SARS pandemic, 2009 H1N1 swine flu pandemic, 2016 Ebola epidemic, and current COVID-19 pandemic (1-4), the internet and social media serve as hubs for the growth and rapid dissemination of information and misinformation globally, often contributing to confusion in public health campaigns. Misinformation is false or inaccurate information that is not supported by scientific evidence and shared without intent to cause harm, and includes conspiracy theories, disinformation, and mal-information (5,6). Misinformation is often conflated with colloquially used terminology such as fake news, rumor, urban legends, spam, and trolling (6,7). This overabundance of information and misinformation contributes to the growing infodemic, where the availability of false or misleading information online and offline creates an environment that can undermine public health responses during a crisis (8). Misinformation has been shown to negatively impact people's COVID-19 health protective behaviors, including spending limited time outside of homes, practicing social distance measures, and washing hands per public health guidance (9). COVID-19 misinformation, specifically the erroneous labeling of COVID-19 as the "Chinese Virus" on social media (10) has led to social discrimination on Asian American communities, contributing to increasing anti-Asian sentiment and anti-Asian hate crimes across the nation. Asian American social networks are distinct from other communities. These networks may include family in other Asian countries and include dissemination of non-English information on social media. However, there is a paucity in literature examining use of social media for COVID-19 information among Asian Americans, types of misinformation disseminated on social media in these networks, and subsequently the impact of such information on Asian Americans' health behaviors. We conducted a narrative review to examine the use of social media platforms to disseminate COVID-19 misinformation across Chinese, Korean, Vietnamese, and South Asian Americans, four of the largest Asian subgroups in the United States (U.S.). We discuss the linkage of COVID-19 misinformation to health behaviors, with emphasis on misinformation about COVID-19 vaccines and vaccination confidence, as well as community-driven responses to address COVID-19 misinformation among Asian Americans. Lastly, we propose recommendations to mitigate misinformation and address the infodemic in Asian American communities. --- COVID-19 Misinformation and Social Media Health misinformation is prevalent across various social media platforms. A systematic review of 69 studies identified Twitter, YouTube, and Facebook as the most common social media platforms for health information dissemination, while less popular platforms included WhatsApp, Pinterest, Tumblr, BK, and Myspace (11). There are six main categories of health misinformation topics including pandemics, vaccines (e.g., antivaccination, side effects of vaccines), medical treatments, non-communicable diseases (e.g., cancer, diabetes, epilepsy), eating disorders, and drugs or smoking (11). Health misinformation poses a public health challenge due to its adverse influence on health behaviors, including panic buying and disregarding health guidance from public health officials (e.g., stay-at-home orders, handwashing, social distancing) (9,12,13). Researchers reported that social media was nationally the third most common information source for COVID-19 information and misinformation (14). Similarly, a recently published scoping review examining the role of social media during the COVID-19 pandemic identified 12 articles that discussed the role of COVID-19 misinformation augmenting the global infodemic (15). The ongoing infodemic has spread rumors and disinformation on social media related to COVID-19 virus etiology, prevention, treatment, interventions, and vaccines (16,17). COVID-19 misinformation is challenging to address and document given its pervasiveness across various social media platforms that are highly utilized and trusted by the global community (18). Greater exposure to misinformation is associated with problematic COVID-19 beliefs, poorer COVID-19-related knowledge, and less preventative behaviors (13). Relatedly, belief in COVID-19 conspiracy theories was found to be associated with lower COVID-19 preventative behaviors (9). Misinformation about COVID-19 may also influence specific health behaviors (e.g., vaccine hesitancy) (9,19). Vietnamese Americans Facebook and YouTube (26,27) Korean Americans KaKaoTalk (26,27) Japanese Americans Line (26,27) Asian Indian Americans WhatsApp (26,27) Social Media Utilization Among Asian Americans Asian Americans, who comprise 5.5 percent of the U.S. population (20), are leading users of internet and mobile technologies, evidenced through higher ownership of smartphones, laptops, and wireless networks (21,22). Approximately 94 percent of Asian American households own a smartphone in the US, which increases access to the internet and social media for entertainment, information seeking, and social connectedness (22). In particular, Asian Americans (many of whom are first generation, foreign-born immigrants) utilize the internet and social media to seek and receive up-to-date health information (23,24). According to the Pew Research Center (25), approximately two-third of Asian Americans speak a language other than English at home, with Chinese (34%) being the most spoken Asian language, followed by Hindi (13%), Tagalog (9%), and Vietnamese (4%). Social media allows for cross-national social connectedness to family and friends in home countries (22), and the efficacious transfer of health, political, economic-related information in native languages. Importantly, for limited English proficient (LEP) Asian Americans, the use of mainstream English-speaking social media platforms may not be appropriate. To mitigate language barriers and receive up-to-date health information, many Asian Americans tend to use social media platforms that enable communication in their native languages. As such, preference for social media platforms among Asian Americans parallels the diversity of Asian American ethnicities, languages, and generation (see Table 1). Despite the prevalent use of social media among Asian Americans, there has been limited research on Asian Americans' pattern of social media use by age, gender, and socioeconomic status (28). Frequently, Asian Americans are not included in national datasets on social media use and engagement. For instance, the Pew Research Center does not report data on Asian Americans in its research on social media use in the U.S. Similarly, detailed race/ethnicity was not reported for "Americans" in an international survey that examined use of social media for health information among Americans, Koreans, and Hong Kongers (29). In the context of the COVID-19 pandemic, there has been no research investigating Asian Americans' use of social media for COVID-19-related information. While recent research has investigated sociodemographic predictors of U.S. adults' use and trust of COVID-19-related information sources (14), researchers have yet to examine race as a factor, limiting what we know about Asian Americans' use of social media for COVID-19 related-information. However, a recent survey of Asian American young adults from March 2021 reported participants used Facebook Messenger (61%) and Instagram (57%) most often with siblings and cousins and WhatsApp (29%) with parents 1. Use of social media platforms also varied by Asian subgroup, gender and country of birth 1. Preference of specific social media platforms may also differ between older and younger generations of Asian Americans. However, there may be in-language social media platforms such as WeChat, KaKaoTalk, and Line that are shared across generations and promote the circulation of information across geographic and generational boundaries. --- COVID-19 Misinformation by Asian American Subgroups Amidst COVID-19-related uncertainty and fears, Asian Americans are turning to the internet and social media for the latest updates on the pandemic, such as confirmed COVID-19 cases, government-issued COVID-19 policies and guidelines, and COVID-19-related health information (e.g., COVID-19 prevention measures, personal protective equipment (PPE), COVID-19 symptoms, diagnosis, and treatment) (30,31), which have been shown to positively influence and improve public health behavioral strategies against COVID-19 (1). Asian Americans' pattern of utilizing social media and other online sources for information during the COVID-19 pandemic is analogous to the use of these platforms during the 2003 SARS pandemic (2), during which the Centers for Disease Control and Prevention (CDC) reported evidence of substantial misinformation that contributed to growing confusion and fear. Likewise, COVID-19 misinformation is circulating on social media platforms utilized by Asian Americans, leading to widespread misinformation across the Asian American communities in the U.S. and the Asian global diaspora. Misinformation on the internet and social media during the COVID-19 pandemic continue to heighten the mistrust, fear, and anxiety stimulated by COVID-19 pandemic and anti-Asian racism (32,33). Extant literature on social media, Asian Americans, and COVID-19 has focused on how contentious COVID-19 misinformation contributed to the increase in discrimination, racism, and violence against Asian Americans (10,34), as well as how social media is used to mediate such effects (35). However, there is limited research on the content of COVID-19 misinformation dispersed across ethnic-specific social media platforms in the Asian American community. We illustrate the heterogeneity of COVID-19 misinformation across social media applications by Asian American subgroups and the potential impacts of COVID-19 misinformation on Asian American subgroups' health behaviors (see Table 2). --- Efforts to Addressing COVID-19 Misinformation in Asian American Communities In response to alarming COVID-19 misinformation in Asian American communities, community and public health leaders are spearheading efforts to combat the impacts of misinformation and to disseminate scientific COVID-19 information. In New York City, in-language virtual webinars were held by the NYU Center for the Study of Asian American Health (CSAAH), in partnership with community and academic partners, for Chinese, Filipino, South Asian, Arab American, and Hispanic communities, to answer questions and concerns about COVID-19. Likewise, the Filipino Young Leaders Program (FYLPRO) developed the Caretaker Program, Tayo, a virtual help desk that provides accurate and timely information on COVID-19 to the Filipino community in Tagalog and English (52). Similarly, the Asian Pacific Islander American Health Forum (APIAHF) has produced in-language videos (e.g., Simplified Chinese, Traditional Chinese, Korean, Vietnamese, and Bengali) to dispel COVID-19 vaccine misinformation in Asian American communities (53). Nationally, the Progressive Vietnamese American Organization (PIVOT) launched the Viet Fact Check Project to release in-language information on COVID-19 vaccination, to dispel common myths, and to encourage higher uptakes of vaccination in the Vietnamese community (46). The Viet Fact Check Project is also collaborating with the University of Washington Center for an Informed Public, to examine the pattern of misinformation and media consumption in the Vietnamese community (54) --- RECOMMENDATIONS Misinformation, especially COVID-19 misinformation, has farreaching impacts on health and wellbeing for Asian Americans and other minority groups that face existing, unique health disparities (9,10,13,56). It is imperative to recognize the abundance of misinformation that exists in these communities and to identify misinformation as a public health crisis (9,12,13,18). The field of public health, with its mission of prevention and education, can informatively identify misinformation and provide accurate scientific facts about COVID-19. Prior research has shown that correction is an effective measure to counter health misinformation on social media (57). In the correction approach, misinformation on social media is identified with a timely and credible explanation of why the misinformation is false, and the explanation is repeatedly reinforced through multiple corrections (57). A collaborative taskforce of public (40) • Consuming curry powder and garlic (40) • KaKaoTalk has been cited as the main source of disinformation by a former staff from Korea's National Assembly Defense Committee (41). • The impact of the rumor in the Korean American community was instrumental, leading to decline of diners and sales at Korean restaurants (38). Identification and correction of misinformation across social media is not enough to end the COVID-19 infodemic. In concordance with the deficit hypothesis (59), public health also needs to improve eHealth literacy among the public to ensure that people are thoroughly assessing the scientific quality and accuracy of the information they encounter on social media (60). eHealth literacy is "the ability to seek, find, understand, and appraise health information from electronic sources and apply the knowledge gained to addressing or solving a health problem (61)." A cross-sectional survey conducted in China found that eHealth literacy is a significant predictor of preventative health behaviors during the COVID-19 pandemic and that eHealth literacy has more impact on health-related behaviors than knowledge (62). Findings from this study suggest that improving and targeting eHealth literacy is critical to manage the COVID-19 crisis (62). In order to improve eHealth literacy and ultimately mitigate COVID-19 misinformation among Asian Americans, in-language training and materials need to be made more available. Through these trainings and materials, Asian American community members can be provided with better tools to decipher between reliable, accurate scientific information on COVID-19 and misinformation on social media platforms. Likewise, public health officials need to improve media literacy in Asian American communities, informing and guiding community members on how to analyze and evaluate the quality of diverse media platforms, including social media, films, newspapers, television shows, etc., for accurate scientific information (63). With improved eHealth and media literacy, Asian Americans can utilize social media to seek accurate health information, which prior studies have shown health education on social media to be effective in addressing prevalent health issues such as hypertension (64), Hepatitis B (65), and obesity (66) among Asian Americans. Lastly, as Asian American is a diverse racial group, it is necessary to hold additional culturally-and linguistically-concordant virtual webinars and create educational materials tailored to addressing the unique COVID-19 misinformation experienced by all Asian American subgroups. In addition, the public health response can utilize and leverage community health workers (CHWs), who are vital members representative of the community, to identify COVID-19 misinformation and to inform community members of ways to be aware of misinformation and provide accurate in-language information. Furthermore, many forms of online misinformation, particularly conspiracy theories, are due to complex psychological mechanisms that need to be further examined to thoroughly combat and contain the infodemic in Asian American and other communities. People are also more likely to have strong conspiracy beliefs that undermine any scientific guidance when they are experiencing high levels of anxiety and uncertainty (67). Research has shown that conspiracy theories are influential and often inimical to health (68), which is concerning during the COVID-19 pandemic as preventative health behaviors are instrumental to abating and containing the spread of COVID-19. Recognizing the prevalence of conspiracy theories and its adverse impact on COVID-19 vaccination acceptance and uptake ( 69) is necessary to develop targeted strategies and interventions to encourage vaccination among low-income and communities of color that are at higher risk for COVID-19. Interdisciplinary research examining the psychology of misinformation can inform public health efforts to stop the spread of COVID-19 misinformation (70). --- CONCLUSION Social media has inevitably promulgated the transnational exchange of COVID-19 misinformation during the pandemic, prompting academic and community action to combat COVID-19 misinformation on social media. We speculate that Asian Americans are likely to utilize social media to receive COVID-19-related health information because they are able to effectively communicate with family and friends, particularly due to the dearth of COVID-19 information by the federal government in diverse Asian languages. We recommend multi-sectoral efforts between public health officials, community leaders, researchers, and social media companies to identify and mitigate health misinformation on social media, as well as the inclusion of CHWs to educate community members on misinformation and provide accurate scientific information. Improving eHealth and digital literacy among Asian Americans could also aid in the evaluation of identifying misinformation and seeking trustworthy and reputable websites. More research on the psychological mechanisms behind conspiracy theories and misinformation could led to effective health promotion communication strategies and interventions that improve acceptance of scientific evidence and public health guidance needed during public health emergencies. Research is also needed to address the origins and mediators of susceptibility to COVID-19 misinformation. For example, the exploration of how social media platforms, usage patterns, and usage of social media for health information by disaggregated Asian American subgroups in the U.S., as well as by other racial and ethnic groups is needed. Increase research efforts are necessary to quantitatively and qualitatively understand the social, economic, and health impacts of COVID-19 misinformation in Asian American communities. The consequences of misinformation on health behaviors in Asian American communities will likely span beyond the impacts of the COVID-19 infection and will require culturally-and linguistically-concordant interventions to address health disparities. --- DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s. --- AUTHOR CONTRIBUTIONS SC, SA, SY, and SK conceptualized the manuscript. SC and SA wrote the first draft of the manuscript. L<unk>, SY, and SK provided critical review of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version. Author Disclaimer: The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the funders. --- Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher's Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Copyright <unk> 2022 Chong, Ali, <unk>oàn, Yi, Trinh-Shevrin and Kwon. This is an openaccess article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Social media has been crucial for seeking and communicating COVID-19 information. However, social media has also promulgated misinformation, which is particularly concerning among Asian Americans who may rely on in-language information and utilize social media platforms to connect to Asia-based networks. There is limited literature examining social media use for COVID-19 information and the subsequent impact of misinformation on health behaviors among Asian Americans. This perspective reviews recent research, news, and gray literature to examine the dissemination of COVID-19 misinformation on social media platforms to Chinese, Korean, Vietnamese, and South Asian Americans. We discuss the linkage of COVID-19 misinformation to health behaviors, with emphasis on COVID-19 vaccine misinformation and vaccine decision-making in Asian American communities. We then discuss community-and research-driven responses to investigate misinformation during the pandemic. Lastly, we propose recommendations to mitigate misinformation and address the COVID-19 infodemic among Asian Americans.

Introduction The current era of globalization allows women to work or be said to be professional workers. Career women are women who because of their education, expertise and skills are involved in professional activities in the public sector such as companies and offices (Alifiulahtin, 2017). This situation can be interpreted that not only men are able to work in the public sector, but women as well. Up to 39.52% or 51.79 million people working above the age of 15 are women, according to data from the Central Statistics Agency in 2021. The data also shows that 10.48% of female workers are professionals and technicians, and 8.65% are jobs in the service sector (Rizaty, 2022). In fact, women who have careers outside the home experience more conflict between work and home life than men because women are housewives in addition to their work duties (Anapratiwi &;Handayani, 2013) The increasing financial needs of the family are a strong reason why mothers choose to work. This condition places a dual role on working mothers in sharing time related to work and family, so work-life balance is very necessary for working mothers to balance time and attention between work and family (Putra, 2021) This condition shows that a person feels comfortable at work and has enough time for family and work. But in individuals who are undervalued in work-life balance will have negative effects, such as feeling bad at work, work matters require more time and attention, so family affairs are not well taken care of (Wardani &;Firmansyah, 2021). Women with low work-life balance will have poor quality communication and interaction with their children (Apriani et al., 2021) This phenomenon can be seen in working mothers who start preparing household needs before going to work, and after returning home from work return to take care of household chores (Ramadhani, 2016) Several previous studies concluded that linking work-life balance to reduce conflict can occur when people are able to balance work and family affairs (Az-zahrah, 2021). This situation can be experienced by all working women, including working mothers with special children (hereinafter referred to as Children with special needs). This means that for many mothers who work and have children with special needs, there is a lot to do. One of them routinely accompanies children through therapy. This condition requires working mothers with special needs to increase focus and time between work and family (Desiningrum, 2017) The limitations faced by children with special needs not only complicate the child's growth and development stage, but also those who take care of them. Parents, especially mothers, must spend most of their time caring, nurturing, and caring for their children with a history of children with special needs (Anggreni &;Valentina, 2015) This situation is a challenge for working mothers because they have to dedicate their time, energy and mind to work and take care of children and family life. These two areas are top priorities, for example, the obligation to work as a career woman and family life, especially parenting issues, must also be a priority for mothers (Erawati, 2016) Raising children is not an easy activity, let alone taking care of children with special needs. Children with special needs need special attention to their education, both financially and with other costs (David Wijaya, 2019). Children with special needs need different treatment from children in general, so mothers definitely need enough time to fulfill their parenting and work duties (Hasdianah, 2013) Especially if working mothers who have children with special needs. The presence of parents is very important in caring for children with special needs, especially mothers, because parents are considered the closest people to their children (Nura &;Sari, 2018) Career and family are two important things for adults that are inseparable from each other. People in a relationship in particular need to find a balance, whether they like it or not, to avoid conflicts of interest between work and family. This role conflict can result in disorders such as depression, emotional and physical exhaustion, and alcohol consumption for individuals (Nohe et al., 2015). For companies, having employees who experience role conflicts can result in absenteeism, affective commitment to the company and turnover intentions (Vignoli et al., 2016) The idea of compensation is at the core of Alfred Adler's theory of personality, which views every human endeavor as a reaction to feelings of inferiority. The theory included in the WLB explains that these efforts are designed to overcome negative experiences in one area through efforts to increase positive abilities in other fields. That is, dissatisfied employees concentrate more on family life than work life, or vice versa (Submitter et al., 2020) Based on the description of the perspective of the theories above, it can be understood that the theoretical perspective of conflict and compensation can be used to develop the concept of work-life balance which initially only makes balance by focusing on work-family balance. But it developed into a concept of balance that is not limited to family members only, but also covers other areas of personal life. --- Method This research uses qualitative research type with instrumental case study design. The reason researchers use qualitative research type with instrumental case study approach to uncover work-life balance problems in working mothers who have children with special needs, because instrumental case studies are research methods that focus and pay special attention to a case in depth and detail through in-depth information collection and analysis. The instrumental case study type research approach allows researchers to collect different data from different sources, in the form of observation, interviews, or research on certain documents, works and products relevant to the case under study. --- Result and Discussion --- Result This study involved two mothers as participants who had children with special needs. The first participant was LS mother, age 47 years, widow status (divorced dead), had one child initials DD, age 22 years, female gender, special needs suffered by DD is Down syndrome. Mrs. LS works as a PAUD teacher in one of the landfills in Surabaya, East Java with a working period of 20 years. Based on data analysis, it is also known that the factors behind Mrs. Mrs. LS and Mrs. LL continue to work despite having children with special needs. The reason Mrs. LS keep working is because they want to help their husbands to meet the family economy and DD therapy costs. According to LS's mother and Mrs. LL, it is not easy to divide attention for work and caring for children with special needs. But all of that can feel easy because Mrs. LS have a special notebook to record all the needs needed by DD and also make detailed work schedules to facilitate LS mother's work activities. --- Discussion Every morning Mrs. LS start the activity with morning prayers, taking care of household needs such as cooking and preparing other needs needed by her husband and children. This routine is also done by Mrs. LL as a form of balancing time for work and taking care of the household. Mrs. LS and Mrs. LL can still spend family time in the midst of busy activities in the office. Commitment of LS and Mrs. LL's involvement in work and parenting. The results of this study found that Mrs. LS from the beginning knew and understood the condition of DD who suffered from Down syndrome were committed to caring for and paying special attention to DD but also continued to work professionally. The form of involvement of Mrs. LS in caring for DD and family activities is to routinely check DD's health every month and can also take time to gather with family at home. The study also found that Mrs. LS and Mrs. LL were happy working and could return home feeling happy. Mrs. LL feels satisfied at work because there are still other activities that can be done besides taking care of children. Meanwhile, Mrs. LS feels comfortable still working and still has enough time to take care of DD. In line with Sirgy and Lee's opinion (Sirgy &;Lee, 2018) that the concept of worklife balance as a high level of interaction and engagement and minimal conflict between roles in the domain of work life and life outside work. Mrs. LS and Mrs. LL both realize that it is not easy to live a working life while caring for children with special needs, but all of that can feel a little light when parents who have children with special needs can immediately reconcile with themselves and accept the child's condition. In addition, cooperation with the husband is also needed in terms of childcare. The stage where individuals can balance their professional commitments and family responsibilities, high work-life balance for the individual (Rahmayati, 2021). Work and family conflicts occur when the demands of work life cause a problem in meeting the demands of life in the family. Mrs. LS decision to continue working because she had the support of her husband and family. The support of her husband, family and people around the neighborhood is a consideration for LS's mother to continue working while caring for DD who suffers from Down syndrome. Meanwhile, the factors that influence Mrs. LL to continue working are because of their commitment to themselves and parental upbringing which emphasizes that girls when they are married must continue to work and there is support from their husbands. This is according to Hooja and Sen (Sen &;Hooja, 2018) that family support is very influential on a person's work-life balance level. High family support tends to be associated with a high level of work-life balance in a worker. A study by Sample (2013) entitled "Work-life balance of parents of special needs children" concluded that parents of children with special needs need more work-life balance (Sample, 2013) (Sample, 2013). Mrs. LS and Mrs. LL always try to work professionally, since knowing the condition of the child suffering from special needs, the step taken by Mrs. LS and Mrs. LL is to commit to good parenting and work together with the husband to take care of household needs and childcare. In line with research conducted by Fitriyani and Humaedi (Fitriyani et al., 2016), it is explained that working mothers tend to have difficulty dividing time and can have an impact on meeting children's basic needs. These basic needs include physical-biological needs, affection, emotions, and stimulation needs. These basic needs are very important for children, especially for children with special needs. Based on the explanation above, the reason why Mrs. LS and Mrs. LL continued to work, it is known that both participants received support from their husbands, families and people around their environment. In addition, Mrs. LS and Mrs. LL have their own ways of balancing time for work and childcare. That working women are expected not to let go of responsibilities in the domestic sector, namely as wives who take care of the household and the public sector as workers (Novenia &;Ratnaningsih, 2017) --- Conclusion Both participants had different ways of improving work-life balance. The first participant, Mrs. LS balanced time and attention to take care of work and take care of children, namely recording children's needs and office work needs in a daily notebook that was used to make it easier for Mrs. LS to carry out activities of daily life. Mrs. LS often take their children to work for the reason that they can always divide their time and attention for work and also to take care of children. While in the second participant, Mrs. LL balanced time and attention to take care of work and childcare, Mrs. LL admitted that her time was spent more at work because she worked from morning to evening. Mrs. LL can only take care of her child when she gets home from work. Advice for future researchers, in order to further explore how much influence working mothers have on the growth and development process of children with special needs

Introduction: Career and family are two important things for adults that are inseparable from each other. People in a relationship in particular need to find a balance, whether they like it or not, to avoid conflicts of interest between work and family. Objective: This study aims to describe work-life balance in working mothers who have children with special needs. Method: This research wants to find out more about work-life balance in working mothers who also care for children with special needs. This study used a qualitative research method with an intrinsic case study approach and involved two participants. The data collection process was carried out through interviews using credibility techniques, namely member checking.The results of this study found that the factors that motivated Mrs. LS and Mrs. LL to continue working were because they had the support of their husbands, family, and people around the environment. Conclusion: Mrs LS and Mrs LL both realize that it is not easy to live a working life while caring for a child with special needs, but all of that can feel a little easier because there is support from her husband and family in terms of taking care of the house and children's needs.

Introduction Successive economic crises occurring alongside the pandemic crisis (SARS-CoV-2 epidemic) have led to a deterioration in the quality of life and working conditions in many member states of the European Union (EU), with an unwanted contribution to the growing areas of inequality between countries and groups of people. Children are more vulnerable to deprivation or social exclusion than other categories of the population in most countries, and the evaluation of the effects on the families in which these children grow up is a necessary condition for the foundation of social policies in the field. Family policies are an important pillar of national public policies and the most significant vehicle for governments to influence the living standards of current and future generations, having particular importance in meeting the Sustainable Development Goals. The analysis of existing public policy documents in Romania that also consider the quality of life of families with children in care (National Strategy on Social Inclusion and Poverty Reduction for the period 2022-2027, approved by Government Decision 440 of 2022, National Strategy for the Protection and Promotion of Rights of the child "Protected children, safe Romania" 2022-2027, project on the approval circuit, the National Strategy for the Sustainable Development of Romania 2030 approved by Government Decision No. 877/2018 and amended by Decision No. 754/2022) reveals an interest in ensuring a minimum standard of well-being in families with children in care, but there are still points of vulnerability that come mainly from the area of monitoring indicators, which are insufficiently developed to assess the quality of life of families with children. Ensuring access to free quality education is a decisive factor that influences the quality of life of families with children in care. Data between 2013 and 2021 reveal a deficit in the allocation of resources in this area in Romania, with public expenditure on education rising to 3.2% of GDP in 2021, 1.6 percentage points below the EU27 average of 4.8% of GDP for the same year. Children who grow up in poverty or social exclusion find it difficult to adapt to the demands of the education system, to enjoy good health, and to reach their full potential later in life. They also face a greater risk of becoming unemployed, having low income, or being socially excluded as adults. Children whose parents have achieved a lower level of education are more exposed to the risks of material deprivation or social exclusion compared to children whose parents have achieved a higher level of education. Education influences the type of jobs a person can access, and the likelihood of poverty or social exclusion increases as educational level decreases. In Romania, in 2022, 80.7% of children living in the same household as their parents, whose highest level of education achieved was mostly lower secondary (ISCED 0-2), were at risk of deprivation or social exclusion. The substantiation of public policies-aiming to reduce poverty among children-needs measurable data regarding the environment in which these children grow up (the family). Studies concerning the role of the family in the upbringing of children affected by extreme poverty reveal that the size of the family structure, which includes the socio-demographic characteristics of the parents, migration, and educational background, has frequently been associated with the nutritional status of children [1]. Studies on quality of life in families that have children in care have mostly focused on the area of families that have disabled children [2,3]. The quality of life (QoL) model proposed by Schalock and Verdugo in 2002 [4] is the most widely accepted QoL model by the international community in the field of disability. The model was originally developed for individuals with intellectual and developmental disabilities, but later it was adapted and used to assess QoL for other disadvantaged groups of people [5]. The model was also adapted for education, with Quality of Life Index for Inclusive Education (QoLI-IE) being a tool designed to monitor the extent to which educational cultures, policies, and practices meet the expectations and needs of children (3-17 years old) with intellectual and developmental disabilities in terms of access, participation, learning, and development [6]. The QoLI-IE is also a tool that can be used for monitoring in the framework of the European Child Guarantee [7]. This tool also supports action guidelines for policy makers, based on the evidence gathered from persons, schools, communities, and social systems [8]. There are also very few studies focused on quality of life in transnational families [9]. The studies completed on the impact of Romanians' quality of life related to emigration [10,11] indicate a general increase in the living standard of those who went to work outside the country's borders, reflected in financial situation, job satisfaction, access to quality medical and educational services, and diverse leisure opportunities. However, the quality of family life (QoLF) of those who remain in the country (children, parents) should be researched, taking into account all dimensions of QoLF (family interaction, emotional and physical well-being, and support related to the care of dependent family members). The migration of parents abroad has profound emotional consequences for family members, especially children [12][13][14]. Quality of life related to children's education in transnational families is influenced by the way in which educational policies are developed and implemented. The studies that evaluate the problems related to the education of children with parents who have left for work outside the borders of Romania are quite poorly developed. On this topic, the present article contributes empirical data to the foundation of some policies in the field. The research question that guides the entire analysis of this article aims to identify some possible solutions in the field of educational policies to ameliorate the quality of life of transnational families with children left in the country of origin. The objectives of the research were as follows: (1) to explore the opinions of public and private providers of social services concerning the effectiveness of public measures and programs aiming to counteract and to alleviate the social exclusion of children left behind, and (2) to identify solutions to improve the quality of life of transnational families. --- Methodological Approach 2.1. Data Source The authors used empirical data from the project "Effects of migration experiences on families, children, and communities left behind: methods of assessment and strategies for mitigating the risk of social exclusion" (PN 19130203 financed by the Romanian Ministry of Research, Innovation and Digitalization under the Nucleu Program Inov Soc implemented by The National Scientific Research Institute for Labour and Social Protection-INCSMPS). --- Methods and Target Groups A mixed methodology was developed by the authors within the previously mentioned project, using a research methods/tools focus group guide and sociological investigation based on a face-to-face questionnaire. The topics addressed in the focus group were as follows: • The mission of public and private organizations active in the field of social assistance for families and children whose parents migrated for work; The analysis carried out in this article only covers the information collected for the dimension "evaluation of measures and programs aimed at preventing and combating the social exclusion of children whose parents have gone to work outside Romania's borders" within the focus groups and for the dimension "possible solutions to improve the quality of life of transnational families, with a focus on the people left in the country" within the questionnaire-based survey. The research was carried out in Macro-region 2 of Romania, which includes the North-East and South-East development regions. These regions are among the most affected by the phenomenon of migration of parents for work. Fifteen representatives of public and private organizations active in the field of social assistance for transnational families with children left behind participated in the two focus groups organized, and 804 questionnaires were completed with parents/grandparents from families with children up to 17 years old and at least one migrant parent (Table 1). --- Methods --- Target Group No. --- Focus group (FG) Representatives of public and private organizations active in the field of social assistance for families and children whose parents migrated for work. --- Sampling and Recruitment of the Participants A non-probability sampling method/quota sampling was used for the recruitment of participants to focus groups. All participants expressed their agreement to participate in the research. The following steps were completed to recruit and to select participants for the qualitative research: --- • Preparation of a list with the contact details of the social assistance public authorities at county-level territorial structures, NGOs that have interventions for the target group of the study, and educational institutions with projects dedicated to the target group of the project in the two development regions (North-East and South-East); --- • In each region, a representative from each social assistance public authority at county level and NGOs in the area was requested; --- • Along with the invitation to nominate a representative for the online focus group, the authorities identified for the focus groups received an official address from the Romanian Ministry of Labour and Social Solidarity regarding the objectives and the ways of using the results obtained from the group discussions. The focus groups with representatives of the authorities/entities involved in the management of the problems of these families were carried out on 19 July 2021. The same type of non-probabilistic random sampling/quota sampling was also used for the sociological investigation based on a face-to-face questionnaire. The reason why a non-probabilistic sampling method was used in the case of the survey derived from the lack of identification data regarding the basic population of parents/grandparents caring for children with migrant parents. First introduced in experimental research, the quota sampling method involves non-probability sampling with the strict application of proportions or controlled quotas [15,16] and uses the selection of a sample after the population has been divided into subgroups [17]. The survey was conducted between June and July 2022 and used a standardized face-to-face questionnaire applied using a TAPI (Tablet-Assisted Personal Interview) technique. Informed consent from all participants in the survey was obtained and, following the field research, data were analysed using statistical software (IMB SPSS Statistics 20). --- Results --- Quality of Life Related to Children's Education in Families Affected by Migration Children left behind in Romania by parents that decide to work abroad face many problems; therefore, quality of life related to the education of children undergoes a series of changes, most of the time in a negative sense. From the perspective of the authorities and institutions involved in managing the problems of families with children whose parents migrated for work, the difficulties encountered by these families are mainly emotional, as well as related to the ability to manage some events that appear in the life of the child left at home, either alone or in the care of other family members, in relation to school, the judicial system, the medical system, or different public authorities. In the case of children cared for by grandparents, communication problems arise due to the differences between generations. Intergenerational communication could become even more deficient during adolescence. "This is a view of life... the grandfather is a grandfather, the young man, at 13-14 years old, already has major problems and does not find that type of support he needs." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) Children with both parents working outside the country's borders, or children from families where the sole breadwinner left, are at the greater risk. Also, if the parent who migrated for work is the mother, it is generally found that these families are more vulnerable and exposed to a decline in the quality of life of the children. Participants in the focus group discussions emphasized the need for training parents to acquire parenting skills or to participate in parental counselling. From this point of view, an important role also belongs to workers in the field of social assistance, but also to educators and teachers. "I tell you from practice... there are many parents who do not know the normal physical and emotional development milestones from that age. And we have to tell them what is a normal behaviour, where the limits are, where they should intervene. This psychoeducation is very important." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) "... a school of parents, for those who are the supporters of parents who have gone abroad..." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) Children left in the country, either alone or in the care of a family member, often experience a sense of abandonment. Emotional problems are diverse and become more and more complex as children enter the adolescence stage. In some cases, behavioural problems are added to these emotional problems. Chain reactions result due to the lack of support, valorisation, and parental support. The main consequences of parents going to work abroad occur both in children and other family members who remain in the country, and the negative consequences are much deeper and longer lasting compared to the impact of the positive consequences (Table 2). The authorities/organizations involved in managing the problem of transnational families consider it important to develop programs in order to increase the support for families, but especially to help children whose parents are working abroad. For all participating entities in the focus group discussions, support for children and families was the main goal, given the context that there is a high degree of awareness of the complexity and diversity of issues that both families and children may face following the labour migration of parents outside the country's borders. "...there are many risks and it is natural that this issue of children with absent parents should be in our attention and to represent a priority for us." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) The available support that the entities participating in the focus group discussions offer to children and families includes the following: social services within community integration centres; emergency intervention services; mobile intervention teams; psychological counselling intended for children whose parents are working abroad, but also for the parents who stayed at home and have children in their care; and advice, including legal advice, for stay-at-home parents to prevent or manage difficult situations that may arise in the situation that one of the parents left to work abroad. "...but we are concerned also with the emotional, psychological part and what is happening with the mental health of the children and the family." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) "...and we try to focus on the psychological aspects. It is very important to prevent trauma, or if it occurs, to try to mitigate it." (Representative of entities involved in the issue of families with parents going to work abroad, North-East region, Romania) "My opinion is that parents' information campaigns should also be intensified, I would see them much more aggressive. They should be more aggressive because they [parents left abroad] don't understand the long-term consequences of parental separation...they don't understand that emotional needs are more important than material needs." (Representative of entities involved in the issue of families with parents going to work abroad, South-East region, Romania) The efficiency of these support measures depends on the existence of a local record of children left at home, and of families whose members intend to migrate for work. The second element of effectiveness for such measures is the existence of early-intervention services to support families and children whose parents intend or are already going to work abroad and who have been identified at the local level. Last but not least, the cooperation and collaboration of all interested parties (educational institutions, social assistance institutions, health institutions) at the local level is necessary to respond to the problem of children whose parents are working abroad. A close collaboration between educators, psycho-pedagogues, school counsellors and teaching staff in general, social workers, local authorities, family doctors, and priests is beneficial. --- Measures/Programs: Assessment and Possible Solutions in the Field of Educational Policies to Enhance the Quality of Life of Transnational Families The main objective of public programs should be to limit the negative effects of separation and to protect children's rights (reducing the risk to which children may be exposed). If both parents and the legal representative are missing, children experience a limitation of their rights to education, health, and legal protection, but they may also experience restrictions in participation in recreational activities (e.g., camps, etc.) The participants in the qualitative research (focus group discussions) were asked to give their opinion on the measures/programs intended to prevent and alleviate the social exclusion of children with parents working abroad, initiated by the Romanian Ministry of Education (MEC) and the Romanian Ministry of Labour and Social Protection (MMPS). It was the inter-institutional Romanian Working Group dedicated to children with parents left to work outside the country's borders (constituted in 2016) who formulated a series of proposals intended to support this category of children (Table 3). We note from the opinions expressed during the focus group discussions that these measures only partially achieved their goals, as a small part of the participants were aware of these proposals. The participants in the focus group discussions from the South-East region also mentioned other measures intended for children with parents that have gone to work abroad (apart from those initiated by MEC and MMPS), initiated/implemented by different entities, for example, the training of specialists at the level of territorial administrative units intended for NGOs that respond this issue. National and international studies on transnational families and the effects of parents' migration on children left at home emphasize the need to adopt complementary measures to support children and family members left behind. --- Measures/Programs in Educational Field Measures/Programs in Social Assistance Field Encouraging the assumption of a "big brother/big sister" mentor role by colleagues under the guidance of a teacher/conductor ("buddy system"). Strengthening the local capacity to provide social assistance services by ensuring human resources (one full-time employee each) in the field of specialized social assistance (preferably social workers). Twinning of schools, followed by camps abroad (financed by the community of residence of the parents), to stimulate the school performance of the children remaining in the country. Professional training of staff in the field of social assistance. --- Continuous communication between the school and the absent parents (e.g., minutes of meetings with parents, communications of various decisions made in the classroom regarding children and parents, Skype meetings with parents). Consolidation of local capacity to provide psychological assessment services, psychological support (individual and group), psychological counselling, and referral to specialists (psychotherapists, clinical psychologists, psychiatrists). Maintaining permanent contact with the children's family for issues related to the psychological condition of these children. The training of teaching staff in managing various aspects related to children with parents that have left to work abroad (based on "best practices" from NGOs). Source: INCSMPS database. The results of the sociological investigation based on face-to-face questionnaires confirmed the conclusions of previous research and support the need to adopt measures in different areas. In the field of education, supervision in completing school homework was considered necessary by 71.8% of parents and 64.5% of grandparents. Education on topics like the negative effects of drugs, alcohol, and other toxic substances was rated as an area where support is needed by most parents and grandparents. In urban areas, the need for education about the negative effects of using drugs, alcohol, and other toxic substances (expressed by 74.3% of parents and grandparents) represents the most necessary measure needing to be developed, followed by support to avoid dropping out of school (with 74.8%) and school counselling (with 74.3%). The need for education about the effects of using drugs, alcohol, and other toxic substances also represents the most necessary measure needing to be developed in rural areas, with 59.1% of parents and grandparents who participated in the sociological survey from the two development regions of Romania mentioning the necessity of this measure (Table 4). --- Discussion --- Limitations of the Research Our results must be analysed in the context of our methodological limits: the research only covered Macro-region 2 of Romania and it used a non-probabilistic sampling method. These limitations are mainly caused by the lack of administrative data regarding the investigated population. However, this study contributes to a better understanding of the main problems that may arise in the education of children from transnational families and the effects of these on the quality of life of all family members. --- Suggestions for Policy Measures There is a growing interest in the situation of children whose parents leave the country for work, due to the negative social, economic, educational, and medical consequences [18]. This concern is transposed at the level of public policy in the implementation of measures aimed primarily at reducing the adverse effects of parent's migration on their child/children [18]. Current research [18][19][20][21] emphasizes that migration influences emotional processes at the family level, directly affecting the psychological and physical state of all family members. Migration is a phenomenon that affects both the person who migrates and their family. The scientific literature [22] emphasizes that interventions are needed in the form of public policy measures to mitigate the negative consequences of such an event in the life of the family left behind in the country of origin. Démurger [20] points out that the effects of parents' migration for work translate into negative effects on the school performance of children who remain in their country of origin. This is also confirmed by the results of other studies [23][24][25][26][27]. Qualitative results from focus group discussions with representatives of public and private Romanian organizations active in the field of social assistance for families and children whose parents migrated for work indicate a major need for intervention through support programs for those families that have at least one parent who has gone to work outside the country's borders. This is also confirmed from the perspective of future beneficiaries of such programs through sociological survey results, which indicate a need for measures in the area of educational policies in order to avoid dropping out of school in percentages of 74.8% in urban areas and 57.6% in rural areas. In this sense, measures to prevent school dropout are necessary, and therefore the collaboration between institutions with an educational role and the family is a prerequisite. Similarly to other previous studies [28,29], the negative effects of parent migration on the emotional wellbeing of children left behind were also highlighted, despite improvements in living standards. For children coming from families that have at least one parent who migrated for work outside the country's borders, the data collected through the survey also indicate a need for the development of program measures in the following areas: supervision in the completion of school homework, support for the development of individual learning skills, and education about the negative effects of using drugs, alcohol, and other toxic substances. Moreover, the representatives of public and private Romanian organizations active in the field of social assistance for families and children whose parents migrated for work indicate difficulties in managing relationships within transnational families (see Table 2), hence the need for intervention in the areas mentioned by the participants in the sociological survey. These effects were reported also in other studies [29][30][31] and draw attention to the implication of this phenomenon for social services providers, as well as for policy makers. In certain situations, the person responsible for the child/children left behind can have different views on education, and this may generate conflicts both with the parent who went to work abroad and with the child/children left in care. Data from our qualitative research (focus group discussions with authorities designated to manage the problems of transnational families) revealed the need to develop parenting skills training courses or parental counselling courses. The educational system should be prepared in this sense with the necessary human resources, but also with logistical resources, especially in rural areas where Romania faces a shortage of qualified workforce. The partnership between schools and families must be promoted more strongly in the case of children from transnational families. The National Action Plan for the implementation of the Children's Guarantee, adopted by memorandum on 31 October 2023, highlights that children from transnational families are affected by vulnerabilities related to the lack of continuity of care and access to various public and essential services, but mainly education. Against the background of the precarious dimension of the phenomenon of children whose parents have gone to work outside the country's borders, caused by the shortcomings of the registration system of these children within public social assistance services and the educational system, the planning of support measures/programs remains extremely difficult. Children who temporarily migrate outside the country or whose parents work abroad are covered by two European documents (the Child Guarantee [32] and the EU Strategy for the Rights of the Child [33]) in the category of children in precarious family situations. National data show that in June 2023 there were 71,152 children in Romania with one or both parents going to work abroad [30,34], with a national distribution showing that the counties in the North-East development region had the most cases. However, the dimensions of the phenomenon are very likely underestimated given the number of Romanians who emigrated, many of them temporarily for work. --- Conclusions The results show that the quality of life related to children's education is affected when a parent goes abroad to work. In addition to the education of children being affected, the influence is visible at the level of the socio-emotional and behavioural development of the child. In order to improve the quality of life related to children's education in families negatively influenced by migration, the main measures necessary to be implemented have the following aims: the correct collection of data on children whose parents work outside the country's borders; the development of support services for the prevention of various negative phenomena associated with education (school dropout, behaviour problems within student groups); the development of remote courses dedicated to parents to strengthen the school-family partnership; and ensuring the human resources necessary to implement an effective educational act in schools, with a special focus on children from transnational families. Future research should consider the development of indicators to measure quality of life related to children's education in families affected by migration. This article offers support to decision makers in the field of educational policies for effectively managing the real problems of children from transnational families, highlighting, at the same time, the need to substantiate the policies in the field on studies designed to evaluate these policies. --- Data Availability Statement: The data are unavailable due to privacy or ethical restrictions. --- Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. --- Conflicts of Interest: The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

The problems of children from Romanian transnational families represent a current issue and one of great interest in the field of public policies that need to be developed in Romania in order to offer solutions that will lead to an increase in the quality of life within these families. This article reflects, based on the results of a survey among parents/grandparents who care for children from transnational families and two focus groups with authorities/organizations with a role in managing the problems of transnational families, the main issues faced by children from these families within education. The results obtained indicate an increased need for intervention in the following areas: support to avoid school dropout; education regarding the negative effects of drug, alcohol and other toxic substance consumption; and supervision in the completion of school assignments. All these problems influence the educational course of children and consequently the quality of life of all family members. This article offers support to decision makers in the field of educational policies to effectively manage the real problems of these families, highlighting the vital role of substantiating public policies based on scientific studies.

Introduction Population ageing has been accompanied by a shift in disease profile to non-communicable diseases (NCD), increased levels of disability, and an increasing loss of physical and cognitive functioning. Most high-income countries have accepted the chronological age of 65 years as a definition of 'elderly' or older person, but like many westernized concepts, this does not adapt well to the situation in Africa. As a result of the changing in legislation via the Social Assistance Amendment Act, 6 of 2008, South Africa aligned itself with the United Nations definition of the 'older persons' as all persons over the age of 60 years. [1][2] South Africa has one of the most rapidly ageing populations in Africa despite the demographic impact of the HIV/AIDS epidemic. The population aged 60 years and older numbered over 4.5 million in 2016, thereby accounting for just over 8% of the South African population [3]. It is expected that, by 2025, the proportion of the older population will increase by 10.5% to reach 5.23 million [4][5]. The latest estimates are that three-and-a-half million South Africans (about 6% of the population) have diabetes, and there are many more who are undiagnosed. This number is anticipated to grow by 30% by the year 2030. [6][7] According to the WHO Global Report on diabetes in 2016, the combination of increasing prevalence of diabetes and increasing life expectancy in many populations with diabetes may be leading to a shift in the types of morbidity that accompany diabetes, such as cancers and cognitive disability. [8] Inevitably, this will place further strain on both healthcare resources and health providers. In South Africa, 80% of the population receive their health care through the government funded public sector. [9] In general, community health centres (CHC) and smaller primary care clinics are the older persons' first point of contact with the healthcare services. These are staffed according to their size and location and provide a comprehensive package of care. The CHCs are overcrowded and poorly resourced due to the multiple disease burden, leaving limited time for the front-line health workers to deal with the management of patients with diseases such as diabetes. [9] A qualitative study found that patients with diabetes in this setting were ill-equipped to play an active role in self-care due to their limited opportunities for education and counselling. [10] In this setting too, poor control of glycaemia and hypertension together with high levels of multimorbidity are commonly encountered. [11] Diabetes self-management practices (DSMP) form the foundation of diabetes care. These involve knowledge, skills and motivation as it requires, amongst others, adjustment of the diet, monitoring of blood glucose levels, where appropriate and an increase in physical exercise. [12,13] Sprague et al. found that the decreased priority given to patient education among older individuals, their support systems, and the healthcare community is a factor that negatively impacts their learning diabetes self-management. [14] A number of barriers associated with ageing reduce the older persons' potential for engaging with traditional self-management education programmes such as lectures/group sessions; for example, hearing and visual deterioration. Further self-management practice may be affected by reduced manual dexterity due to osteoarthritis which is common in this group. [15][16] Therefore, older people might have fewer resources available to manage their condition than younger people and will then have a higher need for self-management support. The loss of friends and family makes them more vulnerable to loneliness and social isolation. [17] Poor social support is associated not only with an increase in mortality, morbidity and psychological distress but a decrease in overall general health and well-being. [18] Several studies have found social support vital to SMP in people with chronic diseases. [18][19] Social support can be either emotional or physical. Emotional, social support is defined as the degree to which interpersonal relationships serve the purpose of providing emotional, informational or influential quality of life for the individual. [20] Physical support is defined as the forms and numbers of social relationships (marital status, the number of friends) and the degree of connection between these relationships (social network). Most frequently social support for persons with diabetes covers aspects of active support and emotional encouragement with taking medications, monitoring blood glucose, foot and eye care, following diabetic meal plans and increasing physical activity. [21] This study was undertaken in older people with diabetes in South Africa to examine their knowledge about living with and managing their diabetes; and to determine the relationship of social support, especially that of family and friends with their self-management. --- Methods --- Study design and selection of participants This cross-sectional study was undertaken in the Cape Town metropole where working class people receive care through a network of primary care clinics. Eighteen community health centers (CHCs) formed the sampling frame for the study based on them being served by Groote Schuur Hospital at the tertiary level. Four CHCs were selected based on population density of the older population in their drainage area as reported in the statistical censuses of 2011 [4]. As a diabetes register system does not exist in the metropole, the population of adults (>60 years) with diabetes within it is not known. Consequently, calculation of a representative sample size was not possible. A purposive convenience sample of (n = 406) was drawn by a random sampling technique from those who attended the four CHCs with approximately (100) participants from each clinic. The study was approved by the Human Ethics Committee of the University of Cape Town. (HEC REF: 21/2013). This research was conducted in accordance with the Declaration of Helsinki and written consent was obtained from each participant prior to their participation. --- Instruments We adopted a validated KAP questionnaire from P & T Journal, Medimedia USA, Inc, which was developed by Palaian et al., 2006 [22] and adapted it for this study. The questionnaire has been used in previous KAP studies among diabetics and has proven to be reliable, Cronbach alpha value 0.72. The questionnaire has been used in KAP studies in older adults in low and middle income (LMIC) including South Africa. [22][23][24][25][26][27][28]. The questionnaire adapted for this study had close ended and Likert scale questions. The questionnaire contained 7 demographic questions, 14 questions about lifestyle and whether they had received various aspects of care when they attended for routine clinical care in the past year, 11 diabetes specific questions relating to diagnosis and management/medication/monitoring, 11 questions on knowledge of diabetes, 6 questions on tangible and 6 questions on emotional social support (S1 File). We piloted the appropriateness, reliability and clarity of the questionnaire in a sample of 40 participants (10%). The English version questionnaire was translated to Xhosa and Afrikaans, the languages of the study area. Specifically, knowledge was measured using a 30-item questionnaire. There were 11 multiple choice format questions which assessed some basic knowledge of diabetes ranging from definition of diabetes, major cause, symptoms, complications, monitoring tools, risks of comorbid conditions such as increased blood pressure, a healthy lifestyle, a balanced diet, foot care, the use of medical therapy and eye screening. The respondent would score one point for each correct answer, and some questions had multiple correct options (e.g. symptoms). All points were summed to obtain a total knowledge score ranging from 0-30 from the 11 questions. For ease of interpretation the total score was transformed to give a range from 0 to 100. The Cronbach's alpha for the knowledge assessment tool was 0.67 (95% Confidence Interval: 0.61 to 0.73). The Diabetic Self-Management Practice (DSMP) was measured using a 6-item questionnaire with mixed type of response ranging from 0 (no practice) to 4 (more or always practice). The assessment took into account the respondent's smoking status, physical activity during the past seven days and following a diabetic eating plan. A respondent who practiced all the three: never smoked, did more exercise and followed a diabetic eating plan scored 2 extra points than those who did not. All points were summed to obtain a SMP score for each respondent, which ranged from 0-11. A score 6 we regarded as good SMP (Poor score (0-5), Good score (6)(7)(8)(9)(10)(11)). Cronbach's alpha for internal consistency was 0.44 (0.31 to 0.57) for the SMP tool. We used the Social Support subscale of the Diabetes Care Profile (DCP) [29]. It includes twelve questions related to family and friend social support by adding up the six tangible support variables (Follow meal plan, Take medicine, Take care of feet, Get activity, Test sugar and Handle feelings) and the six emotional support variables (Accept me, Feels uncomfortable, Encourage me, Discourage me, Listen to me, Nag me). The level of social support was assessed using a five-point Likert scale (strongly Disagree, Disagree, Neutral, Agree or Strongly Agree). For the positively worded variable, a score of Agree or Strongly Agree was coded as 1 and that of Strongly Disagree, Disagree, Neutral was coded as 0. For the negatively worded variable, the reverse coding applied. The raw scores ranged from 0 to 12. To simplify interpretation of the scoring, the scores were transformed to a score ranging from 0 to 100, i.e., a score of 12/12 was 100. A higher score meant greater social support. A score >50% that is 7/12 and above was considered as good social support and a score 6/12 or less was considered poor support. The Cronbach's alpha for physical support was 0.763 and for Emotional support was 0.623. The afore mentioned transformation of the scores for K, DSMP and SS done for a participant were as follows: K-index = K score / 30 x 100; DSMP-index = DSMP score / 11 100; and SS-index = SS score / 12 100. A review of clinic notes to record the HbA1c readings of the last three clinic visits was conducted to assess glycaemic control on patients who had completed the questionnaire. An overall HbA1c level was calculated as the average of the three separate readings. The interval between the readings varied for each participant, dependent on the appointment schedule. Glycaemia control was considered good, acceptable or poor when the HbA1c level was lower than 7.5 between 7.5-8.5% or greater than 8.5%, respectively. [30] --- Data collection Six fieldworkers were responsible for data collection using questionnaires and a review of medical records for HbA1c and blood glucose results, at the diabetic clinics in the four community health centres. They were trained in the administration of data collection tools, research ethics and an approach to interviewing older persons. Patients with diabetes aged 60 years and over attending the four clinics on the date of data collection were approached for participation in the study. The study was conducted from April to October 2015. Questionnaires were completed for those who agreed to participate. Signed consent to participate in the study was obtained before administration of the study questionnaire. The number of those who refused to participate in the study was not recorded. The fieldworkers were closely supervised by the research team to guarantee the quality of data collection. A random sample of ten questionnaires was checked for completeness and correctness. A total of 413 questionnaires were completed but seven were excluded from the analysis as the participants were below the age of 60 years. --- Data analysis Data was managed and analysed using SPSS Statistics version 23, [31] and Stata 15.1 (Stata Corp College Station, TX, USA). Categorical data was summarised as frequency and proportions, and continuous data as mean and standard deviation (SD). Unpaired t-tests and oneway analysis of variance (ANOVA) were used to compare knowledge, self-management practice and social support scores between two and three (or more) group variables respectively. Ordinal logistic regression was used to determine associations between outcomes (knowledge, self-management practice) with components of social support scale. Multivariable linear regression was used to evaluate the associations between outcomes (knowledge, self-management practice) and sociodemographic variables, glycaemia control, and social support (S3 and S4 Tables). Regression estimates were reported with 95% CIs. Statistical significance was indicated by p<unk>0.05. --- Results --- Descriptive and bivariate analysis Patients' demographic profile and clinical characteristics. Socio-demographic and clinical characteristics of the study sample are presented in Table 1. Of the 406 participants, 68.5% were females, 60.5% were living with a family member, and almost half were married. Two hundred thirty-three (57.4%) had less than 7 years of education. Unsurprisingly most of the participants were pensioners 374 (92%) and 348 (84%) reported family income R 1500 (US $107) per month. The mean duration of diabetes from diagnosis was eight years. Sixty-two percent were using oral therapy for glycaemic control, 31% combined insulin and oral agents and 7% insulin alone. The majority were taking medication for other conditions: 306 (75.4%) for hypertension, 138 (34%) for other chronic diseases and 61 (15%) for heart problems. The HbA1c was higher than 8.5% in nearly 185 (47.7%) of the participants, <unk>7.5% in 112 (28.9%) and 7.5-8.5% in 91 (23.4%).). Women were more likely than men to self-report experiencing low and high blood sugar (21% vs 11%, p = 0.016) and (53% vs 37%, p = 0.004) respectively. Knowledge, diabetes self-management practices, social support scores. A 'good knowledge' score was for those whose score for correct answers was 50%. Overall the level of knowledge was poor in 204 (50.2%) (Table 2). The deficiencies were particularly noticeable in relation to symptoms of diabetes and complications of diabetes and hypertension. There was a better level of knowledge about aspects of self-management e.g., a healthy diet and foot care; with 61.3% and 64.8% of respondents correctly answering questions about walking barefoot and daily foot inspection respectively (Fig 1A and1B). Similarly, 233 (57.5%) were assessed as having a good level of physical exercise and almost two-thirds (262 (64.5%)) of the participants were following a diabetic eating plan (Table 3). Over three-quarters of participants (312 (76.8%)) had good social support (Table 2). The majority of participants agreed that their family supported them to follow all aspects of selfcare included in the questionnaire and encouraged them in managing their diabetes. Only 74 (18.4%) reported that their family nagged them about their diabetes and 71 (18%) reported that their family felt uncomfortable about them because of their diabetes (Table 4). The mean diabetes knowledge, self-management practice and social support scores of the participants by socio-demographic and clinical characteristics are presented in S1 and S2 Tables. The mean knowledge score was significantly lower for a single (41.2 (SD = 2.9)) compared to married respondents (46.7 (SD = 12.5)) or those in a companionship (49.4 (SD = 12.7)). The mean knowledge score was greater for participants with higher education level and those who had experienced high or low blood sugar (p = 0.005 and p = 0.001 respectively), and those receiving medications for chronic diseases other than hypertension (p = 0.038) than their counterparts. The social support score means also differed significantly with living arrangements; it was lowest in those living with a spouse (67.7 SD = 21.0), intermediate in those living alone (75.0 SD = 21.5) and highest in those living in multiple household members (82.3 SD = 18.0 p = 0.001). On the other hand, there were no significant associations between diabetic knowledge, self-management practice and social support score with duration of diabetes, the type of medication used to treat diabetes, or receiving treatment for hypertension. The multivariable regression models. The ordinal logistic regression models of knowledge, self-management practice and the components of social support scale are given in Table 5. The table shows the effect of the K and SMP indices on the 12 individual SS components, which were measured on a Likert scale. Social support was positively associated with the self-management practice score for following a diabetic meal plan, taking care of feet, physical activity, handling participants' feelings about being diabetic and testing blood sugar, but not for taking medication. Family and/or friend emotional support (nagging, encouraging /reassuring and handling feelings about being diabetic) were positively associated with knowledge score. Multivariable linear regression analysis (S3 and S4 Tables) showed older age was negatively associated with knowledge (1: -1.893, 95% CI-3.754; -0.031) and higher income was positively associated with self-management practice (1: 3.434, 95% CI 0.797; 6.070). There were no significant associations between socio-demographic variables, HbA1c and social support with knowledge or self-management practice. --- Discussion In this study, half of the participants had poor knowledge about diabetes and its complications. Just under two-thirds were assessed as having a good level of physical exercise and two-third of the participants were following a diabetic eating plan. Three quarters perceived that their family supported them to follow all aspects of self-care management. Being in the high-income group was associated with good level of self-management practice. Finally, social support was positively associated with both knowledge and a number of self-care aspects. The deficiencies noted in the participant's knowledge relating to the complications of diabetes and hypertension are alarming. Education of patients including those with diabetes is essential for self-management. It suggests in this study that whatever diabetes educational opportunities participants, particularly the older group, have been exposed to, have not been effective. There are many potential reasons for this. For instance, the high patient numbers and multiple disease burden in primary care clinics, are likely to negatively impact on the time available for patient education by health promoters, nurses or doctors [32]. Other factors to be taken into consideration include lack of attendance at educational sessions when they take place, communication barriers, such as poor hearing, lack of concentration, inability to engage with the material presented and use of didactic modes of communication [33]. However, the participants seemed to have a better knowledge of self-management practice such as foot care and healthy eating. Whether this is because these messages are practical and easier to convey or that the information comes from multiple sources and not only health care workers is uncertain. To date, there is a scarcity of evidence regarding diabetes self-management education and support in older adults [34]. Some studies that included older persons suggest that this group needs diabetes self-management education that stresses problem-solving skills rather than "rules" to follow [35]. For example, The Diabetes Education and Self-Management Ongoing and Newly Diagnosed (DESMOND) educators observed that older persons contributed to the group and brought valuable experience, but that they may have required a different approach at times [36]. However, no specific examples of such approaches were given in their study. Sinclair et al. reported that older people benefitted more than middle-aged people from a highly structured group diabetes self-management education intervention with embedded cognitive behavioural strategies compared to standard group education or individual sessions with dietitians and nurse educators [37]. As older persons may have difficulties concentrating and understanding abstract concepts, there is a need for educational material to be provided in the form of simple messages, delivered in a style that engages the person with diabetes and is personalized to their needs, with the emphasis on what they need to know, rather than all there is to know about diabetes [38,39]. Notably, these concepts to enhance knowledge and self-management practices are not unique to the older person with diabetes and are relevant in all societies. Income and financial issues are possible barriers to optimum self-management for many older diabetic patients because of the costs of blood glucose testing, medication and following diet recommendations [39]. Management of chronic disease require support from a patient's network. We found that almost 75% of participants perceived that their family supported them to follow all aspects of self-care management. Earlier studies have shown that social support and social networks influence health behaviours and health outcomes [40][41]. For example, a study of family behaviours and relationships to adherence and metabolic control, individuals with diabetes negative perceptions of support from family were associated with lower adherence to diabetes management areas (i.e. glucose testing, diet adherence, and insulin injections). On the other hand, positive impacts have been shown to affect an individual's management of the disease. For instance, Sinclair et al. found that adherence to self-care regimens (i.e. insulin treatments, monitoring blood glucose, exercise, and self-care away from home) was associated with emotional and instrumental support from friends and family [42]. This suggests that the perceived availability and knowledge of friends and family as being present positively impacts self-management efforts of individuals with diabetes. Furthermore, Connell et al found that social support had only a positive association with general morale among women, while there was a direct correlation between social support and adherence to treatment among older men with diabetes [43]. Among older persons, it has been found that women tend to exhibit better self-care behaviour, are less likely to be married, and are more likely to discuss personal issues with friends than men are [43]. This is in line with our study that showed that women tested their glucose levels more frequently than men. In contrast, men are more likely to have a family member who assists with various aspects of their self-care regimen. The self-care behaviour of older women with diabetes is also influenced by social role obligations, and this is especially true of certain communities like the South African community, where women often bear a bigger responsibility as the caregiver for the whole family [43]. Unsurprisingly, such women also report a lower quality of life as well as encountering more barriers to the self-management of their diabetes [43]. Weaver et.al reported that both symptoms of diabetes and difficulties achieving socially important roles contribute to poor mental health among Indian diabetic women. [44] Diabetes self-management education (DSME) is an integral part of diabetes care "for all people with diabetes who want to attain successful health-related outcomes," irrespective of age [44]. It is important to be aware of current DSME guidelines for older persons and how these guidelines can be implemented in a clinical setting. However, older persons are underrepresented in DSME research studies, so evidence-based guidelines specifically targeted toward older individuals are challenging to formulate [45]. The American Association of Diabetes Educators (AADE) and the American Geriatric Society (AGS) have formulated guidelines for DSME in the older adults mainly based on expert consensus [46,47]. For example, Older persons who are experiencing difficulties with daily tasks such as hearing loss, vision problems, decreased mobility and falls, will need individual rather than group DSME. Indeed, learning new skills will take longer and may require referral to a visiting nurse/CHW to make sure the task is fully integrated into the person's selfcare regimen. If needed, family members or other caregivers should be included in DSME [47]. Additional information about the influence of social support on chronic illness self-management has been supported by research. A systematic review reported evidence for a modest positive relationship between social support and chronic illness self-management, particularly for diabetes [48]. The finding was that a large information network is beneficial for self-management capabilities, especially in low education populations. This may be of an advantage in many cultures such as in South Africa where strong family relationships and family caring are important and highly valued [49][50][51]. A cohesive and supportive family may provide older diabetic patients with an opportunity to express feelings and fears. When DSMP is reviewed as a shared responsibility with the whole family, older persons may adopt DSMP activities more easily and feel more self-confident in managing diabetes [52]. As family-focused interventions may be more effective in improving DSMP performance than individual-focused interventions, including family members or friends in education programmes should be considered [53][54][55][56]. However, the shortage of professional health care workers in South Africa highlights the need to develop alternative delivery models for education and self-management for people with diabetes who attend primary care services. These include using the services of community health workers (CHWs) and peer supporters and should draw on previous lessons learnt [56]. For example, while a pragmatic trial of a group diabetes education programme led by health promoters in Cape Town improved blood pressure, but not self-efficacy, locus of control or glycaemia control; process evaluation suggested numerous problems. These included, finding suitable space for group education, with patient attendance and with full adoption of a guiding style by the health promoters. Thus, groups held outside of primary care clinics in the community and led by well-trained CHWs or peers may be a better option, so too may the active participation by family members in these groups [57]. In addition, the emphasis on diabetes prevention programs in middle-aged people must be highlighted, because it will enable the next generation of older persons to live with a reduced diabetes burden. For these reasons, South Africa's health care system needs to transform its services offered to older persons to reduce health care costs and ensure quality of life [57]. The various initiatives currently underway to re-engineer the primary healthcare system in SA to more effectively deal with NCDs, will go some way to meeting the identified needs of older diabetic patients and to addressing their barriers to care [58][59][60].However, as part of this re-modelling exercise, it is perhaps opportune for the health department to consult older chronic care patients and involve them in decision making and the planning of services. This study alerts policymakers and clinicians to some of the specific issues considered to be pertinent and important in the care and management of older persons with diabetes. As our results show that weak social support is a predictor of both poor knowledge and poor SMP, consideration should be given to health practitioners assessing social support when people with diabetes are reviewed clinically. However, interventions need to be put in place to enhance the level of support. This could include recommending that the person be open to accepting support from family and friends, suggest that the main carer attend some clinic visits with the person or referral of the person to a support group. --- Study strengths and limitation This study contributes to an understanding and fills a gap in the current knowledge, relating to diabetes self-management practices, and perceived social support from family and friends and diabetes care for older people in South Africa. However, the study has some limitations. First, as a cross-sectional survey design, our study could not assess cause and effect. Second, the measurements of self-report rather than direct observation of self-care practices are recognised as a limitation. In addition, the use of a convenience sample drawn from a population who attend a diabetes clinic excludes those who did not attend. Third, younger participants age <unk>60 years were not included in this study to enable comparison of outcome measures to older participants. Fourth, our study was limited to one region and may not be representative of all older South Africans with diabetes. Lastly, as an assessment tool, we have used a diabetes-related social support scale which we believe was a more specific tool in identifying diabetes specific social support than a more general social support scale. However, it would have been strengthened by adding an open-ended component following the social support scale, (for example asking the participant to list the top 3 ways that family and friends help in managing diabetes, and the 3 ways they help least in diabetes management). This might have provided better insight into what role family and friends play in the management process and inform more appropriate measures and/or items on social support scales. Future research should focus on developing and evaluating family/friends focused community-based multi-disciplinary education programmes to improve DSMP among older individuals attending primary care clinics with a view to enhancing the quality of life and to reduce disability. --- Conclusions Consideration needs to be given to developing and evaluating education programmes that focus on the needs of older people with diabetes mellitus and emphases the role of family and friends. However, it is imperative to introduce programmes at a younger age so that diabetes self-management strategies are embedded as a life course perspective to enhance positive outcomes for persons living with diabetes. --- All relevant data are within the manuscript and its Supporting Information files. The link can be accessed on ZIVA HUB (Open Data UCT) at the University of Cape Town https://zivahub.uct.ac.za/s/ bfa0fd85e001a074980f

In South Africa with one of the most rapidly ageing populations in Africa despite the demographic impact of the HIV/AIDS epidemic, diabetes is a major cause of morbidity and mortality. Self-management is challenging for all those with the condition but is likely to create a higher demand for those who may have existing co-morbidities associated with age, and long-standing chronic diseases.To determine the relationship of social support, especially that of family and friends with their self-management.This cross-sectional study was undertaken in the Cape Town metropole primary care clinics. The sample comprised 406 people drawn from four community health centres (CHC) that are served by Groote Schuur Hospital at the tertiary level.Of the 406 participants, 68.5% were females, 60.5% were living with a family member, and almost half were married. The mean duration of diabetes from diagnosis was eight years. More than half (57.4%) had no or only primary education. Half the participants (50.2%) had poor knowledge level in relation to symptoms and complications of diabetes. Multivariable linear regression showed older age was associated with poor knowledge (®: -1.893, 95% CI-3.754; -0.031) and higher income was associated with self-management practice (®:

What do we find on examination? These children are likely to be tired, hungry and have little interest. Their activity levels may be abnormal, their impulse control poor and they may lack empathy. There is a reduced capacity to learn, which accentuates feelings of failure. Social skills are poor and recourse to violence is frequent. Dreams are absent or die early. Their life takes place in a blighted urban environment that offers little contact with nature or cultural activities. What do we find on follow-up? The child's story is likely to be one of school failure, resulting in a lack of education and poor skills that lead to early unemployment. Then there is frustration, rebellion and violence. Disappointment, exclusion and a poor self-image lead to drug abuse, illness and antisocial behaviour. There will be a huge need for public services -health, support, rehabilitation, securityand a shortened lifespan. We know that social conditions have a strong impact on the health and longevity of people according to the stress level under which they live (2). We have a better understanding of the basic science underlying the relationship between the environment and the development of the human being. An infant's brain develops progressively from simple structural frameworks to more complex internal relationships in response to the intensity, frequency and nature of sensory and perceptual experiences. Positive experiences, by their frequency and intensity, strengthen pathways, increase interconnections between brain cells and build capacity (3). The brain's development is influenced most strongly before the age of three or four years, after which, its capacity to change reaches a lower steady state by approximately six or seven years of age (4). We have learned that stress, when caused by persistent adversity, becomes toxic because of its severity and duration. The resulting overstimulation of the hypothalamic-pituitary axis leads to a prolonged increase of steroid secretion, resulting in the death of cells in certain areas of the brain, to a delay in forming connections with higher brain centres as well as to a loss of connections already made. What is the result? The result is a smaller brain (5), a lower IQ and an abnormal function (6). This altered Commentary <unk>2010 Pulsus Group Inc. All rights reserved function is expressed as a decrease in attention span, control of activity and emotions, ability to learn, and empathy. Impulsivity is increased, as is an early recourse to violence. By the time the child becomes an adult, the continued high steroid levels cause an increased incidence of chronic diseases: diabetes, cardiovascular disease and cancer (7). Life ends early. We have known for some time that environmental and family stress have an influence on the development of early menarche and pregnancy in girls (8). Now we understand the mechanism. Toxic stress changes the expression of our genes. Research has shown that pregnant animals, if stressed, give birth to female offspring that become fertile earlier than normal, behave differently and become pregnant when very young. These stress-induced biological changes seem to be inherently stable and persist into the next generation (9). Does this apply to human beings? In Montreal, the teenage pregnancy rate in the Centre Local de Services Communautaires area of Lac St-Louis is 10%. In the Centre Local de Services Communautaires area of Hochelaga-Maisonneuve, being one of Montreal's poorest, where social stresses are high and where the incidence of children's lack of school readiness stands at 46%, the teenage pregnancy rate is 65% (1). conclusion What have we learned about child development? We have learned that genes and experience determine brain structure; that brain structure and capacity develop progressively from the simple to the complex in response to the intensity, frequency and nature of sensory and perceptual experience; that toxic stress during pregnancy and in early childhood leads to damage of the brain's structure, and to persistent changes in the function of the nervous and hormonal systems that lead to potentially lifelong problems with learning, behaviour, mental and physical health, and to a shorter life; and that such stress during pregnancy can lead to a change in the phenotypic expression of the genome. We have learned that assuring favourable conditions for development in early childhood is better than early intervention, which is better than late intervention. We know that normal child development leads to a capable adult, to a productive and civilized citizen, and to a prosperous, sustainable society. On the other hand, we have learned that abnormal child development leads to an incapable adult who is likely to be a dependent, nonproductive citizen prone to exhibit antisocial behaviour and to be both a burden and a danger to society. The root cause of these problems is the nonrespect of the rights of the child. The social determinants of health find their equivalent expression in the United Nations Universal Declaration of the Rights of the Child (10). We have not respected them and yet we claim peace, order and good government to be our touchstone. Are we living up to our ideal? Every child comes with the message that God is not yet discouraged of Man. -Rabindranath Tagore

O ur central responsibility as human beings is to equip our children for life, to enable them to become capable, civilized adults, and to be better than we are. We are not doing a good job in Canada. In Montreal, Quebec, as in many other Canadian cities, approximately one-third of children grow up in poverty and as many are not ready for school because of developmental delays. In 2008, approximately 35% of Montreal children were not ready for school. In the most disadvantaged areas, this rate rose to 46% (1). These findings represent a human and a social disaster. However, after one day's headlines and another of comment, this was, like yesterday's news, forgotten. Especially in the present economic climate, how can we sideline one-third of our future population from fruitful participation in a society that needs them? Whom do we find behind these numbers? We find children who have developmental delays in language, motor and social skills; they have a reduced attention span and abnormal activity levels. What is their family history? Parents of these children often have a small family circle and lack a support network. They feel isolation, exclusion, insecurity and discrimination. They have few opportunities or choices. They lack access to transportation and are often un-or underemployed. Their housing is poor. They have a high incidence of chronic illness. As a consequence, the parents do not feel good about themselves -they feel 'down' and are often depressed. They are tired of struggling, often afraid and frustrated, and their minds are filled with worries. Things appear hopeless, and their dreams die. Substance abuse is common. In these circumstances, their behaviour toward their children is characterized by impatience and an inability to listen. They never have time to pay attention to their children. Bedtime stories are rare, but neglect and abuse of all kinds are common. What is the child's experience? A child growing up in this environment is likely to lack attention, love, food, sleep, positive stimulation, good experiences, opportunities to test him-or herself, consistency, praise and a feeling of success. Abuse is frequent. The physical environment is poor. It is noisy, dirty, unsafe and filled with traffic. There are few public sports facilities or parks to play in.

Introduction Farzindar & Inkpen (2015) describe social computing as "an area of computer science that is concerned with the intersection of social behaviour and computational systems" (p.109). Similarly, Facebook (2016) defines social computing as "computational techniques and tools to study human social behaviour". Social computing is thus comprised of two essential elements: social behaviour and the computer systems that enable this behaviour. We believe that social behaviour is closely tied to norms of cultural behaviour in specific populations and in this research, we propose to use social computing perspectives to study how and why indigenous organizations use Facebook to engage and disseminate information with urban indigenous youth at risk in Ontario. Our assumption is that indigenous organizations can reduce risks among urban indigenous youth in Ontario by engaging and disseminating information on Facebook. Given that urban centers pose significant risks for urban indigenous youth (Miller et al., 2011), it is critically important to develop an understanding of how indigenous organizations use social media like Facebook to disseminate information to their targeted populations, particularly urban indigenous youth, who use Facebook and other social media heavily and daily. One problem is that help for vulnerable indigenous youth frequently arrives too late. It normally takes some time for family, friends, community members, and government agencies to identify indigenous youth in danger. It also usually takes some time for people voluntarily seeking help to inform their family members and friends about their situation. Besides these barriers, some matters are associated with social stigmas especially drug addiction and mental health. The delayed identification of indigenous youth at risk poses great challenges for how to provide assistance and treatment. We explore how the indigenous organizations use Facebook to engage with urban indigenous youth at risk in Ottawa through conducting in-person interviews with two communication managers and two youth program managers of three indigenous organizations based in Ottawa. The interviews were conducted during the spring and summer of 2017. Our contribution is an analysis of how indigenous organizations could reduce risk among urban indigenous youth by engaging and disseminating information on Facebook. We have found that many indigenous youth living in Ottawa experienced remote-to-urban forced migration. Many of them came involuntarily to Ottawa in order to receive medical care not available in their remote home communities. Urban indigenous youth in Ottawa are using Facebook to combat their loneliness by contacting their families and friends who reside in rural and remote communities. --- Background UNESCO (2016) defines youth as the population group ranging from 15 to 35 years of age. Youth, according to that definition, are the most at risk age group in populations! 99 across societies and cultures. This stage of life comes with changes in hormones, physical state, mood, and social behaviour while having to find one's identity and become autonomous. Youth in vulnerable and impoverished populations face an even a higher risk since dealing with poverty and vulnerability add to the transitional challenges. In this research, the term "at risk" means that individuals face a high probability of being exposed to physical and/or psychological harm or danger. For this research, we focus on urban indigenous youth because they face increased risks in terms of health care, education, housing, unemployment, racial and cultural discrimination, substance addiction, murdered and missing indigenous women, sexual abuse and domestic violence, and youth suicide (Boksa, Joober, & Kirmayer, 2015); (Drache, Fletcher, & Voss, 2016). We chose to further narrow the geographic scope of this study to urban populations in the city of Ottawa, Ontario given that data from Statistics Canada shows that the majority of indigenous peoples in Canada live in the province of Ontario and mostly in urban areas rather than on reserves which are often more rural and remote. Census data collected in 2011 indicates that out of Canada's total indigenous population of 1.4 Million,21.5 % (301,425) reside in Ontario (Statistics Canada, 2015a). This amounts to 2.4% of the population of Ontario, considerably lower than the national percentage of the indigenous population in Canada at 4.3%, (Statistics Canada, 2015b). The percentage of indigenous peoples with registered indigenous status living offreserve in Canada stands at 50.7% (Statistics Canada, 2015b). Many of them now live in urban areas of Ontario such as Toronto, Ottawa, Thunder Bay, Sudbury, Hamilton, and Sault Ste. Marie (Ontario Federation of Indigenous Friendship Centres, 2013). Ottawa is the capital of Canada and has a population of approximately one million people (City of Ottawa, 2014). It is the seat of Parliament and as the main location of the federal government, it is where important policy decisions are made that have an impact on all indigenous communities across the country. According to the 2011 statistics from the City of Ottawa there are about 18,180 indigenous peoples living in Ottawa (Statistics Canada, 2015c). The number is quickly growing and many of the indigenous peoples in Ottawa are young with 4 in 10 being under the age of 25 (Statistics Canada, 2017). Therefore, Ottawa is a good location for this study since the city attracts indigenous youth from many communities within Ontario and from other provinces as well. The city of Ottawa has a number of diverse indigenous organizations; no other city in Ontario has the same characteristics. Ottawa's indigenous organizations look after the basic needs and wellbeing of indigenous residents, including their housing, education, health, and culture. These functions are similar to the indigenous organizations in other cities in Ontario but Ottawa is only the city in which many national-level indigenous organizations have chosen to locate their headquarters. At the time of writing this article, there are 15 indigenous organizations located in Ottawa. Of these entities in terms of focus 9 are policy-oriented and 6 are community outreach organizations. Policy-oriented indigenous organizations negotiate and participate in political matters with the Canadian federal government and parliament who are also located in the capital. Community outreach-oriented indigenous organizations focus on providing! 100 support to all indigenous people living in Ottawa. The characteristics of indigenous organizations in Ottawa are unique and position this study well to gather both regional and national information about how the indigenous organizations use Facebook to engage with urban indigenous youth at risk. In recent years, researchers have looked at social media and Internet usage by indigenous people who live in rural and remote areas around the world, mainly in Australia, Canada, United States, and New Zealand (Gauvin, Granger, & Lorthiois, 2015;Filippi et al., 2013;Ormond-parker & Sloggett, 2012); (McMahon, Gurstein, Beaton, O'Donnell, & Whiteduck, 2014;Kopacz & Lawton, 2011;McMahon, 2014;Singleton, et al., 2009;Watson, 2015). The purpose of social media usage among indigenous peoples include cyber-activism (Warf, 2009;Elwood & Leszczynski, 2013), digital channels to record and promote their culture (Huang, Chen, & Mo, 2015;Dalseg & Abele, 2015), connecting and maintaining relationships with other people (Watson, 2015), and seeking health information and establishing virtual health support groups (Stephens-Reicher et al., 2011;Carlson, 2013;Laakso, Armstrong, & Usher, 2012). There is no academic literature documenting the voices of the indigenous organizations in urban areas, who work diligently to offer services, run outreach programs, and advocate for indigenous rights, and how they use Facebook to connect with youth at risk. --- Research methodology --- Research design We interviewed two communication managers and two youth program managers who are responsible for social media campaigns in the three indigenous organizations in Ottawa. Their social media activities are followed by many of the indigenous youth in Ottawa. We conducted four interviews in total. Each interview lasted between 45 and 60 minutes. These interviews aimed to discover their Facebook strategies and practices for engaging with urban indigenous youth at risk. --- Recruitment We recruited participants by approaching them directly with a formal email that explains the research project. The information sheet, informed consent form, and interview guidelines were also attached. We used email addresses that were found on indigenous organization websites with office addresses in Ottawa. We waited for their email responses in order to schedule the interview and we followed up by telephone. --- Data collection We held our interviews at the participants' offices in a private meeting space. In this research, the interviews were semi-structured and used interview guidelines. This allowed interviewees to freely express their personal opinions and to add their own comments beyond the prepared questions. The interviews conducted with communication managers, whose roles included generating and sharing content on their organization's Facebook page, and youth program managers whose roles are to run! 101 cultural and social activities for indigenous youth in Ottawa and act as the first responder in identifying indigenous youth at risk and help them to receive or access necessary assistance. This was followed by inquiries focused on their professional use of Facebook, the types of content they read, generate, and share on Facebook for their organizations, whether their Facebook content targets urban indigenous youth who feel at risk physically and/or psychologically, their observations when an urban indigenous youth is at risk through the language or verbal tone of the content that he or she has generated or shared on Facebook, and the support of indigenous culture toward generating and sharing content on Facebook. There was no financial compensation for their participation. --- Data analysis Our interviews were audio-recorded and done in-person. The data generated from the interviews were manually transcribed in English and coded based on patterns identified through thematic analysis (Braun, & Clarke, 2006). Based on our research question, how and why indigenous organizations use Facebook to engage with urban indigenous youth at risk, and interview guidelines, we developed predetermined codes and then created additional codes to capture specific emergent themes when we were more familiarized with the data. Our data analysis codes include Facebook usage frequency, account control and access, type and format of information generated and shared on Facebook, motivations, challenges, indigenous attitudes toward Facebook, and their observations on Facebook usage among urban indigenous youth at risk. Then we did the second and third level content analysis to identify relationships between themes. As the number of our interviews was not large, we were able to make comparisons between each interview to find the similarities and differences to further improve our data analysis with special attention to the type of indigenous organizations (1. National; 2. Regional) and their work mandates (1. Advocacy and political issues; 2. Social programs and outreach). We presented our preliminary findings to our interviewees in this research in September 2017, which helped to enhance the quality of our findings (C. Intahchomphoo, personal communication, September 19, 2017). --- Ethics This research received ethics approval from the Health Sciences and Science Research Ethics Board of the University of Ottawa (File# H02-17-01). We asked research participants to give written consent in willingness to participate in the interviews. They were given time to read the consent as much as they needed and prior to the interviews taking place. We also read the consent aloud to the participants. In the results section, we omitted all personally identifiable information including participants' names, organizations, addresses, and physical appearances. We used pseudonyms for our participants in this paper. No data from this research was stored online in a cloudcomputing environment. Data will not be shared with any third parties. In terms of the assessment of risks, we only publish results where we have explicit permission from the! 102 indigenous organizations and the communication managers and youth program managers from which we obtained data. --- Results Indigenous organizations in Ottawa frequently use Facebook and their Facebook accounts are being controlled by communication managers and management level personnel. "I, the executive director, and one of the managers are pretty much the people who focus on communications through Facebook."-Alinga All of our four interviewees (whose pseudonyms are: Adoni, Alinga, Alba, and Alkina) said that their indigenous organizations use Facebook on a daily basis. We discovered that many indigenous organizations in Ottawa do not have a high capacity to constantly generate new content for their Facebook accounts due to human resources limitations. As a result, they also read and share other organizations' and individual's posts to display on their organizations' Facebook walls. Generating new content of their own is usually done internally with a small team consisting of members from administrative staff, community outreach staff, and communication managers. Those people have full access to the organizations' Facebook accounts. Indigenous organizations in Ottawa use Facebook to share information about the important events and meetings that staff attend, to listen to public opinion about emerging indigenous peoples' issues and concerns, and to promote their outreach campaigns. --- "A couple weeks ago, he (organization leader) was in New York at the United Nations. I will post pictures of him there."-Adoni Two interviewees mentioned that their indigenous organizations use Facebook to share information about the attendance by their directors or staff at prominent social events and work-related meetings taking place in Ottawa or other urban cities and rural areas in Canada, as well as abroad. Furthermore, we learned that sharing information about such events on Facebook is being done at two out of four indigenous organizations we interviewed in Ottawa. Moreover, from our interviews we discovered that indigenous organizations in Ottawa use Facebook to listen to public opinion about issues and concerns arising in indigenous communities. These are some examples of quotes from our interviewees on this matter: "I think that we found a lot of strength on Facebook. We were able to connect with other people in term of understanding indigenous issues. I find that there are power in mini voices and sharing information" -Alba "I use Facebook often to find news stories that people shared from other parts of the country that might be interested... for example, the missing and murdered indigenous women inquiry"-Adoni! 103 The national inquiry into missing and murdered indigenous women and girls was discussed in two of our four interviews. We were told that information and opinions about the inquiry are also being shared and discussed on the indigenous organizations' Facebook accounts. When the inquiry was starting, there were a lot of different opinions about whether they were doing a good job or whether an inquiry should be held at all and a lot of people shared their thoughts on Facebook. Some indigenous organizations in Ottawa would use the peoples' thoughts on Facebook to get a sense where the organizations should stand on this issue. In addition, we noticed another aspect of Facebook usage among indigenous organizations in Ottawa especially the ones with a mandate to provide communitybased programs and services: they put a very strong emphasis on using Facebook to promote their outreach campaigns. Below are some sample quotes from our interviewees: "We have an upcoming campaign about childhood sexual abuse. This campaign will disproportionately be on Facebook."-Alinga "We post information (on Facebook) of our programs and any helpful information about safety or knowledge about abuses." -Alba "Through Facebook we actually reach [out] to some of the youth that are on [the] street and our outreach people [are able] to access them and find them and get them into shelters." -Alkina Alinga, Alba, and Alkina advised us that in Ottawa as well with other parts of Canada, there are conversations being held about indigenous youth's mental health, suicide crises, and substance abuse. These are also significant social problems among urban indigenous youth in Ottawa. Substance abuse has a strong relationship with mental illness and suicide (Agrawal et al., 2017). Our interviewees commented that there really should be more public discussion and Facebook campaigns on those specific topics. Interestingly, in our interviews we also were briefed that they have to ensure that whatever content they post on Facebook targeting urban indigenous youth must not be too manipulative, controversial, or emotionally taxing. This is particularly the case when the Facebook post targets urban indigenous youth who are at risk. Otherwise, it is possible they might see the organizations' Facebook outreach campaigns and react with psychological or emotional discomfort, which may have social repercussions particularly since urban indigenous youth in Ottawa do not necessarily know where they can get help. Therefore, indigenous organizations in Ottawa are extremely mindful of the sensibilities of urban indigenous youth when creating Facebook outreach campaigns.! 104 Other campaign materials such as posters and press releases that the indigenous organizations have created also go through their Facebook pages. Some of our interviewees indicated that their organizations also use Twitter to disseminate information about their outreach campaigns, but remarked that it is not as popular as Facebook. They found that Twitter is difficult for the indigenous organizations' outreach campaigns in part because of the limited word count of tweets. They also believe that Facebook functionality is better suited to oral expression, which aligns with the indigenous peoples' rich oral history traditions. Indigenous youth can use Facebook video functionality to verbally express and share their thoughts and feelings disseminated in the form of moving images and sounds to their circles of friends on Facebook. Sharing a video does not require an indigenous youth to input any text. Furthermore, Facebook videos can be produced and uploaded very easily on smartphones, which are the devices that most urban indigenous youth in Ottawa are using. Many urban indigenous youth in Ottawa are unaccompanied and they were forced to migrate to the city. They use Facebook to connect with their families and communities in the rural and remote areas from which they came. --- "It is very overloaded in the urban setting, it is very challenging. It does compound with race and class... I think when you are here alone your connectivity is so important and that why they are relying on Facebook to get in touch to their home community."-Alinga The data from our interviews helped us to recognize the crucial relationship between remote-to-urban forced migration (Darling, 2017) and urban indigenous youth in Ottawa. Unfortunately, many of them are also living in Ottawa without any family members accompanying them from their home communities. In many cases, they were forced to migrate to Ottawa to receive medical care or better education, which are not available in their more remote communities. So, those indigenous youth often end up living alone for a long period of time in Ottawa, as some medical care can be very lengthy. Because they reside in Ottawa on their own, the connection with their families and communities in rural and remote areas is often done through Facebook on their personal cellphones. For the unaccompanied urban indigenous youth in Ottawa, it is very important to maintain ties with their home communities, so their Facebook connectivity is not only about maintaining social connections but it is also about staying connected to their indigenous cultures. Here we see the use of Facebook by unaccompanied urban indigenous youth is at the intersection of personal relationships and cultural identity. When they use Facebook to connect with remotely located family and friends, they are also being updated with a lot of social news what is happening in the community, which nourishes and sustains a very important part of their cultural identity. One of our interviewees said that urban indigenous youth in Ottawa who wish to return to their rural and remote indigenous communities face many challenges because of the high cost of transportation to their homes. In addition, there is also now a housing crisis! 105 in many indigenous reserves in Ontario and other provinces and territories (Anaya, 2015). For urban indigenous youth in Ottawa to return home, they must find a place to live and their families often experience overcrowding with multiple families in one house. In many cases this situation would have worsened since they left their communities. Therefore, many urban indigenous youth who leave remote communities for southern cities like Ottawa do not have much choice but to continue to stay in the city. They therefore have to live alone for a longer period than they had expected in a place that has a very different culture with different social interactions than their home communities. Unfortunately, these circumstances are contributing factors to the struggles that many urban indigenous youth faces. The circumstances increase risks such as drugs and alcohol, which are readily accessible in urban settings. Using minimal text and simple language with images and short videos on Facebook posts. --- "The number one thing that people (indigenous youth) said is to cut back on the text (on Facebook)." -Alinga "I was editing some videos.. ---. I just promoted the event on Facebook. The video I did is about 30 seconds." -Adoni Indigenous organizations in Ottawa prefer to use short videos to communicate on their Facebook pages. For them, it is essential to use minimal text with simple language. This is due to the fact that neither English nor French are the first language of many urban indigenous youth living in Ottawa although many do speak English or French as a second language depending on where they were raised. Therefore, indigenous organizations in Ottawa prefer to use audio and visual media to communicate with indigenous youth. We observed that indigenous organizations in Ottawa use symbolic imagery on their Facebook posts such as the ancient traditional indigenous arts and paintings with the cultural representations about the concepts of spirituality, human beings and animals, and the relationship to the nature. Normally, they cut back on text as much as they can and they keep the English or French text to a reading level that is appropriate to their readership. They also ensure that the images they use represent indigenous peoples. In terms of being culturally appropriate, our interviewees indicated that this is very important because otherwise indigenous youth will just scroll right past it on their Facebook's News Feed if they have non-indigenous themed designs. In our interviews, the research participants discussed at greater length the fact that Facebook posts are fast paced and its News Feed has a rapid turn-over rate. However, they have to ensure whatever they post is eye catching, colorful, and not too wordy, in order to capture indigenous youths' attentions. Adoni added in her opinion, Facebook has contributed to shortening indigenous youths' attention spans. A study on the future of on-line news video by Kalogeropoulos, Cherubini, & Newman (2016) suggests that, to have an impact, even videos on Facebook have to be short and tight.! 106 Indigenous organizations in Ottawa have different motivations for their Facebook strategies. --- "We are trying to figure out how to get into the indigenous peoples' Facebook algorithm for them to see our posts on their Facebook walls. That is one of our motivations-how to get ourselves out there more." -Alinga Some indigenous organizations in Ottawa explained that they want their Facebook posts widely shared to help their organizations to better connect with urban indigenous youth. One way to evaluate social media strategies and to measure the public reach of Facebook posts is to count the number of "shares". All of our interviewees want to maximize the extent to which their messages reach the public on Facebook. However, the indigenous organizations in Ottawa with a Facebook outreach mandate sometimes find it difficult to get across information about programs and services that they have available to assist indigenous youth. On the other hand, there are other indigenous organizations in Ottawa who are working on the political side and advocating for indigenous peoples' rights. Their Facebook motivation is more about how to quickly inform all indigenous peoples about forthcoming laws, amending legislation, the House of Commons and Senate debates, and the Prime Minister messages to the members in the Chamber and the public that will be introduced in the Canadian Parliament, which they believe may have a great negative impact on their people. Their motivation for posting on Facebook is more to address political and legal changes and to raise awareness, which in some cases might lead to public mobilization (Wood, 2015). --- Challenges with Facebook usage by urban indigenous youth in Ottawa include: language barriers, personal interests based on age, content creation during slow periods, and cellphone data plan limits. "When you are 13 years old, you are usually not thinking of the term'mental health '. So, interest is difficult". - --- Alinga "Not all the youth have a data plan on their cellphones... a lot of them do not own [a] cellphone. It is very important that we actually reach out to somebody." -Alba We found the first main challenge of reaching out to urban indigenous youth in Ottawa via Facebook is the language barrier. Across the board, language is the most common challenge for indigenous organizations in Ottawa. Language barriers among urban indigenous youth create a sizeable communication gap. For example, a study by Arnaert & Schaack (2006) in a different context suggests that Inuit patients in the emergency department at hospitals in Montreal should have a direct access to a nurse or interpreter who can speak their language. Overcoming language barriers helped to improve their experiences with medical care. Alinga mentioned in our interview that urban indigenous youth speak a wide variety of languages and dialects. Their understanding of English or French is often very basic, which creates a significant language barrier between them and the indigenous organizations who are trying to reach out.! 107 Personal interests based on age are another challenge. Alinga thinks that youth of all societies usually do not think much about the complex truths of their lives, especially their own mental health. Therefore, it is difficult to engage urban indigenous youth in Ottawa with Facebook posts related to mental health. Another challenge described by Adoni is about creating new content when there are few significant events happening, especially within indigenous organizations that work on politics and advocacy at times when there is not much political activity. All the interviewees were not inclined to visit their Facebook pages unattended and they like to update these pages as often as possible. This is another challenge for these organizations: finding new content to post about when there is not necessarily much going on, such as when the Canadian Parliament breaks for summer. One last challenge shared by Alba was that urban indigenous youth in Ottawa who own a cell phone typically do not have a data plan on their cellphones. Many of them rely on free Wi-Fi offered at public libraries, schools, and indigenous organizations. Interestingly, Alba also told us that a lot of indigenous youth in Ottawa do not own a cellphone. They access to their social media accounts by using public computers or their friends' cellphones. Alba faces this technological barrier in reaching out to this population and this challenge becomes more acute when reaching out to urban indigenous youth who are actually at risk and require emergency services. Indigenous culture supports generating or sharing content on Facebook. Facebook. Indigenous peoples still have powwows but you will find information about powwows on Facebook, like the electric powwow from the Tribe Called Red."-Adoni "We are able to reach more people through Facebook. We put out our survey and we put out other things how people can access our services." -Alba Adoni further explained his quote above by referring to the Idle No More movement that began in 2011, the indigenous rights demonstration that was organized through Facebook. Nowadays, it seems any kind of social event, gathering, or movement of indigenous peoples is certainly driven by social media. --- "It is interesting how indigenous cultures have embraced Our interviewees suggested in several group meetings that the indigenous organizations in Ottawa should consider Facebook as a main source of communication for giving notices and sharing other valuable information. From our interviews, we were told that indigenous youth claim that they check email but in fact they do not like to do so and they are more easily reached on Facebook since they already log-on to Facebook multiple times every day. In conclusion, indigenous communities are supportive of generating or sharing content on Facebook. Facebook is the easiest way for many of the urban indigenous youth to connect with others at home without the restrictions of geography and hence a natural channel for urban indigenous organizations to connect with them. "I can see how certainly it (Facebook) will help someone who is feeling with suicide or depressed to be reminded the good things in life when everything seems bad." -Adoni "I think a lot of youth who are having a hard time will end up throwing all that out. They vent. They put it in text and a lot of youth are putting music. They put out videos of different songs that are sad." -Alkina All of our interviewees agreed that Facebook is now a big part of their organization's social media strategy. In term of indigenous youth at risk, our interviewees think Facebook has at least shined a light on many of the issues faced by urban indigenous youth at risk. Adoni gave an example of somebody who was living in a fly-in community in northern Ontario and feeling depressed. Prior to the era of social media, nobody in the world would have known about this person's state but because people can express their feelings and their thoughts more easily on social media and reach a wide audience, there is a higher likelihood that such people could be found and helped. Our interviewees felt strongly that Facebook could help urban indigenous youth at risk especially the ones who are struggling with mental illness and suicidal ideation. --- Indigenous organizations have urged Adoni explained to us why so many of the urban indigenous youth are facing mental illness, as evidenced by the language and verbal tone of the content that he or she has generated or shared on Facebook. One of the main causes, he says, is the intergenerational trauma of the residential schools' policy implemented and enforced by the Canadian Government between 1880 and 1996 (Miller, 2012). Adoni began by referring to the Truth and Reconciliation Commission and commented that it had done a good job of showing how residential schools stole a generation of indigenous peoples. In that one generation, children were sent to schools away from their families and never learned what it was like to have a parent. Therefore, they often did not learn to parent themselves and it is hard to break that chain of dysfunction when they have children of their own. They do not know how to be parents for their children and pass on this inability to their own children thus perpetuating intergenerational trauma (see: Bombay, Matheson, & Anisman (2014) and Elias et al. (2012)). Whereas another interviewee remarked that Instagram has recently become popular among urban indigenous youth, it would not necessarily be a good enough medium to be able to express or identify their mental health, in part because Instagram has different purposes and modalities of expression. Our interviewees indicated that urban indigenous youth feel more of a community base with Facebook whereas Instagram is more like Twitter, less engaging with a community. Instagram is more about publishing user-generated content mostly pictures and it is less about news (Phua, Jin, & Kim, 2017). They know their people on Facebook and they can configure privacy controls for their groups and communities. It is also a more group platform with more ways to share thoughts, opinions, and feelings. So, it was noted the youth feel that they can talk about their mental health issues more freely on Facebook including depression and suicide. --- Discussion --- Study findings! 109 Our interviews revealed the unexpected fact that many indigenous youth living in Ottawa experienced remote-to-urban forced migration. Many of them came involuntarily to Ottawa in order to receive medical care not available in their remote home communities. Unfortunately, they are not able to return to their reserve communities because of the length of their treatment. Often, this forced migration also made them unaccompanied youth migrants. They had to migrate to Ottawa without being under the care of a parent or legal guardian. This situation is a result of the poor access to health care in rural and remote indigenous communities in Canada and the health care funding inequalities between rural, remote, and urban settings in Canada (Marrone, 2007). Consequently, urban indigenous youth in Ottawa are forced to live alone for long periods of time without family interactions, which adds to their existing risks such as drug use, dropping out of school, and lack of access to good quality food. We learned from our interviews with indigenous organizations that unaccompanied urban indigenous youth living in Ottawa use Facebook to connect with their families and communities in rural and remote areas. Facebook is a therefore a tool for them to stay connected to their roots while they live in the city. Facebook is being used as a place where urban indigenous youth seek dignity, love, and care from their own people and to stay connected to their culture, things that they cannot find in the city. They can also gain spiritual support when they learn about what their friends, siblings, parents, and other community members are doing from Facebook News Feeds. Even simply seeing short texts, images, and videos on Facebook, generated and shared by their circle of Facebook friends can help these youth cope with their experience of loneliness in Ottawa. Our interviewees indicated that encouragement and support are what these young people benefit from the most in their Facebook usage. When indigenous organizations in Ottawa create Facebook campaigns for local indigenous youth they address the broad diversity of indigenous cultures, languages and dialects by creating Facebook posts with few words and focus more on images and videos. We observed that English is the lingua franca that indigenous peoples use to communicate with other groups of indigenous peoples whose first languages are different from their own. This explains why our interviewees mentioned their use of simple English syntax on their Facebook posts. Also, these organizations' use of images and videos is not straightforward because they have to come up with the materials that are culturally appropriate while avoiding possible misinterpretations caused by cultural differences. For example, when an organization envisages running a campaign about HIV prevention and safe sex practices among indigenous youth, they have to carefully consider how to disseminate the campaign information including images and videos on Facebook in a manner that is culturally appropriate given that sex is still a taboo topic in many indigenous communities (Mill et al., 2008). We also noticed that there are differences in the ways in which Facebook is used by indigenous organizations in Ottawa, differences that depend on their mandate and their target population. For organizations who are working on outreach programs, Facebook accounts directly target urban indigenous youth in Ottawa and many of those youth are also following the organizations' Facebook posts. These outreach programs can identify! the urban indigenous youth at risk of drug addition, homelessness, unemployment, HIV, health care access hardship, and dropping out of school and their posts are aimed at this population with these issues. On the other hand, in the case of indigenous organizations whose mandate is advocacy and political issues, their Facebook accounts seldom target urban indigenous youth in Ottawa. They look more broadly at the governing bodies of indigenous peoples across the country in order to work and communicate with the Canadian Federal Government. Their Facebook pages target all indigenous people across the country, NGOs, and government. --- Limitations Our study has some limitations. We restricted our interviews only to communication managers and youth program managers of indigenous organizations with offices in Ottawa, the capital city of Canada. We did not conduct this research in other cities in the province of Ontario, nor did we conduct research in other provinces and territories of Canada where many other indigenous youth also reside. Thus, we do not aim to report findings of all indigenous organizations on the national level. --- Conclusion Indigenous organizations use Facebook for two main reasons. The first reason is to promote the work of these organizations to the public and for them, in turn to listen to the public's opinions about positive and negative news related to indigenous peoples' wellbeing. Secondly, Facebook is also used to engage urban indigenous youth at risk with indigenous organizations that provide social programs and outreach. Indigenous organizations use Facebook because many urban indigenous youth in Ottawa are using Facebook, and it is the fastest way to connect with them when they are or feel at risk. Communicating with urban indigenous youth in Ottawa via Facebook is difficult because there are several cultural and language differences among the diverse indigenous peoples in Ottawa. Indigenous organizations have solved this issue by using a minimal amount of simplified English as well as images and videos in their messaging. We learned that urban indigenous youth in Ottawa are using Facebook to stay in contact with their families and friends in remote communities in order to combat their loneliness and other risks that they are experiencing in the city. Moreover, we discovered some unexpected realities about the forced migration and unaccompanied youth migration of indigenous youth in Ottawa. Warf, B. (2009). Diverse spatialities of the Latin American and Caribbean Internet. Journal of Latin American Geography, 8(2), 125-145.

A qualitative study in which we conducted four interviews with two communication managers and two youth program managers of three indigenous organizations with offices in Ottawa, the data generated from the interviews were coded based on factors identified through thematic analysis. Indigenous organizations use Facebook for two main reasons. The first reason is to promote the work of these organizations to the public and for them, in turn to listen to the public's opinions about news related to indigenous peoples' wellbeing. Secondly, Facebook is also used to engage urban indigenous youth at risk with indigenous organizations that provide social programs and outreach. Indigenous organizations use Facebook because many urban indigenous youth in Ottawa are using Facebook and it is the fastest way to connect with them when they are or feel at risk.

over half (54.0%) believing that perpetrators faced no consequences at all. Respondents described personal and professional consequences for reporting; system weaknesses and perceived inaction left many with disappointment, distrust, and disdain for MSF itself. Cumulatively, our data suggests abuse is occurring at a significant scale and points to fundamental distrust in both the organisation's accountability systems, and leadership's ability to address this crisis. These results are difficult, if not impossible, to contextualize because aid organisations rarely measure-or disclose-abuse or discrimination perceived by their workforce. Often there is no baseline: 2020 marked the first in MSF's 50-year history in which it compiled formal complaints from global offices in addition to those from medical projects in the field [11]. As recently as October 2021, an institutional response to a racial justice questionnaire by The New Humanitarian stipulated that MSF lacks data for the one-year period since May 2020 on changed human resources practices and does not report on the percentage of leadership coming from under-represented groups [12]. In a context of inaccessible information on matters of the public interest, effective activism may require the direct production of research, even with limited means and methodological weaknesses, to advance understanding and future action when an institution cannot, or will not, study these questions. Such research may therefore be considered inherently agitational, requiring a mutual responsibility for diligence and safety by organisations and activists. Our report-created with free resources and volunteer labour-openly acknowledges its limitations, including possible self-selection despite our expansive outreach, ambiguities in question phrasing, our inability to verify employment status, and the lack of a suitable peer reviewer. We were not officially authorized to do this work but chose to disclose our authorship to promote full transparency and feedback. We nonetheless had to carefully navigate a maze of painstaking efforts to minimize our own legal and professional risks, including attacks by current MSF stakeholders and continual revision to ensure our report wording could not be framed as defamatory. This type of research forces us to query the assumptions coded into the roles of the researcher and researched. Critically engaged activist research "provides an important approach to addressing the practical and ethical dilemmas of knowledge production" [13]. It confronts us with how we-as academics, activists, and practitioners-may dislocate ourselves from the problems we seek to change by looking forever outside rather than within, leaving us blind to how the same systems, hierarchies, and problems entangle and trap us from change [14]. As survivors of abuse within the organisation, we would argue that our lived experienceour entanglement-allowed us to formulate specific questions currently not captured by the institution, gain necessary trust, and better level power imbalances intrinsic to research processes. Our reflexivity served a process to unknot how power manifests towards and through us as individuals and institutions, resulting in a re-examination of the classical research model itself. Global health transformation requires relatable and personal discourse through reflexive dialogue between individuals across geographic, economic, and epistemic divides [15]. The report's emphasis on storytelling, artwork, and counter-perspectives is consistent with this view. As former staff with racialized backgrounds employed on international contracts, we have tried to embrace our "responsibility to speak up commensurate to position and privilege" [16]. Our report attempts to challenge institutional resistance by showcasing the pressing need for additional investigation, including a global independent assessment engaging survivors and patients, the vast majority of whom exist outside MSF's associative and consultative structures. We hope it will also support fellow resistors at Decolonise MSF in tracing a path forward for future research and grassroots action. If it succeeds, Dignity at MSF represents activist research in the spirit of the "generative power of protest", which recognizes the "power of protest events themselves to shape internal movement dynamics" [17]. In our case, the vision is a movement for anti-racism and anti-discrimination at MSF that is self-sustaining, locally-led, heterogenous, and diverse in opinion and approaches. This requires the creation of a space where everyone can safely speak out and act. In crafting this report, we highlight the urgent need to reduce the structural barriers that contribute to the stifling of agency in pursuit of discovery-a process central to the long-term march to justice and equity in aid and global public health.

A torrent of public attention has surfaced myriad allegations of exploitation, discrimination and abuse by humanitarian, development, international aid, and global agencies. Amid these, Me ´decins Sans Frontières (MSF), our former employer, has been thrust into the spotlight, owing in large part to the témoignage and activism of current and former employees mobilizing in a growing movement to "Decolonise MSF". In June 2020, a staff-led open letter containing more than 1,100 signatures and 200 testimonies of abuse and discrimination went public [1]. MSF's international leadership, in the aftermath, welcomed "the current debate on racism" but did not formally acknowledge the letter or its specific demands [2]. Since then, media attention has amplified insider concerns on racism and segregation within the organisation [3]. Calls to 'decolonise' global health by academics, practitioners, activists, and students differ in their approach and focus. While they jointly seek to address and dismantle racist and discriminatory structures and norms, few have delved into the more complex analysis of the "monopoly, misuse, or abuse" of power [4], including, crucially, the ways in which it is experienced. Abuse-be it in the form of retaliation or rape, derogatory comments or discriminatory contracts, nepotism or neglect-stems from systemic asymmetries that foster, invite, and enable the misuse of power and dominance. Abimbola et al have explored facets of supremacy, encompassing coloniality, patriarchy, racism, white supremacy and saviorism, that together maintain power and privilege within global health [5]. These facets also underpin aid systems, resulting in internal inequalities that compound and spill over into how the mandate of an organization is implemented [6]. Discussions at MSF have centred, for example, on its two-tiered staff structure and the glaring over-representation of the "Global North" in management and governance [7,8], more than 16 years after it committed to equal opportunities for all staff [9]. As Decolonise MSF organisers, we contributed to urgent demands for structural reform and transparency by publishing Dignity at MSF in September 2021, a movement-wide report on abuse and discrimination based on findings of an online survey with 359 current and former staff and stakeholders [10]. It is the first volunteer and survivor-led initiative to publicly assess and disclose findings on these subjects and the first to involve MSF's global and historical staff base, as well as community partners. MSF's external accounting on abuse focuses on a relatively small number of formal complaints logged each year with ethics and behavioral units [11]. Our results present a startling contrast to this narrative. One in two MSF respondents (54.0%) witnessed or experienced one or more types of abuse (including discrimination) between July 2020 and July 2021. A majority (59.4%) have at one point reported abuse when reporting to line management is included. However, of these respondents, only 8.7% indicated feeling fully satisfied with outcomes, with

Introduction Type 2 diabetes is one of the leading causes of cardiovascular morbidity and mortality and contributes to an increasing disease burden in all European countries [1]. Since the 1990s, the prevalence of type 2 diabetes has increased steadily in Europe, and estimates suggest that the age-adjusted prevalence rate is expected to rise from 6.3% in 2019 to 7.8% in 2045 [2]. Evidence indicates large inequalities in the rates of diabetes and related complications between migrant and ethnic minority groups (henceforth,'migrants') and the host European-origin populations (henceforth, 'Europeans') [3]. Europe is ethnically diverse, especially in the urban centres, due to international migration. In the last few decades, political instability, particularly in low-resource regions of the world, has forced millions of people to seek a more stable future outside their home countries. Europe remains one of the main recipients of migrants from low-resource regions of the world, and migration is expected to increase in the coming decades [4]. To prevent early disease onset and diabetesrelated complications in migrants in Europe, special attention is needed to identify high-risk populations and the underlying determinants to guide prevention and treatment efforts. Drawing on evidence from the existing literature, this review discusses the burden of type 2 diabetes and its related complications and the potential explanatory mechanisms in migrants in Europe. The review also discusses evidence on prevention and treatment of diabetes in migrants and presents recommendations for addressing future challenges in Europe with special emphasis on migrants from low-resource countries. Currently, there is no universally agreed definition of what constitutes a migrant population. As a result, countries in Europe define migrants in different ways such as by nationality, country of birth and/or parental country of birth, or world region [4]. Such variations in definitions of migrants create ambiguity in comparing the health of migrants across countries. This review draws on evidence from the existing literature and we applied the various definitions used in the original studies in European countries where necessary. We used the complementary term 'host European population' to refer to people who originate from the host country. Where necessary, second-generation migrants who were born in Europe are distinguished from first-generation migrants and the host European population --- Migrant and host populations --- Type 2 diabetes burden in migrants The prevalence of type 2 diabetes differs importantly between ethnic groups in Europe. A meta-analysis of its prevalence among migrant groups in Europe, of papers published between 1994 and 2014, shows that generally, these groups had higher rates of type 2 diabetes compared with Europeans [3]. Migrants from South Asia had the highest rate, with the pooled OR being nearly fourfold higher, followed by those from the Middle East and North Africa, Sub-Saharan Africa and South and Central America, compared with Europeans (Fig. 1). Type 2 diabetes prevalence also differed importantly between people from a similar region of origin. For example, among the South Asian subgroups, the pooled OR of type 2 diabetes was sixfold higher in Bangladeshis, fivefold higher in Pakistanis and fourfold higher in Indians compared with Europeans [3]. The meta-analysis also shows a clear sex difference, with the magnitude of the ethnic differences in type 2 diabetes being greater in migrant women than in men, relative to Europeans [3]. Furthermore, several studies have found a higher prevalence of prediabetes or impaired fasting glucose (IFG) in several migrant groups, including migrant children, compared with Europeans [5,6]. Middle Eastern migrants in Sweden have been shown to have higher prevalence rates of insulin resistance [7] and combined IFG&IGT than Europeans (Fig. 2), predisposing them to a high risk of developing diabetes. Data on intergenerational differences in type 2 diabetes prevalence among migrants are scarce but seem to suggest that the burden of diabetes cuts across different generations of migrants. A recent study among three generations of Moluccans resident in the Netherlands found higher odds of type 2 diabetes across all the three generations compared with European Dutch [8]. Evidence indicates that migrants develop type 2 diabetes 10-20 years earlier [9][10][11] and at a lower level of BMI [12,13] than Europeans. The United Kingdom (UK) Biobank cohort data, for example, show that, for the equivalent prevalence of diabetes at 30 kg/m 2 in European descent participants, BMI equated to 22.0 kg/m 2 in South Asians, 26.0 kg/m 2 in African descent populations and 24.0 kg/m 2 and 26.0 kg/m 2, respectively, in Chinese women and Chinese men [12]. Diabetes prevalence was also found to be higher in migrant children (<unk>16 years) than in European children in the UK [14]. Diabetes precursors, including HbA 1c, insulin, triacylglycerol and C-reactive protein levels, have also been shown to be higher in South Asian and African-Caribbean children than in European children, even after adjusting for differences in adiposity in the UK [15,16]. The prevalence of type 2 diabetes among migrants in Europe is also far higher than among the populations in their countries of origin. For instance, findings from the Research on Obesity and Diabetes among African Migrants (RODAM) study show that age-and education-adjusted prevalence of type 2 diabetes in men ranges from threefold higher in Ghanaian migrants in London to nearly fivefold higher in Ghanaian migrants in Berlin compared with their rural Ghanaian peers [17]. In addition, Bhatnager et al. found that Punjabi people living in West London had higher fasting blood glucose than their siblings living in the Punjab [18]. Furthermore, Mbanya et al. found among populations of African descent an increased gradient of type 2 diabetes prevalence, from rural Cameroon (0.8%) to urban Cameroon (2.0%) through Jamaica (8.5%) to Manchester, UK (14.6%) [19]. --- Diabetes-related complications in migrants Complications The prevalence of diabetes-related complications also differs between migrant groups and Europeans [10,20]. Evidence indicates a high risk of microvascular and macrovascular complications in migrants with type 2 diabetes than their European counterparts [21][22][23] Dutch. The rate of coronary heart disease (CHD) was also higher in all migrant groups, with ORs ranging from nearly threefold higher in Ghanaian migrants to nearly sevenfold higher in Turkish migrants compared with European Dutch in the fully adjusted model [21]. In an 11-year follow-up of the Southall Diabetes Study, South Asians with type 2 diabetes were almost fourfold more likely than Europeans to report a history of myocardial infarction, but there was no significant difference in the risk of hypertension, stroke or amputation [10]. Furthermore, a 20-year follow-up of the UK-based Southall And Brent REvisited (SABRE) study showed that the risk of stroke was almost twice as high in South Asians and over twofold higher in African-Caribbean individuals with diabetes compared with their European peers [24]. A higher CHD rate was also observed in Iraqi migrants compared with Europeans in the All New Diabetics In Scania (ANDIS) study, a 10-year follow-up of individuals with new-onset type 2 diabetes in Sweden [25]. In contrast, the same study found a considerably lower incidence of chronic kidney disease in Iraqi migrants compared with European Swedes. --- Mortality risk The data on ethnic differences in diabetesrelated mortality remain complex. In one study that examined differences in causes of death across different migrant groups and Europeans living in six European countries, the diabetes mortality rate ratio was higher for most migrant populations born in North Africa, Sub-Saharan Africa, the Caribbean, South Asia and Turkey but was lower for those born in East Asia and Latin America (Fig. 3) [26]. These findings corroborate those of earlier studies showing the relatively higher risk of death among migrants with type 2 diabetes compared with Europeans [10,27]. By contrast, more recent studies show lower mortality risks in migrants with diabetes compared with Europeans [28]. In a nationwide 10-year longitudinal study of type 2 diabetes in Sweden, the adjusted risk of all-cause mortality and cause-specific mortality (due to CVD or cancer) was lower in first-generation migrants born in the Middle East, Asia, Africa and Latin America compared with European Swedes [11]. However, second-generation migrants with type 2 diabetes, especially those with both parents born abroad, had a 28% higher risk of all-cause mortality than the European Swedes [11]. In another recent cohort study using Clinical Practice Research Datalink data from 383 general practices in England, the adjusted risks for mortality from CVD, cancer and respiratory diseases were lower in people of South Asian and African descent with diabetes compared with European with diabetes [28]. A lower risk of death has also been observed among South Asian and Chinese people with diabetes in Canada [29]. The emerging mortality advantage among migrants with type 2 diabetes is surprising and difficult to explain especially given their relatively high risk of microvascular and macrovascular complications [21][22][23]. It is probable that this mortality advantage may be explained, at least in part, by biological or lifestyle factors that contribute to low mortality, such as low epigenetic age acceleration [30] or beneficial effect of lifestyle intervention and medication at a young age in migrants as they appear to develop type 2 diabetes earlier than Europeans [28]. --- Potential explanatory mechanisms for the type 2 diabetes burden in migrants Despite the high prevalence of diabetes and its related complications among migrants in Europe, the key underlying drivers Nevertheless, it is well acknowledged that the potential factors underlying the high risk of type 2 diabetes and its related complications in migrants are multifaceted, including postmigration factors [31], pre-migration factors and genetic predispositions [32]. The conceptual model of hypothesised causal pathways leading to the high risk of type 2 diabetes in migrants is shown in Fig. 4. It is hypothesised that premigration factors (e.g. intrauterine growth, parental socioeconomic status [SES], health behaviour), post-migration factors (e.g. host countries contextual factors, health systems and policies and lifestyle changes), genetics and shared family habits can influence socioeconomic circumstances, behaviour and biological factors, access to healthcare, physical and psychosocial stress and epigenetics upon migration, which in turn affect insulin secretion and action and subsequently type 2 diabetes risk [33]. Because of the limited data on genetics and gene-environment interactions in relation to type 2 diabetes among migrants in Europe, we discuss only premigration and post-migration factors in detail. Post-migration factors Migration is associated with huge changes in SES, lifestyle and wellbeing of migrants. One of the important changes upon migration to a new country is adjustment to the new environment. This involves changes in dietary habits, physical activity levels, psychosocial stress and socioeconomic circumstances, which are important risk factors for type 2 diabetes [33]. While migration can be beneficial in terms of improvement in SES, it can also be disastrous in terms of unhealthy lifestyle, increased psychosocial stress and limited access to preventive and curative care, with a major impact on biological factors such as obesity, IFG and subsequent type 2 diabetes [34]. Western diet, characterised by high consumption of red meat, processed meat, sweets, desserts and high-fat dairy products, has been shown to be associated with an increased risk of type 2 diabetes [35]. Evidence indicates changes in dietary behaviour upon migration [36,37], which may contribute to the high prevalence of type 2 diabetes among migrants. Gardu<unk>o-Diaz and Khokhar's study on the association between dietary patterns and the metabolic syndrome among South Asians living in the UK found a direct association between the western-pattern diet and the overall risk of the metabolic syndrome [36]. Furthermore, a sedentary lifestyle reduces energy expenditure, promotes weight gain and increases type 2 diabetes risk [38]. Changes in lifestyle (e.g. changes in body weight) upon migration can influence health, including type 2 diabetes. The RODAM study, which assessed African migrants living in three European countries (England, the Netherlands and Germany) and their nonmigrant compatriots living in Africa, showed that the prevalence of obesity among African (Ghanaian) migrant men in Europe was up to 15 times higher than in their compatriots living in rural Ghana [17]. The high prevalence of obesity in Ghanaian migrants was consistent with the higher prevalence of type 2 diabetes in migrants compared with their rural Ghanaian peers [17]. Migrants in European cities are frequently congregated in deprived neighbourhoods, where opportunities for social integration, as well as access to healthy food options, a safe walking environment, social support systems, healthcare providers and quality of care are often limited. This disadvantageous position makes it difficult for many migrants to engage in healthy lifestyles and can subsequently impact their health risk. White et al's natural experiment among refugees in Sweden showed that refugees who were dispersed to highly deprived neighbourhoods had a 22% higher risk of diabetes compared with refugees who were dispersed to lowdeprivation neighbourhoods [39]. Psychosocial stress and depression are associated with an increased risk of type 2 diabetes [40]. Many migrants experience discrimination, risk of deportation, separation from families, poor social circumstances including '3D' (dirty, dangerous and demeaning) jobs, poor housing, poor quality of sleep and pressure from their home countries to remit money to support those left behind; such difficulties may expose them to a high risk of psychosocial stress and depression. This, in turn, increases their risk of type 2 diabetes [41,42]. In the HELIUS study in the Netherlands, the prevalence of self-reported perceived discrimination ranged from 28% in Ghanaian migrants to 31% in Turkish migrants, compared with 2% in European Dutch [41,42]. The high prevalence of perceived discrimination among these migrants groups was associated with obesity [43] and the metabolic syndrome [44], important risk factors for type 2 diabetes. There is marked variation between host countries in terms of food environment, politics of migration, ethnic relations and organisation of healthcare and preventive services, and this can have a profound impact on the health of migrants both directly and indirectly. These differential contexts can influence migrants' health behaviour, socioeconomic circumstances, psychosocial stress and access to preventive services and healthcare and subsequent risk of diseases including type 2 diabetes [45]. Studies comparing similar South Asian Indian and African-Caribbean residents in England and the Netherlands found very important differences in obesity and type 2 diabetes among these groups living in these two countries [45,46]. Furthermore, the risk for type 2 diabetes is higher in women who have had gestational diabetes than in those who have not, due to defects in both insulin secretion and insulin action [47]. Evidence suggests that the prevalence of gestational diabetes is higher in most migrant groups than in Europeans, which may contribute to their high risk of type 2 diabetes [48]. Pre-migration factors Pre-migration factors may also contribute to the high prevalence of type 2 diabetes among migrants. Evidence suggests that adverse early life factors such as low birthweight and malnutrition have a profound effect on several health outcomes including type 2 diabetes in adulthood. Low childhood SES and undernutrition during early development have been shown to be associated with an increased risk of type 2 diabetes [49]. This is thought to be related to the theory of 'thrifty genotype' in which insulin resistance might improve survival rate during states of energy scarcity but would lead to type 2 diabetes in states of energy excess and the theory of 'thrifty phenotype' in which malnutrition during foetal and early life may be associated with an increased risk of insulin resistance, glucose intolerance and type 2 diabetes in adulthood [50]. Many migrants from poor-resource countries have to some extent been exposed to poor nutritional circumstances during early life, such as low birthweight, malnutrition and stunting due to poor nutrition driven by poor socioeconomic circumstances, which may expose them to high risk of type 2 diabetes upon migration to a foodabundant environment in Europe [51]. Furthermore, premigration psychosocial stress (often resulting from poverty, wars and human rights violation, which force many to flee their home countries) can have a negative impact on health, including type 2 diabetes, after migration to Europe. The migration process itself can also be treacherous, with a devastating impact on psychological wellbeing among migrants, especially among asylum seekers. According to the United Nations High Commissioner for Refugees (UNHCR), migrants routinely face horrors during their journeys such as starvation, dehydration, lack of access to medical care, arbitrary detention, kidnapping, trafficking and sexual abuse [52,53]. The horrible experiences of some migrants en route can leave psychological scars and subsequently have an impact on their physical health, including type 2 diabetes, directly or indirectly through adoption of risky behaviour such as smoking, alcoholism or sedentary lifestyles as coping strategies. A study among asylum seekers in the Netherlands found that individuals with post-traumatic stress disorder (PTSD) had higher odds of diabetes than individuals without PTSD [54]. --- Prevention and treatment of type 2 diabetes in migrants Prevention Though the lifetime risk of developing type 2 diabetes is high, it is difficult to predict and prevent type 2 diabetes in the general population. Nevertheless, individuals at high risk of type 2 diabetes, such as those who are obese, those with IFG or impaired glucose tolerance (IGT) and some migrant groups originating from South Asia, the Middle East and Africa, are appropriate candidates for preventive interventions [55]. Lifestyle intervention involving weight loss and increasing physical activity levels can improve insulin sensitivity and glucose tolerance and can prevent progression from IGT to type 2 diabetes [56]. However, evidence indicates that it is difficult to achieve sustained weight management with or without increased physical activity [57]. Furthermore, it has been well demonstrated that community engagement interventions have a positive effect on health behaviours, health consequences, self-efficacy and perceived social support outcomes across various health conditions [58]. However, most intervention trials have been conducted among nonmigrant populations, and it remains uncertain whether the results from these trials can be extrapolated to migrants with different lifestyle and cultural traditions. Thus, evidence on how best to deliver effective culturally adapted health promotion interventions to prevent type 2 diabetes, such as physical activity and healthy eating, to migrant groups in Europe is limited [59,60]. In the last few years, however, culturally adapted lifestyle modification intervention trials to prevent type 2 diabetes in high-risk South Asian adults have been conducted in Norway [61], Scotland [62], the Netherlands [63] and in Iraqi migrants in Sweden [64]. The results suggest more modest effects [65] compared with trials in Europeans [66]. In addition, evidence on the clinical effectiveness or cost-effectiveness, as well as the long-term effect, of these adapted health promotion interventions among migrants in Europe is currently lacking [67]. These observations clearly demonstrate the need to identify the key facilitators and barriers for migration-related lifestyle changes, taking into account national context [68], to support cultural adaptation and evaluation of the effectiveness of these interventions among migrant groups. Moreover, other major migrant groups, such as those of African and East Asian descent, that have an increased risk of type 2 diabetes should be included in these intervention trials. Treatment Evidence suggests that culturally appropriate education intervention can lead to improvement in glycaemic control among migrants with diabetes. A meta-analysis of 28 randomised controlled trials, including four studies in South Asians from the UK and one study from the Netherlands, on culturally appropriate diabetes health education interventions showed significant improvements in glycaemic control and diabetes knowledge over a period of 24 months among the intervention group compared with those receiving usual care [60]. However, evidence suggests that there is poor glycaemic control among migrant groups in Europe despite their high levels of diabetes awareness and treatment [69][70][71]. In one Dutch study, the awareness and medical treatment of type 2 diabetes were two to five times higher in migrants than in European Dutch individuals, although levels of awareness/ treatment in those with glycaemic control (HbA 1c levels on target <unk>52 mmol/mol) ranged from 37 to 53% in migrant men compared with 67% in European Dutch men [9]. A study in Scotland also found higher rate of suboptimal glycaemic control in migrants compared with European Scots despite migrants being generally younger and having lower BMI [72]. The poor glycaemic control among migrants has been suggested to be due to poor adherence to treatment and lifestyle recommendations, possibly due to low health literacy, poor care standards, poor quality of care or ineffective response to glucose-lowering agents [9]. Furthermore, a large proportion of individuals who migrate from the Middle East and South Asia to Europe have insulin-deficient diabetes, which may also contribute to their poor diabetes control [13,25]. These observations suggest the need for greater efforts to improve the effectiveness of type 2 diabetes treatment in migrant groups in Europe, taking into account the migrants' personal perspectives on diabetes. This is highly relevant because a systematic review of various ethnic minority patients' views on self-management of type 2 diabetes identified several analytical themes that highlight the complex nature of self-management of type 2 diabetes among these populations [73]. How patients identified themselves and the feeling of being understood by health professionals about their culture and by family members regarding their condition were of great importance to successful self-management, while patients' own limited understanding of diabetes was a barrier to self-management of their condition [73]. --- Conclusions and recommendations Type 2 diabetes and related microvascular and macrovascular complications remain a major burden among migrant populations in Europe. However, the extent of the burden varies across migrant groups and the countries in which they now live. Earlier studies found higher mortality rates among migrants, but recent studies seem to suggest a shifting trends in favour of migrants. Explanations for the high risk of type 2 diabetes among migrants in Europe are mainly speculative. Evaluation of diabetes interventions among migrants are limited and mainly focus on South Asian populations. The level of awareness of diabetes among migrant populations is high, but glycaemic control remains suboptimal relative to Europeans. These observations call for investment in prospective studies and basic scientific research to gain insight into the causal pathways linking migration to the development of type 2 diabetes. The research should include the role of genetics, epigenetics, early life factors, key specific migration-related lifestyle changes and psychosocial stressors. There is also a need for intervention trials of longer duration to investigate clinical outcome measures such as development of microvascular and macrovascular complications and to evaluate the cost-effectiveness of the adapted interventions. The current data on the culturally adapted lifestyle modification intervention trials to prevent type 2 diabetes are mainly based on South Asian adults in Europe, suggesting the need for more work among other major migrant groups such as African-Caribbeans, Sub-Saharan Africans, North Africans, East Asian and Middle Eastern populations, such as Turkish and Iraqi populations, in whom type 2 diabetes is also highly prevalent. Lastly, in order to improve type 2 diabetes treatment outcomes among migrant groups in Europe, more work is needed to gain better understanding of factors driving the poor treatment outcomes and to help develop cultural tailored interventions for migrants with type 2 diabetes. --- Supplementary Information The online version contains a slideset of the figures for download, which is available to authorised users, available at https://doi.org/10.1007/s00125-021-05586-1. --- Authors' relationships and activities The authors declare that there are no relationships or activities that might bias, or be perceived to bias, their work. Contribution statement All authors were responsible for drafting the article and revising it critically for important intellectual content. All authors approved the version to be published. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

European populations are ethnically and culturally diverse due to international migration. Evidence indicates large ethnic inequalities in the prevalence of type 2 diabetes. This review discusses the burden of type 2 diabetes and its related complications, and the potential explanatory mechanisms among migrants in Europe. The current available data suggest that the rate of type 2 diabetes is higher in all migrant groups and that they develop this disease at an earlier age than the host European populations. The level of diabetes awareness among migrant populations is high, but glycaemic control remains suboptimal compared with Europeans. The culturally adapted lifestyle modification intervention trials to prevent type 2 diabetes mainly focus on South Asian adults in Europe. Diabetes-related microvascular and macrovascular complications remain a major burden among migrant populations in Europe. Earlier studies found higher mortality rates among migrants, but recent studies seem to suggest a shifting trend in favour of first-generation migrants. However, the extent of the burden of type 2 diabetes varies across migrant groups and European countries. Despite the higher burden of type 2 diabetes among migrants, the key underlying factors are not well understood mainly due to limited investment in basic science research and development of prospective cohort studies. We hypothesise that the underlying risk factors for the high burden of type 2 diabetes and its related complications in migrants are multifaceted and include pre-migration factors, post-migration factors and genetic predispositions. Given the multi-ethnic nature of the current European population, there is a clear need for investment in research among migrant populations to gain insight into factors driving the high burden of type 2 diabetes and related complications to facilitate prevention and treatment efforts in Europe.Ethnic minority groups . Europe . Migrants . Review . Type 2 diabetes Abbreviations CHD Coronary heart disease HELIUS Healthy Life in an Urban Setting IFG Impaired fasting glucose PTSD Post-traumatic stress disorder RODAM Research on Obesity and Diabetes among African Migrants SES Socioeconomic status

Introduction A positive correlation between the fertility of children and their parents has been observed in many populations of humans and in several other species (Heyer et al., 2005). This phenomenon of fertility transmission (FT) is highly variable, with significant differences found among populations, at different time periods and between sexes (Murphy, 1999). As fertility is one of the major life-history traits, the relative contribution of genetic, social and environmental factors on fertility has been a subject of considerable debate for over a century among geneticists, demographers and economists (see e.g. Kohler et al., 1999). In particular, FT is known to reduce effective population size and to increase inbreeding (Robertson, 1961 ;Nei & Murata, 1966 ;Wray & Thompson, 1990) and to have a strong impact on the frequency of genetic diseases in some human populations (Austerlitz & Heyer, 1998). FT can be caused by genetic or cultural factors. An example of the cultural transmission of fertility (CTF) comes from the Saguenay-Lac Saint Jean (SLSJ) population in Quebec, where intergenerational correlations in effective family size (EFS) are much higher than correlations in census family size (CFS), which was interpreted as evidence of a cultural rather than genetic basis for FT (Austerlitz & Heyer, 1998). Indeed, EFS only considers individuals who reproduced at least once in their native population and is thought to be strongly influenced by cultural factors, such as transmission of wealth or migration behaviour. In other human populations, evidence of CTF is identified through the simultaneous observation of specific allelic frequencies distributions and of transmission of cultural traits linked to fertility. For instance, the transmission of female social ranks observed in Maoris was proposed to explain the peculiar frequency distribution of mitochondrial DNA (mtDNA) haplotypes (Murray-McIntosh et al., 1998). Similarly, cultural transmission of higher male social status initiated by Genghis Khan and his relatives was suggested to explain the increase in gene frequency of some closely related Y-chromosome haplotypes in modern Asia (Zerjal et al., 2003;Dulik et al., 2011). Thus, CTF is observed in several human populations, even though it is not a general phenomenon (Lansing et al., 2008). Finally, note that CTF was also observed in non-human species, such as matrilineal whales (Whitehead, 1998) and cheetahs (Kelly, 2001), where migration strategy, foraging techniques and vigilance against potential predators are culturally transmitted through the generations. Contrary to the above studies, which emphasize the role of social and cultural determinants, other research has instead advocated a significant genetic basis for FT. The intergenerational correlations in sibship size observed in Huterrites, for example, probably results, in part, from genetic factors, because of the high social and environmental homogeneity of this population (Pluzhnikov et al., 2007;Kosova et al., 2010). Similarly, a Danish study of twins found moderate genetic heritability for fertility (Rodgers et al., 2001). Actually, any pattern of selection on a quantitative trait occurring during several generations will lead to FT in any species (e.g. in wheat, Goldringer et al., 2001). Note that in all cases, this genetic FT is not linked to selection at a single locus but on a quantitative trait coded by many loci located throughout the genome. Finally, FT may even be linked to both genetic and social factors in some cases (Fre<unk>re et al., 2010). The impact of FT on neutral genetic diversity can be theoretically studied using a parametric model simulating the transmission of various degrees of reproductive success. Using haploid models without recombination, Sibert et al. (2002) and Blum et al. (2006) showed that, for neutral genes, coalescent trees under FT are expected to differ substantially from standard coalescent trees (Kingman, 1982) in the following ways : (i) the time to the most recent common ancestor (TMRCA) is reduced, resulting in lower genetic diversity ; (ii) the ratio between external and internal branch lengths is increased, yielding more star-like trees ; (iii) coalescent trees tend to be highly imbalanced, i.e. for a given binary node, the distribution of tips is not balanced between the two subtrees. While a star-like tree can also be observed in cases of population expansion (Slatkin & Hudson, 1991), tree imbalance for neutral sequences is much more specific of FT (Sibert et al., 2002 ;Blum et al., 2006). Estimating the imbalance of coalescence trees reconstructed from samples of neutral DNA sequences, therefore, constitutes an unambiguous method by which to infer FT (Blum et al., 2006). This is quite helpful because genealogical data of sufficient depth are rarely available, while current population genetics data can be obtained more easily. Using mitochondrial HV1 sequences, Blum et al. (2006) could thus show that matrilineal FT is more frequent in hunter-gatherer populations (HGPs) than in foodproducer populations (FPPs). This approach requires the absence of recombination, restricting the application of the model to the uniparentally inherited non-recombining sections of the Y-chromosome, and the mitochondrial genome. While these haploid markers provide information on a unique gene history, using several independent markers increases the ability to detect genome-wide processes (Nielsen, 2001). Another limitation of nonrecombining uniparentally transmitted loci is the difficulty to discriminate between FT and selection at a single locus. This is important, because non-neutral evolution is common for mitochondrial DNA (Elson et al., 2004 ;Ruiz-Pesini et al., 2004 ;Bazin et al., 2006 ;Stewart et al., 2008 ;Balloux et al., 2009). Maia et al. (2004) demonstrated theoretically that purifying selection also leads to tree imbalance at the locus under selection. Interestingly, Kivisild et al. (2006) suggested a purifying selection process that would differ between populations with a diet based on grains or not. This may be a part of the explanation of the more imbalanced genealogical trees inferred in HGPs (Blum et al., 2006). Analysing many independent autosomal neutral markers is obviously a more powerful strategy to infer FT than one that relies on a unique gene history, possibly affected by selection. A further limitation of previous theoretical studies on FT is that they do not distinguish between sexes. Yet, empirical data show that fertility can be transmitted by the mother alone (matrilineal FT) as in the British (Pearson et al., 1899 ;Murphy, 2006) and Icelandic populations (Helgason et al., 2003), by the father alone (patrilineal FT) as suspected for the Mongols (Zerjal et al., 2003) or by both parents (biparental FT) as in the French Canadian (Austerlitz & Heyer, 1998), Hutterite (Pluzhnikov et al., 2007) and Danish populations (Rodgers et al., 2001). A joint analysis of biparentally and uniparentally inherited markers (autosomal, X-linked, mitochondrial and Y-linked markers) should help to differentiate the sexspecific contributions to FT, in the same manner as sex differences were found for other demographic forces such as migration (Se <unk>gurel et al., 2008). Moreover, other demographic phenomena may occur simultaneously with FT. One such factor is an increased heterogeneity in family size among couples relative to expectations under the Wright-Fisher model. Such heterogeneity may stem from socioeconomic conditions and behavioural characteristics of the parents (Ronsmans, 1995 ;Sastry, 1997) or from other sources of heterogeneity such as individuals'' frailty'at birth (Vaupel et al., 1979). In the SLSJ population in Quebec the observed high frequencies of severe genetic disorders can only be explained by the conjunction of FT and heterogeneity in family size (Austerlitz & Heyer, 1998). Finally, one may question whether spouses tend to marry more often with partners from families with sizes similar to their own family. Mate choice is generally not random and may depend on social factors such as material resources, cultural factors (e.g. religion) and biological factors (e.g. indicators of good genes in the context of sexual selection) (Geary et al., 2004). As social factors may affect both mate choice and number of siblings, homogamy in sibship size is a possible consequence of assortative mating practices, and has been observed in several human populations (Imaizumi et al., 1970;Bocquet-Appel & Jakobi, 1993 ;Murphy, 2006). In this paper, we study the joint impact of all these factors on demographic features and on the shape of gene trees inside populations. For this purpose, we simulated a population of diploid individuals with separate sexes, carrying genes belonging to the four genomic compartments : autosomal genes; X-linked genes ; Y-linked genes; and mitochondrial genes. These loci have biparental, maternally biased, paternal or maternal inheritance, respectively. We investigated the joint impact of the level of FT, heterogeneity in family size, homogamy in sibship size, and the population size, on the variance in reproductive success among individuals in a given generation together with the correlation of this reproductive success between parents and offspring. We next investigated how these phenomena jointly affect the level of imbalance of coalescent trees, as measured by an index that we modified from Fusco and Cronk (1995) to allow for non-binary nodes. This enabled us to establish the differences between our diploid model and the haploid model previously studied by Sibert et al. (2002) and Blum et al. (2006), and to assess the conditions that resulted in a maximal level of imbalance. We also evaluated the impact of FT on the different genomic compartments (autosomal, X-linked, Y-linked and mitochondrial), under biparental, matrilineal and patrilineal transmission, and determined the amount of independent nuclear loci that would be necessary to detect FT as a function of the intensity of this phenomenon. --- Materials and methods --- (i) Simulations Populations were simulated using a forward individualbased approach where the pairing rules of individuals and the Mendelian rules of gene transmission from parents to children were explicitly implemented. This allowed us to incorporate a flexible model of FT into the classical Wright-Fisher framework. All simulations were performed assuming non-overlapping generations and a constant-size population of N individuals. In order to study the properties of the resulting coalescent trees, we stored the genealogical links that were progressively built between gene copies in each simulation. In other words, we did not directly simulate coalescent trees with a backward algorithm, but we considered the gene trees that resulted from our forward-in-time simulation procedure. For each simulation, we randomly sampled n=100 gene copies in the last simulated generation and we extracted their coalescence tree. We performed first a set of simulations in a haploid model as in Sibert et al. (2002) and Blum et al. (2006). In this model, each child had a single parent, which was drawn randomly among the individuals of the previous generation according to a probability distribution modelling FT. As in Sibert et al. (2002), the probability p i for a given individual i to be chosen as parent was p i = c i (a)rs a i g M j=1 c j (a)rs a j,(1) where M was the number of individuals (M=N), s i the sibship size of individual i (i.e. the progeny size of its own parent), a the intensity of FT, and c i (a) was a random deviate drawn independently for each individual i from a gamma distribution with shape parameter a and mean 1. We considered a wide range of intensities of FT, from no FT (a=0) to very high FT (a=2). The a-parameter is a way to control the variance in reproductive success among individuals, as this variance increases when a decreases. As in Sibert et al. (2002), we only considered the cases (a=') and (a=1), which correspond respectively to a Poissonlike and a geometric-like distribution of the offspring number, the latter having a higher variance. Then we modified this model to consider a population of diploid individuals with separate sexes assuming a constant 1 : 1 sex-ratio and simulated the transmission of autosomal, mitochondrial, X-linked and Y-linked genes. We assumed that each individual produced offspring with one partner only (strict monogamy). In each generation, couples were formed either randomly or under a model of homogamy for sibship size adapted from McKusick et al. (1990). In this latter case, we first ranked the individuals according to their sibship size. An individual coming from a family of rank k (i.e. with k-offspring) randomly chose a mate among the individuals coming from families of rank between (kxs/2) and (k+s/2) (s thus being the 'homogamy window size'). Therefore, the choice of mates was limited to individuals with similar sibship sizes. When such mates were not available, the actual mate was chosen among families with the nearest rank. Unlike in McKusick et al. (1990), the ranking was not circularized in order to avoid mating between individuals with opposite rankings (an additional set of simulations confirmed that results were not sensitive to this hypothesis, results not shown). This parametric pairing rule allowed us to quantitatively model homogamy for sibship size ; i.e. the level of correlation between sibship sizes increased as the homogamy window size (s) decreased (Fig. 1). Once the couples were formed, we then drew from them the parents for each individual in the next generation. The probability p i for a given couple i to be chosen as parents was computed with eqn (1) except that M was the number of couples (M=N/2) and s i was either the mean sibship size of the two parents, the sibship size of the mother or the sibship size of the father; chosen in order to simulate biparental, matrilineal or patrilineal FT, respectively. As in the haploid model, the parameter a controlled the level of FT and the parameter a controlled the level of heterogeneity in family size. In order to be consistent, we performed the comparisons between haploid and diploid models for the same number of gene copies (N c ). In the haploid model, we have N c =N, since each individual carries one allele for each locus. However, in the diploid model, the number of gene copies for a given locus in a population of N individuals varies according to the compartment : N c =2N for autosomal loci whereas, because we assumed a 1 : 1 sex-ratio ; N c =N/2 for Y-linked and mitochondrial loci ; N c =3N/2 for X-linked loci. For each simulation, we recorded the variance in reproductive success, the intergenerational correlation in this reproductive success, and the level of imbalance of the coalescent tree, as described below. --- (ii) Variance and correlation in reproductive success Comparing the reproductive success in haploid and diploid models is not trivial, as in the first case the number of offspring of an individual is the same as the number of copies for which a gene of this individual will be found in the population in the next generation, whereas in a diploid model a gene of a given individual is transmitted on average to only half of its offspring. In order to make valid comparisons between haploid and diploid models, we considered in all cases the'gene reproductive success'(GRS), defined for each gene in a given generation as the number of copies of this gene occurring in the next generation. For all simulations, we computed the variance in GRS among genes in the final generation and the correlation between the GRS of a gene in the final generation and the GRS of its parent. In diploid models, we computed these values for each genomic compartment (mtDNA, X-linked, Y-linked and autosomal). As variances and correlations stabilize rapidly, each simulation was run until the 100th generation and we computed the mean values of variance and correlation in GRS over 1000 replicates for each set of parameters. (iii) Tree imbalance analysis FT reduces the height of coalescent trees (Sibert et al., 2002) and thus increases the frequency of polytomies (i.e. when a node has more than two direct subnodes). Therefore, in order to measure tree imbalance in this context, we developed a modified version of the imbalance index I of Fusco and Cronk (1995) used in its unbiased form (Ik) in Blum et al. (2006), as this index does not permit the analysis of non-binary nodes and is biased for nodes with an even number of tips (Purvis et al., 2002). For each node in a tree, this modified index (I nb ) was computed as: I nb = Bxm k, t Dxm k, t, where k is the number of direct sub-nodes of the studied node (e.g. 2 for a binary node, 3 for a tertiary node, etc.) ; t is its number of tips ; B is the maximum number of tips across all its sub-nodes ; D=t-k+1 is the maximal possible value for B ; m k,t was a correction factor allowing the expected value of I nb to be 0. 5 in a standard population (without FT). It is computed as m k,t =2 B k,t,coal -D, where B k,t,coal is the expected value of B in a standard population (without FT). B k,t,coal was obtained in practice by performing repeated simulations of a random Kingman's coalescent for t tips that was stopped when only k nodes remained in the genealogy. In each simulation, we recorded the B-value as the number of tips of the node with the highest number of tips among these k nodes. B k,t,coal was obtained by averaging the B-values over 1000 independent simulations. To handle polytomies, we only considered the nodes with a minimum of t=k+1 tips (Supplementary Fig. S1, available at http://journals.cambridge.org/grh). As for Fusco and Cronk's (1995) I index, the expected value of I nb was 0. 5 for each node in the coalescent for a standard population, whereas it was greater than 0. 5 when the nodes were imbalanced, as for instance under FT. The imbalance measure of the whole tree was simply the mean of I nb across all its nodes on which it could be computed, only removing the most terminal nodes (i.e. the nodes occurring less than four generations before the present), as a preliminary study had shown us that using these nodes was biasing downward the I nb value of the tree (result not shown). Each retained node was given an equal weight in the final measure of total tree imbalance, as in Blum et al. (2006). Finally, we estimated the expected value of I nb for each set of parameters by averaging 500 independent replicates. (iv) Impact of the number of loci on the possibility to detect FT We also studied the number of independent loci necessary to detect FT for a given level of transmission (a). In order to do so, we performed sets of simulations in which we simulated a given number (x) of independent loci and computed the mean I nb value over these x loci ; for x=1, 2, 5, 20 and 40. We performed 500 replicates for each value of x and a and computed the 10th percentile of the mean imbalance statistics as a function of the intensity of FT a. This allowed us to obtain for each value of x, the minimal value of a for which at least 90% of the replicates showed an I nb greater than 0. 5. --- Results --- (i) Haploid model Considering first the haploid model without heterogeneity in family size (a='), the variance in GRS and the level of correlation in GRS between parents and offspring increased with the level of FT (a), whatever the assumed population size (N). Starting from a value of 1. 0 as expected in the standard Wright-Fisher model (Fig. 2a), variance increased at first slowly with a for values up to 1. 2, but then increased more rapidly to reach values of about 3. 0 for a=2. The strength of correlation showed the opposite pattern (Fig. 2b) ; starting from the expected null value for a=0, it increased first rapidly until a=1. 2, but much more slowly afterwards. Starting from the expected value of 0. 5 for a=0, the level of imbalance I nb increased with a according to a sigmoid shape with an inflexion point for a=1. 0 (Fig. 2 c). Regarding the impact of population size (N), for a<unk>0. 8, the levels of variance and correlation were similar for all N values (Figs 2 a and b). Nevertheless, for higher values of a, variance increased substantially with N, whereas the correlation slightly decreased with N. The pattern was more complex for I nb ; the higher the population size N, the steeper the slope at the inflexion point (Fig. 2 c). The initial increase of I nb was thus delayed for high N, while the opposite pattern was observed after the inflexion point. The final pattern also was complex as I nb reached a plateau for the highest N, whereas it kept increasing for the lowest N. These parameters were also affected by the level of heterogeneity in family size (Fig. 3). Variance in GRS was larger when heterogeneity in family size was higher (a=1) than when it was lower (a='), especially for high a-values (Fig. 3a); the ratio between the variances obtained in these two cases increased from 2. 0 for a=0 to 3. 3 for a=2. Conversely, the correlation in GRS between generations was lower for a=1 than for a=' (Fig. 3b). Finally, I nb was higher for a=1 than for a=' as a result of a left-shift of the curve in the case of high heterogeneity in family size (Fig. 3c). The difference was particularly noticeable for small values of a: for instance for a=0. 8, I nb was only y0. 53 when a=', whereas it was y0. 64 when a=1. (ii) Diploid model We then investigated whether the levels of variance and correlation in GRS were different between the haploid and the diploid models (Fig. 4). The levels of variance and correlation in GRS were much lower for the diploid model with biparental transmission than they were for the haploid one (Figs 4 a and b). Regarding I nb, it increased rapidly in the haploid model, until it reached a plateau, whereas it increased slowly but continuously in the diploid model (Fig. 4 c). In this diploid model, we found no differences among the different genetic compartments (autosomes, X-linked, Y-linked and mtDNA) in their levels of variance and correlation, as long as FT was biparental (Figs 4 a and b). Nevertheless, we observed that I nb values were slightly higher for the haploid compartments (mitochondrial and Y-linked) than for the autosomal and X-linked compartments (Fig. 4 c). When considering matrilineal transmission, the level of variance in GRS increased substantially with a in a similar manner for all four compartments, even for the strictly paternally transmitted Y-chromosomes (Fig. 5 a). This contrasted with the patterns observed for the correlation (Fig. 5b) and I nb (Fig. 5 c), which were very strong for mitochondrial genes, but reached lower values for X-linked genes, even lower values for autosomal genes and null for Y-linked genes. A symmetric pattern was observed for patrilineal transmission : variance was the same for all compartments (Supplementary Fig. S2 a, available at http://journals.cambridge.org/grh), whereas the correlation and I nb values were the highest for Y-linked genes and then were lower for autosomal genes, even lower for X-linked genes and null for mitochondrial genes (Figs S2b andc). Finally, as in the haploid model, variances in GRS and imbalance I nb were higher in the diploid model with high heterogeneity in family size (a=1) than in the standard diploid model (a='), while intergenerational correlations in GRS were lower in the former than in the latter (Supplementary Figs S3-S5, available at http:// journals.cambridge.org/grh). --- (iii) Homogamy in family size We observed that both the correlation and the variance in GRS increased as the level of homogamy increased (i.e. as s decreased, Supplementary Figs S6a andb, available at http://journals.cambridge.org/ grh). Regarding I nb, we observed no difference for low a values (a<unk>0. 6) between the situations with or without homogamy (Fig. 6). For intermediate a values (0. 6<unk>a<unk>1. 4), the addition of homogamy yielded an increase in I nb, which increased as s decreased. However, for large a-values (a>1. 4), I nb was larger without homogamy and decreased as the level of homogamy increased. (iv) Impact of the number of loci used on the possibility of detecting FT We studied the impact of increasing the number of loci on the power of detecting FT in populations. We observed that the threshold value for a above which at least 90% of the I nb values are larger than 0. 5 decreased with the number of loci used, from 1. 65 with one locus to 1. 16 with 40 loci (Table 1). --- Discussion --- (i) Interaction between parameters in the haploid model Using an individual-based model of FT, our simulations have provided a theoretical framework to detect and understand FT and its impact on the demographic patterns of the population (namely the variance and correlation between parents and offspring in reproductive success) and on the level of imbalance of genealogical trees, as measured by our new index (I nb ) that can account for polytomies in the tree. When demographic data are available, FT can be directly measured through the correlation between parents and offspring in reproductive success. In most cases, however, demographic data are not available ; so FT cannot be inferred directly. Therefore, Blum et al. (2006) developed a method to detect FT from genealogical tree imbalance. We showed here with our simulation study that the imbalance index (I nb ) increases in a population submitted to FT. This index, however, is affected by other parameters, in particular population size (N) and heterogeneity in family size (a). Regarding the latter parameter, I nb increased when heterogeneity increased (Fig. 3), which is a consequence of the increased variance in GRS. By definition, a genealogical tree can be imbalanced only if some of its nodes have more tips than others, and this is only possible if some individuals have a larger reproductive success than others. Thus, an increased level of heterogeneity in progeny size will increase the potential to observe tree imbalance, but this imbalance will only occur in the presence of FT, because the nodes with more direct sub-nodes will always occur in the same part of the tree. In other words, variance is the prerequisite for the creation of imbalance in the genealogical tree, but this imbalance can only be maintained in subsequent generations when GRS is correlated between generations. It is interesting however to note that increasing the level of heterogeneity in progeny size resulted in reductions in the level of correlation for a given level of FT (a), while I nb simultaneously increased. This decrease in correlation results naturally from the increase of variance, which appears in the denominator of the formula used to compute the correlation coefficient. However, this decrease in correlation does not preclude an increase of I nb, demonstrating that the relation between I nb and correlation is not completely straightforward. This is because of the increased variance in reproductive success, which means that a higher value of I nb can be reached when variance is increased, even if the correlation is lower. The other factor that affected the relation between I nb and FT is population size (N). An increase in population size led to an increase in variance and correlation for a given value of a (Fig. 2). Hence, in small populations, variances and covariances between generations were limited by the total size of the population. As a result, I nb was higher for large population sizes than for small ones, for moderate FT values (1<unk>a<unk>1. 6). However, the opposite pattern was observed for lower levels of FT (a<unk>1), indicating a rather complex interaction between the parameters. From a theoretical perspective, it is interesting to observe that under FT the topology of the tree consequently depends on population size. This contrasts strongly with the classical Kingman's coalescent process, in which population size does not affect the shape of the coalescent tree, but is only a scaling factor for its branch lengths (Kingman, 1982;Hudson, 1990). Thus, under FT, trees cannot be simply normalized by population size as in the classical Kingman's model. Regarding the possibility to detect FT in natural populations from the inferred coalescent trees, without heterogeneity in family size, the I nb value hardly deviates from 0. 5 for a-values below 0. 8 (Fig. 2 c). Thus, this value of 0. 8 appears as the minimal value for which FT could be detected. This would correspond to the correlation of 0. 394 (Fig. 2 b), which is higher than the highest correlation ever observed in human populations, namely the Hutterite population (Pluzhnikov et al., 2007). Here, our conclusions differ from those of Blum et al. (2006), who stated that even without heterogeneity in family size, FT could be detected in human populations. However, their inferences were based on the simulations of Sibert et al. (2002), which considered populations of 50 individuals. In such small populations, when a<unk>1. 0, correlations are expected to be lower and I nb values higher than in larger populations. Thus, the conclusion that FT can be detected in populations without heterogeneity in family size does not hold in larger populations (No500). However, when there is heterogeneity in family size, I nb deviates substantially from 0. 5 for a-values as low as 0. 5 (Fig. 3c), corresponding to a correlation value of 0. 22, which is realistic for human populations. Thus, it is probable that the populations, in which FT was detected through tree imbalance estimations such as HGPs (Blum et al., 2006), were also subjected to other factors such as heterogeneity in family size, which is a common phenomenon in human populations (Vaupel et al., 1979 ;Ronsmans, 1995;Sastry, 1997 ;Austerlitz & Heyer, 1998). In this context, it is interesting to note that the high frequency of rare diseases alleles in Saguenay-Lac-Saint-Jean was also explained by an interaction between FT and high heterogeneity in family size (Austerlitz & Heyer, 1998). (ii) Diploid model Furthermore, simulating diploid individuals with autosomal, X-linked, Y-linked and mitochondrial genes allowed us to predict the level of variance and correlation in GRS, as well as gene tree imbalance, for the different kinds of genes under biparental, patrilineal or matrilineal FT. It is clear that the haploid model is a good approximation of the impact of FT on mitochondrial genes for matrilineal scenarios (Fig. 5) and Y-linked genes for patrilineal scenarios (Fig. S2). This approximation does not hold, however, for autosomal and X-linked genes, for which the pairing process of individuals has to be taken into account. Here we have demonstrated that the variance, correlation and tree imbalance associated with the different compartments depended on the characteristics of FT. Firstly, under matrilineal transmission tree imbalance and correlation in GRS were higher for mitochondrial genes than for autosomal and X-linked genes, and were absent for Y-linked genes as expected. However, it is interesting to note that even when FT is strictly matrilineal, as is probably the case for the Maoris (Murray-McIntosh et al., 1998) and the cetaceous species (Whitehead, 1998 ;Fre<unk>re et al., Table 1. Minimum levels of FT (a) for which 90 % of the imbalance index values (I nb ) are higher than 0. 5 as a function of the number of loci used to compute this index. We assumed autosomal loci in a diploid model without homogamy in family size (s=0) under biparental transmission of fertility and without heterogeneity in reproductive success (a='), assuming a population size N=2000 and a sample size n=100. 2010), the variance in GRS is the same for all compartments. Although the Y-chromosome is not directly affected by

Fertility transmission (FT) is a phenomenon with a cultural and/or genetic basis, whereby a positive correlation exists between the number of offspring of an individual and that of his/her parents. Theoretical studies using a haploid individual-based model have shown that FT increases the variance and intergenerational correlation in reproductive success and results in an imbalance in the coalescent tree of sampled genes. This phenomenon has been documented in several demographic studies conducted on the correlation in fertility between generations, or through the reconstruction of the genealogical trees of mitochondrial DNA sequences. However, as mtDNA is a single locus, potentially subject to other forces (e.g. natural selection), it is of interest to extend the theory of FT to nuclear loci. We show that because random mating between individuals leads to a mixing of their fertility profiles, FT in these cases will have less influence on the variance and intergenerational correlation of reproductive success. This, in turn, results in less impact on the shape of the coalescent trees. Nevertheless, in the presence of FT, high heterogeneity in reproductive success and homogamy for family size will increase the imbalance in the coalescent tree. Thus, FT should be easier to detect when occurring in conjunction with these other factors. We also show the utility of analysing different kinds of loci (X-linked, Y-linked, mitochondrial and autosomal) to assess whether FT is matrilineal, patrilineal or biparental. Finally, we demonstrate that the shape of the coalescent tree depends upon population size, in contrast to the classical Kingman's model.

matrilineal transmission tree imbalance and correlation in GRS were higher for mitochondrial genes than for autosomal and X-linked genes, and were absent for Y-linked genes as expected. However, it is interesting to note that even when FT is strictly matrilineal, as is probably the case for the Maoris (Murray-McIntosh et al., 1998) and the cetaceous species (Whitehead, 1998 ;Fre<unk>re et al., Table 1. Minimum levels of FT (a) for which 90 % of the imbalance index values (I nb ) are higher than 0. 5 as a function of the number of loci used to compute this index. We assumed autosomal loci in a diploid model without homogamy in family size (s=0) under biparental transmission of fertility and without heterogeneity in reproductive success (a='), assuming a population size N=2000 and a sample size n=100. 2010), the variance in GRS is the same for all compartments. Although the Y-chromosome is not directly affected by matrilineal transmission, the application of strict monogamy in our model (pairing each unique male with a unique female) results in the variance in reproductive success being the same for both sexes. The effective population size will, therefore, be decreased for the Y-chromosome due to FT occurring between mothers and daughters. The reason for this was pointed out by Sibert et al. (2002), who noted that individuals are no longer equivalent in the presence of FT because, unlike in the classical Wright-Fisher model, they no longer have the same propensity to reproduce. Some male reproductive strategies, such as serial or simultaneous polygamy should, to some extent, decouple this effect, but the variance in reproductive success of males will still be affected by the variance in the reproductive success of the females they marry. Conversely, for a patrilineal FT transmission, such as in the Mongol population in Eurasia (Zerjal et al., 2003), we expect higher correlation and imbalance for Y-linked genes than for X-linked and autosomal genes, and neither correlation nor imbalance for mitochondrial genes. However, the effective population size of these mitochondrial genes should still be decreased as a result of the increased variance in GRS. In contrast, biparental FT, as for example described in the French Canadian population (Austerlitz & Heyer, 1998), will have a similar impact on all the compartments. Investigating different genetic compartments, therefore, could allow researchers to assess whether FT is sex-biased or not in their study population, provided that the criteria for the detection of FT are met. Contrasting the autosomes and X-chromosomes should be particularly relevant in this context, as they harbour many independent neutral loci ; while the non-recombining parts of the Y-and mitochondrial chromosomes may be subjected to selective processes, which may affect their tree imbalance. One of the prerequisites to detect FT is the occurrence of heterogeneity in family size, as it increases the level of imbalance for a given level (a) of FT. However, the random mating of individuals in the standard diploid model leads to a decrease of imbalance, rendering it potentially more difficult to detect FT in populations. The introduction of homogamy in family size to our model, however, reduced the strong averaging effects between very dissimilar mates, leading to an increase in the variance and correlation in GRS in proportion to the level of homogamy (Fig. S6). This effect also led to an increase of I nb for intermediate levels of FT (0. 6<unk>a<unk>1. 4) (Fig. 6). Interestingly, homogamy in family size has been observed in several human populations such as the Arthez d'Asson village in France (Bocquet-Appel & Jakobi, 1993), the English (Murphy, 2006) and the Uto in Japan (Imaizumi et al., 1970). Combined with heterogeneity in family size, it can substantially increase I nb and their presence should facilitate the improved detection of FT in human populations. (iii) Detecting FT in practice We showed that increasing the number of loci allows one to detect lower levels of FT (Table 1). If only one locus is studied, the threshold value of a above which FT can be detected is of 1. 65, which corresponds to a correlation of 0. 32 in GRS (Fig. 4b), which is quite a high value for human populations (Pluzhnikov et al., 2007) ; whereas for 20 loci this correlation drops to a more realistic value (y0. 21). Studying as many loci as possible, therefore, is crucial to an increase in the power of detection for the presence of FT in human populations. The analysis of many autosomal neutral markers could provide a means to distinguish between FT and selection on a single locus, as for instance in the case of HGPs, where FT was inferred from mitochondrial HVR1 sequence data only (Blum et al., 2006). FT should affect all neutral loci in the same way, whereas single-locus selection should only affect the target locus. Inferring tree imbalance for several neutral nuclear DNA sequences could, therefore, facilitate the separation of these two phenomena. Finally, it should be noted that Blum et al. (2006) demonstrated that population structure can lead to an imbalance of the coalescent tree in rather extreme cases. Therefore, it may be helpful to perform an analysis with a clustering program, such as Structure (Pritchard et al., 2000), in order to detect any spurious structure within a population under study. (iv) Future studies One practical problem envisaged is that nuclear DNA sequences are prone to intragenic recombination. This can lead to some distortions when reconstructing the genealogical tree using phylogenetic methods (Schierup & Hein, 2000), necessitating an appropriate design for the analysis of nuclear DNA sequences. One solution to this could be to restrict the analyses to 'haploblocks ', i.e. portions of sequences that have not undergone any recombination events. Detecting these haploblocks can be achieved through different methods, such as the level of linkage disequilibrium (measured for instance by D or Dk), or the four gamete test (Hudson & Kaplan, 1985). If the number of mutations is sufficiently large on any given haploblock to build a phylogenetic tree, we may expect to be able to analyse the imbalance of these trees and thus detect FT. The haploblocks identified by the HapMap project (The International HapMap Consortium, 2003) show, on average, between 19 and 25 single-nucleotide polymorphisms (SNPs) (depending on the population), with some blocks showing much higher numbers. There should be sufficient information in such polymorphic haploblocks to reconstruct the phylogenetic trees. Otherwise, the process of recombination may need to be taken into account, through the development and testing of alternative methods, such as Approximate Bayesian Computation (Beaumont et al., 2002). Additional theoretical studies could also be performed, in particular on the impact of FT on parameters such as the reproductive value, which was shown to be a central parameter in the prediction of the genetic contribution of individuals to their population (Barton & Etheridge, 2011).

Fertility transmission (FT) is a phenomenon with a cultural and/or genetic basis, whereby a positive correlation exists between the number of offspring of an individual and that of his/her parents. Theoretical studies using a haploid individual-based model have shown that FT increases the variance and intergenerational correlation in reproductive success and results in an imbalance in the coalescent tree of sampled genes. This phenomenon has been documented in several demographic studies conducted on the correlation in fertility between generations, or through the reconstruction of the genealogical trees of mitochondrial DNA sequences. However, as mtDNA is a single locus, potentially subject to other forces (e.g. natural selection), it is of interest to extend the theory of FT to nuclear loci. We show that because random mating between individuals leads to a mixing of their fertility profiles, FT in these cases will have less influence on the variance and intergenerational correlation of reproductive success. This, in turn, results in less impact on the shape of the coalescent trees. Nevertheless, in the presence of FT, high heterogeneity in reproductive success and homogamy for family size will increase the imbalance in the coalescent tree. Thus, FT should be easier to detect when occurring in conjunction with these other factors. We also show the utility of analysing different kinds of loci (X-linked, Y-linked, mitochondrial and autosomal) to assess whether FT is matrilineal, patrilineal or biparental. Finally, we demonstrate that the shape of the coalescent tree depends upon population size, in contrast to the classical Kingman's model.

Introduction Sub-Saharan African immigrants and refugees (SSAIRs), often referred to as Black Africans, come from about 40 ex-colonies of mainly Britain (e.g., Kenya, Nigeria and Uganda) and France (e.g., Mauritania, Senegal and Côte d'Ivoire) and form the majority of African newcomers [1]. SSAIRs have attracted a relatively low profile in Canada, in part, because they are usually lumped with either the broader immigrant population or the equally growing pool of Blacks, which, at 1,198,540 in 2016, made up 3.5% of Canada's population with 71% (aged 25 to 59) being foreign-born [2]. Interestingly, the rise in Canada's Black population is largely fueled by SSAIRs in newer destinations such as the Prairies. In Alberta, for instance, the number of Blacks has grown significantly from 39,955 in 1996 to 174,655 in 2016 [3]. The proportion of SSAIRs grew from just 1.9% of newcomers to Canada before 1971 to 13% in 2016 [3], with Africa placing second among Canada's immigrant-sending regions. This study, emerging from three years of research on SSAIRs in Alberta, Canada, contributes to addressing the gap in the literature on SSAIRs' extended family experiences. The purpose of this study was to explore the experiences and viewpoints related to the extended family relations and family life of African immigrant women living in Canada. Although they are socially diverse, SSAIRs share an affinity for community life rooted in religion and culture, both of which endow a deep sense of belonging and affirm collective responsibility [4]. Gender relations within the immediate and extended family form a bedrock on which the polity structures and executes collective responsibilities [5]. SSAIRs' history and immigration trajectories differ notably from those of other Canadian Blacks and barriers to rebuilding their lives raise serious concerns. Unlike previous Black immigrant cohorts, such as those from the Caribbean, SSAIRs do not exhibit the high rates of singleparent, female-headed households, or of economically marginalized men [6], stemming from longstanding systemic inequities rooted in slavery and colonization. Similarly to the broader population of more recent immigrants, however, SSAIRS not only maintain their ties to countries of origin but also find ways to nurture these ties as a crucial facet of support, in a new homeland where they have arrived with little or no culturally appropriate support system. The extant literature on the impact of transnational ties on newcomer transition and integration in Canada is relatively small but steadily growing [7]. The little we know suggests that a critical understanding of these ties is crucial to the development of public policy and practice and meaningful research agendas to better serve immigrants [8]. SSAIRs, one of the newest and fastest growing Canadian immigrant groups, remain highly under-researched. The concept of family connectedness is a prominent cultural feature in African Black families [7,8]. Immigration changes not only the family structure but also the relationships between household members. In the context of immigration, the importance of extended families is reflected in clustered patterns of migration settlement. Immigrant people often live in the same neighbourhoods or even the same apartment buildings in order to provide family support such as sharing meals and childcare responsibilities [9]. Married couples often share both paid work and housework in a more equal way than they would in their country of origin. African immigrants extend their lives back across the Atlantic, remaining in constant contact with their families on the continent. Most are very honest about longing for home and those they left behind. They usually keep in touch with their home countries in many ways [10,11]. For many immigrants, the telephone is the preferred channel of communication with home. Most immigrants call home frequently and many even reach their distant relatives back home. According to the participants, Immigrants also stay in contact with their families at home by sending "remittances", or sums of money that they wire abroad using several different services. Overall, family support through the exchange of remittances is essential to African immigrants, particularly among working-class families [12]. The role of gender is prominent in regard to the immigration experiences of African immigrants [13]. In general, among SSAIRs, gender roles are traditional and patriarchal, with men and women assigned to separate scopes of work in which males are breadwinners and females are homemakers. This idealized gender role pattern prevails despite high rates of women participating in the labor force. Studies have noted that women's employment post-migration tends to change gender role patterns, causing a relatively more unequal distribution of domestic tasks within immigrant families [13]. For African men, domestic activities are often seen as being unsuited with the notion of masculinity. The lack of support from a male partner within the household leads to extra and burdensome workloads for women and increased dependence on female relatives and female-based support networks. In this study, female study participants identified specific ways in which ties to their families "back home" undermined their wellbeing within the nuclear family in Canada but emphasized, for the most part, the linkages they had established as mostly being key enhancers in the process of rebuilding life in a new homeland. Gender relations, the experiences and viewpoints of participants show, not only shape the dynamics of family life in Canada but also interact with transnational relations in different ways for men and for women. --- Theoretical Framework: Postcolonial Feminism and Transnationalism We adopt both a historical and an intersectional approach in analyzing the experiences and viewpoints of our participants, drawing primarily on two feminist theoretical perspectives: postcolonial feminism and transnationalism. A postcolonial perspective rejects the homogenization of marginalized populations within and outside the West, a strategy that often subsumes their diverse histories and circumstances [14]. Thus, we recognize the diversity of histories, cultures, social contexts and circumstances of migration and settlement that define African immigrants. We focus, however, on the commonalities, which uniquely shape their personal goals, as well as their settlement and integration trajectories. These commonalities include (1) collectivist cultures that place community above the individual, valuing family, immediate and extended, as a crucial engine for enacting collective responsibility [15]. (2) Gendered and racialized by culture, colonization and capitalist expansion prior to their migration, albeit in different ways, women's status remains socially subordinate to men's in the transition into a postcolonial era [16]. (3) Due to tradition, women's latitude to pursue economic ventures is curtailed, given their primary responsibility to provide for their children. The postcolonial perspective also highlights women's cultural knowledge, agency, and resilience in engaging life's challenges as crucial actors within and outside the immediate and extended families [17]. Although men's social status as decision-makers and critical agents is similar to the case in most societies, the role men could play as allies with women in the process of establishing a home in a host society is crucial [18]. Transnational approaches to migration recognize two key trends that shape the lives of immigrants, particularly women. First, the flow of people and goods across national boundaries, propelled by forces of global capitalism and the exigencies of political upheaval, has compelled the reappraisal of life in Western host societies as geophysically stable and mediated only by forces within them [19]. Immigrants not only hold on to but seek ways to nurture ties to their homelands as they anchor themselves in new soil. Transnationality highlights aspects of an immigrant's life that extend beyond the host society's geophysical space and consider how they intersect with other social trends and categories [13]. Transnationals interact with cultural identities and practices in homelands, diasporic communities and their host societies. Transnational feminist approaches challenge a longstanding presumption that men and women are equally involved in making decisions about when and where to migrate, which has only been recently questioned [20]. Ties to countries of origin remain a crucial source of support, the sparse extant literature suggests, but for male and female spouses, relationships with extended family members could present unequal challenges [21]. These challenges could also be mediated by the highly gendered nature of international migration and settlement; decisions about the destination, timing and logistics of migration are highly gendered, often resting with male partners. Shifts in gender roles within the family can also easily upset gender relations, placing women in vulnerable situations [22]. However, these patterns are not uniform across cultures and contexts; thus, there is a need for further research that links structures of gender evolving from specific histories and changing social contexts to specific immigrant contexts. --- Methodology This study was conducted between 2017 and 2020. Our data collection approach relied on the participatory action research (PAR) framework. PAR engages researchers and participants as co-constructors of knowledge who work together to examine an issue to change it for the better [23]. PAR differs from most other approaches to public health research because it is based on reflection, data collection and action that aims to improve health and reduce health inequities through involving the people who, in turn, take actions to improve their own health [24]. The reflective process is directly linked to action, influenced by an understanding of history, culture and local context and embedded in social relationships. The process of PAR is empowering and leads to people having increased control over their lives [25]. A purposeful sampling strategy was used to recruit participants through self-administered online surveys developed by a research team, and assistance was provided by community workers and an established network of community organizations. The rationale for the use of purposive sampling is to select a group of individuals that are especially knowledgeable about or experienced with a phenomenon of interest. Particular efforts were made to include women and men; people from different religions, such as Christians and Muslims; and immigrants from various parts of Sub-Saharan Africa. We conducted individual interviews with 20 women and 20 men who identified as Sub-Saharan African immigrants living in Edmonton or Calgary at the time of the study. This paper only focuses on the experiences of women participants. In an interview, participants were asked to share their experiences about extended family relationships and the challenges they faced as an African immigrant. The three main questions asked during the interview were (1) Tell me briefly about your experience with extended family own your own side (Probes: view/nature of extended family, responsibilities, influence on nuclear family as support/stressor). (2) Tell me briefly about your experience with your spouse's extended family (Probes: view/nature of extended family, responsibilities, influence on nuclear family as support/stressor). (3) Based on how you have fared with extended family, comment on how African families should manage their extended families (Probes: own/spouse's family, comments for families in general). The inclusion criteria included participants who were Canadian citizens or permanent residents, spoke English and were at least 25 years of age. We excluded people who were international students, temporary foreign workers or refugee claimants due to unique policies that influence their circumstances. We conducted face-to-face or phone semi-structured interviews as per participants' preferences. Informed consent was obtained from the participants and all the interviews were audio-recorded and transcribed. Confidentiality and privacy were maintained by using encrypted computer-based files, and documents (i.e., signed consent forms) were stored in a locked file cabinet and all personal identifiers were removed from study documents. Interview transcripts were thematically analyzed using NVivo software [26]. Rigor and trustworthiness were maintained through ensuring credibility, transferability, dependability and conformability in the research process [27]. Ethics approval for this study was obtained by the University of Alberta Research Ethics Board. --- Results The following themes emerged from the manual coding and analysis of the study transcripts: (1) Extended family as a source of crucial support, a sense of belonging and collective responsibility; (2) managing extended family as a challenge in a new homeland; and (3) the burden as well as the privilege of extended family according to men. In addressing this challenge, participants highlighted (4) improved gender relations through open communication and the need to set boundaries. --- Extended Family as a Source of Crucial Support, A Sense of Belonging and Collective Responsibility Participants spoke about the importance of support systems and networks, particularly related to maintaining contact with family members and establishing communities with regards to parenting. The most important issue for immigrant Africans was balancing nuclear family obligations with extended family obligations and navigating the tensions that this can create within immigrant families. In this regard as well, participants noted the importance of clear communication between partners to ease potential tensions from the perceived sense of obligation in the African culture to maintaining extended family support networks and providing financial support to those networks. One noted the team effort between her and her spouse: "Yeah my husband's side like for me, for us both families, like we don't divide, me and my husband we work together, like he looks at my family as his own, and even his family there are very nice to me and they are my family; but unfortunately for my husband he lost all his siblings, he's the only one surviving, so the siblings left their children that's-and he's the firstborn of the family... so there is an obligation, yes". (CAF013) Participants also spoke on the sense of obligation in general to extended family networks and support systems: "The extended family members always want that you should take care of-Like I said, at home I have so many people in my home. Some of them were extended families... What they expect is they expect you to help them in either learn a trade or put them in school. So I've been experiencing that. It's still my passion of helping out, so at times I help out in my family and my extended family. That burden is always there. So you help out with your family and help out with the extended family all the time... I think the problem is we just grew up with this thing of we enable each other. We call it Ubuntu in Southern Africa... " (CAF009) "I think most of the Africans though I'm not the spokesperson for all African but like the way I see from even our people because we have a sense of family, right, we have a sense of family and we like tending to each other and we like holding each other up, we like pulling each other up... so you have to do what you need to do to help your people back there, you know, you have to share the little you have to share with them from time to time; so it is not necessarily sometimes every month, right, so you, of course, interested in maybe I need to help my people properly so you find that you may be a little bit, right, because of finances because you have your own responsibilities here also to take care of". (CAF012) "... we as Africans as well, we have to look after everybody back home, you know. We are the breadwinners, we are overseas, everybody's expecting too much from you. The funeral, everything is you. So I think yeah, most of the tension comes from that. Which makes us who we are. We are African, like we look after each other so, you know. In a way it's bad but in anyways that's who we are". (EDF011) Study participants also expressed the importance of clearly communicating to family members the shift in gender roles (if/when it occurred), in which the woman had become the primary earner, contradicting the African notion of the man as the main provider for families. This was noted as a way to manage expectations from extended family members and highlighted some of the communication challenges that may exist within immigrant families and their support networks. A participant explained: "Until Africans begin to know that they must have sincere transparent conversation around their finances, they will always have a problem. Because for example, if my husband brought me to Canada and his brothers heard he's the rich one, but I could come to Canada and I'll have a better job. It is my husband's place to let his family know, oh I brought my wife to Canada, but she's been lucky, she's the one who has a job, do you understand, she's the one. But how many Nigerian families are going to tell their families in Nigeria that their wife is feeding them, it's a matter of communication, transparent communication, open communication, put it in the table and talk about it, and then there will be no problem". (EDF006) Participants also noted that although African immigrant families in Canada tended to provide financial support to their family networks, they also benefited from the emotional and community networks maintained through interactions with their support systems in their countries of origin. Participants noted praying together, sharing experiences and seeking guidance/discussing issues. A participant put it this way: "in terms of relationship, we are still there, you call your sisters, you speak to them, you pray together and now it doesn't change much, it's just that you try to share the good experiences and the things you have learned here with them. And somehow, for example, like my own family, they listen and they understand... " (EDF006). Another participant, discussing the benefits, stated: "Oh my gosh. We [extended family] have phenomenal, phenomenal communication. We [extended family] have WhatsApp groups and just my sisters, my mom and I and then other extended families, my aunts, because as I said, Zimbabwe is different from Canada. All my aunts are my little mothers. So even when I'm having a problem they're the first person I talk to. So that's-my cousins are my sisters and my brothers, so we have that relationship. A really good relationship". (CAF013) The importance of community was also emphasized in regard to parenting and creating similar communal support systems, enjoyed in African countries for immigrant families in Canada as a way to ease and address the challenges of raising children in a different culture. Regarding parenting challenges and conflicts with children, a Sudanese participant explained the importance of a support system and community in avoiding such challenges: "[O]ne of the issue that I see, that caused most of these things to be happening not right isn't for the African child, it's from the bigger people, it's the parents. We don't like ourselves, we don't come together... Some of them because of the barrier, the language barrier, and then until-you cannot stop children from growing, they keep growing and the parents don't have the language, you know, because like our people so many of them came because of the war and they became maybe directly from villages, they just ran... and they come here... they can't be able to help their own child even with the homework, right, so the community should be able to, you know, they should be able to have groups where they have women that we can discuss not only women, even with men, how or what they want to see in the lives of their children because children they grow". (CAF012) Aside from challenges based on expectations, participants expressed how Canadian ideals have influenced this relationship as well and allowed them to encourage family members to become self-sufficient and communicate the same message to each spouse's network to avoid tensions within the family. "But in terms of relationship, we are still there, you call your sisters, you speak to them, you pray together and now it doesn't change much, it's just that you try to share the good experiences and the things you have learned here with them. And somehow, for example, like my own family, they listen and they understand and sometimes they, you know, they give me credits, they applaud what I have learned here and then they try to use it and to, you know, they try to use it there, back in Nigeria. Try to apply it to their lives as well". (EDF006) Some other participants noted similar experiences of encouraging family members to be self-sufficient and supporting them in that endeavour: "... one of the things that I did was that I send some money for my sister to get some training, so she's doing, like, a nurse aid role and she's hoping that she'll have a job now. That would mean that I can cut down on sending her money and start putting money aside for me for my retirement. Because my husband is already retired... " (CAF013) "... whatever we send from here, I think it's supposed to be enough for them to start their life, so that they will not be dependent. They are supposed to do businesses, do something that will generate income for them. So, not to rely totally on those who are abroad". (EDF004) --- Managing Extended Family as a Challenge in a New Homeland In addition, participants touched on the challenges that come with interactions with extended family in countries of origin, particularly relating to the notion of improved economic status based on relocation to Canada and how this creates unrealistic expectations and pressure for support: "So in terms of my relationship with my own extended family, I feel like we are pretty close, although we are in different countries. I'm obviously very far away. Most of them are in South Africa and Zimbabwe. The challenge that I would say I have had since moving here is most of them just think that now I'm in Canada I have a lot of money... So I've been getting a lot of messages, voice messages, phone calls, about people who are just asking for money for some reason or another. That's the only challenge, that people now think I have a lot of money". (CAF007) In addition, another also stated: "The drawback is the constant demands, they cannot-they don't get it really and I don't blame them because back home they-we support each other. They were there when I was in school, I went to my sisters and they will always give you pocket-I expected that from them. If they didn't give me I didn't feel happy. I felt it was their duty to help me while I was in school, and they did that very well. Some of them didn't eat, one of my sisters, her husband was even-she will ask me, "Go to this store, go and take all the provision you need for going back to school, I will give you money"; and she will do that from the little pocket money. She has some money from the food money her husband had given her and she could pay for that for two months. I looked after the time I could help her but unfortunately, I'm here and my finances can't do even for my family not to talk of helping-. Right now I don't even talk with her.; I don't feel happy that I can't help her". (EDF015) Most participants agreed that while maintaining a connection with families in their countries of origin had the tendency to create undue tensions, managing communications and setting boundaries helped to manage any potential conflicts that may have arisen between partners in this regard. They also noted prioritizing their family/spousal needs before attending to maintaining ties with their extended family in the country of origin and supporting them. A participant commented that, "I think they [partners] should not feel responsible for every need and I think they first take care of themselves before they invest in taking care of other people because sometimes you can take care of other people and you realize that you neglected certain aspects of your life. You did not invest in yourself and sometimes it can affect even your health or your relationships so I think it's good to first take care of themselves and do it in agreement together with the-like if they are married, the husband and wife, to be in agreement to do that then it will not cause you tension". (CAF014) Participants discussed the various circumstances that contributed to the need to set boundaries so as to ease financial burdens within the home which had the potential to strain gender relations, stemming from providing support to extended family and community networks. Generally speaking, there was a perception of constant demand and dependence: "What I find is that when we come to the diaspora we tend to over-help our families back home. And I find that this is not good. Because if it's a case of giving them the fish instead of giving them the fish or teaching them to fish. And I'm only learning that now. You know, in the last year when I've been dealing with my sister and trying to get her to start becoming more independent and looking after herself and her family back home in Zimbabwe. But I think that's what we really need to do is start becoming a lot more firmer with the people back home. Because they think that money grows on trees when-when they hear from us it's all about money, or if they phone us it's all about money... " (CAF013) "So really a lot of things we have had to do to support the extended family, helping my mom-I only have my mom-and then my siblings, paying school fees for their kids and everything. Sometimes from, you know, even providing money for food or things like that. So yeah, it has been cumbersome". (CAF014) Participants also elaborated how the overdependence of extended family networks may sometimes hinder their development and growth, which further reinforced the need to set boundaries while maintaining those network systems. One participant noted that some of the support requests often seemed unreasonable based on the shared sense of obligation Africans have to support their community: "I think what we do is a lot of Africans or a lot of African families, they expect too much from one another... Sometimes some of the requests are genuine like maybe someone is sick and stuff. But most of the requests, to be honest, are just quite unreasonable. And they hinder us Africans from reaching our own personal goals, then you end up also having resentment towards some of your family members". (CAF009) Beyond establishing boundaries, participants noted the need to enforce the boundaries as a united front and communicate this clearly to extended family back in the country of origin, as well as to help manage expectations and relationships and ease the burden. One participant explained: "Like everything is different but setting limits sometimes kind of helps. Sometimes you need set limits to say "You know what, maybe this year this is what I can give, what I can afford to give" but everybody has to look after themselves. "You can't all depend on me all the time because I have my own things to do as well." But setting limits I think, you have to set limits, you have to explain to them because most of them they think we are in Canada "Oh my goodness, money rains everywhere" but you just have to set limits. You say "This is this" or "This is who I'm helping." That's my thing, setting limits". (EDF011) Participants also noted how boundaries were important in maintaining the support system and relating with extended family while managing family finances and gender relations: "So in that regard, I always tell people you can't do without your extended family back home, but you have to set the boundary. You need to let them understand that you are not making a million dollars here; if they are placing an unnecessary financial burden on you, you need to set the boundary and let them know, "I will do what I can within my means, but I can't steal", I'm sure they don't want you to end up in jail. But if you don't set those boundaries and be truthful about it, then people's expectations of you are higher than what you can meet, and then they feel you are not helping them; and this is something you have to talk with your husband, you have to agree first as a family and you are both passing the same message to both families. So it's not like you are doing so much for your husband's family when your own family needs so much and you are neglecting them". (EDF006) --- The Burden as Well as the Privileged of Extended Family for Men Participants also spoke about the pressures their partners experienced to be providers for both nuclear and extended families. The general perception was that although men enjoyed some privileges based on the African culture as the head of the home, this position also carried considerable burdens. Given the changing dynamics of immigrant families and women taking more roles in sharing family responsibilities as providers, the perceived burden on men and their privilege was diminished to an extent. One participant commented on the burden and how her involvement upon migration to Canada eased some of the tensions: And like I said, my husband carries the, most of the economic burden, financial burden and I just support with a few bills here and there, you know, but at least we now, we both know what we earn and then how to save when children's school, when it's time to send the children to school, we know what's going to happen and we kind of know what we have. "Yeah. When I got here, I was just really to work, like I worked with DynaLIFE, in their laboratory in the specimen department. I know that I, I used to work through the agencies, I worked in the bank, but they were very little money like I wasn't even contributing, my husband was running the whole finances in the house, because he was a professor in the university, so. But I just used to support him very, in a very little way, unspoken support, like you know, he gives you money and then you add your own or whatever. So, that was how it was until I got my job until I went back to school; I took out a student loan anyway, so I went back to school. And then when I got here, I now took over some of the expenses, like some of the bills I had to pay, but generally, my husband pays most of the, carries most of the financial burden, so, in the house". (EDF006) Similarly, other participants noted the perceived financial burden and pressures of responsibility African men felt to manage their nuclear and extended families; "I think as a man, oh yeah absolutely. I think men, yes, because as a man your family they're looking at you, you know, for support especially where families where the dad has passed on and you are the eldest son or you are the one who's-we had the opportunity to leave, you know, Africa and go somewhere. So everybody's like looking to you as a man. But yet you the woman too who's saying, you know, "You have to look after me" and she has her own people to look after. But I find it more with men than women. Extended families, they expect more". (EDF011) "[S]o it is my husband only who is working to support the family. So sometimes it is tough, you don't have money, even if they request. He said, "that's our relationship". Sometimes they are not happy because for them, we are here, they ask us for money, we don't have money, they spend like two/three months without calling him because they didn't get the money they asked for. That is okay. We-we try to live with that. And when we have money, we just send it to them". (EDF003) "... With my husband's side, now I feel like they also want us to take responsibility of taking care of his brothers, meaning that they have to come here to Canada. Not sending money, this one now different. They want to send like his brothers also to come here so that we can stay with them. Which I think is not a bad thing, but we might just not be ready financially". (CAF007) --- Improved Gender Relations through Communication The study revealed that improved communications between spouses have been a large contributor to improved gender dynamics within the family. Participants spoke about the difference between the culture of communication between Africans and the level of transparency and conversation that was required while settling in a foreign country. It was noted severally that based on the African cultural dynamics and gender roles, men often took the lead, with less transparency and communication with their spouses. A participant explained: "I think that's the basic problem is communication; that is what African's lack, that's where the Caucasians are okay because they talk, Africans don't talk" (EDF006). The participant elaborated further on how the need to be transparent about finances in Canada changed this dynamic and allowed for more open conversation between spouses: "... if there is a good communication network between a wife and a husband everything should be put on the table because it is a nuclear family. Like my husband's family is my family, my family is his family, and of course, if we know the needs, if we are able to be transparent with the needs of all our families. And then we come together with our income, with what we can afford, what we can give and we talk about it, we know how much we have, we know what my family needs, he knows what his family needs and then we have an agreement; there won't be a problem. But because of lack of trust and because of how African families have always been, the man has always been the provider". (EDF006) In addition, the study revealed the importance of general open communication with regards to personal, family or career tensions that may affect spousal relationships in order to maintain good gender relations within the home. For instance, conversations covered the importance of openness, such as sometimes not being aware of a spouses' work tensions and challenges which may be affecting their behaviour in the home, which may cause friction unless it is discussed. One participant elaborates: "but the thing is openness-be open about it. If something is bothering you, talk about it. And I usually tell my husband like if you have a challenge at work or I have a challenge at work, telling you might not help but letting you know it's good, because sometimes I might be talking to you full of anger, not because you've done anything to me but because of what is going on at my workplace. But if you know, you might be able to understand me or deal with the situation much better. But if you don't know, then you'll blame me for everything, and you'll also react in the same way. So I don't know-we are still working around this area, but it is a very sensitive area, and openness and discussion is really key. Yeah". (EDF 005) Participants also noted that the communication approach was important in maintaining and improving upon communications. They noted the importance of being sensitive to spouses' trigger points and being respectful in conversation approaches so as not to contribute to the stress and pressures of carrying family responsibilities and the roles each partner played in the home. "know what your partner likes, even in talking about finances you have to respect your partner's likes and dislikes, you understand. Like you can have your wife, who really likes to dress up, she likes to spend money, yes, she pays, respects that. And the wife, you who likes to spend money on clothes also know how much your family finances are, you understand, come to an agreement, let there be a balance. As long as we don't do that, give problem, we have, trust me it's communication, financial communication, just communicate. It doesn't mean that I'm going to take all my money and give to you or you're going to take all your money and we are all together in this thing, it's a marriage, right... " (EDF006) --- The Need to Set Boundaries Beyond establishing boundaries in relation to providing financial support, the need to also set boundaries in regard to the influence and interference of the extended family in the nuclear family dynamics was frequently mentioned by participants. Given that the African culture is largely communal, there was a tendency for interference with issues amongst spouses, which more often than not reinforced gender stereotypes. Participants noted that setting boundaries within the context of their relationship and the involvement of extended family members, as well as communicating this clearly to all parties, also improved gender relations. A participant described how establishing boundaries worked in her home: "I mean if I have issues with my husband, my mother is not the first person I am calling; and this is because for me I do believe strongly in God... and we had an agreement that if we ever have issues, we're going

African immigrants are increasingly migrating to high-income countries, including Canada, in search of a better life. These immigrants often face several challenges, such as keeping healthy ties with their extended families back home and in a new socio-cultural context. We present findings from a participatory action research (PAR) study of Sub-Saharan African immigrants and refugees (SSAIRs) living in Alberta, Canada. Using the theoretical framework of postcolonial feminism and transnationalism, in this study we investigated how cultural roots and transnational ties affect newcomer transition and integration to provide information on the female perspectives of SSAIRs. The results of the study indicate that maintaining relational ties with one's extended family in the homeland has been highlighted as both a source of support-providing moral, social, religious, and cultural support during the integration processes-and strain, with participants noting its impact on their integration processes, such as delaying the ability to restructure life and to successfully plan their life financially. Our findings outline various implications of the existing gaps and recommendations for policymakers and community stakeholders for future improvement. Overall, our study findings affirm the importance of extended family relations for African immigrants living in Canada.

it's communication, financial communication, just communicate. It doesn't mean that I'm going to take all my money and give to you or you're going to take all your money and we are all together in this thing, it's a marriage, right... " (EDF006) --- The Need to Set Boundaries Beyond establishing boundaries in relation to providing financial support, the need to also set boundaries in regard to the influence and interference of the extended family in the nuclear family dynamics was frequently mentioned by participants. Given that the African culture is largely communal, there was a tendency for interference with issues amongst spouses, which more often than not reinforced gender stereotypes. Participants noted that setting boundaries within the context of their relationship and the involvement of extended family members, as well as communicating this clearly to all parties, also improved gender relations. A participant described how establishing boundaries worked in her home: "I mean if I have issues with my husband, my mother is not the first person I am calling; and this is because for me I do believe strongly in God... and we had an agreement that if we ever have issues, we're going to talk it through between ourselves first, so before-we rarely bring third parties into our issues, even our extended family, so the boundaries are there. And from my own part and his own part, I think we-both families understand that, that those boundaries are there. We might be quarreling and somebody from our families comes in, you can't tell we're quarreling because it's not for you to know, we'll deal with that. So that way there's not been much difference for me when we got to Canada, because from Nigeria, we've not really allowed too much influence from extended family". (EDF013) Another participant narrated how family influence had resulted in tensions and the break-up of her friend's marriage, while noting that she had not had any experience as family relations often differed, but also highlighting the importance of boundaries: "The experiences are different, different from each family. Most families do not even like that the extended families come. Because my friends have experienced very bad experiences in their own marriages and they're broken up. So the thing is only that if I had a broken home I will have a good example of that, but now that I don't have a broken home, that's why I don't have a good example of those extended family influences". (CAF009) This shows that while family networks are crucial in the diaspora, it is important that they are managed properly and that communications with all parties are transparent regarding their dynamics, capacity and needs. --- Discussion In this study, we explored how cultural roots and transnational ties affected/influenced newcomer transitions and integration in Canada and the results highlighted four major intersecting points of influence and confluence, focusing primarily on the female perspectives of SSAIRs. From this perspective, it appears that transnational ties and post-colonial constructs play critical roles in relationships with family and the communal approach to support, as well as in gender relations between SSAIR couples in managing the burden of care and privileges associated with men and communication [16]. Based on the transnational perspective, the importance of maintaining relational ties with the extended family in the homeland has been highlighted as both a source of support-providing moral, social, religious and cultural support during the integration processes-and strain, with participants noting its impact on their integration processes, such as delaying the ability to restructure life, financially plan and save [28]. Although the degree of impact and specificity differs based on slight cultural differences across the SSA region, the commonalities and the range of impact on their lives from managing external family relations were the same for all participants. Our findings are in alignment with these earlier works. Overwhelmingly, SSAIR females have reported relatively positive changes in the relationships between spouses and with extended family relations upon migration. Notable among these were increased financial independence, improved communications stemming from the need to jointly provide for and manage the family's affairs as opposed to the culturally imposed presumption of that being a privilege and burden for men, as well as the ability to better manage family interference and interaction due to distance and setting boundaries [29,30]. Although the conclusions are relatively positive, two main challenges were highlighted repeatedly: (1) the strain of shifting gender roles, and (2) the financial burden of maintaining transnational ties with extended family. These two factors contributed to the slowness of the integration process for SSAIRs who were unable to rebuild their lives at a much faster pace. The study shows that the sense of responsibility towards the extended family that SSAIRs felt was deeply rooted. While many highlighted the importance of setting boundaries, they still considered providing financial support to extended family in their home country an inescapable duty [31]. Additionally, although the study showed shifting gender roles within the family towards more balanced roles, it also revealed a reluctance to embrace this shift. Many participants went on to note that, upon immigration, they have had to play a more prominent role in providing for their families, yet they also spoke about how most of the responsibilities were still the burden of the men [32]. This further illustrates the difficulty in moving past postcolonial ideals of gender relations to transition to more equal gender roles and responsibilities, as well as integration into Canadian society. The socio-economic pressures experienced by extended families allow immigrant Africans to rethink the distribution of gender roles in the family [11]. This redistribution of roles often requires improved and open communication, resulting in women taking on more financial roles, which thereby provides a sense of responsibility and participation in the decisions regarding the family [33]. The result of this is improved communication between spouses, which contributes to shifting gender roles, as most women, based on their economic contributions and more prominent role in the family finances, have more input in family management, unlike what was the case when in Africa. This is a reinforcing cycle that creates a more balanced gender relation in the family. Furthermore, the increased participation of women in the labour market and as income earners also contributes to the shifting dynamic. Traditionally in most African countries, the women are the homemakers and often run businesses (shops) as their economic activity, which often mean that income may or may not be a steady stream for females. However, as Canadian immigrants, most African women embark on salaried employment, which gives them a steady stream of income that allows them to contribute to and participate in family finances and decisions [34]. Thus, clear communication between spouses helps in this regard and reduces tension that may arise from shifting gender roles. Nonetheless, it appears that although many African immigrant women in Canada contribute more to the family upkeep financially, most final decisions and the heavier burden of finances are still made/shouldered by the men. The majority of programmes and policies provided to ease integration into Canadian society often focus on economic integration-jobs and financial literacy regarding credits, bills and banking, etc. [35]. Given that a significant challenge to the economic integration of SSAIR women in particular stems from cultural and social ideals from their home countries, perhaps policies and programmes also need to address these issues by upgrading their policies and programs to ease cultural integration and improve gender relations that may arise due to assimilation in Canadian society. Often, when shifting gender roles are not properly managed, this creates further strain on the overall integration processes of SSAIRs, which might create dependence on government support [18]. There need to be enough programmes to prepare immigrants for what to expect culturally, socially and beyond. Most of our female participants acknowledged the importance of having strong relations with their extended families and feeling responsible for financially supporting them. The findings of our study lead us to recommend that African immigrant families should create open and trusting relationships between partners to avoid conflict involving extended family relations [36]. Moreover, being transparent with family members regarding finances and the challenges of employment in Canada can avoid creating conflict among extended families. Furthermore, there is a need to set up boundaries with extended family members and help them financially within sustainable means and based on the availability of resources to make families self-sufficient, rather than making them dependent [37]. Although this study provided initial information about extended family relations among SSAIRs from the female perspective, future research should focus in greater depth on several of the findings noted in this study. In particular, immigration history (foreign vs. Canadian born, years in Canada) and socioeconomic status (income and education differences) are especially central to issues of immigration and family support and these warrant additional study. Furthermore, the findings of our study underscore the need to strengthen informal support systems for women, such as elders and community leaders. Further investigations can provide insights into the impact of extended family support on family relations and functioning. --- Conclusions Globalization has fostered new forms of migration as Africans seek better economic opportunities in different parts of the world, predominately in North America. In this study, we examined African immigrants and refugee families' experiences of extended family support systems from back home through female participants' perspectives. Our findings lead us to propose the need for interventions focused on strengthening extended family relations in this population. Potential strategies of such interventions include addressing open and trusting relationships among partners, attending to the role of gender, and equipping African immigrant females with skills for dealing with families back home, for example, by setting boundaries, all of which have the potential to improve family harmony and help to avoid conflicts. Support from socially and culturally appropriate programs and support from community partners that attend to these challenges will likely improve the extended family relations of African immigrants living in Canada. --- Data Availability Statement: Not applicable. --- Conflicts of Interest: The authors declare no conflict of interest.

African immigrants are increasingly migrating to high-income countries, including Canada, in search of a better life. These immigrants often face several challenges, such as keeping healthy ties with their extended families back home and in a new socio-cultural context. We present findings from a participatory action research (PAR) study of Sub-Saharan African immigrants and refugees (SSAIRs) living in Alberta, Canada. Using the theoretical framework of postcolonial feminism and transnationalism, in this study we investigated how cultural roots and transnational ties affect newcomer transition and integration to provide information on the female perspectives of SSAIRs. The results of the study indicate that maintaining relational ties with one's extended family in the homeland has been highlighted as both a source of support-providing moral, social, religious, and cultural support during the integration processes-and strain, with participants noting its impact on their integration processes, such as delaying the ability to restructure life and to successfully plan their life financially. Our findings outline various implications of the existing gaps and recommendations for policymakers and community stakeholders for future improvement. Overall, our study findings affirm the importance of extended family relations for African immigrants living in Canada.

Introduction Despite the antecedent of the bubonic plague in the Middle Ages or the Spanish Influenza in the early years of the 20th century, for the first time in history, humankind was able to witness the worldwide spread of a new disease in real-time. The coronavirus disease 2019 (COVID-19) broke out in Wuhan, China, in December 2019. The virus showed a high infection rate, fast transmission, and considerable mortality. As no vaccine or pharmacological treatment was available, authorities throughout the world considered it a serious threat to the population. After the disease reached Asia it spread to Europe and the Americas. Governments declared a sanitary emergency in almost all countries with a restriction on mobility and social isolation as common strategies to avoid or delay the spread of the disease. In Mexico, the conditions needed to declare the sanitary emergency were reached in the last week of March 2020. By this time, most of the population was familiar with the high levels of people infected or deceased in other countries, so the sanitary indications were neatly followed. Mobility decreased significantly, while the social isolation conditions 2 of 11 reflected an extreme concern of the population regarding health and safety. Thus, within a short time, a large number of families began to experience stressful conditions in their daily lives. The lack of social interactions, restrictions on human mobility, imposed fulltime interactions with family members, and the new conditions of working from home or homeschooling may have influenced the modification of circadian rhythms (sleep-wake cycles). In addition, the alarming daily news and the uncertainty of the new economic conditions induced fear and agitation in big sectors of the population. Brought together, all of these conditions could have induced sleep disturbances and the onset of mental health impairments such as anxiety or depression. Previous reports indicated that the 2003 pandemic of severe acute respiratory syndrome induced severe mental health disturbances, mainly among Asian populations [1]. As most of the current health efforts have been oriented towards medical attention for those infected and health workers, the psychosocial effects on the general population have been overlooked [2]. In a recent review of the psychological impacts of quarantine, some authors reported negative psychological effects [3]. In addition, recent reports from China indicated the presence of sleep disturbances, generalized anxiety disorders, and symptoms of depression in the population, all associated with quarantine conditions [4]. It is well known that isolation facilitates the onset of psychopathological manifestations, including anxiety, depression, sleep disturbances, hallucinations, and suicidal behavior [5]. Recently, the trigger of psychopathological manifestations was identified and includes boredom, stress, and sleep restriction [6]. Furthermore, it is well known that sleep and mental health disturbances present reciprocal causation [7]. Sleep disturbances often generate a negative vicious circle, worsening the quality of life of patients with mental health impairments. In addition, stressful situations such as social isolation often generate sleep disturbances that negatively influence mental health balance, mainly worsening the state of anxiety or depression [8][9][10]. In the last decades, a great volume of scientific papers has reported the negative effects of sleep restriction (for a review see [11]). These effects are observed both in the short and long term [12,13]. The short-term effects of sleep restriction are characterized mainly by an impairment of executive functions, including verbal and non-verbal working memory, attention deficits, and emotional and motivational regulation, as well as difficulties in problem-solving, among others (for a review see [14]). The negative effects of sleep restrictions can be observed in a broad range of ages and are particularly critical in children's day-time functioning [15]. Furthermore, mood alterations due to sleep restriction also have been repeatedly reported [16]. These mood changes are particularly in stressful working situations, such as in hospitals. The effects of sleep restriction on the increase of conflicts within groups have been repeatedly reported [17]. Medical residents display increasing impairments in their ability to show empathy towards patients and towards other members of the group [18]. Thus, stressful conditions within the family group due to the sudden lockdown have an impact on the circadian rhythms and, as consequence, on sleep, resulting in a significant level of sleep restriction. As shown in preceding paragraphs, quarantine and isolation conditions to safeguard human beings from COVID-19 could produce stressful circumstances that cause negative mental health effects, such as depression, anxiety, and sleep disturbances [3][4][5]. An additional factor that could be presented during quarantine is the increase of exposure to light stimulation at night by using bright screens, such as cell phones, TVs, or tablets, which may suppress the release of melatonin, resulting in major disturbances of circadian rhythms and sleep disorders [19]. Knowing the effects of quarantine on sleep and mental health of the world's population, it is essential to design psychological and social interventions during and after the current pandemic. In the case of Mexico, results that could guide such interventions and contribute to national and international scientific knowledge have not yet been reported. Thus, the purpose of the present study was to explore the presence of sleep and/or mental health disturbances, as well as, the use of bright screens at night in the Mexican population while living in quarantine conditions to prevent COVID-19 infections. We hypothesized that adult Mexican people living in quarantine to prevent COVID-19 infection manifest sleep disturbances and symptoms of mental disorders such as anxiety and depression. --- Materials and Methods --- Participant Recruitment An open invitation was sent via Internet to participate in the present survey. Google forms focused on questionnaires about sleep habits, sleep quality, and mood were sent. We used our social networks and those of the researchers and specialists affiliated with our group to invite people who voluntarily wanted to fill out the forms. Subjects included in the study and those who filled out the questionnaires were older than 18 years. Incomplete or incoherent responses were discarded. In return, participants received the results of their questionnaires accompanied by particular recommendations for improving their sleep quality, as well as suggestions for sleep hygiene and relaxing anti-stress indications to improve their mood and decrease their anxiety levels. The protocol followed the principles outlined in the Declaration of Helsinki and it was reviewed and approved by the Institutional Review Board. All the participants provided their written informed consent to participate in this study and their data were protected and kept in strict privacy. --- Electronic Instruments Questionnaires were electronically available from 28 March to 26 May 2020. Although social isolation officially started on 24 March, several families, particularly in the big cities of Mexico, decided to decrease their mobility two or three weeks before. Thus, the period we defined allowed us to obtain data regarding the initial week of isolation. Participants filled out demographic information (age, sex, profession, and place of residence), a sleep diary for the final week, the Pittsburgh Sleep Quality Index (PSQI), the Generalized Anxiety Disorder 7-item Scale (GAD), and the Patient Health Questionnaire (PHQ-9). Filling out these questionnaires took about 10 min. The PSQI includes 19 items that assess sleep schedules, sleep latency, frequency, and the duration of awakening from sleep and nocturnal signs that suggest sleep disorders. The version we used was the one validated and already used among Mexican samples, which considers a set point >5 to indicate low sleep quality [20]. The GAD scale explores the presence of symptoms representing the generalized anxiety diagnosis according to the Diagnostic and Statistical Manual of Mental Disorders, 4th Ed. (DSM-IV). It includes 7 items concerning anxiety data in the two weeks before when the participant answers. We used the Spanish version [21] previously validated among Mexican samples [22] to identify the presence of mild, moderate, or severe anxiety. The PHQ-9 questionnaire assesses the presence of depressive symptoms during the 2 weeks before the test, according to the DSM-IV criteria. With 9 specific items, the Mexican validated version [23] indicates no signs, depression that could require treatment, and depression that needs pharmacological or psychological therapy. As bright light exposure can alter circadian rhythms and induce sleep restriction, the daily hours of exposure to electronic screens or devices were analyzed. To this aim, a special questionnaire was designed to explore how many hours the subjects were exposed daily to a bright screen as well as the moment in which this exposure took place. Additionally, in the same questionnaire we explored the activities the subjects performed before going to bed. --- Data Analyses Databases were generated, and statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) 21st version. The chi-square test was used to analyze proportions, and one-way ANOVA with the Duncan test, Binomial exact test, and t-test for independents variables were used for group comparisons. A p-value of 0.05 was established as a maximum to consider significant differences. --- Results A total of 1230 questionnaires were received from people living in 31 of the 32 federal regions in Mexico; most of them (62.4%) came from the central region, most of the participants were women (864, 70.2%), and most of them were around 40 years old or younger (713, 58%). Concerning sleep quality to explore sleep disturbances, Figure 1 summarizes the results about the PSQI items. A high percentage of the sample reported a sleep latency of more than 15 min (A), sleep duration for most of the participants was less than 7 h (B), subjective sleep quality reported by the participants showed no major differences between good/very good compared with bad/very bad (C), sleep efficiency was mainly reported as less than 85% (D), a high percentage of presence of any signs of sleep disorders and a very low percentage of the absence of sleep disorders was observed (E), and more than 80% of the participants reported daytime dysfunctions (F). --- Data Analyses Databases were generated, and statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) 21st version. The chi-square test was used to analyze proportions, and one-way ANOVA with the Duncan test, Binomial exact test, and ttest for independents variables were used for group comparisons. A p-value of 0.05 was established as a maximum to consider significant differences. --- Results A total of 1230 questionnaires were received from people living in 31 of the 32 federal regions in Mexico; most of them (62.4%) came from the central region, most of the participants were women (864, 70.2%), and most of them were around 40 years old or younger (713, 58%). Concerning sleep quality to explore sleep disturbances, Figure 1 summarizes the results about the PSQI items. A high percentage of the sample reported a sleep latency of more than 15 min (A), sleep duration for most of the participants was less than 7 h (B), subjective sleep quality reported by the participants showed no major differences between good/very good compared with bad/very bad (C), sleep efficiency was mainly reported as less than 85% (D), a high percentage of presence of any signs of sleep disorders and a very low percentage of the absence of sleep disorders was observed (E), and more than 80% of the participants reported daytime dysfunctions (F). Table 1 summarizes the statistical analysis by using the chi-square test for the items included in the PSQI. Sleep latency was different between sexes, with women showing longer sleep latencies than men. Sleep duration also differed between sexes; a higher percentage of women slept less than 5 h. Regarding sleep quality, a higher proportion of women (55%) reported bad and very bad sleep quality, whereas men under the same parameter showed a lower proportion (37.9%). Sleep efficiency showed a similar trend and differences were represented in extreme values. The percentage of women with a sleep efficiency of 85 or more was lower compared to men. On the contrary, the percentage of Table 1 summarizes the statistical analysis by using the chi-square test for the items included in the PSQI. Sleep latency was different between sexes, with women showing longer sleep latencies than men. Sleep duration also differed between sexes; a higher percentage of women slept less than 5 h. Regarding sleep quality, a higher proportion of women (55%) reported bad and very bad sleep quality, whereas men under the same parameter showed a lower proportion (37.9%). Sleep efficiency showed a similar trend and differences were represented in extreme values. The percentage of women with a sleep efficiency of 85 or more was lower compared to men. On the contrary, the percentage of women with sleep efficiency below a value of 65 was higher than the percentage of men. The presence of sleep disorders was also different between sexes since a high proportion of women experienced these problems once or twice a week. Concerning the use of sleep medication, no significant differences were observed between sexes. Differences between sexes were observed regarding daytime dysfunctions, with a higher percentage of women affected. Table 2 shows the results for the chi-square test when participants were grouped by age: young = 18-40 years and old <unk> 40 years old. The younger group showed longer sleep latencies, a higher proportion of young participants reported a sleep duration of fewer than 5 h, more than a half of the younger participants reported bad or very bad sleep quality, a higher proportion of young participants reported low sleep efficiency <unk>65%, younger participants reported a higher frequency of sleep disorders once or twice a week, the older group reported a high percentage of participants using drugs several times, and daytime dysfunctions showed a proportion of more than half of younger participants reporting this issue as moderate or severe. Figure 2 summarizes the overall results concerning the PSQI evaluation of sleep quality grouped by ages. The <unk>40-year-old women seem to be the most affected group and showed differences from the other age groups. Similarly, <unk>40-year-old men reported more sleep disturbances than the >40-year-old participants. No significant differences were observed between men and women who were older than 40. Figure 2 summarizes the overall results concerning the PSQI evaluation of sleep quality grouped by ages. The <unk>40-year-old women seem to be the most affected group and showed differences from the other age groups. Similarly, <unk>40-year-old men reported more sleep disturbances than the >40-year-old participants. No significant differences were observed between men and women who were older than 40. Regarding anxiety, 34.8% of women reported mild symptoms, 18.3% moderate, and 18% severe. In the case of men, 38.1% reported mild symptoms, 16.9% moderate, and 19% severe. No differences between genders or ages concerning anxiety were identified. Regarding depression, results for the PHQ-9 indicated no differences between genders or age, with p = 0.664.However, 24.5% of women and 18.6% of men reported moderately severe and severe depression symptoms. In addition, the results show that younger females reported a higher percentage of depression symptoms (26.4%) than older males (14.7%) Furthermore, when the use of electronic devices (TV, cellphone, or any kind of computer) was analyzed, the results indicate that more than half of the sample (68.9%) was exposed to bright lights for at least 5 h daily. Women reported a higher percentage of screen use at around 10 h daily (17.4%) and reported a longer used of screens compared to men (13.4%) p <unk> 0.01. When activities related to preparing for bed were explored, a high percentage of the participants (62.3%) reported the use of some kind of electronic device as part of their activities; prior to sleep no differences were observed regarding age. --- Discussion The purpose of this study was to explore the presence of sleep and/or mental health disturbances in the Mexican population while living under quarantine conditions to prevent COVID-19 infections. Our results included the evaluation of the first eight weeks of social isolation in Mexico and focused mostly on big cities. Similar to an Italian report, a high percentage of females responded to the survey (more than 70% in both studies) [24,25], which suggests that women manifest a greater concern for their own and their family's issues. Regarding anxiety, 34.8% of women reported mild symptoms, 18.3% moderate, and 18% severe. In the case of men, 38.1% reported mild symptoms, 16.9% moderate, and 19% severe. No differences between genders or ages concerning anxiety were identified. Regarding depression, results for the PHQ-9 indicated no differences between genders or age, with p = 0.664. However, 24.5% of women and 18.6% of men reported moderately severe and severe depression symptoms. In addition, the results show that younger females reported a higher percentage of depression symptoms (26.4%) than older males (14.7%) Furthermore, when the use of electronic devices (TV, cellphone, or any kind of computer) was analyzed, the results indicate that more than half of the sample (68.9%) was exposed to bright lights for at least 5 h daily. Women reported a higher percentage of screen use at around 10 h daily (17.4%) and reported a longer used of screens compared to men (13.4%) p <unk> 0.01. When activities related to preparing for bed were explored, a high percentage of the participants (62.3%) reported the use of some kind of electronic device as part of their activities; prior to sleep no differences were observed regarding age. --- Discussion The purpose of this study was to explore the presence of sleep and/or mental health disturbances in the Mexican population while living under quarantine conditions to prevent COVID-19 infections. Our results included the evaluation of the first eight weeks of social isolation in Mexico and focused mostly on big cities. Similar to an Italian report, a high percentage of females responded to the survey (more than 70% in both studies) [24,25], which suggests that women manifest a greater concern for their own and their family's issues. According to our results, a high percentage of the participants' sleep was severely affected, with reports of poor quality of sleep. Furthermore, the percentage of women affected was higher than men. Additionally, younger people seemed to be more sensitive to conditions of isolation and their sleep was more affected than that of older people. Women under 40 proved to be the most sensitive population to sleep disturbances. These results are similar to those reported in early studies on Italian [25] and Chinese samples (4). These results can be explained by the role that young women usually play within the family home. In addition, the increase in domestic violence, mainly affecting young women, has been recently reported [26]. These results may reflect important sleep restriction conditions presented among the Mexican population while facing quarantine conditions. Thus, the present data should be taken as an extreme concern since current knowledge about sleep restriction suggests deficits in behavioral performance and mood swings [27,28], and research conducted on medical students with sleep restriction reported an impairment of empathy and an increase of in-group conflicts [29]. In addition, similar to what was reported in other countries during quarantine, the calls to 911 reporting domestic violence and violence against health workers in Mexico have considerably increased [30,31]. Moreover, and as reflected in our results, disturbances of the circadian rhythm play a role in the onset of sleep disturbances. In this sense, the purpose of the study included the exploration of the use of bright screens at night. Participants declared a frequent use of these instruments with a clear trend of using them at night. As has been reported, the excess of blue light stimulation provided by electronic devices interferes with the onset of sleep [32]; particularly in school children and adolescents, the excessive use of electronic devices induces a delayed bedtime, shortened sleep duration, and longer sleep latency [33,34]. Thus, it seems that quarantine conditions facilitate the use of TV and electronic devices among the entire family, which may affect circadian rhythms, resulting in a cascade of adverse events that includes sleep restriction and reduced sleep quality [19]. Moreover, the association between the use of electronic devices, sleep disturbances, and depression has been confirmed for adolescents. Thus, the presence of depressive symptoms observed in our sample can be due, at least partially, to sleep disturbances induced by the excessive use of electronic devices [35], since there is a clear relationship between young women using electronic devices for 10 h daily and sleep disturbances. Concerning ages, Yu et al. reported in 2018 [36] that social isolation was the main factor for sleep disturbances in older people; when comparing their results obtained from 639 participants in Taiwan, the authors indicated that social isolation and not loneliness was the main factor for sleep disturbances. Controversially, our results do not confirm this report. Rather, we observed that younger people manifested sleep disturbances included in the PSQI, whereas older people showed only slight changes in sleep. Furthermore, young women appeared to be more vulnerable to lockdown conditions, showing an increased frequency of sleep and mental health disturbances. It is possible that the role and responsibilities that young women have within Mexican families impose additional pressure on their mental stability. It remains to be elucidated why young women seems to be more vulnerable to sleep loss in lockdown conditions. Some studies reported sleep disturbances in pregnancy and during premenstrual syndrome. However, to date, endocrine factors cannot fully explain the increased frequency of sleep and mental disturbances in young women. Furthermore, social isolation induces a decrease in solar light stimulation that negatively affects both circadian rhythm synchronization and mood [37]. Mexico City is the most crowded region of the country with nearly 20 million people, and its authorities decided to implement measures of social isolation to avoid spreading the virus. Health authorities reported that more than 70% of the people decided to stay at home with minimal social interactions. The present results are in accordance with other recent publications reporting mental health disturbances due to lockdown conditions all over the world [38]. Concerning generalized sleep restriction, public health strategies should be discussed to avoid future adverse health outcomes. Since 2015, the American Academy of Sleep Medicine and the Sleep Research Society addressed the relationships between sleep duration and the risk of suffering from a wide variety of diseases including diabetes, stroke, endocrine dysfunction, mental illness, and cancer. A panel of experts in sleep medicine reviewed the available scientific evidence in more than 5300 scientific articles. After more than 12 months of review and discussion a final statement was disclosed. The consensus statement indicated than sleeping less than 7 h regularly increases the risk of death. This death risk was associated with the presence of obesity, diabetes, heart diseases, and depression. In addition, there is also a greater risk of accidents [39]. Thus, there is an increased probability that this prolonged period of sleep restriction leads to an increase in the frequency of dangerous diseases among the general population. Our exploratory study involved samples reflecting the first weeks of quarantine. The most significant results indicate that a high percentage of the participants reported poor sleep quality and that women and young people were more sensitive to such effects. In addition, quarantine conditions seemed to facilitate the use of electronic devices at night, which can affect the quality of sleep and favor the presence of depression, particularly in young people. These results are important for designing mental health intervention strategies during quarantine that include sleep hygiene habits and practices to maintain or improve the quality of sleep in the population, and to prevent the presence of symptoms of depression, with emphasis on women and youth. The present results should be taken with caution because the population evaluated came only from the country's large cities and not from other regions in which quarantine conditions and related information may vary. Furthermore, our sample was limited to participants who could access digital devices to answer the questionnaires, leaving out a large part of the population that does not have access to such resources. Importantly, we do not have previous measures on sleep quality, depression, anxiety, or the use of electronic devices in equivalent samples before quarantine conditions, so, although our results coincide with previous reports in Italy and China, they cannot be strictly associated with only quarantine conditions. Future studies can be conducted at different times in the pandemic as a comparative measure. The pandemic and lockdown conditions have lasted more than expected, so new surveys should be done to explore the effects of several months of social isolation on sleep, anxiety, and depression, to understand in more detailed the effects of quarantine conditions on mental health. --- Conflicts of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The coronavirus disease (COVID-19) that broke out in China in December 2019 rapidly became a worldwide pandemic. In Mexico, the conditions requiring the declaration of a sanitary emergency were reached by the last week of March 2020, and health authorities' limited mobility and imposed social isolation were the main strategies to keep the virus from spreading. Thus, daily living conditions changed drastically in a few days, generating a stressful situation characterized by an almost complete lack of mobility, social isolation, and forced full-time interactions with family members. Soon, complaints of sleep disturbances, anxiety, and symptoms of depression were reported. The present study reports the results of an online survey performed during the first two months of isolation. Questionnaires exploring sleep disturbances, anxiety, and depression were sent to people who responded to an open invitation. A total of 1230 participants filled out the sleep questionnaire, 812 responded to the anxiety questionnaire, and 814 responded to the depression questionnaire. Both men and women reported poor sleep quality, but women showed a higher proportion (79%) than men (60%); young women were more likely to be affected by social isolation. Concerning anxiety and depression, both sexes reported high similar symptoms. These data suggest that stressful conditions related to social isolation and the economic uncertainty caused by the pandemic may induce mental health disturbances, which may become worse with sleep restriction.

Introduction With a quarter of Nepal's 29 million citizens living below the poverty line (Asian Development Bank, 2016), Nepal is facing health and socioeconomic disparities which have compelled millions of Nepalese to look for work away from home. Currently, over 2.2 million Nepalese, mostly men, are engaged in migrant work (Central Bureau of Statistics, 2012a, 2012b). These workers annually send approximately 3 billion dollars in remittance to their country, comprising 15% of Nepal's GDP (Government of Nepal, 2014). Due to the large number of Nepali men spending extended periods of time working away from home, it is possible that Nepal's large pool of male migrant workers is engaged in risky sexual behaviour, putting their wives at risk for reproductive tract infections (RTIs) when they return home. According to World Health Organization (WHO) female RTIs can be endogenous, iatrogenic or sexually transmitted (WHO, 2005). They are often asymptomatic and represent a significant cause of maternal morbidity and mortality in low and middleincome countries (LMIC). According to the Nepalese Ministry of Health and Population (MOHP) and based on previous studies conducted in Nepal RTIs are prevalent but not equally distributed across different geographic regions (MOHP, 2012). Studies have suggested that physical and social differences between local environment conditions, such as lack of access to clean water or high prevalence of squatter settlements can cause an increase in RTI rates (Hernandez et al., 2006). A cross sectional study of 282 women in seven randomly chosen health centres in Nepal's Kaski District found that 78.9% of women had at least two symptoms of RTIs (WHO, 2011). A study of 104 women from the outpatient department of gynaecology at Tribhuvan University Teaching Hospital in Nepal's capital, Kathmandu, found that 30% of women reported RTI symptoms (Bohara et al., 2012). Women surveyed during a gynaecological camp conducted in Bajhang District in the Far-West Development Region (DR), found that 13.9% of the 530 women surveyed had an RTI (Tuladhar, 2005). Among women attending a reproductive health camp in the ethnically homogenous municipality of Khokana, 23.5% had clinical cervicitis (Johnson et al., 2013). Given geographic variations in RTI prevalence, understanding how geospatial and sociocultural factors are related to RTIs can help anticipate priorities in the health of Nepali migrant workers and among their wives or other sexual partners. --- Geospatial Nepal has a diverse geographic landscape that could potentially result in the creation of clustered subpopulations at high risk for RTIs. Nepal is divided into five DRs (Eastern, Central, Western, Mid-West, and Far-West) and three ecological zones (Mountain, Hill and Terai). The two former regions represent over 77% of Nepal's land mass and 52% of its population. In contrast, the Terai region (consisting of tropical grasslands, savannahs, forests and river valleys in the South) represents 23% of Nepal's land mass and 48% of its population (Central Bureau of Statistics, 2012a, 2012b). Approximately 19% of Nepal's population lives in urban areas where 40% of urban adolescents have comprehensive knowledge of sexually transmitted infections (STIs) (Khanal et al., 2013), but in the rural areas of Nepal only one-third of young men and one-fourth of young women have any substantial knowledge of STIs (Khatiwada et al., 2013). Residents of Nepal's Terai ecological region report some of the lowest rates of condom use in Nepal (Central Bureau of Statistics, 2012a, 2012b). The prevalence of male migration in Nepal varies by region and this variation could potentially contribute to geographically different RTI rates. A majority of migrant workers who travel to India for employment come from Nepal's western DRs. However, out of the Nepali migrant workers registered for labour permits in countries other than India, 64% come from the two non-western DRs (Eastern DR, Central DR) (Government of Nepal, 2014). Fifty-seven percent of households report at least one member leaving home for work within the last five years, even higher (66%) in the Far-West DR (Central Bureau of Statistics, 2012a, 2012b). Knowing that RTI rates vary by region, the uneven regional distribution of migration patterns in Nepal is potentially a RTI risk factor. RTIs include STIs (Kim le et al., 2012) and previous research has shown that men who engage in migrant work are at higher risk for STIs, transmissible to their wives when they return home (7.8). Migrant workers represent <unk>10% of Nepal's population but 41% of the reported human immunodeficiency virus (HIV) infections (Bam et al., 2013). A study of male migrant workers in the Nepali district of Doti reported that 8% of returning migrant workers was infected with the human immunodeficiency virus (HIV) and 22% were infected with syphilis (Poudel et al., 2003). Circular migration, where individuals leave their partners for work and return only periodically, is one of the most common forms of migrant work in Nepal and it is correlated with an increased STI risk (Zuma et al., 2005). Complicating the migration-STI association, migratory men sometimes establish parallel sexual partnerships at their distant workplaces, forming long-term bridges that facilitate the spread of STIs (Nguyen et al., 2011). Geospatial and sociocultural factors might be associated with the sexual health of Nepali migrant worker's spouses. The objective of our exploratory study was to investigate the geospatial differences between the prevalence of RTI symptoms reported by married Nepali women enrolled in the 2011 Nepal Demographic and Health Survey (NDHS) and assess the impact of their husbands' migration status. --- Materials and Methods --- Demographic and health survey data The NDHS programme was established in Nepal by the United States Agency for International Development (USAID) in 1984 (USAID, 2006). The goal was to provide decision makers and government officials with improved information and analyses useful for making health policy choices on marriage, fertility, family planning, reproductive health, child health, and HIV/AIDS. The standard version of the programme is conducted every five years and began in 1996. However, an in-depth assessment of fertility and family planning in Nepal was conducted already in 1987. The 2011 NDHS programme included 12,674 women of reproductive age (15-45), but our study sample was restricted to all 9607 married women in the NDHS. Data for the 2011 NDHS are based on a two-stage random sample design. During the first stage of sampling, a stratified sample of enumeration areas (EA) is selected with a probability of selection proportional to their size. An EA can be a city block or an apartment building in urban areas, while in rural areas it is typically a village or group of villages. The population and size of sampled clusters vary, but on average, clusters contain 100 to 300 households, 20 to 30 of which are randomly selected for survey participation. During the second stage of sampling, a number of households are selected for surveying by equal probability systematic sampling in the EAs. In our study, a household questionnaire was completed for each selected household to identify women of age 15-49, men of age 15-59 (15-54 or 15-49 in some surveys) and children under the age of five. Every eligible woman was interviewed with an individual questionnaire (USAID, 2012). --- Nepal Demographic and Health Survey geographic information systems data The NDHS has been collecting spatial and demographic data since 2001 in Nepal; detailed methodology is described elsewhere (Burgert et al., 2013). Briefly, coordinate locations (latitude and longitude) represent the population centre of a sampling cluster. Cluster coordinates are linked to geographic information system (GIS) administrative unit shape files. Latitude and longitude positions of all clusters contain random errors according to the follow- ing criteria by the NDHS to maintain respondent privacy: i) urban clusters contain between 0 and 2 km of error; ii) rural clusters contain 0 and 10 km of error. The error, however, is restricted so that clusters stay within the country and within the NDHS survey region where the surveys were conducted (Burgert et al., 2013). GIS data for the 2011 NDHS are represented at three levels: i) Nepal's five DRs; ii) Nepal's three ecological zones; and iii) a subregional level containing 13 combinations of the development and ecological zone (Central Hill, Central Mountain, Central Terai, Eastern Hill, Eastern Mountain, Eastern Terai, Far-Western Hill, Far-Western Terai, Far/Mid/Western Mountain, Mid-Western Hill, Mid-Western Terai, Western Hill, Western Terai) (Burgert et al., 2013). --- Reproductive tract infections outcome, migration status and variables of interest A woman is considered to have a RTI symptom if she answered yes to either of the following NDHS questions: Had abnormal genital discharge in last 12 months or Had a genital sore or ulcer in last 12 months (Khatiwada et al., 2013). This study used the definition of male migration found in the NDHS report Impact of Male Migration on Contraceptive Use, Unmet Need, and Fertility in Nepal (Khanal et al., 2013). The report defines male migration as a non-resident husband who has migrated internationally or within the country. This definition does not specify the length of migration time, nor does it differentiate between reasons for migration. However, given the high rate of male labour migration in Nepal, the report assumes the vast majority of non-resident husbands are away for work. The NDHS survey question Is your husband/partner living with you now or is he staying elsewhere? is asked to assess male migration. If a woman states that her husband/partner is living elsewhere, then the amount of time her husband/partner is away from home is determined by a follow-up question For how long have you and your husband not been living together? We categorised this time as either <unk>1 year or alternatively as away >1 year in order to follow the categorisation of migrant time used in the standard national reports conducted using the NDHS data (Khanal et al., 2013). Other covariates were classified as demographic, socioeconomic status (SES), sexual behavioural or geographic variables. Demographic variables included age (used as a continuous or a categorical variable for descriptive tables) and urban/rural residence. SES variables included an asset index based wealth (categorised as lowest, low, middle, high, highest) and the educational level (no education, primary, secondary or higher education) for each gender. Sexual behaviour variables included questions such as contraception currently used (condom, other or none); ever heard of a STI; and age at first marriage. The geographic variables included the five DRs, the three ecological zones, the 13 development and ecological zone combinations (Central Hill, Central Mountain, Central Terai, Eastern Hill, Eastern Mountain, Eastern Terai, Far-Western Hill, Far-Western Terai, Far/Mid/Western Mountain, Mid-Western Hill, Mid-Western Terai, Western Hill, Western Terai) as well as urban/rural residence. --- Ethical considerations --- Statistical analysis We described the frequencies of potential risk factors for women with and without RTI symptoms, using analysis of variance and Chi-square statistical tests. Multivariable logistic regression models stratified by DR investigated the association of the husband's migration status and reporting RTI symptoms while adjusting for asset-based wealth index, contraception use, age at first marriage, urban/rural status, and husband's education. All variables included in the multivariable models were selected based on: i) a review of the RTI risk factor literature from developing countries or ii) a P value of <unk>0.1 in univariate analyses. Due to the complex survey design of the NDHS, specific statistical procedures were used to correctly estimate variance. Variance estimation for univariate and multivariate statistical methods contained in the SAS PROC SURVEY procedures in SAS version 9.4 (SAS Institute, Cary, NC, USA) were used to perform all statistical analyses and account for the NDHS survey design. ArcGIS version 10.1 (ESRI, Redlands, CA, USA) was used to produce choropleth maps of the percentage and change in percentage of women reporting RTI symptoms according to a husband's migration status. --- Results The average age of women surveyed was 31.3 (SD<unk>0.13) years. Almost one third of married women reported that their husbands were currently migrating for work, (not currently migrating, 67.9%, migrating for <unk>1 year, 25.7%, >1 year, 6.3%). Of the women whose husbands were migrating for work, 12.4% of them reported RTI symptoms, while 14.9% of women whose husbands were not migrating for work reported RTI symptoms (P=0.35). Of the 12.6% of women who answered yes to at least one question about RTI symptoms, 19.9% said they had a genital sore or ulcer and 92.7% a genital discharge in the last 12 months. Univariate analysis of these married women indicated that the 1). The women who reported RTI symptoms were more likely to reside in a rural area if their husbands were currently migrating for work (rural residence with migrant husband 91.7%, and non-migrant husband 84.2% P=0.001). Women who reported RTI symptoms and whose husbands were currently migrating for work were more likely to reside in the Terai ecological zone than women who reported RTI symptoms and whose husbands were not migrating for work (Terai ecological zone: migrant husband 57.3%, non-migrant husband 45.4% P=0.008). Residence in the DR was significantly associated with a husband's migration status (husband's migration status: Eastern 32.4%, Central 26.9%, Western 39.7%, Mid-Western 29.6%, Far Western 34.6%, P=0.001) (Table 2). Multivariate logistic regression models investigated the association between the husband's migration status and reporting RTI symptoms by DR. We found that women whose husbands were migrating for <unk>1 year were more likely to report RTI symptoms than women whose husbands were not currently migrating for work if they lived in the Mid-West region (OR 1.93 95%CI 1.02-3.67) or if they lived in Nepal's Far-West region (OR 2.89 95%CI 1.24-6.73; Table 3). The prevalence of male migrant workers' wives reporting RTI symptoms was less than 10% in two NDHS sub-zones (RTI symptoms: Central Terai 9.4% and Far Western Terai 9.0%; Figure 1). --- Article Married women whose husbands were currently migrating for work were more likely to report RTI symptoms than married women whose husbands were not currently migrating for work in 7 of the 13 NDHS Sub-regions (Central Mountain, Far Western Hill, Eastern Terai, Central Terai, Western Terai, Mid Western Terai, and Far Western Terai). Married women whose husbands were not migrating for work were more likely to report RTI symptoms than married women whose husbands were currently migrating for work in 6 of the 13 NDHS Sub-regions (Eastern Mountain, Far/Mid/Western Mountain, Eastern Hill, Central Hill, Western Hill, and Mid Western Hill). Married women from the Central mountain Sub-region reported the highest percentage of RTI symptoms regardless of whether their husbands were migrating or not (Central mountain: migrant husband 21.9%; non-migrant husband 17.3%) (Figure 2). --- Discussion Among the 32.0% subset of women in our study who reported that their husbands were currently migrating for work, 12.6% Article [Geospatial Health 2017;12:513] [page 161] reported RTI symptoms. Multivariate logistic regression models indicated increased the odds of a woman reporting RTI symptoms if she resided in the Mid-Western or Far-Western DRs and her husband was currently engaged in migrant labour for more than one year. Our results suggest that the sub-region where a woman lives in conjunction with her husband's migratory status is associated with her risk of having RTI symptoms. However, the migration association was not robust and the overall finding did not suggest an association between male migration and RTI symptoms in his wife. RTI rates in other South Asian countries are known to vary by geographical region (Begum et al., 2003;Rehan et al., 2009;Yasmin and Mukherjee, 2012). For example, the prevalence of RTIs in women varies by region in India, where the rates of RTIs for Uttar Pradesh, West Bengal, and Ludhiana are 34, 23.6, and 17.3%, respectively (Nandan et al., 2001;Yasmin and Mukherjee, 2012;Philip et al., 2013). An eighteen months long study conducted in Pakistan's Punjab Province found that rural women were 32% more likely to have a STI identified through syndromic management compared to women living in urban areas (Khan et al., 2014). The likelihood of women reporting RTIs also differs by urban/rural status in Bangladesh (Bogaerts et al., 2001;Hawkes et al., 2002). While there is sparse information on RTIs in relation to migration in Nepal, regions with high levels of STI risk factors are often the same regions characterised by high levels of migration. Nepal's Eastern and Central DRs have the largest percentage of migrant workers and some of the strongest indicators of STIs in Nepal (MOHP, 2012). The 2011 NDHS report suggest that men from Central DR reported a higher percentage of sex worker usage than the national average. Women from the Central DR were the least likely to know where to obtain a condom, while women from the Eastern DR were the least likely to report condom usage when hav-ing sex (MOHP, 2012). While in the overall univariate analysis, women without husbands away reported more RTIs, the prevalence differed by development regions and ecological zones. If both labour migration and STI rates are known to be associated with geospatial variables, then it is possible that RTI rates among the wives of migrant workers could correlate with these variables as well. Of note, certain geographical regions only had few husbands migrating, so the analyses could not be conducted. Results from our study indicate that women whose husbands are currently migrating for >1 year and who reside in either the Mid-Western or Far-Western DR are more than twice as likely to report RTI symptoms than women whose husbands have never migrated for work, even after adjusting for wealth, contraception use, age at first marriage, urban/rural status, and the level of husband's education. Our results suggest that if a woman's husband is currently engaged in migrant work, her regional residence influences the likelihood of her reporting RTI symptoms. More studies are needed to assess whether some of these disparities are due to either women themselves or their husbands, in certain geographical regions, are having extramarital affairs that are the cause of infection. Previous reports indicate that most male Nepali migrant workers residing in the Mid-Western and Far-Western DR migrate to large Indian cities, such as Mumbai or New Delhi (Poudel et al., 2004) where HIV prevalence among sex workers is high and where Nepali migrant workers often engage with commercial sex workers (Poudel et al., 2003(Poudel et al.,, 2004)). Infectious diseases contracted by male migrant workers who migrated to India may be transmitted to their wives when they return, notably in Mid-Western or Far-Western Nepal. The DRs with high rates of migrant activity and RTI symptoms could represent entry points of STIs into Nepal. Previous studies have reported that, although the percentage of individuals infected with HIV in Nepal is <unk>1% of the population (Government of Nepal, 2014), >40% of individuals who test positive for HIV have engaged in migratory work (Poudel et al., 2003). One study determined that 22% of migrant workers returning to the Far West Doti District tested positive for syphilis (Thapa et al., 2014). Our study is compatible with these other studies suggesting that migratory work is a potential mode of entry for STIs into Nepal. Nepali truck stops and highway areas are well known for high STI rates (USAID and FHI Report, 2011), suggesting that the trucker-sex worker risk axis is relevant for Nepal as is true in many other nations. Migratory work represents a potential way in which new STI infections are entering Nepal. Though the overall rates of STIs in migrant workers were similar to the rates in non-migrant workers, the reference to the geographic region suggest complex dynamics within the migrant population. --- Article Study strengths included the nationwide representativeness of the 2011 NDHS and the reasonable sample size. Study limitations included the ecological study design and our inability to translate group characteristics to the individual level. For example, only current migration is captured in the study and not the previous history of migration of the husbands or the women themselves are accounted for. Our outcome relies on self-reporting of RTI symptoms; syndromic identification of STIs and RTIs is known to have low sensitivity (Yin et al., 2008), though WHO considers syndromic identification of STIs and RTIs to be acceptable for screening in South Asia (WHO, 2011). --- Conclusions Our study suggests possible regional differences in RTI symptoms reported by married women in the NDHS. It is possible that a husband's migratory work is associated with the rates of RTI symptoms reported by married women in specific regions of Nepal. Additional studies providing laboratory-based survey data are needed to assess the role of labour migration on the rates of RTI symptoms in Nepal more definitively, to assess migration patterns over time in general, and to evaluate regional behavioural patterns and other factors associated with RTIs and STIs.

Nepal has a diverse geographic landscape that could potentially create clustered subpopulations with regional socio-cultures that could result in differential health outcomes. With an alarming rise in married male populations migrating for work, it is possible that these migrants are engaged in risky sexual behaviour, putting their wives at risk for infectious disease outcomes, including reproductive tract infections (RTI), when they return home. The prevalence of male migration varies by geographic region in Nepal and this variation could potentially contribute to different RTI rates. Using a cross-sectional dataset (the 2011 Nepal Demographic and Health Survey) including 9607 married women, we investigated geospatial and socio-cultural factors associated with the symptoms of RTIs with a focus on the husbands' migration status. Choropleth maps were created to illustrate areas with high percentages of RTIs that correlated with migration patterns. Overall, 31.9% of the husbands were migrating for work. After adjusting for wealth, contraception use, age at first marriage, urban/rural status and husband's education, women whose husbands had been absent for a year or more in Nepal's Mid-West region (OR 1.93 95%, CI 1.02-3.67) or Far-West region (OR 2.89 95%, CI 1.24-6.73) were more likely to report RTI-like symptoms than others. Our results suggest a potential association between husbands' migration status and Nepali women reporting RTI symptoms by geographic regions. However, further research is needed to put this outcome on a stronger footing with respect to this under-studied population, specifically in the context of geographical variation.

Introduction Alternate Reality Game (ARG) is a term often used to describe a game, or narrative, that is delivered to players and participants via a variety of different forms of media, for example via the Internet, social media, newspapers and physical artefacts or telephony, using voice or SMS, with the aim that this mixed use of media allows story, characters and interaction to develop to an extent that would not be possible in a mono-media context. ARGs may make use of concepts or technologies more commonly found in so-called pervasive games, for example spatial, temporal or social expansion -operating in large physical areas, lasting for an extended or apparently undefined length of time, and involving an unspecified or unclear cast of characters and players [20]. ARGs use these expansions in order to create the illusion of literally an alternate reality, or the illusion of not being in fact a game, while also framing themselves as being games, even through explicit denial. A fundamental tenet of many ARGs is "this in not a game". These acts allow players to immerse themselves in the characters that they are playing within the game and, as Montola [20] describes, to pretend and perform belief collectively contributing to the social expansion and the feeling of alternate reality. For this reason ARGs provide a compelling demonstration of performance as a social process, one that transcends narrative forms and professional creative practice and is manifest in everyday interactions and identity formation. In collaboration with a community of alternate reality gamers and developers, McGonigal developed the following definition: "alternate realities are the antiescapist game... designed to make it easier to generate the four intrinsic rewards we crave -more satisfying work, better hope of success, stronger social connectivity, and more meaning... In other words, ARGs are games you play to get more out of your real life, as opposed to games you play to escape it." She argues that "because ARGs are played in real-world contexts, instead of in virtual spaces, they almost always have at least the side effect of improving our real lives." [17] Regardless of the form of the media, ARGs often involve players performing a number of feats to uncover or gain access to an understanding of the underlying narrative. These might include puzzle solving or code breaking, interaction with in-game characters potentially played by live actors, or socalled scavenger hunts, searching online or physical venues for clues or artefacts. The spatially expansive, or physically distributed nature of these tasks, for example, necessitates the inherently communal nature of ARGs, with each player contributing their own skills, or even simply geographic location, towards solving the common goal. I Love Bees [18] was an ARG that was developed in order to serve as a marketing campaign for the computer game Halo 2. The ARG's narrative was not immediately apparent to players, instead consisting of a distributed narrative consisting of many hundreds of small fragments, emails and MP3 recordings, scattered across the Internet or slowly released to different players, requiring the community of players as a whole to operate together in order to piece together and understand the narrative of the game. Ultimately the game involved physical interaction, including assembling at particular GPS coordinates, or organised face-to-face meetings between characters and the players who travelled to visit them. Project APE [23] was an ARG used to promote the film Planet of the Apes, which built itself upon the existing GeoCaching phenomenon, existing solely as a number of caches that players had to physically locate -in this case the locations of the key places within the game are known, however the details of the narrative are unknown until the caches are discovered. "Alternate Reality Games take the substance of everyday life and weave it into narratives that layer additional meaning, depth, and interaction upon the real world", functioning by means of "the insertion of additional slices of reality into our own, and the only demand is that you interact with these as yourself" [15]. Of course, performance (the adoption of a persona in relation to the story world) is central to playing an ARG, but the rootedness in (physical, temporal) reality that the connection to individual lived experience establishes is stronger than with other forms of game (e.g. MMORPG) or interactive narrative (e.g. Choose Your Own Adventure books). Although ARGs are relatively new, and sparsely populated as a genre compared to existing forms of game play, their communal nature, potentially large audiences, and highly pervasive nature notably support rich elements of performance and interaction by players, and consequently new opportunities for entertainment and academic communities, particularly theatre studies, HCI and CSCW, making them an apposite topic of discussion in this special issue. This paper presents a study of The Malthusian Paradox (TMP), an ARG developed and deployed in September 2012 by the artists group Urban Angel, which is directed by Dominic Shaw and Adam Sporne. The authors collaborated with them on the project in a research capacity, from the planning stage, throughout production and beyond the conclusion of the game. We offered technical support to the artists, documented their process and players' activities, and collected qualitative accounts of the experience from both perspectives. The game was hosted simultaneously in arts venues across four cities in the East Midlands region of the United Kingdom. 300 participants played it over a period of three months. TMP cast players as agents of a radical organisation attempting to uncover the truth behind a kidnapping and a sinister biotech corporation. It had a strong episodic narrative driven by professionally produced short films that attempted to respond to the actions of players. Players participated in the game via a broad spectrum of interaction channels, including performative group spectacles and 1-to-1 engagements with game characters in public settings, making use of low-and high-tech physical and online artefacts including bespoke and third party websites. Players and game characters communicated via telephony and social media in both a designed and an ad-hoc manner. We will explore both the strategies employed by Urban Angel to create an ideal coherent pervasive journey for their players, and the consequences of those strategies for the actual journeys that resulted. Rather than see agency residing with either the producers or the audience, this article will explore how different forms of agency, which related to different components of the narrative, were fluidly and reactively negotiated throughout TMP's evolution. Such agency, and shifts in who could claim it, created an ambiguous but still important frame around the experience for both producers and players. --- "It's Their Thing Now" As above, central to much work that focuses on ARGs, transmedia and pervasive narratives is the notion of audience agency and power. Early precursors to the narrative forms demonstrated by TMP such as invisible theatre explicitly used their hybrid spaces to encourage political engagement and generate revolutionary movements. Augusto Boal identifies the potential for such forms to act in this way precisely through the shift in the audience's position, and so their agency: "The spectator is less than a man and it is necessary to humanize him, to restore to him his capacity of action in all its fullness. He too must be a subject, an actor on an equal plane with those generally accepted as actors, who must also be spectators" [3]. For Boal, the potential to bring the audience into a more participatory role, to shift them from spectators to subjects as actors with representational agency over their own experiences, allows a reconfiguration of their engagement with the narrative and, for him, its essentially political message, with liberating consequences. More recently, Lévy's concept of 'collective intelligence' [13] in which the players act not as individuals but as a group, combining brainpower and skills to complete tasks that they are unable to complete alone, has become a central component to understanding of transmedia narratives and ARGs. Henry Jenkins appropriates Lévy to highlight the importance of agency in such a process: "For Lévy, at his most utopian, this emerging power to participate [in knowledge communities] serves as a strong corrective to those traditional sources of power, though they will also seek ways to turn it toward their own ends. We are just learning how to exercise that power -individually and collectively -and we are still fighting to define the terms under which we will be allowed to participate. Many fear this power; others embrace it" [11]. In both models, the collective act of interpretation becomes political, a display of power via participation. It can be argued that ARGs draw on a trajectory of participatory art that stretches back to the early twentieth century. Certainly, by offering a fragmented transmedia narrative, often with an explicitly anti-authority tenor, the games invite players to question the logic and boundaries of the alternative reality represented. And, when they provide opportunities for physical involvement, ARGs reduce the distance between actors and spectators, giving the latter access to the process of production. This is not to argue that ARGs necessarily redefine power relations; the critique of participatory art is that it is "no more intrinsically political or oppositional than any other" [2]. By sharing some element of authorial control with players, the producers of ARGs encourage collaborative creativity and attempt to foster exciting and inclusive experiences. ARGs explore the equality of players as co-producers, through their interpretation of, and enriching responses to, the alternate reality, giving them agency in the unfolding of the game. The appropriation of spaces for performance outside convention theatre buildings is a practice that historically "represented a reaction to crises generated by bursts of modernization in society" [6]. In a period of rapid digital innovation, ARGs have features that make them akin to a festival, "the genre of cultural performance most capable of leading to the formation of new communities". In the case of most ARGs to date, including TMP, the game is a singular, unique event [10]. Players of TMP did not have to pay anything to participate, potentially opening up the experience to those who might not be able to access live performance for economic reasons. The producers of TMP looked to involve large numbers of people in a situated narrative linked to particular locations in the East Midlands. However, the community formed by ARGs is ephemeral, in the sense that it is unlikely to persist on an extensive level beyond the scope of the narrative itself. The importance of agency does not always have overt political overtones and is similarly picked up elsewhere on work on ARGs that explores the importance of a'suspension of disbelief'. Jane McGonical argues that immersion occurs in an ARG via players choosing to maintain the narrative's integrity. This occurs through what she terms a 'Pinnochio effect'. In this case, agency manifests through a will to believe the game's reality, even when faced with its artificial nature, to "wink back at the puppetmasters and pretend to believe" [16]. O'Hara, Grian and Williams extend McGonical's work to argue that such suspension of disbelief is "at times delicate and vulnerable. There were times when there were breakdowns in this behaviour and through these the real importance of this collective responsibility is revealed" [21]. The communal experience enshrined in models of collective intelligence is entwined with a desire to uphold the integrity of an ARG's narrative. Solving the narrative and ensuring that it remains intact are both matters for deliberate collective action on the part of the audience. If agency is one key way to articulate the experience of ARGs, a second emerges from the fields of sociology and gaming. The work of sociologist Erving Goffman has been applied separately in the fields of HCI and theatre studies; it theorises the cognitive work involved in making sense of situated social experiences of various types, the 'frames' that determine the meaning of human interactions. According to Goffman, frames are "principles of organization which govern events -at least social ones -and our subjective involvement with them" [8]. The frames within which different forms of behaviour occur and are understood are not immutable, they are socially contingent and rely upon learnt rules and conventions, which can evolve or be deliberately contravened. Goffman uses theatre-going as a metaphor for analysing everyday behaviour, one of the examples to demonstrate many elements of his theory, which is indicative of the performative nature of social structures and relations. ARGs such as those described above transform activities "already meaningful in terms of some primary framework... into something patterned on this activity but seen by the participants to be something quite else", a process Goffman refers to as 'keying'. According to Goffman, play is an example of a keying: play fighting, for instance, resembles real fighting in many respects because the two share common features. In order for play to be engaged in participants must appreciate the original meaning of the activities and that alteration is taking place, and there need to be cues to indicate the period to which the transformation is restricted. A keying can serve to integrate very different activities into a coherent experience of play, analogous to the concept of meaningful play found in game design literature [25]. This concept of framing has also evolved through discussion of more explicitly 'game' scenarios. Huizinga's concept of the'magic circle' has been central to game studies theorisations of the gaming experience [25]. The magic circle conceptualises the gaming experience as bounded, with a division between the rule-dominated game space and the wider, non-game space. Management of this boundary acts to ensure the clarity of the diegesis and so becomes a key part of the production of such narrative forms. Jesper Juul [12] argues, "It is a hallmark of a coherent world game that the bounds of the game space are reasonably motivated by the fictional world". Juul equally argues for the relative solidity of this boundary: "Soccer is played within a designated playing field; a board game only takes place on the board". This solid boundary is even clearer in a digital game that "only takes place on the screen and using the input devices (mouse, keyboard, controllers), rather than in the rest of the world; hence there is no 'ball' that can be out of bounds". Elsewhere in mainstream narrative forms, some kind of spatial, temporal, technological or social boundary similarly tends to exist between the narrative world and its audience delimiting the start and end of that audience's engagement -the cinema or television screen, book cover, the edge of the stage (even if that edge is indistinct or moveable), the curtain rising, the performers' bows or closing credits. However, the boundaries of pervasive ARGs are by definition unclear, malleable or even absent. They are designed to bleed into and across the activities of daily life, appearing as if they were part of the players' everyday lives. As we shall go on to see, TMP offers examples of how this occurred. Their very nature as pervasive comes into conflict with otherwise established needs to provide a frame or circle around a narrative or gaming experience. The increasing mobility of gaming technology similarly blurs any clear distinction between game and not-game [1][19] [24]. Montola [20] describes pervasive games as breaking, or blurring the conventional magic circle, either through seemingly unbounded movement in space, unconventional or unending timelines, or most controversially socially expansive games in which the players' and characters' identities are not clearly revealed or believable. However, the framing of agency, or the balance of power, arguably can be thought of as being established and designed in the sense that the 'creators' are ultimately and completely responsible for the game structures in which players are operating. While Boal suggests that giving players agency within a game may give them political agency outside of the magic circle, and McGonigal describes players' unspoken awareness of their defined agency within the magic circle, we argue that Montola's metaphor of breaking and blurring the magic circle is also applicable to the framing of player agency when contrasted with the agency of an ARG's producers in collectively shaping the boundaries of an unfolding game. It is then necessary to ask who is responsible for this framing in an ARG, or, with respect to game design, who is responsible for ensuring that the "actions and outcomes in a game are both discernable and integrated into the larger context" [25] -and the larger question of how this responsibility is negotiated and ultimately framed itself, if indeed it is. This article will continue this discussion by considering the production strategies employed by Urban Angel in TMP, and how these strategies led to consequences, but most notably examining this by framing the relationship between players and production. By allowing movements through different spaces between the virtual and the physical, the public and the private, the individual and the collective, ARGs offer the chance to interrogate how the boundaries of narrative experience are formed. In particular, the fluid and hybrid nature of these boundaries can lead to potential problems for the production team in that they can lead to confusion or disengagement with the narrative. We will explore how this potential danger was managed by Urban Angel; and the role the players, and an unspoken negotiation of agency, played in this management. --- Playing Experience of The Malthusian Paradox TMP began with a lecture, which was given in a physical venue in four cities over the course of four days, and also broadcast live online. The lecture purported to be an opportunity to see Dr Solomon Baxter, a world-renowned environmental scientist, speak on a new discovery. The lecture was advertised via posters and fliers that also indicated that it was the opening of a new Alternate Reality Game, linking to themalthusianparadox.com, although the site initially revealed little additional information. The lecture itself was advertised in an exaggerated manner that hinted towards conspiracy theories and whistle blowing, with the tagline find out what THEY don't want you to know. The lectures were attended by between 3 to 150 players across the four venues. Each lecture began with Dr Baxter, played by an actor, speaking for around twenty minutes on the evils of biotechnology, including pseudo-scientific jargon, in particular regarding a company known as TFT who are developing genetically modified crops, pesticides and herbicides. Three men wearing suits and brandishing guns suddenly interrupted the lecture, threatening the audience to remain in their seats, while they knocked Dr Baxter to the ground before bundling him out of the theatre. Two more game characters reacted to this disturbance in the auditorium -Rachel, Dr Baxter's daughter, and Alex, the founder of a resistance group known as AMBER, which is campaigning against the activities of TFT. While Rachel was distraught, Alex addressed the audience, imploring them to join AMBER and to help reveal the truth of Dr Baxter's discovery, ensure his safe return and to destroy TFT. Consequently, the audience were given the opportunity to officially begin the game by 'enlisting' in the AMBER organisation. Each player provided details such as a code name, email address and phone number, and in return was given a number of postcards that foreshadowed and contained information required to solve a number of puzzles that would be included later in the game. Players could subsequently join via AMBER's website, a separate website from the official game site, in keeping with the notion that AMBER is a real organisation. In total approximately 300 players joined AMBER over the course of the game. Following the lectures, the game proceeded with a largely regular structure. Approximately every fortnight, a new episode of the narrative would be released to players, with each episode consisting of a short film of between 10 -20 minutes documenting the continuing struggles of Rachel and Alex in trying to find Dr Baxter, and the on-going activities of the sinister TFT, but also containing hidden clues to forthcoming puzzles. Each episode was accompanied by a separate call to action from AMBER, requesting that players engaged in some form of activity, either online or by visiting one of the participating venues at a particular time. There were 6 filmed episodes in total, along with a number of long and short running and concurrent tasks to be performed. The first episode following the lectures took place during a large, weeklong games festival, GameCity, based in Nottingham city centre. The large throughput of visitors to the event provided an opportunity for more players to sign up for the game, and existing players were invited to participate in a number of small activities during the festival. Alex and Rachel were present in character for players to interact with face-to-face and to receive more of the back-story of the game. A live graffiti demonstration revealed a second, visual clue towards the relevance of the previously received postcards, while players collectively hunted for Scrabble tiles and placed them on a board to uncover a further clue in the form of a message. In order to imbue the players with a sense of agency, belonging and ownership of the AMBER organisation they are incited into forming a very public protest march through the city, demonstrating against the activities of TFT. Similarly in a later episode players, or AMBER agents as they are referred to, were requested to create a public blog documenting their own activities, and further demonstrating their membership and association with AMBER. The episodes continued in a similar manner, with episode films and activities giving fragments of information to solve puzzles that led to further understanding of the on-going narrative, or uncovered a new direction or activity to take. These largely took the form of visiting a venue to engage in a faceto-face conversation with Rachel, Alex or one of the additional performers, or collecting a physical artefact, but also involved visiting a number of websites, sending SMS messages or making phone calls. As with previous ARGs, not all players were expected to, or even able to, attend all of the physical engagements or individually solve all of the clues, and as such were expected to work collectively and document and share their activities. Puzzles and clues in TMP were often deliberately abstruse, and required trial and error in order to understand them, or to know what subsequent action should be taken. Further clues were hidden in public blogs owned by the characters, or on their profiles on social media sites. The on-going activities of AMBER, both the characters and the players, were retrospectively documented via the AMBER website, in order for less engaged or active players to follow the narrative. Often material received physically had to be used to solve puzzles digitally. For example, during one episode players who attended a meeting with a performer were given a small box containing several shredded documents (inspired by the DARPA shredding reconstruction challenge). Reassembling the shreddings revealed several partially redacted letters between TFT executives; the contents that had been redacted were revealed in a later activity. An additional document was an image containing a phone number, and an image of the map from the book Treasure Island. Players needed to realise that the seemingly encoded numbers on one of the postcards indicated the phrase 'dead men don't bite' from the book, and if this phrase was sent via SMS to the phone number they received a further clue via an automated phone call. Other episodes attempted to engage the players with the'reality' of the episodes, with Rachel and Alex observed obtaining some information in one of the films, which was physically given to players during face-to-face performances to subsequently use in an online task -for example receiving a DVD containing a video, a password with which to hack the TFT website and release sensitive documents, or a PIN to access the voicemail of TFT operatives, again via an automated phone system. Conversely, some activities required players who were available to make use of information that the community of players had uncovered. One of the final payoffs for the players was to unlock one of four locked safes, one in each city, purporting to contain information regarding Dr Baxter's discovery. Again, previous clues provided ambiguous suggestions towards the code that would open the safe, ultimately requiring some degree of brute forcing a large number of potential codes. Once open, the safe revealed a mobile phone that automatically played out a video message from Dr Baxter, but also recorded the reaction of the player opening the safe via its camera. As the game began to reach its conclusion it took on a darker, more sinister tone as the narrative was brought to a resolution. Dr Baxter is ultimately killed, and Rachel revealed in a final twist as being a double agent. Players were more deeply integrated with the theatrical nature of the performance and challenged on the nature of their involvement with AMBER. On visiting a venue in order to meet one of the characters, players were individually 'kidnapped' by TFT operatives, bundled into a waiting car before being driven to a sparsely furnished office, where they were interrogated as to what they knew regarding the story so far. Players subsequently boasted online about how little they gave away, essentially attempting to demonstrate their skill as players of the game. The following sections present a study of TMP that covers the three-month period during which the game was played. We focus on how Dominic and Adam of Urban Angel, the producers of TMP, addressed the challenges of coming to terms with and understanding the hybrid, pervasive space that the game created. We also focus on the consequences of this space for the agency of Urban Angel as producers of the game, and the tensions of giving players space to perform and play, while simultaneously maintaining control of the unfolding game. Our study draws on multiple sources of data, from questionnaires and interviews with both the producers and players over the course of the game and afterwards, which give insights into their experience, to notes and observations that reveal the activities of the same, in order to provide a holistic account of TMP from multiple perspectives. --- Controlling Hybrid Space As already discussed, established media are framed by particular social, temporal, spatial or technological conventions of production and reception, from which expectations about the boundaries of a performance event and the role that audience members will play are derived. However, an ARG composed of various types of media is an unrestricted experience that can encompass different contexts and modes of engagement. TMP offered players a narrative experience that explored and experimented with multiple dimensions of transmedia storytelling: information was presented, but also had to be sought out; a mixture of individual and group interactions with the game took place in public and in private and blended fact with fiction; the substance of interactions was both physical and digital, both live and recorded; content was sometimes authored and sometimes improvised; the production created original material and appropriated existing resources; parts of the player experience were personalised and parts were generic. Given the inherent breadth of the game, it is important to consider how the producers of TMP attempted to manage an unbounded process of performative exchange over an extended period of time. The lecture(s) that opened the game encapsulated the creative intention of exploding the boundary around performance space, through a violent rupture of that space that released the story into'reality' and invited players to pursue it. The initial promotional materials for TMP presented it as a unique ARG with broad parameters, which would involve clues, interactions with characters, live events and a global audience. In some ways the scope of this original proposition reflected the narrative setup of multinational biotechnology interests and a central character (Rachel) returning from travels to reconnect with her father (Dr Baxter), but it also suggests the extent of the opportunities for engagement envisaged by the project, which was designed to unfold in an online environment as well as in the host venues in the East Midlands. The performative nature of this opening interaction was to some extent made clear, and a film-style trailer available on the game's website established this as a story that would conform to patterns familiar from mainstream media. At the same time, however, there was a strong intimation that the generic elements of mystery and suspense being introduced might develop in ways that would challenge normal conventions and invite players to participate in ways often unsanctioned in other performance spaces, as the publicity information stated: If you attend be prepared for the unexpected -filming and photography is encouraged. This suggested that the game would transgress another established boundary: that erected around copyrighted creative content, which gives producers a significant measure of ownership and authority over consumption of their work. There was a distinct tension between fact and fiction in relation to the performance space in which TMP began. The flyer (Figure 1) produced by Urban Angel was double-sided: one side promoted Whose Holy Grail? A talk by the world renowned Dr Solomon Baxter; while the other side introduced the ARG as detailed above. This meant that, while the intention was for the audience to be cognisant of the nature of the experience as a performance of an alternative reality, people who picked up or were handed a flyer might have believed that they were being invited to attend a lecture by an expert environmental campaigner, if they did not read the reverse or made no conceptual connection between the two. The venues were plausible environments for a public presentation of this sort and Dr Baxter was described as a respected and prolific authority with more than 20 years experience writing features and news stories on everything from climate change and wind farms, to oil spills and the decline of honeybees. The content and delivery of the lecture was such that its status as a performance of science was uncertain, which contributed to a sense of ambiguity about its 'true' purpose prior to the sudden invasion of the auditorium by the armed gunmen. Once this breach had occurred the performance space became porous and fluid with scripted character interactions, player registration and vox pops going on in the public areas of the venue. This was therefore a point at which audience members were asked to take on the role of player, embracing the fiction as something that would intersect with the reality of their lives and would take place outside the demarcated boundaries of traditional performance space. For audience members perhaps unfamiliar or uncomfortable with being given this type of agency over a narrative experience, this moment of transition could potentially present an obstacle to engagement, something that we will discuss further. In subsequent elements of TMP the physical spaces of performance became even more difficult to distinguish from the surrounding context. The protest that was organised as a central activity at GameCity required players (and performers) to physically assemble in the centre of Nottingham, surrounded by shoppers and people going about normal weekday business, and to deliberately draw attention to themselves in a performative fashion, according to the conventions of public protest, by marching, chanting and carrying placards. The choice of this particular type of live group activity on the part of the game's producers went beyond performing an alternate reality in public space because the nature of a protest is to increase the visibility of a cause and garner support for that cause. In this sense the protest served as a means to promote TMP and reach a wider audience of potential players. At the same time it also endowed the fiction of AMBER with apparent veracity by giving the impression that this was an organisation to which a group of people were committed on an ideological level and for which they would openly declare their conviction. As such it made an intervention into the wider world outside the alternate reality by demonstrating (in both sense of the word) the potential for public protest and the collective agency of UK citizens. As a consequence, the protest had the potential to have ramifications beyond the boundaries of the narrative and the community of players. Within the game, the creation of anti-TFT and pro-AMBER placards, which was supported by online 'propaganda', increased the game's central dramatic tension by requiring hands-on investment on the part of players. And of course enacting a live group performance in public space entailed very real health and safety considerations, about which the project manager needed to brief participants beforehand, an incursion of usually concealed backstage concerns into audience experience, which framed that experience in terms of lived reality rather than performative illusion. The expansive physical space that TMP inhabited was a source of creativity and improvisation, but also presented challenges to the game's producers. During the GameCity event'special assignments' were rapidly devised and promoted on a daily basis, alongside the pre-planned activities throughout the week, in order to maintain the interest of those playing as part of the festival and to draw in the new recruits signing up each day. This fluid and responsive mode made it difficult to gauge the nature of individual player's involvement in TMP. The method devised to counter this was to issue each registered player with a unique QR code that could be scanned by performers at the start of interactions, ostensibly to verify the AMBER agent's 'identity' and affiliation, thereby providing a record of that interaction. A code was assigned at registration and if players then brought this in person to the AMBER recruitment desk in the GameCity tent they could exchange it for a dog tag that featured the AMBER logo on one side and QR code on the reverse. As well as serving a practical purpose for overseeing engagement with TMP, the idea was that the dog tags would provide players with a tangible and personal connection to the game world, a token that would be imbued with symbolic value. In this sense they seem to have achieved their purpose. However, as a means of monitoring and mapping live player interactions with TMP the tags proved to have limited effectiveness because the code label was quickly worn away through handling and rendered illegible. This technological frame (of handing over your dog tag to begin an interaction) ultimately failed to materialise and had consequences for the negotiation of agency, which we shall return to in the below section. The producers of TMP experimented with providing different opportunities for player interaction with the game in public performance spaces. At regular intervals during the game, assignments were arranged in which players were invited to meet characters face-to-face in a public place (usually the host venues). These encounters allowed players to embody their AMBER agent personas, but they also provided a means for the game's producers to extract information about players' responses to the narrative and experience of TMP, to find out what they knew or suspected at that point in the process. Gleaning these insights required the performers to improvise carefully according to a predetermined brief in order to discover more than divulge whilst maintaining character consistency and sensitivity to each player's particular understanding of the game. As such, these were difficult moments to control, an issue that we will return to presently. A more indirect form of interaction with TMP in public settings was offered by AMBER nodes, which were installed at the four host venues, and constructed a virtual frame for the circulation of digital game-related content. Utilising an existing technology, the PirateBox, these broadcast a network in the immediate vicinity to facilitate file sharing by anyone with a Wi-Fi device. The origins of the underlying PirateBox technology as an illicit means of circumventing copyright restrictions, with connotations of covert activity and anti-establishment stance, fit with the story world being created. The chat function of the nodes could be used for initial, surreptitious contact between characters and agents without the latter having to immediately approach the former in person. The nodes were designed to function as a channel for free and open communication that players could utilise and appropriate in whatever ways facilitated their engagement with the story world and playing of the game. In this way the producers ceded some control over content generation and creative agency, in return for insights into their players. As it transpired, however, the nodes were not extensively used by players, seemingly because of the effort required to be in a particular location (which was "just too busy... rather difficult to access") and go through the files looking for relevant information ("it's quite confusing to find the right thing"). The nodes offered players one chance to engage in digital exchange with each other and the producers of TMP, in a way that neatly imitated a 'dead drop' in the physical world, but there were also a host of online spaces associated with the game that did not have the spatio-temporal restrictions on digital interaction that were attendant on the nodes. Much of the narrative material made available to players of TMP online was quite conventional in terms of its mode of address: it was one-to-many authored content expressed and delivered by means of digital formats. Some

The Malthusian Paradox is a transmedia alternate reality game (ARG) created by artists Dominic Shaw and Adam Sporne played by 300 participants over three months. We explore the design of the game, which cast players as agents of a radical organisation attempting to uncover the truth behind a kidnapping and a sinister biotech corporation, and highlight how it redefined performative frames by blurring conventional performer and spectator roles in sometimes discomforting ways. Players participated in the game via a broad spectrum of interaction channels, including performative group spectacles and 1-to-1 engagements with game characters in public settings, making use of lowand high-tech physical and online artefacts including bespoke and third party websites. Players and game characters communicated via telephony and social media in both a designed and an ad-hoc manner. We reflect on the production and orchestration of the game, including the dynamic nature of the strong episodic narrative driven by professionally produced short films that attempted to respond to the actions of players; and the difficulty of designing for engagement across hybrid and temporally expansive performance space. We suggest that an ARG whose boundaries are necessarily unclear affords rich and emergent, but potentially unsanctioned and uncontrolled, opportunities for interactive performance, which raises significant challenges for design.

des were designed to function as a channel for free and open communication that players could utilise and appropriate in whatever ways facilitated their engagement with the story world and playing of the game. In this way the producers ceded some control over content generation and creative agency, in return for insights into their players. As it transpired, however, the nodes were not extensively used by players, seemingly because of the effort required to be in a particular location (which was "just too busy... rather difficult to access") and go through the files looking for relevant information ("it's quite confusing to find the right thing"). The nodes offered players one chance to engage in digital exchange with each other and the producers of TMP, in a way that neatly imitated a 'dead drop' in the physical world, but there were also a host of online spaces associated with the game that did not have the spatio-temporal restrictions on digital interaction that were attendant on the nodes. Much of the narrative material made available to players of TMP online was quite conventional in terms of its mode of address: it was one-to-many authored content expressed and delivered by means of digital formats. Some of the material made available online was bespoke, created specifically for the game, such as the six video episodes and websites for the in-game organisations like AMBER and TFT. In other cases, the game's producers appropriated third party online platforms and services to serve their storytelling purposes. That the TMP characters had a presence in the digital public sphere via LinkedIn, Twitter, Flickr and Blogger, meant that the boundary between the alternate reality of the game and the wider world was indistinct in digital interactions, just as it was in physical interactions. Of course TMP players also drew on available online resources to facilitate and support their own interactions with the game independent of the digital framework that had been constructed by the production team. In the early stages of the game both a TMP wiki and a Facebook group were established and populated by players. These were digital spaces of paratextual [7] [9] activity that needed to be folded into processes of narrative development and interaction design; this necessitated ceding and distributing creative control in ways which were, as we shall see, demanding to negotiate. One way in which the game's producers sought to control the distribution of information among players in digital space was to absorb this process within the 'official' structure of the experience via the assignment that asked agents to create a blog under their AMBER codename. The producers of TMP also deliberately created new digital resources themselves in order to expand opportunities for interaction for the community of players, which was the function that the wiki and Facebook group served in a supplementary capacity by making publically visible personal in-game experiences through digital documentation and discussion. The assignment involving recovery of shredded documents included the rapid, parallel development of an online, interactive document assembler, which was made available to the all players online a few hours after the final opportunity to obtain the material in person had passed. In this way the digital aspects of the game served to broaden its potential to engage. On other occasions technology was used to enable dynamic digital interactions that were planned but also had flexibility inherent in their execution, for example the phone-based system to which players had to text the phrase 'dead men don't bite'. In attempting to unlock the secrets of the system, players tried texting a variety of different things. The system was designed to automatically produce hostile responses if prompted in the 'wrong' way, but Adam realised that the players' experimentation presented an opportunity for additional interaction and began to manually input messages tailored to the players' tactics. So, what began as a chance to utilise another mode of narrative communication (telephony), became an opportunity for dialogic improvisation on the part of both producers and players. The finale of TMP combined physical and digital elements for maximum impact, in a manner that proved to be just as adaptable as the 'automated' text messages. The safe installed in each of the venues would simultaneously play the video message from Dr Baxter and record anyone watching it. These reactions were collected and posted on the game's YouTube channel. While the intention was to surprise players and assimilate their individual experiences (in the form of unwitting performance) into the public record of TMP, these intentions were subverted in some instances. Players recorded the in-safe recording(s) on their own phones, physically concealed their identity, and even in one case presented a written response to the camera: 'STFU TFT! AMBER FTW!'. This particular interaction epitomises the complex performance dimensionality of the TMP experience, as a recorded fragment of narrative, delivered in a public space, elicits a situated response from a player, which is aimed at the producers, and is then disseminated by the producers online to the player community. --- The Consequences of Hybrid Spaces for Production Design Despite employing the strategies discussed above in an attempt to control players' pathways through TMP's hybrid space, and so the frame around the experience, Dominic and Adam of Urban Angel created a number of unintended consequences both for their role as experience designers and for the players. Throughout the game, there were a number of moments in which the game became a site of struggle for agency between the producers and its players. The producers were forced to negotiate over both logistical agency (how the game was run) and narrative agency (how the game's narrative would progress). Examining how this negotiation played out offers the opportunity to explore the consequences of constructing fictional worlds across hybrid space, where the player is positioned in different roles and in a different relationship to the text and its production. In particular it becomes possible to identify how agency itself functions as a frame around mediated experiences by examining how that frame of agency -who is the creator of the experience and who is not -broke down during TMP. The key consequence for the production team in terms of the game's live interactions was as a result of the blurring of boundaries between the fictional world and the wider world. Although digital spaces allowed a certain amount of such blurring, with narrative websites such as those for TFT and AMBER sitting alongside legitimate websites, and little indicating their fictitious nature, such hybridity came to the fore in the interactions that took place in public spaces. Despite the above described attempts to manage players' shifts between the hybrid spaces of the games, the fact that these live interactions were layered over the public spaces of daily life led to a number of complications and, ultimately, significant changes to the game's design. The opening event of the game, the lecture by Doctor Baxter, offers an initial, dramatic example of the fluid boundaries of the game's hybrid spaces and raised a number of issues for the production team concerning the way it was framed. A number of audience members attended with the belief that it was a genuine lecture about genetically modified crops. At one of the lecture performances, one member of the audience had an explicit, performative response to Dr Baxter's kidnapping, rushing out after the kidnapping scene and attempting to intervene before production personnel could calm him down. Other audience members had a more sedate, but still negative response. As the following discussion demonstrates, this led to frustration within the production team who felt that they had made sufficient efforts in the promotional material they designed, such as flyers and posters, to establish the lecture as fiction: DOMINIC: At Nottingham Contemporary people had really gone there thinking it was a real lecture about GM crops, rather than actually reading the flyer and seeing it said, right at the top, Augmented Reality or Alternative Reality game... ADAM: Oh they were so pissed off. They stormed out of the bloody room... Owen, who was doing the lecture as Dr Baxter, said [the one in Sleaford] was the hardest one he did, because he came up, he started, and he said there were only, like two people in the audience, really. He said, the bloke turned to his wife and said 'that doesn't look like a professor'. And then basically all the way through they were just going 'that's not right', 'that's not right'. The response of these audience members highlights the consequences of creating a game that exploits hybrid spaces and the fluidity such spaces can give to the boundaries between fiction and reality. Whilst the status of the lecture as fiction was presented in the paratextual material surrounding it such as leaflets and posters, such material can be missed [5], with potentially significant and dangerous consequences. The ambiguous framing of the game led to similar ambiguity in the agency afforded to those involved. By perceiving the lecture as genuine, the above affected participants acted in accordance with the agency such an event would give them -to try and stop an apparent kidnapping or to leave -even if this was not the actual agency offered by the producers. How welcome that agency was to those participants in the short or longer term is questionable. --- Expanding the Scope of Performance These responses to the lecture were particularly dramatic examples of the consequences of manipulating space into a hybrid alternate reality. However, other examples of such moments actually led to more complex responses from the production team and had a greater impact on the development of the game. The majority of these related to player responses to individual members of the production team, some of whom were clearly established as characters within the narrative and some who weren't. It is perhaps unsurprising that it was through character that the negotiation of agency appeared so explicitly. In traditional narrative forms, characters have long been considered they key point of audience engagement [5][14] [22]. Within an ARG such as TMP, characters serve an even more fundamental role, being the player's key point of stability across the hybrid spaces of the game. Alex and Rachel appear in the live interactions, the episodes and in digital online spaces and so offer important coherence and consistency as the player moves through the game's hybrid space. However, the greater agency offered to, and taken by, the players led to a manipulation of the characters, only some of which was within the power of the production team. The fluidity with which the game moved through public spaces and the ambiguity of its boundaries led to players ultimately expanding the game's story world to include additional members of both the production and research teams. A number of the volunteers who served a logistical role during the GameCity events, were integrated into the narrative by the players. One in particular, Loz, ultimately became a key figure in the latter stages of the narrative. Adam and Dominic described the process: ADAM: The audience, sort of, didn't like Loz. It was really funny because Loz came to volunteer at first...he is a performer but he didn't come as a performer...He wasn't a character, he didn't exist in the game at all. DOMINIC: Because Loz is really only meant to be there for GameCity and it's just that the players latched onto his character so much and...it would just be mad not to use him. Despite merely being present in game-related spaces in order to facilitate the smooth running of the live interactions, the fact that Loz was present at all meant that he was available to the players as a narrative figure. A similar situation occurred when one of the research team was transformed from objective observer during the protest march in an enemy agent in the player-created wiki's report on the event posted on the wiki. The players used their personally controlled digital spaces (blogs, wiki and the Facebook page) to shape the narrative of the live events. At several points in the narrative, the players even expanded the boundaries of the narrative world to include members of the general public. --- Figure 5: A member of the public (identity obscured by the authors) is incorporated into the narrative by a player The picture in Figure 5 accompanied the wiki report on the protest march written by a player: "Upon arrival at the Arts Centre, I circled the building and discovered two suspicious characters, one signalling our arrival to the other before disappearing under the carriageway. The signalled person was holding a distinctive orange and black backpack". As the producers described "they [the players] created a load of characters that didn't exist and there's, like, photos of that guy with a backpack outside of Contemporary". Figures that would otherwise have only served a logistical purpose within the game's design were suddenly transformed, given back-stories and integrated into the diegesis. The creation of additional online spaces shifted the frame of agency away from the producers as the players began to create the game narrative as well. This hybridity had an impact on the daily lives of the actors appearing in TMP. The performer playing Rachel found that her role within the game bled into her daily, non-acting work life, with players discovering her outside of the game's official narrative and engaging with her in character. She described how: RACHEL: There was one day I was at work and a player came in and was 'oh my god, it's Dr Solomon Baxter's daughter!' And I was like 'Yes, don't tell anyone where I am so TFT doesn't find out. So tell me about the clues you've been figuring out, you've not worked that one out yet? That's a shame isn't it?'... it was just very, sort of, wow! This accidental meeting with one of the players, in a context that was outside the frame of the game for the performer but not for the player, demanded that her performance expand beyond the pre-conceived boundaries of the game (and her acting contract). She was required to immediately shift from her real self to her performance self in order to maintain the integrity of the game narrative. However, whilst on this occasion the performer was able to successfully navigate the blurred boundaries of the game and the wider world, elsewhere this became more problematic and led to the game's most significant narrative change, when the character of Rachel was transformed into a double agent working for a previously unseen third organisation. Rachel's transformation from distraught daughter to malevolent double agent began with a live interaction halfway through the game in which players met with the character face-to-face. Whilst the aim of the interaction from the production's point of view was to give players the next puzzle, in order for the performer playing Rachel to properly interact with them she needed to determine what parts of the game they had taken part in, and therefore which narrative points they already knew. The producers turned to the performers as a way to manage this: ADAM: It was my brief to [the performer playing] Rachel and she followed my brief exactly, and my brief was to try and get as much information out of them as possible so we knew what was going on, but that sort of backfired. The unintended consequence of this was that players grew suspicious of the number of questions Rachel was asking and, in line with the generic conventions of spy and conspiracy dramas that rely on deception [4], she was identified as a double agent: PLAYER: One thing I noticed with Rachel was at the start she seemed to know a lot about what was going off and then when she said 'well what do you remember' she couldn't remember anything, you know? And it was from that [that we began] to say maybe we can't trust her. Rather than resisting this transformation of one of their central characters, the producers decided to change her character arc and re-film the final episode: ADAM: Initially Rachel was good and was the real Rachel but with everything going on it just seemed more fun and [the audience] had identified it and we thought, yes. And it gave us more stuff to play with in that respect. In this case, agency over the interpretation of in-game characters shifted from the producers to the players, who re-interpreted Rachel's character against the way she was initially established. The established frame of creator/player partially broke down to the extent of significantly changing the game's narrative. Although this was ultimately the decision of the producers, it was motivated by the players asserting themselves in a position of greater control over their interpretation of the narrative. --- Communicative Agency The collective interpretation of Loz as an important character or Rachel as a double agent emerged from the second form of agency that was negotiated through TMP. This form can be termed communicative agency, or control over the conversations around the game, rather than the actual development (or transformation) of the core narrative. The production had established a framework of digital spaces in which the players could discuss the game's puzzles and events. These spaces functioned to facilitate the 'collective intelligence' of the game, allowing players to share solutions to puzzles or offer descriptions of events to those who had missed them. They also established Urban Angel as the key creative agent shaping the game experience. The blogs were 'authorised' by them and promoted via the AMBER website, Twitter and email list, and so maintained a conventional frame of agency with Urban Angel as the producer/creator and the players as operating within the parameters they had set. Not long after the game's launch, however, the players began to seek control over these digital spaces of collaboration. One player created a wiki where players could upload crucial information or reports on the game's activities, whilst another created a Facebook page. Initially these were both set up to allow alternative spaces in which to engage with the game, with both players later reflecting on the websites as spaces to'support' or simply 'talk to each other'. At the same time, both the wiki and Facebook page allowed the players to discuss the game within personally created spaces that were in the control of the players, rather than the producers. When it became clear that the production equally had access to these spaces, players took a number of steps to prevent such access. One player created a cipher that would allow them to encode their messages to each other and the Facebook page was immediately made private. They explained the reasoning for doing this as an attempt to ensure that the producers did not disrupt the game by making it too difficult: PLAYER: When all these ideas came about that they, you know, were using the Facebook group to deliberately make things more difficult, or they were going to plot things, it was kind of like we thought, okay, this might actually be a problem. We might be making it worse for ourselves communicating so openly. Facebook became a site of negotiation over the game's paratextual conversations, with the players denying too much control for the producers by limiting their access to their gameplay processes. Again, the established frame of who is creator and who is not, who has power over the experience and who does not, broke down as the players asserted greater agency over their play. For the producers, this lack of access was initially far more problematic than the reinterpretation of characters discussed in the above section. Being cut off from the discussions around the game made the producers feel they were missing large sections of the game's narrative development and, potentially, its quality: ADAM: There's a lot of stuff we're missing, there's a lot of private conversations going on on Facebook chat and stuff that we just can't access. Like, when they doing all the text messaging they were all talking on Facebook, chatting, deciding what to text in. But we couldn't see that. And if I'd been there to see that I could have done the responses better. Elsewhere, the producers displayed frustration at their inability to truly shape the experience and how the players created their own paths through the game. When referring to player responses to two events -one the accidental appearance of random letters on an episode description and the other material the production team made available on a DVD -Adam demonstrated some of this frustration: ADAM: They've gone and found all of this weird stuff that I haven't done on purpose, some accidental letters that have gone on the YouTube video description, that's taken on massive significance, but they don't look on the bloody disk we give them. That's weird to me. The change to Rachel's character, or the expansion of Loz's, remained within the control of the production team (however inspired by the players they may have been). They maintained a more knowledgeable and authoritative position. However, losing access to the conversations around the game, and any ability to steer them, limited that knowledge. They had no way of knowing how well the individual puzzles were being received, whether they were being solved too easily or risked alienating players. It was this loss of agency that was the cause of greater anxiety to the producers than players reinterpreting the characters or the core narrative. Ultimately, however, both sides reached a necessary truce by accepting a limited loss of agency on both sides. The players acknowledged that they could never really shut the production out: "Once we made the group private we just accepted that, you know, if the production were going to do anything based on our Facebook group, they were going to do it". The producers accepted that they could lose a certain amount of knowledge about the gameplay whilst still creating a successfully responsive narrative: "Now they're generally talking between themselves on Facebook and not telling us anything, really...it's their thing now, rather than our thing". Adam went on to say at a later interview that "I'd always hoped that we'd get a wiki. Then I was like, oh no we're not in control of that. Then it was like, we don't need to be. That's the point". Within the design of the game, these player-produced digital spaces were integrated into the official narrative. Links to Facebook and the wiki were added to the homepage and the production team began using the cipher for official messages (one of the players independently programmed an online tool to automate the cipher, and a link to this was shared via the Facebook group), legitimising the players' control over their paratextual conversations. Any clear delineation between who was within the frame of creative agency and who was not was broken down. Despite the potential for the kinds of moments discussed above to disrupt the game's narrative, the integrity of the diegesis was maintained via two key factors. The first was that rather than resist the players' attempts to maintain some level of control over their experience, the production team was happy to sacrifice a certain amount of creative control. They ultimately found value in "A sort of symbiosis, and we're feeding off them, and they're feeding off us. Their ideas are changing our ideas, and vice versa... It's exciting, it's tiring though". The second factor was a clear desire on the part of the audience to maintain a narrative bubble around the game, the suspension of disbelief identified within previous work on ARGs [16][21]. The fluidity of the boundary between real and fictional space ultimately acted to the advantage of the production team, allowing that fluidity to explain any potentially disruptive clashes between the real world of production process and the narrative world of the characters. The very reason for Rachel evolving into a double agent in the eyes of the audience, as discussed above, was precisely the audience's desire for her behaviour to be justified within the narrative itself, rather than as part of the production process. Similarly, one player described how he would see several of the cast around the city, but was able to justify their presence in the city within the terms of the narrative: "the only times I bumped into characters was when they were supposed to be in the city. For example, Rachel and Alex obviously left town at various points -I think they went to the Yorkshire Coast or somewhere -and during that time I didn't bump into them". Rather than explain such moments through the practicalities of the actor's having non-TMP-related daily lives, he chose to explain their presence within the boundaries of the narrative. Similarly, moments that were the result of the producers necessarily intervening in the running of the game were ultimately explained by the players equating the production team with TFT and AMBER. During the kidnappings the production team used information from players' Facebook profiles in order to create a more personalised experience. One player had set up a fake Facebook account for the purposes of the game and so was disturbed when the kidnapping team used information from his real account. However, his response was to explain such an occurrence through the narrative world of the game, rather than break the diegesis: "it made me think immediately, well, there must be some kind of leak in AMBER". They similarly equated the production team with the narrative organisations: "I think the problem was the fact that [the producers] knew about it, meant that TFT knew about it". Another described making a conscious effort to ignore the production structure behind the game: "I sort of deliberately didn't want to see that part of it". In line with McGonical's theory of the 'Pinnochio effect' [16], players expressed a desire to ensure that even when the fluid boundaries were tested, they would help the production stretch the narrative world over those boundaries. There was only one occasion in which this diegetic bubble was broken. A player was sent a text message purporting to be from AMBER that there was a charge to receive. Concerned, the player contacted AMBER to determine if it was legitimate. Although they sought to maintain the narrative illusion of AMBER, the production team immediately responded outside of that narrative frame to confirm the text's illegitimacy and reassure the player of the game's aims and processes. Whilst other occasions allowed for in-narrative responses, the potential threat of expensive unsolicited text messages forced them to break the bubble and respond as a production team rather than as in-game characters. Throughout the majority of the game, however, players were keen to maintain the appearance of an agency hierarchy. In focus groups, players would deny that they had impact on the game narrative. One described how "the players could, you know, solve the puzzles and find out things but they couldn't really feed that information back into the game" and later said that "it didn't ever feel like the plot was changing because of what we knew". Even though the actual location of agency became more and more complex throughout the game, the players maintained a frame around their experience in which agency was primarily located with the producers, rather than themselves. --- Conclusions: Agency as Frame Our study of a live-action and performance focused ARG, The Malthusian Paradox, has revealed how professional artists worked to create an experience that aimed to be compelling and engaging, mixing live-action and face-to-face interactions with performers with a disparate collection of online and digital resources. We have seen how they developed a distinctive approach of constant assessment and reflection that was required to react to the fluid boundaries of the hybrid game space, which were constantly shifted due to the sometimes misguided rationalisation of the perceived narrative by players, or elements of the production process. We have also seen how the narrative and development of the game became an iterative process of game mastery rather than authorship, a requirement of the necessary collaborative agency afforded to players of an ARG, with control shifting from game producers to players and back again constantly. ARGs offer challenges to the simple social, spatio-temporal or technological framing of narrative or mediated experiences by deliberately blurring where and when the games exist. This case study, however, reveals the need to also consider agency as a framing device. TMP began with a clearly established 'frame of agency': Dominic and Adam as Urban Angel were responsible for the creation of the narration, the devising of the puzzles and the management of information; the players were responsible for solving the puzzles and facilitating the forward progression of the narrative. The experience was therefore framed by who was in a position of greater creative power (Urban Angel) and who was not (the players). As the game progressed this frame began to breakdown. At some points the players took greater agency over their own experiences, expanding the boundaries of the game by creating online spaces that they perceived to be outside the producers' control, if not their awareness. At other points the producers explicitly handed agency over to the players, allowing them to create new characters and ultimately choose how the game would end. However, throughout the game, players worked to maintain the integrity of the producers' agency by denying their own. For the players, their own agency, and their understanding of the limits of that agency, functioned as a way to frame their experiences, just as the television screen, proscenium arch or book cover does. While TMP itself is arguably a unique experience in terms of narrative and game content [10], it embodies general principles that can be used by the developers and designers of future ARGs. The space of performance is distinctly hybrid, both in the sense of symbiotic and unclear relationships between players and producers that challenge conventional media and game design, and with performance occurring in public settings, both in ad-hoc online spaces, as well as designed physical interactions. Whilst real-world interactions can be the most engaging or visceral, the significant driving force of an ARG can be said to be the performance space that players create for themselves as a community collectively reaching a consensus regarding the unfolding story, and this can be challenging to monitor. Understanding how the players' understanding of their own agency frames their experiences is essential for ensuring that conflict does not emerge and undermine the integrity of the game narrative. It is through constant attention to detail in terms of both content and on-going events, more than just system maintenance, that the artists have been able deliver a successful experience.

The Malthusian Paradox is a transmedia alternate reality game (ARG) created by artists Dominic Shaw and Adam Sporne played by 300 participants over three months. We explore the design of the game, which cast players as agents of a radical organisation attempting to uncover the truth behind a kidnapping and a sinister biotech corporation, and highlight how it redefined performative frames by blurring conventional performer and spectator roles in sometimes discomforting ways. Players participated in the game via a broad spectrum of interaction channels, including performative group spectacles and 1-to-1 engagements with game characters in public settings, making use of lowand high-tech physical and online artefacts including bespoke and third party websites. Players and game characters communicated via telephony and social media in both a designed and an ad-hoc manner. We reflect on the production and orchestration of the game, including the dynamic nature of the strong episodic narrative driven by professionally produced short films that attempted to respond to the actions of players; and the difficulty of designing for engagement across hybrid and temporally expansive performance space. We suggest that an ARG whose boundaries are necessarily unclear affords rich and emergent, but potentially unsanctioned and uncontrolled, opportunities for interactive performance, which raises significant challenges for design.

Introduction Women represent a growing proportion of all older people, but added survival means increasing disability associated with chronic conditions such as cardiovascular and respiratory disease, cancer and diabetes, and functional limitations that impact on daily living [1][2][3][4]. What is not well understood is the extent to which, apart from age, social and economic factors differently contribute to disability in older men and women. This paper analyses differences in disability prevalence between men and women aged 50 and over, using multicountry data from the World Health Survey (WHS). Disability spans physical, mental and psychosocial functioning, encompassing a complex suite of conditions, activities and relationships [2,5]. This study adopts the most recent definition of disability based on the International Classification of Functioning, Disability and Health (ICF) developed by the World Health Organization (WHO) [6]. Under the ICF "disability" includes limitations in functioning resulting from interactions between the individual's "health condition" (e.g. diseases, injuries, and disorders) and "environmental factors". These decrements in body function and capacity are not necessarily linked to a particular health condition [5]. At a population level, "social determinants" or the conditions in which people are born, live, work and age, influence morbidity and mortality [7]. Differences in health between men and women are attributed to both biological and social factors [8][9][10][11][12]. Social determinants contribute to gender inequalities and inequitiesunfair and avoidable differences in health [11]. In many countries age adjusted disability prevalence rates are higher for women and those in the poorest wealth quintile [3]. A study of gender differences in health at ages 50 and older in 11 European countries, England and the United States showed that, after adjusting for age, women were more likely than men to have disabling, non-lethal conditions including functioning problems and depressive symptoms [13]. Socioeconomic disadvantage, measured by occupation, lower levels of income and education in higher-income countries is associated with greater disability among older women [14][15][16]. Evidence from studies in lower-and middle-income countries also shows association between social factors and disability in older women [10,11,[17][18][19][20]. The aims of this study are to: identify and assess how disability in males and females aged 50 and older is separately associated with a range of social factors; measure and evaluate the effects of sex (i.e. being male or female) on disability in people aged 50 and older after adjusting for the effects of age and the social determinants, and decompose the extent to which these different factors explain male-female differences in disability in older adults. This study fills several gaps in the existing literature with regard to the determinants of disability in older adult populations for several reasons. Firstly the data derive from a large multi-country survey that used the same instrument and methodology to collect health and other information at the individual level across high-, middle-and low-income countries. Secondly, using an internationally agreed definition of disability based on the ICF, an advanced statistical method is applied to create a measure of disability which ensured cross country comparability. Thirdly, the study focusses specifically on the social determinants of health in men and women [7]. Finally, we apply a decomposition method to show how age and social factors contribute to the disability difference between men and women in this large international dataset. --- Methods --- Sample and data collection The WHS was conducted by the WHO in 2002-2004 to provide representative and comparable population data on the health status of adults, aged 18 years and older, in 70 countries from all regions of the world [21] (see http://www.who.int/healthinfo/survey/en/index.html). All country samples were probabilistically selected. In China, Comoros, the Republic of the Congo, Côte d'Ivoire, India, and the Russian Federation, the WHS was carried out in geographically limited regions. To adjust for the population distribution represented by the United Nations Statistical Division (http://unstats.un.org/unsd/default.htm) and also non-response, post-stratification corrections were made to the sampling weights [22,23]. Our study includes 57 countries drawn from all income groups as defined by the World Bank [24]. The remaining 13 countries in the WHS were excluded here because either information on sampling weights was not available (11 countries) or there were insufficient data to create the household wealth index, which is one of our independent variables and a measure of socioeconomic status. The final sample comprises 63638 respondents aged 50 and older. Two pooled datasets of 28568 males and 35070 females are analyzed. (Additional file 1: Table S1 shows each country's final sample by sex.) --- Dependent variable For the purposes of our study "disability" is defined as decrement in functioning beyond a specific threshold. The construct is based on 16 questions in the WHS, grouped into eight health and functioning domains: vision, mobility, self-care, cognition, interpersonal activities, pain and discomfort, sleep and energy, and affect [25,26]. Self-reported response categories to these questions were: no difficulty, mild difficulty, moderate difficulty, severe difficulty, and extreme difficulty. See Additional file 2: Table S2 for list of WHS questions. An Item Response Theory (IRT) partial credit model [27] was used to construct a composite measure of disability for each individual. The measure, which was previously developed for the WHO Report on Disability [3], allows a parsimonious measurement of the disability construct using a constellation of items from the eight domains of health in the WHS. The scale ranged from 0 (=no difficulty) to 100 (=complete difficulty). The average of scores from respondents who reported extreme difficulties or total inability in any of the eight domains and who reported having been diagnosed with a chronic diseasesuch as arthritis, angina, asthma, diabetes, and depressionwas around 40. A cut-point of 40 was used to divide the population into "disabled" vs. "not disabled". Scores at or above this threshold identify those individuals experiencing significant difficulty in their everyday lives, irrespective of the underlying etiology [3]. This difficulty may be due to a range of health conditions such as arthritis, angina, alcohol dependence, depression or low vision. --- Independent variables In addition to sex, the independent variables (all categorical) are: participants' age (using categories 50-54, 55-59, 60-64, 65-69, 70-74, 75-79, and 80+ years); marital status (married/cohabiting, never married, divorced/ separated/widowed); educational level (no education/incomplete primary, complete primary, secondary/high school, college completed or above); employment status (currently in paid employment, not working for pay); area of residence (rural, urban), and country of residence. The selection of these covariates was guided by the findings of the Commission on Social Determinants [7,28]. A dichotomous hierarchical ordered probit model was used to develop an index of household economic status based on owning selected assets. The premise is that wealthier households are more likely to own a given set of assets. Asset-based approaches avoid some of the reporting biases that arise from using self-reported income. The effects of asset ownership and household characteristics on household wealth were simultaneously estimated with the hierarchical error term at the household level. The model produced asset cut points representing the threshold on the wealth scale above which a household is more likely to own a particular asset. This "asset ladder" was then applied to every household in each survey to produce adjusted estimates of household wealth. The asset-based index [29] has previously been used in equity-analyses in low-and middle-income countries [30], in analyses of the WHS, and in other studies [31,32]. The index was divided into wealth quintiles within each country. --- Analysis We first computed the overall prevalence of disability for males and females, and the prevalence of disability by each socio-demographic factor. We then used multiple logistic regression to estimate the adjusted effects of each social determinant on disability by sex. Finally, we used a variant of the so called Blinder-Oaxaca decomposition technique [33] to partition the measured inequality in disability between males and females into two components. This first component is the "explained" part that arises because the two groups, on average, have different values for the known characteristics (i.e. the characteristics that were used as determinants in the underlying logistic regression model). The second component is the "unexplained" part which is attributed to the differential effects that the characteristics have on each group as well as other factors not included in the logistic regression model [34]. Decomposition results were computed employing the "Oaxaca" command for Stata [35] with the "pooled" and the "logit" options specified. The "pooled" option uses the coefficients from a pooled model over both groups (including a group indicator) as the reference coefficients [36]. The "logit" option causes a nonlinear decomposition based on logistic regression to be computed, as proposed by Yun [37]. Sampling weights that take into account the selection probability of the individual were included in the analysis. These weights reflect each country's population, in such a way that if the sample size for two given countries are the same (but the population sizes of the countries are different), more weight is given to the country with a larger population when calculating the pooled estimates. Allowance was made for the non-independence of observations within each survey cluster. All analyses were carried out using Stata 11 (StataCorp, 2009). --- Results Table 1 shows that prevalence of disability among women aged 50 years and older was nearly double that of men (40.1% vs. 23.8%). Table 1 also shows the sexspecific prevalence of disability by different levels of the independent variables. Disability prevalence increases with advancing age. Both for men and women, married/ cohabiting individuals are less prone to be disabled than those who are divorced/separated/widowed. Disability is negatively associated with higher household economic status and higher educational levels for both men and women. Furthermore, disability is less prevalent among those in paid employment and among respondents living in urban areas. Table 2 shows the adjusted effects of the social determinants on disability resulting from the logistic regression models for males and females. There is a positive and significant association between age and disability. However, marital status has no significant effect once the other variables are controlled for. Lower levels of education are significantly associated with disability both among men and women. Likewise, the likelihood of being disabled is significantly higher among people who do not participate in the labor force, although this effect is stronger for men than for women. Also economic status is significantly related to disability, with increased disability prevalence associated with lower economic status, especially for men. In contrast, area of residence is not significantly associated with disability for either men or women once the effects of the other factors have been controlled for. The decomposition in Table 3 shows how the social determinants influence the difference in disability between older men and women. Approximately 45% of the inequality between men and women in this population can be attributed to differences in the distribution of socio-demographic factors. This is the socalled "explained" part of the inequality, meaning that this results from differences in the characteristics between older men and women. Of this "explained" inequality, 81% of the contribution comes from social determinants including employment (49%), education (15%), marital status (12%) and household economic status (4%). The remaining 19% of the "explained" inequality is attributed to differences in the distribution of age (10%) and country of residence (10%). Approximately 55% of the sex difference in disability prevalence results from differences in the effects of the determinants on disability (Table 3). This includes factors in the model (age, country of residence, social determinants) as well as others not in the model. This is the "unexplained" part of the inequality in disability between men and women. The differential effects of age and country of residence contributed most to the "unexplained" part, although these effects were not statistically significant. With the exception of employment, the differential effects of the social determinants included in the model (i.e. education, marital status, urban-rural residence and household economic status) were smaller and not statistically significant. --- Discussion --- Prevalence of disability in men and women aged 50 and older This analysis of household survey data from 57 countries demonstrates inequality in disability prevalence between men and women aged 50 and over. Prevalence of disability is higher among women than among men across all age groups. Prevalence of disability among women aged 50 to 54 years is higher than that among men aged 60 to 64 years. Other studies reported in the literature show that older aged women are more likely than men to become disabled and remain disabled, thus having longer duration of disability, particularly at very old ages [11,13,[38][39][40]. In populations aged 50 and older, women fare worse in their health than men with regard to the social determinants. Specifically, women who completed their education to college level or above have higher disability than men who have no formal education. Women in the highest wealth quintile have higher disability than men in the second lowest wealth quintile. These results are also consistent with the literature showing that the health of older women in many parts of the world is worse for unemployment, lack of education and lower income [3,10,[14][15][16][17]40,41]. --- Multivariate analysis of social factors associated with disability in men and women aged 50 and older The multivariate analysis for males and females shows that, after adjusting for marital status, education, employment, household economic status and area of residence, age is significantly positively associated with increasing disability, with a stronger gradient across age groups for older women. Compared to those with college education or above, men and women with no education or incomplete primary education are twice as likely to be disabled. Studies have examined associations between "disability" (measured in relation to difficulty performing common daily activities), education [14] and other socioeconomic factors [15,16]. Here we use a measure of disability based on the current ICF definition. This conceptualization captures people who are experiencing difficulties in executing tasks or actions on a continuum of functioning, regardless of whether this is due to chronic conditions such as heart disease or arthritis, or physical impairments such as blindness or paralysis. Employment remains a major factor associated with the health of people aged 50 and older, and this is particularly important for older men [14]. We would expect stronger associations between unemployment and disability at the younger end of this age range (e.g. 50 to 64 years) [42] although these patterns are likely to vary between countries. The associations between employment and household economic status and disability are stronger for older men than older women, possibly reflecting differences in gender roles. A study of educational and income inequalities in morbidity among the elderly (60 and over) in eleven European countries showed inequalities in all age groups and countries and for both sexes [42]. --- Decomposition of difference in disability between men and women aged 50 and older The decomposition showed that approximately 45% of the inequality in disability between men and women aged 50 and over is attributed to the distribution of determinants between the sexes. Employment, education, marital status and household economic status, together and individually, made a significant contribution, and age and country were also major contributors to the explained component. Employment was the largest single contributor to the "explained" component, this being due to the fact that a higher proportion of men than women were in paid jobs (63.2% vs. 24.2%). Analysis of WHS responses to a question in which people stated reasons for not working for pay showed that only 6.4% gave "ill health" as a reason. The main reasons given by women for not having a paid job were associated with being homemakers or caring for the family followed by being retired. Being retired was the single main reason for men. Education also contributed to the "explained" component resulting from the fact that more women than men had no education (41.6% vs. 28.8%). The contribution of marital status to the "explained component" is due to the fact that a much higher proportion of women than men were divorced, widowed or separated (40.0% vs. 10.6%). For details see Additional file 3: Table S3. We hypothesize that country specific social factors that we have not measured here (e.g. differences in occupational opportunities between men and women [43], differential access to social protection mechanisms, religious beliefs, family arrangements and cultural norms [18,44,45], also contribute to our measured inequality in disability among older men and women [9]. Our "country" variable is possibly a proxy for these factors although more research is needed to better understand these patterns. More than halfapproximately 55%of the inequality in disability between these older men and women is due to the differential effects of the determinants we investigated as well as factors not included in our model. Employment made a statistically significant contribution to the "unexplained" component, because being unemployed is associated with higher odds of disability for men -2.20 (95% CI: 1.89 to 2.57)than for women -1.44 (95% CI: 1.26 to 1.66). Biological (genetic, physiological and hormonal) differences between males and females determine malespecific responses to morbidity and co-morbidity. --- Strengths and limitations Our research has several strengths. The study is the first to use a multi-country dataset to decompose the determinants of the inequality in disability among older men and women. The use of a large WHS dataset (57 countries) which is based on a consistent set of measures, ensures comparability in measuring disability prevalence and inequalities between men and women. Importantly, the decomposition has allowed us to explore social factors that compromise the health of older men and women. Research such as this is needed so that evidence-based policy responses can be developed [46]. Much of the literature on gender differences in disability in older populations focuses only on limitations in ADL and IADL [47]. The measurement of disability among adults aged 50 and older, was based on a parsimonious set of eight health domains (vision, mobility, self-care, cognition, interpersonal activities, pain and discomfort, sleep and energy, and affect) in order to give a more complete picture of disability. While the "unexplained" component of the inequality suggests that there are factors that contribute to the inequality that were either not assessed in the WHS or were not included in the present analysis, this is a strength rather than a limitation. Unlike other studies of inequalities in disability between men and women, we have used a decomposition method to identify the "explained" and "unexplained" components of the inequality. Breaking down the inequality in this way provides a platform for further research to better understand policy relevant factors that contribute to the inequality. However this study has some limitations. Firstly the analysis is based on self-reported data, and so incurs the possibility of report bias whereby response is influenced by people's understanding of questions, their experiences, expectations, and culture. There are also differences in the way in which men and women describe their health. Women are more likely to report poorer functioning and worse overall health than men [48]. Future studies should include physical assessment of functioning in multiple domains to minimize report bias and calibrate self-reports. Secondly, the countries included here are not necessarily representative of all countries in the world, or of groups of countries defined by income or geographical region. Thirdly, the position that older women have in society in each of these countries is likely to vary from country to country and we were not able to explain how this occurred. Cultural and societal factors that place women in a subordinate position to men underlie and contribute to gender inequalities in health. It is possible that differences in disability result from interactions between sex (biological) gender (social), race, ethnicity, and other social and behavioral factors but we were unable to identify these influences here [9]. --- Conclusions Global ageing has a major influence on disability trends and national populations are ageing at unprecedented rates [3]. The important demographic and epidemiological shifts that are occurring internationally are being accompanied by increasing feminization of older adult populations due to differences in life expectancies between men and women. These trends are occurring in both developed and developing countries and they require policy actions that respond to the needs of disabled older adult populations [3]. Underpinning this policy imperative is the need for data and methodologies that can identify how social, biological and other factors separately contribute to the health decrements facing men and women as they age. While attention is often paid to health inequalities in younger populations, this study highlights the need for action to address social structures and institutional practices that impact unfairly on the health of older men and women. --- Additional files Additional file 1: Table S1 Additional file 3: Table S3. Distribution of determinants in men and women aged 50 and older. Pooled analysis of 57 countries, World Health Survey, 2002-2004. --- Competing interests The authors declare that they have no competing interests. --- Authors' contributions AH designed the study and undertook the statistical analysis. JSW undertook the literature review and drafted the manuscript with input from AH and SC. BJ, AP, PK and AO read the draft and provided input and critical comment. All co-authors read and approved the final draft. --- Support This paper uses data from WHO's World Health Surveys (WHS). The analysis for this paper was supported by the US National Institute on Aging through Interagency Agreements (OGHA 04034785; YA1323-08-CN-0020; Y1-AG-1005-01) and through a research grant (R01-AG034479). Additional support was provided by CBM International.

Introduction: Women represent a growing proportion of older people and experience increasing disability in their longer lives. Using a universally agreed definition of disability based on the International Classification of Functioning, Disability and Health, this paper examines how, apart from age, social and economic factors contribute to disability differences between older men and women. Methods: World Health Survey data were analyzed from 57 countries drawn from all income groups defined by the World Bank. The final sample comprises 63638 respondents aged 50 and older (28568 males and 35070 females). Item Response Theory was applied to derive a measure of disability which ensured cross country comparability. Individuals with scores at or above a threshold score were those who experienced significant difficulty in their everyday lives, irrespective of the underlying etiology. The population was then divided into "disabled" vs. "not disabled". We firstly computed disability prevalence for males and females by socio-demographic factors, secondly used multiple logistic regression to estimate the adjusted effects of each social determinant on disability for males and females, and thirdly used a variant of the Blinder-Oaxaca decomposition technique to partition the measured inequality in disability between males and females into the "explained" part that arises because of differences between males and females in terms of age and social and economic characteristics, and an "unexplained" part attributed to the differential effects of these characteristics. Results: Prevalence of disability among women compared with men aged 50+ years was 40.1% vs. 23.8%. Lower levels of education and economic status are associated with disability in women and men. Approximately 45% of the sex inequality in disability can be attributed to differences in the distribution of socio-demographic factors. Approximately 55% of the inequality results from differences in the effects of the determinants. Conclusions: There is an urgent need for data and methodologies that can identify how social, biological and other factors separately contribute to the health decrements facing men and women as they age. This study highlights the need for action to address social structures and institutional practices that impact unfairly on the health of older men and women.

The Marmot Review laid out how public expenditure on policies, through the life course, could act on the social determinants of health to reduce health inequalities. The Review may have been welcomed in theory but in reality, under the banner of austerity, public expenditure was cut from 42% of national income in 2009-10, to 35% in In a new report, Health Equity in England: the Marmot Review Ten Years On, we show that austerity has taken its toll -in almost all of the areas identified by the Marmot Review as important for health inequalities -from rising child poverty and closing of children's centres, to declines in education funding, to a housing crisis and a rise in homelessness, to people with insufficient money to lead a healthy life and resort to food banks in large number, to left-behind communities with poor conditions and little reason for hope. Given the strength of the evidence on social determinants of health inequalities, it is likely that this rolling back of the state had an adverse effect on health and health inequalities. I cannot say with any certainty which of the changes associated with austerity may have been responsible for the increase in health inequalities, and which may cause damage in the future, but the figures reveal adverse impact across the whole range of domains in the Marmot Review on the lives people are able to lead. --- Inequalities in health -region and deprivation The flattening of the upward curve of life expectancy was marked. From 1981 to 2010 life expectancy increased about one year every five and a half years among women, and about one year every four years among men. From 2011 to 2018 this slowed to one year every 28 years among women, and one year every 15 years among men. The slowdown is real. It is the explanations that have been disputed. I claim, based on a general view of health trends globally, that if health stopped improving society must have stopped improving. There are more prosaic explanations that, having considered, we can reject. First: perhaps we've reached peak life expectancy. It has to slow some time. Plausible, but it is given the lie by comparison with other European countries. Countries with longer life expectancy than England, or other countries of the UK, continued to increase. We have a way to go before we hit peak. Second, perhaps we had bad winters and bad flu outbreaks. Mortality goes up in a bad winter. Analyses that we did for the Ten Years on Report showed that improvements in mortality slowed for non-winter months, as it did for winter months. At most, winter effects could account for between one sixth and one eighth of the slow down in improvement. In support of my contention that the real causes of the failure of health to improve are social, are the growing inequalities in health according to deprivation and region. The relation between deprivation and health is graded. Classifying areas of the country by the index of multiple deprivation reveals a strong and consistent gradient: the more deprived the area, the lower the life expectancy, and healthy life expectancy -Figure 1 in the Marmot Review of 2010. The Figure here shows what has happened to the gradient in life expectancy. Inequality increased quite markedly -life expectancy is improving for the top 60%, not for the bottom 40%. More than that, for females in the bottom five deciles of deprivation, life expectancy declined. --- Figure Gain in life expectancy by sex, England, 2012-14 to 2015-17 Source: based on PHE, 2019 (13). There are well-known regional differences in mortality and life expectancy -sicker in the North. Deprivation and geography come together in important ways. For people living in districts in the least deprived decile, there is little regional difference in life expectancy. If you are among the most fortunate ten per cent it doesn't much matter in which region of the country you make your home, and your demographic sub-group will have experienced improvement in life expectancy. The more deprived your district the greater the disadvantage of living in the North compared to London and the South East. Particularly for women, life expectancy in the most deprived ten percent declined in the North East, Yorkshire and Humberside and other areas outside London and the North West. Life expectancy is used as an indicator, but health is more than expectation of length of life. The life expectancy figures apply to healthy life expectancy, only more so. The social gradient in healthy life expectancy is steeper than that of life expectancy. And years spent in poor health increased between 2009-11 and 2015-17: from 15.8 years to 16.2 in men; from 18.7 years to 19.4 years in women. --- Cutting back in the role of the state may indeed have played an important part The governments elected in Britain in 2010 and 2015 had austerity as a central plank of policy. The stated aim was to restore the economy to growth, by restricting public expenditure. By one measure, at least, it didn't work: wage growth. International comparisons of wage growth between 2007 and 2018 show that Britain, with minus 2%, was the third worst of 35 rich (OECD) countries, only beaten to bottom place by Mexico and Greece. ( ) If pushed, the governments would probably have denied that the purpose of policy was to make the poor poorer and allow the top 1% to resume the trajectory -briefly interrupted by the global financial crisisof garnering a larger share of national income and wealth. That was, however, its effect. Changes to taxes and benefits introduced in 2015 were neatly regressive. The lower the income the greater the reduction in income as a result of changes to taxes and benefits. Cuts to services and to local government hit more deprived areas and families hardest. In one sense, it is hardly surprising that if spending on welfare for families went down by more than 40%; if cuts to local government expenditure went down by 31% in the most deprived decile, compared to 16% in the least deprived decile; if funding for sixth form and further education fell by 12% per pupil; then there should be ill-effects. If the architects of these policies imagined that all that money had been going to waste, the evidence suggests that they were mistaken -our Ten Years On Report documents substantial harmful effects Which of these might be most responsible for an increase in health inequalities is difficult to say because they are inter-related. In the Ten Years On Report we examine impacts in five of the six domains of recommendations made in the Marmot Review: <unk> Give every child the best start in life <unk> Education and lifelong learning <unk> Employment and working conditions <unk> Having enough money to lead a healthy life <unk> Sustainable places and communities Here I highlight two, in addition to the cuts stated above: early childhood and housing. Giving every child the best start in life and continuing through the life course Investing in early childhood creates hope for the future. Adverse trends over the last decade will not be the explanation of the health trends of the last decade but may indeed be a harbinger of things to come. Early childhood is crucial; not just health of children, but level of development -cognitive, linguistic, social, emotional and behavioural development. Good early child development predicts good school performance, which in turn predicts better educational and occupational opportunities and more salutary living conditions in adulthood. Conversely, adverse child experiences (ACEs) cast a long shadowpredicting later life mental and physical illness. More generally, both the good things that happen in early childhood, and the bad, influence individuals' abilities through the life course. We emphasise the importance of 'agency' -having control over one's life, empowerment -as central to health and health inequalities. Agency starts with good early child development and protection from ACEs. It continues with having enough resources in adulthood to lead a life that one has reason to value. Having insufficient money to pay rent or feed children is deeply disempowering. Early childhood contributes to health inequalities in adult life because of the social gradient. Good early child development is less common with increasing deprivation and adverse child experiences are more common. Therefore, two approach to reducing inequalities in health must be to improve services that break the link between deprivation and poor outcomes for children and to reduce deprivation in childhood. The signs are not good. It has been estimated that, as a result of cuts to local government, 1000 Sure Start Centres have had to close. The welcome support for child care for older pre-school children does not make up for these closures. A much-used measure of child poverty is the per cent of children living in households at less than 60% median income. In 2009-12, child poverty was 18%, rising to 20% in 2015-18. But the cost of housing can drive people into poverty. After housing costs are taken into account, child poverty rose from 28% in 2009-12 to 31%. Shortage of housing contributes to increased costs. The proportion of people paying more than a third of their income on housing costs, unsurprisingly, follows the social gradient, but has risen in all income groups. In the lowest decile of income in 2016/7, 38% of families were paying more than a third of their income in housing, up from 28% a decade earlier. --- Time for action In the Marmot Review in 2010, we wrote: health inequalities are not inevitable and can be significantly reduced. They stem from avoidable inequalities in society. A report from the Royal Society for Public Health suggested we got the evidence and approach more or less right. RSPH surveyed their members and a panel of experts on their views on the major UK public health achievements of the 21 st Century to date. The top three were the smoking ban, the sugar levy, and the Marmot Review. We have now laid out a new set of recommendations following the intellectual framework of the Marmot Review. It is our firm view that such recommendations are vital to creating a society that is just, and sustainable for the current and future generations.

Britain has lost a decade. And it shows. Health, as measured by life expectancy, has stopped improving and health inequalities are growing wider. Improvement in life expectancy, from the end of the 19th Century on, slowed dramatically, beginning in 2011. Now in parts of England, particularly among women in deprived communities and the North, life expectancy is falling, and the years people are spending in poor health may even be increasing -a shocking development. In the UK, as in other countries, we are used to health improving year on year. Bad as health is in England, in Scotland, Wales and Northern Ireland, the damage to the health of the country is nearly unprecedented. Put simply, if health has stopped improving it is a sign that society has stopped improving. Evidence, assembled globally, shows that health is a good measure of social and economic progress. When a society is flourishing health tends to flourish. When a society has large social and economic inequalities there are large inequalities in health(1). The health of the population is not just a matter of how well the health service is funded and functions, important as that is, but health is closely linked with the conditions in which people are born, grow, live, work and age and inequities in power, money and resources -the social determinants of health(2). This damage to the nation's health need not have happened. In 2010 there was concern both by the Labour government, and the Conservative-led Coalition government that followed it, that health inequalities in England were too wide and action to reduce them had to happen. To inform that action the Government in 2008 commissioned me to review what government and the wider society could do to reduce health inequalities. With colleagues at what later became the UCL Institute of Health Equity, we convened nine task groups of more than 80 experts to review the evidence and assembled a distinguished Commission to deliberate on that evidence. The result was the Marmot Review, Fair Society Healthy Lives, published in 2010(3). The Review was commissioned by the Labour Government and welcomed by the Coalition Government in a Public Health White Paper.

Background The school environment is an arena where many dietary norms and habits are established which potentially affect the individual throughout their future lives [1]. Owing to the considerable amount of time adolescents spend at school during the average weekday, it has been estimated that approximately one third of their food and drink is consumed in the school environment [2,3]. Environments which encourage a high energy intake and sedentary behaviour amongst adolescents are termed obesogenic environments, and such environments are considered to be one of the main elements behind the rapid increase in overweight and obesity among children and adolescents [4]. In this regard, the local food environment of schools, including arenas such as supermarkets and convenience stores close to the schools, is an environmental influence potentially affecting the quality of the food intake of attending adolescents [5]. Providing healthy food and drinks to adolescents in schools via canteens or vending machines plays an important role in modelling a healthy diet, particularly for those who may not have access to healthy food outside school hours, thereby making school nutrition policies a powerful tool for improving students' nutritional status and academic achievement [6]. Yet in the school environment, foods consumed are not always obtained from on-campus sources. Research upon supermarkets and convenience stores located in the vicinity of schools has reported that these venues provide an increased accessibility to unhealthy foods and drink for school-going adolescents [7]. The <unk>vre Romerike region, located in the eastern part of Norway, has a total area of 2,055,550 km 2, and composed of 6 municipalities housing approximately 100,000 people [8]. The 2016 average net income for all households in the region was 456,667 NOK, compared to the national average of 498,000 NOK for the same period [9]. In our recent investigation upon adolescents in <unk>vre Romerike, we reported that 33% of participants purchased food or drink in their school canteen at least once a week [10]. In addition, 27% and 34% of participants reported purchasing food and drinks from shops around schools one or more times a week, either during school breaks or on their way to or from school, respectively [10]. Investigations on adolescent behaviour in Norway and elsewhere have reported similar results, whereby approximately 30% of school-going adolescents visit local food stores for nourishment, whilst the majority are consuming their lunches at school [11,12]. In Norway, the average school day includes a lunch period in the middle of the day [13], and most students travel to school with a home packed lunch, usually consisting of bread slices with various toppings [14,15]. School canteens are often run by catering staff, with students in need of more practical education sometimes included in food preparation and selling. It is not uncommon for the canteen to be managed on a daily or occasional basis by students together with a teacher as a part of their education. School canteens most commonly offer baguettes, waffles, milk (regular or chocolate), juice, cakes and, perhaps, fruit [16,17]. The Norwegian Directorate of Health regularly publishes guidelines concerning school meals and eating environments, with the most recent published in 2015 [18]. The latest guidelines offer suggestions regarding topics such as length of meal times, hygiene, fresh water accessibility, the absence of sugar-rich foods and drinks, and the reduction of saturated fats on offer. The guidelines are published as a tool to assist school administration in their management of school canteens. Eating behaviour amongst adolescents is a complex theme often involving an interplay of multiple influences and factors such as peer influence [19] and a desire to socialise whilst eating [20], a combination which often leans toward unhealthy eating practices. Furthermore, it is not uncommon for young Norwegian teens to receive pocket money [21], and this emerging autonomy aided by pocket money increases the prospect for a disruption of dietary behaviour established in the home [22]. As the school food environment has such a significant impact on food choices [23,24], a better understanding of adolescent's consumption behaviour demands further attention. In particular, understanding student's shift away from home packed lunches and canteen foods towards the appeal of off-campus shop food is necessary for implementing the successful promotion of healthier lunch alternatives at school. The aim of the present study was to gain a better understanding of the consumption habits of adolescents in the Norwegian school lunch arena. Unlike previous ESSENS studies, here we use quantitative data combined with qualitative interviews among adolescents and school administration, in order to explore the purchasing behaviour and lifestyle demographics of the sample grouped as frequent and infrequent school canteen users compared to those never or rarely using the canteen. --- Methods --- Design and sample The participants in this study were students and staff from eleven secondary schools participating in the Environmental determinantS of dietary behaviorS among adolescENtS (ESSENS) cross-sectional study [10,25]. Recruitment of students and staff was initiated by our making contact with principals of the twelve secondary schools in the <unk>vre Romerike district, after first having received permission from district school leaders. The school principals were each sent a letter detailing key elements of the proposed intervention, as well as information regarding the ESSENS study, together with a permission form requesting their school's participation. Of the twelve secondary schools invited to participate in the study, eleven accepted the invitation. In this mixed method approach, our sample were grouped as being part of either a quantitative or qualitative data source. --- Recruitment of sample --- Quantitative recruitment In October 2015 we recruited 8th grade adolescents for participation in a questionnaire survey. An informative letter was sent home with all 1163 adolescents in the 8th grade (average age of 12-13 years) from the 11 participating schools, containing a consent form for signing and with additional questions relating to parental education levels. A total of 781 (67%) received parental consent for participation. As the range of ages of the sample represents the lower end of the adolescent scale (10-19 years), the use of the term 'adolescent' here implies 'young adolescent'. A total of 742 adolescents (64% of those invited and 95% of those with parental consent) participated in the survey. Quantitative data collection took place between October and December 2015. --- Qualitative recruitment Recruitment of adolescents to participate in the qualitative part of the study was also facilitated by approaching principals of district schools as described above, and was completed between October 2015 and January 2016. Six of the 11 participating schools were selected for qualitative data collection based upon criteria such as location (being in one of the six municipalities of <unk>vre Romerike), and size (based upon number of students attending). The aim was to include schools with a varied profile, with proximity to city centers, shops, and collective transport as determining factors. Thereafter a selection process for participation in the focus groups was conducted, whereby two students per class were sought after, representing both sexes. Further inclusion criteria stipulated that the students be in the 9th grade, had attended Food and Health classes, and currently lived in the <unk>vre Romerike area with either one or both parents. --- Data collection --- Quantitative data A web-based questionnaire was used to collect data from the adolescents, using the LimeSurvey data collection tool. The questionnaires were answered at school, taking approximately 30-45 min to complete, and queried respondents about their nutritional intake, parental rules regarding food and drink consumption, students' school canteen and surrounding shop use, physical activity, and sedentary behaviour habits. Research group members were present during data collection to answer questions and make sure the adolescents responded independently from each other. The questionnaire relating to food behaviours completed by the sample is available online (see Additional file 1: Appendix 1 ESSENS questionnaire relating to food behaviours). A pilot test of the survey was conducted parallel with this process in a neighboring municipality with similar age students from the 8th grade (n = 23). The students spent approximately 30-40 min to complete the survey, and then provided feedback regarding comprehension. The questionnaire was subsequently shortened and some questions rephrased for clarity. The results of the pilot test were not included in the final results. --- Qualitative data Focus group interviews were conducted over a period of 10 weeks, from November 2015 to January 2016. Focus group settings were favoured as they provide a more relaxed setting for data collection, facilitating the flow of a natural conversation amongst peers, especially when adult researchers interact with young subjects [26]. Six focus group interviews including a total of 55 students (29 girls, 26 boys) from the 9th grade with an average age of 13-14 years were conducted. Interviews had a duration of approximately 60 min. In addition, interview sessions with headmasters and teachers for the 9th grade students from the participating schools were also conducted. Interviews with 6 teachers (4 women and 2 men) and 6 headmasters (3 women and 3 men) were conducted from October 2015 to January 2016. The interviews with principals and teachers were each conducted separately. Qualitative data collection took place at the selected schools using an audio recorder, with a semi-structured interview guide used for the interviews, partially inspired by a previous study conducted amongst 11-13 year old Norwegian adolescents [27]. The main themes explored by the focus group sessions were students' eating habits, their definition of healthy and unhealthy food, attitudes towards and their impact upon diet and physical activity, as well as the student's assessment of opportunities and barriers attached to health-promoting behaviour. School administration interviews probed food availability and meals served at the school, as well as physical activity options available for students at the schools. The interview guides used for the focus groups and the school administration are available online (see Additional file 2: Appendix 2 Interview guide for focus group interviews, and Additional file 3: Appendix 3 Interview guide for headmasters and teachers). Interviews were transcribed verbatim, with names of the participants and of the schools anonymised. Interviews were analysed using a thematic analysis approach [28]. Codes were developed after an initial reading of all the transcripts and were based on the main interview questions, prior research, and emergent concepts from the current data. The initial codes were discussed among researchers and a codebook was developed. The codes were further refined during coding of subsequent transcripts. Codes were then successively grouped into general themes. The data analysis was supported by the use of NVivo software (version 10.0; QSR International, Cambridge, Mass). Pilot testing of the intended focus group question guide was performed in October 2015 in a school belonging to a neighbouring district. After written consent was obtained from the principal of the school, 6 students from the 9th grade were selected by a 9th grade teacher from the school. Three girls and 3 boys were included in the focus group pilot test. A moderator conducted the focus group following an interview guide in order to test comprehension and flow of the planned themes. The pilot test proved effective and consequently no changes were made to the interview guide. Data from the pilot testing was not included in the results of the study. Recruitment of school staff for participation in indepth interviews was also facilitated by the agreement with administrative school leaders as described above. A written invitation was sent to principals and teachers of the 9th grade classes from the same 6 schools participating in focus group interviews. Those agreeing were later contacted by phone to arrange a place and time for the interview. Pilot testing of school staff interviews was performed in October 2015 in a school belonging to a neighbouring district. Two interviews were conducted with one headmaster and one teacher separately in order to assess the comprehension and flow of the various themes probed, as well as the time used for the interview. Data from the pilot testing was not included in the results of the study. --- Measures The following measures obtained from the questionnaire were used in the quantitative analyses of the present study. --- Sociodemographic measures Two questions assessing parental education (guardian 1 and guardian 2) were included on the parental informed consent form for the adolescent. Parental education was categorised as low (12 years or less of education, which corresponded to secondary education or lower) or high (13 years or more of education, which corresponded to university or college attendance). The parent with longest education, or else the one available, was used in analysis. Participants were divided into either ethnic Norwegian or ethnic minority, with minorities defined as those having both parents born in a country other than Norway [29]. --- Dietary behaviours Frequency of carbonated sugar-sweetened soft-drink intake (hereafter referred to as soft-drinks) during weekdays was assessed using a frequency question with categories ranging from never/seldom to every weekday. Weekday frequency was categorised as less than three times per week and three or more times per week. The questions assessing the intake of soft-drinks have been validated among 9-and 13-year-old Norwegians using a 4-day pre-coded food diary as the reference method, and moderate Spearman's correlation coefficients were obtained [30]. Consumption of fruits and vegetables (raw and cooked) were assessed using frequency questions with eight response categories ranging from never/seldom to three times per day or more. These were further categorised as less than five times per week and five or more times per week. The questions assessing intake of fruits and vegetables have been validated among 11-year-olds with a 7-day food record as the reference method and were found to have a satisfactory ability to rank subjects according to their intake of fruits and vegetables [31]. The consumption of snacks [sweet snacks (chocolate/ sweets), salty snacks (e.g. potato chips), and baked sweets (sweet biscuits/muffins and similar)] was assessed using three questions with seven response categories ranging from never/seldom to two times per day or more. These were further categorised as less than three times per week and three or more times per week. Acceptable to moderate test-retest reliability have been obtained for these measures of dietary behaviours in a previous Norwegian study conducted among 11-year-olds [27]. Self-efficacy related to the consumption of healthy foods was assessed using a scale with six items [e.g. Whenever I have a choice of the food I eat..., I find it difficult to choose low-fat foods (e.g. fruit or skimmed milk rather than 'full cream milk')]. Responses were further categorised as those with 'high' self-efficacy (score of 3.5 or higher, from a scale of 1-5) or 'low' self-efficacy (under 3.5, from a scale of 1-5). The scale has been found to have adequate reliability and factorial validity among 13-year-olds [32]. Adolescents' breakfast consumption was assessed using one question asking the adolescents on how many schooldays per week they normally ate breakfast. The answers were categorised as those eating breakfast 5 times per week or less than 5 times per week. This question has shown evidence of moderate test-retest reliability (percentage agreement of 83 and 81% respectively for weekday and weekend measures) and moderate construct validity (percentage agreement of 80 and 87% respectively for weekday and weekend measures) among 10-12 year old European children [27]. --- Food/drink purchases in school environment The adolescents were asked how often they purchased foods or drinks from school canteens and on their way to and from school (answer categories ranging from 'never' to 'every day'). The frequency of purchase of food/drinks at the school canteen were then re-categorised into 'never/rarely', 'once per week', or 'two or more times per week'. The frequency of purchase of food/drinks at off-campus food stores were re-categorised into 'never/ rarely', or 'one or more times per week'. They were also asked about the presence of food sales outlets (e.g. supermarket, kiosk, or gas station) in a walking distance from their school (with answer categories 'none', 'yes, one', 'yes, two', and 'yes, more than two'), with results categorised as 'less than 3' or '3 or more'. Further details regarding data collection and methodology in the ESSENS study have been described previously [10]. Ethical clearance for the study was obtained from the Norwegian Social Science Data Service (NSD 2015/ 44365). Written informed consent was obtained from all parents of participating students. --- Statistical analyses The study sample was divided into three groups, those who reported 'never or rarely' using the school canteen (NEV), those using the canteen once per week (SEL), and those reporting use of the school canteen 'two or more times during the week' (OFT). Results are presented as frequencies (%), with chi-square tests performed to examine differences in sociodemographic, behavioural, and dietary characteristics between the three groups. A further logistic regression analysis was performed to assess the adjusted associations between canteen use and dietary habits (salty snacks, baked sweets, soft-drinks, and home breakfast frequency). Adjustment was made for significant sociodemographic and behavioural characteristics (gender, parental education, self-efficacy) and shop use (during school break and before/after school). Logistic regression was also used to explore the adjusted association between visiting shops during school breaks or before/after school ('never/rarely', 'one or more times per week'), and use of canteen (NEV, SEL, OFT). Results are presented as crude odds ratios (cOR) and adjusted odds ratios (aOR) with 95% confidence intervals (95% CIs). Cases with missing data were excluded from relevant analyses. Because schools were the unit of measurement in this study, we checked for clustering effect through the linear mixed model procedure. Only 3% of the unexplained variance in the dietary behaviours investigated was at the school level, hence adjustment for clustering effect was not done. A significance level of 0.05 was used. All analyses were performed using SPSS 24.0 (IBM Corp, Armonk, NY, USA). --- Results --- Sample demographics The mean age of the survey sample was 13.6 years <unk>0.3 standard deviation, 53% of participants were females, and 60% had parents with a high level of education (<unk>13y, Table 1). The proportion of adolescents who never or rarely used the school canteen was 67.4%. When comparing demographics and behavioural characteristics for the sample grouped as those using the school canteen never/ rarely (NEV), those using the canteen once a week (SEL, 19.7%), and those using the canteen two or more times a week (OFT, 12.9%), we found a significantly higher proportion of the NEV group were female, having parents with a high education, and with a high self-efficacy. --- Canteen use and dietary habits When analysing the dietary habits for the sample grouped by frequency of canteen use, a significantly higher proportion of the OFT group reported consuming salty snacks, baked sweets, and soft-drinks <unk>3 times per school week, and a significantly higher proportion of the NEV group reported eating breakfast 5 days in the school week compared to the SEL and OFT groups (Table 2). A multiple logistic regression was conducted to assess whether these significant associations between canteen use and dietary behaviours persisted after adjustment for gender, parental education, self-efficacy, NEV: adolescents never or rarely using the school canteen (n = 485); SEL: adolescents using the school canteen once a week (n = 142); OFT: adolescents using the school canteen <unk>2 times a week (n = 93) c Chi-square test between frequency of canteen use groups and use of shops (both during and before/after school). The difference between NEV, SEL, and OFT adolescents regarding baked sweets thereafter became non-significant. However, the difference between NEV and OFT adolescents regarding salty snacks, soft-drinks, and breakfast consumption remained significant, indicating that adolescents using the canteen <unk>2 times per week had increased odds for consuming salty snacks and soft-drinks (aOR 2.05, 95% CI 1.07-3.94, p <unk> 0.03, and aOR 2.32, 95% CI 1.16-4.65, p <unk> 0.02, respectively, data not shown). Additionally, the OFT group had reduced odds of consuming breakfast at home daily (aOR 0.48, 95% CI 0.28-0.80, p = 0.005, data not shown). No significant differences between the three groups were found for the other food items explored. --- School environment When comparing the frequency of food purchases at shops during school breaks or on the way to/from school for the NEV, SEL, and OFT groups, we found that a significantly higher proportion of OFT adolescents reported purchasing food/drink from a shop near school either during school breaks or else before or after school, one or more times during the week (Table 3). Logistic regression analyses revealed that the OFT group had significantly higher odds of purchasing food/drink from a shop near school, either during school breaks or else before or after school, than the NEV group (aOR = 1.80, 95% CI 1.07-3.01, and aOR = 3.61, 95% CI 2.17-6.01, respectively, Table 4). --- Results of focus group and interview analyses The data from the focus group interviews indicated that students were aware of issues related to food and health. A number of the relevant themes which emerged are outlined below. --- Student's lunch habits The majority of students confirmed that most foods consumed at school were brought from home. Some students, however, stated that the other option was to purchase foods from either the canteen or local shops: Interviewer:....do you bring a packed lunch from home regularly? Boy2: We usually tend to buy something from the canteen. Girl5: It's kind of both in a way. NEV: adolescents never or rarely using the school canteen (n = 485); SEL: adolescents using the school canteen once a week (n = 142); OFT: adolescents using the school canteen <unk>2 times a week (n = 93) Girl5: Yes. Ehm, it is usually both, there are many who have food with them also. Also you are free to buy something. Boy1: Yes, that's common...there are quite a few who tend to buy food at the canteen and, yes, the shop. One teacher suggested it was the presence of pocket money that determined the source of a student's lunch: Teacher1: It is an incredibly large amount of money they have to buy canteen food with, especially in the 8th grade...so that means they do not have so much food with them from home, but buy it instead. --- Types of foods purchased at school canteen, students' impression of canteen In response to the types of foods available for purchase at the canteen, student's representing different schools reported similar food items. Overall, the students at all schools expressed a level of dissatisfaction with the healthiness of the food/drinks offered by the canteen: Interviewer: What is the most popular items people buy [at the canteen]? Boy2: Mainly toasted sandwiches Boy2: And wraps Boy3: Eh, maybe a baguette with ham and cheese Boy1: Whole-wheat bread with cheese and ham. Capsicum maybe. Boy2: There are many different drinks one can buy, as well as yoghurt of various kinds. There is also a main thing available too, such as a baguette, pizza, or something similar. Boy2: There are many who buy toasted sandwiches and wraps. Interviewer: What can be done better in order to make other students or yourselves eat healthier from the school's part?. Girl3: They can begin to sell more fruit and such at the canteen. Boy4: We could have healthier drink offers [from the canteen]...such as smoothies... Girl2:...and switch chocolate milk with plain milk. Boy3: [The canteen] should have healthier alternatives, not just unhealthy white-flour baguettes...with a little cheese, bit of ham and a little butter..... --- Peer influence, perceived peer self-efficacy regarding healthy eating There were questions designed to assess if students perceived other students as being more concerned with healthy eating. Those bringing food from home or considered'sporty' were often perceived as eating healthy food, with the overall impression that those perceived as eating healthy tended to not purchase food at the canteen: Interviewer:...do you think there are some in your class then, that are more concerned with eating healthy than others? Boy3: Yes, there are. Interviewer: Who are they then? Boy3: Those who ski. NEV: adolescents never or rarely using the school canteen (n = 485); SEL: adolescents using the school canteen once a week (n = 142); OFT: adolescents using the school canteen <unk>2 times a week (n = 93) Interviewer: How do you know that? Or, what is it that makes them stand out? Boy2: They....don't buy food at the canteen. Boy4: They eat healthy food Boy1: Those that eat relatively healthy food as a rule usually prepare food themselves. A number of school staff commented upon the influence some students' lunch habits had upon others: Teacher6:...if there is one who begins to drop home brought food because it is boring, it become contagious over other's behaviour I think, and then it isn't cool to eat home packed lunches. They are at a very vulnerable age, and very affected by such things I believe. Teacher2:...(food choices are affected by) what food they have at home, how much money they have in their pocket, and what their friends eat. I think it is these three things. And I think some....won't bring out their home packed lunch because it is not cool enough. --- Prices, timing, and permission for visiting shops In many instances, it was reported that although leaving school grounds was not allowed during school hours in individual school policy, many students frequently did so in order to visit local food shops during breaks. There were reports of shop visits outside school hours as well (before/after school). Some students also discussed the cheaper prices at the shops, as compared to the school canteen, as being an incentive to purchase from shops. Girl2: We have some in the class that shoot off to the shops to buy some sort of fast food every day. Interviewer: So you are allowed to leave the school in your free time to buy food? Girl2: No, but after school or right before. Girl4:......They go over [to the shops] when the lunch break starts, then you see them come back when everyone has to go outside then. Boy4: Because then there are no teachers out......and then it is easy to take a trip to the shops and... Boy1: Buy cheaper things. Because they sell at a high price here. The paradox between students visiting shops in school hours, although not allowed, was also pointed out by school staff: Teacher1:...no, it is not allowed (to go to the shops), but there are some that do it anyway. Headmaster6:...of course the schools must represent counterculture in some way....so our students go to the shops...and then they make use of the offers that are there...as long as they have money from home. Teacher2:...and they prefer to go (to the shops) in a group at the same time, because it is social and fun. --- Types of foods purchased in shops When questioned about the types of items purchased at the shops, the majority were in consensus that unhealthy snacks such as sweets, baked goods, and soft-drinks were mainly purchased. No participant mentioned the purchase of healthy food from the shops. Interviewer: What do people mostly buy there then? You mentioned sweet buns..[Looks at Girl1] Boy2: Both sweet buns and doughnuts. Girl1: There are many that buy candy after school and such. Boy4: There are always some who always have money and always buy candy and such. Just like one I know who bought 1 kg of gingerbread dough here after school one day and sat down and ate it. Girl2: Mostly those....soft drinks Girl1: Soft drinks Boy1: Candy and ice-tea. Boy2: People don't buy food at the shop...most buy themselves candy. --- Adherence of school administration to guidelines for school meals When school staff were asked about the implementation of the latest guidelines from the Norwegian Directorate of Health, most pointed out that they already offered the suggested timespan suggested for lunch, whilst others had yet to read the document. Teacher1: We have heard there is something new that has come, but we have not spent a lot of time discussing it amongst ourselves. Teacher2: No, no relationship with them (new guidelines). I'm not sure. We do not sell sodas and juice in the cafeteria, but they [students] have it from home. Teacher3: Hehe, I don't think I've seen them, no...(laughs). Headmaster1: So, what we do is to make sure that they have a good place to eat and that they have peace....we offer supervision and they do have a long enough lunch break, is it 20 minutes they should have? Headmaster2: I just have to be honest, I do not think we have come far with these. --- Discussion We found the NEV group were mainly female, having a high self-efficacy regarding the consumption of healthy foods, and with parents having an education over 12 years. By contrast, the OFT adolescents had a significantly higher proportion of males consuming salty snacks, baked sweets, and soft-drinks 3 or more times a week, as well as consuming breakfast less than 5 times a week when compared to the other groups, also when controlling for gender, parental education, self-efficacy, and use of shops (both during and before/after school). When comparing the frequency of purchasing food and drink from local shops for these groups, we found the OFT group had a significantly higher proportion purchasing food/drink from shops near the school, both during the school break as well as before or after school, one or more times per week. Logistic regression analyses revealed the OFT group had nearly twice the odds for visiting shops during the school break, and significantly higher odds for visiting shops before/after school than the NEV group of adolescents. Of the adolescents featured in this sample, females were revealed as more likely to never or rarely use the school canteen, a finding supported by previous research amongst adolescents [33,34]. That females have been previously reported as having a greater self-efficacy related to healthy eating [35] may help to explain this result, although another study involving over 1200 students of comparable age found no significant difference in self-efficacy regarding gender [36]. As 67% of the sample stated that they never or rarely use the school canteen, this then begs the question of what form of lunch this group are consuming. Many of the interviews have mentioned the consumption of home packed lunches, and studies of school lunch habits amongst Norwegian adolescents have previously detailed the importance and predominance of the home packed lunch in Norwegian culture [37,38], with over 60% of young Norwegians reporting a packed lunch for consumption at school, a proportion similar to the results we present here. This figure is also consistent with global reports examining school lunch eating practises [39]. Our results profile the OFT group as being mostly male, skipping breakfast, with a high frequency of shop visits during and on the way to/from school, and with a higher frequency of snacks, baked sweets, and soft-drinks, elements which have featured in previous studies regarding adolescent consumer behaviour [12,[40][41][42][43]. A clear association between adolescents skipping breakfast and subsequent purchases of foods from shops and fast food outlets, usually on the way to or from school [42,[44][45][46], in addition to other health-compromising behaviours [47] have been previously reported. Although direct questions regarding pocket money were absent from our study, its role in the behaviour of this sample is evident from statements mentioning money use in the school administration interviews as well as alluded to in focus group interviews. Additionally, it stands to reason that adolescents using the school canteen often (i.e. the OFT group) would be equipped with money in order to make such purchases, as financial purchases are the norm in Norwegian secondary schools [48]. Research directed upon adolescents and pocket money has presented a number of findings that support our results regarding the OFT group, whereby access to spending money was associated with an increase of nutritionally poor food choices by adolescents, such as the increased consumption of fast-foods, soft-drinks, and unhealthy snacks off campus [40][41][42][43][49][50][51][52][53]. These results may also be indicative of a gender imbalance in regards to pocket money provisions, where some studies report upon more males than females receiving pocket money [54,55]. The mean age of this sample previously has been described as a stage in life of an emerging autonomy for young individuals, an autonomy which is exercised in terms of disposable income use and consumption of foods away from home [42,56,57]. This period of emerging autonomy may also manifest unhealthy eating behaviours as a strategy to forge identity amongst adolescents [58]. Frequent mention by students and staff in this study of themes relating to peer influence and defiance of school rules support the link between rebelliousness and unhealthy eating. Moreover, it has been reported previously that foods independently purchased by adolescents are often unhealthy, forbidden or frowned upon by parents, and express a defiant period of appearing 'cool' among peers, especially amongst males [37,[59][60][61], all of which support our findings here, particularly regarding gender, self-efficacy, and peer influence. Value for money and dissatisfaction with the school canteen were frequently mentioned in the focus group interviews, and are elements that may be affecting choices made by the groups in this study. Statements concerning student dissatisfaction with canteen prices and/or the limited healthy options available have also appeared in previous research [35,37,38,42]. That many of the school administrators interviewed seemed barely aware of the guidelines published by the Norwegian Directorate of Health is an alarming result, and likely adds some degree of weight upon student discontent with the school canteen. Although nearly all reports from the focus groups indicate the shops were used for unhealthy purchases, the possibility that shop purchases are a result of some adolescent's need for healthier lunch alternatives cannot be dismissed completely. The focus group interviews together with the quantitative data support the notion of healthy eaters avoiding the school canteen, opting instead for a home packed lunch. This view is further supported by previous reports that home prepared lunches help contribute to a healthy dietary pattern [39,62,63]. Furthermore, it has been reported that students consuming a lunch from home have significantly lower odds of consuming off-campus food during the school week [41], which further concurs with the results presented here. By contrast, those often using the canteenwhich, by all reports, could improve the healthiness of items offeredare using the off-campus shops often, purchasing mainly unhealthy snacks and drinks. The strengths of the study include a large sample size with a high response rate at the school level, and moderate response rate at the parental level. Using a mixed method approach also provides a more comprehensive assessment of adolescent school lunch behaviours, allowing a fuller understanding of this and other adolescent food-behaviour settings by contrasting the adolescent's own experiences with quantitative results. That the quantitative material, based on cross-sectional data, precludes any opportunity for causal inference to be made may be one of the prime weaknesses of this study. Quantitative data regarding adherence to national policy regarding school canteens, pocket money and what items it was spent upon, as well as data regarding the content and frequency of home packed lunch consumption, were also lacking from the study, where inclusion of these elements in the various analyses would have considerably strengthened the quality of results. Furthermore, reliance upon self-reported data may have

Background: Food/drinks available to adolescents in schools can influence their dietary behaviours, which once established in adolescence, tend to remain over time. Food outlets' influence near schools, known to provide access to unhealthy food/drinks, may also have lasting effects on consumption behaviours. This study aimed to gain a better understanding of the consumption habits of adolescents in the school arena by comparing different personal characteristics and purchasing behaviours of infrequent and regular school canteen users to those never or seldom using the canteen. Methods: A convergent mixed methods design collected qualitative and quantitative data in parallel. A crosssectional quantitative study including 742 adolescents was conducted, with data collected at schools via an online questionnaire. Focus group interviews with students and interviews with school administrators formed the qualitative data content. Quantitative data were analysed using descriptive statistics and logistic regression; thematic content analysis was used to analyse qualitative data. Results: Sixty-seven percent of adolescents reported never/rarely using the school canteen (NEV), whereas 13% used it ≥2 times per week (OFT). When the two groups were compared, we found a significantly higher proportion of the NEV group were female, having parents with a high education, and with a high self-efficacy, whilst a significantly higher proportion of the OFT group consumed salty snacks, baked sweets, and soft-drinks ≥3 times per week, and breakfast at home < 5 days in the school week. The OFT group had significantly higher odds of purchasing food/drink from shops near school during school breaks and before/after school compared to the NEV group (adjusted odds ratio (aOR) = 1.80, 95% CI 1.07-3.01, and aOR = 3.61, 95% CI 2.17-6.01, respectively). The interviews revealed most students ate a home packed lunch, with the remainder purchasing either at the school canteen or at local shops. Conclusions: Students using the canteen often are frequently purchasing snacks and sugar-soft drinks from shops near school, most likely owing to availability of pocket money and an emerging independence. School authorities must focus upon satisfying canteen users by providing desirable, healthy, and affordable items in order to compete with the appeal of local shops.

here. By contrast, those often using the canteenwhich, by all reports, could improve the healthiness of items offeredare using the off-campus shops often, purchasing mainly unhealthy snacks and drinks. The strengths of the study include a large sample size with a high response rate at the school level, and moderate response rate at the parental level. Using a mixed method approach also provides a more comprehensive assessment of adolescent school lunch behaviours, allowing a fuller understanding of this and other adolescent food-behaviour settings by contrasting the adolescent's own experiences with quantitative results. That the quantitative material, based on cross-sectional data, precludes any opportunity for causal inference to be made may be one of the prime weaknesses of this study. Quantitative data regarding adherence to national policy regarding school canteens, pocket money and what items it was spent upon, as well as data regarding the content and frequency of home packed lunch consumption, were also lacking from the study, where inclusion of these elements in the various analyses would have considerably strengthened the quality of results. Furthermore, reliance upon self-reported data may have led to issues regarding validity and reliability, particularly with a sample of young adolescents. --- Conclusion We found the majority of adolescents (67.4%) in this sample rarely or never used the school canteen. Those adolescents using the school canteen two or more times a week were also the group most likely to be purchasing food/ drink from a shop near the school, either during school breaks or before/after school. This group also tended to skip breakfast and consume snacks and soft-drinks more frequently compared to the adolescents who rarely or never used the school canteen. These findings highlight a lack of satisfaction of items available for consumption at the school canteen, with adolescents intending to use the school canteen preferring instead the shops for foods that are cheaper and more desirable. Future strategies aimed at improving school food environments need to address the elements of value for money and appealing healthy food availability in the school canteen, as well as elements such as peer perception and self-identity attained from adolescent food choices, especially in contrast to the competitiveness of foods offered by nearby food outlets. --- Availability of data and materials The datasets generated during and/or analysed during the current study are not publicly available due to ongoing project work but are available from the corresponding author on reasonable request. --- Additional files Abbreviations aOR: Adjusted odds ratio; CI: Confidence interval; cOR: Crude odds ratio; ESSENS: Environmental determinantS of dietary behaviorS among adolescENtS study; NEV: Adolescents never or rarely using the school canteen; NOK: Norwegian kroner; OFT: Adolescents using the school canteen two or more times a week; SEL: Adolescents using the school canteen once a week Authors' contributions AC conducted the data analyses and wrote the first draft of this manuscript. MKG2 designed the study, led the project planning and implementation of the intervention, and participated in data collection and analyses. LT1, SH, MG1, LET2 and MKG2 substantially contributed to the conception, design, and implementation of the study, as well as providing content to the final manuscript. MG1 recruited participants, conducted and transcribed focus group interviews, and contributed to data analyses. All authors have critically read and given final approval of the final version of the manuscript. --- Ethics approval and consent to participate Ethical clearance for the study was obtained from the responsible institutional body, the Norwegian Social Sciences Data Services, which is the data protection official for research. Written informed consent was obtained from all parents of participating adolescents; adolescents provided assent. School administrators also provided consent for the study. Consent for publication 'Not applicable'. --- Competing interests The authors declare that they have no competing interests. --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background: Food/drinks available to adolescents in schools can influence their dietary behaviours, which once established in adolescence, tend to remain over time. Food outlets' influence near schools, known to provide access to unhealthy food/drinks, may also have lasting effects on consumption behaviours. This study aimed to gain a better understanding of the consumption habits of adolescents in the school arena by comparing different personal characteristics and purchasing behaviours of infrequent and regular school canteen users to those never or seldom using the canteen. Methods: A convergent mixed methods design collected qualitative and quantitative data in parallel. A crosssectional quantitative study including 742 adolescents was conducted, with data collected at schools via an online questionnaire. Focus group interviews with students and interviews with school administrators formed the qualitative data content. Quantitative data were analysed using descriptive statistics and logistic regression; thematic content analysis was used to analyse qualitative data. Results: Sixty-seven percent of adolescents reported never/rarely using the school canteen (NEV), whereas 13% used it ≥2 times per week (OFT). When the two groups were compared, we found a significantly higher proportion of the NEV group were female, having parents with a high education, and with a high self-efficacy, whilst a significantly higher proportion of the OFT group consumed salty snacks, baked sweets, and soft-drinks ≥3 times per week, and breakfast at home < 5 days in the school week. The OFT group had significantly higher odds of purchasing food/drink from shops near school during school breaks and before/after school compared to the NEV group (adjusted odds ratio (aOR) = 1.80, 95% CI 1.07-3.01, and aOR = 3.61, 95% CI 2.17-6.01, respectively). The interviews revealed most students ate a home packed lunch, with the remainder purchasing either at the school canteen or at local shops. Conclusions: Students using the canteen often are frequently purchasing snacks and sugar-soft drinks from shops near school, most likely owing to availability of pocket money and an emerging independence. School authorities must focus upon satisfying canteen users by providing desirable, healthy, and affordable items in order to compete with the appeal of local shops.

Introduction As an important form of education, vocational education is not only related to individual career development, but also directly affects the economic prosperity and innovation ability of the society. With the continuous development and change of social economy, vocational education has gradually evolved from the traditional "spare choice" to an important way to cultivate practical skills and professional knowledge. In vocational education, students can gain knowledge, skills and experience related to practical work, preparing them for a smooth entry into the workplace. Although the importance of vocational education is gradually being recognized, family prejudice and discrimination against vocational education still exist in China. Many families are more inclined to see academic education as a more valuable option, while vocational education is often labeled as "inferior". At the same time, some foreign countries highly recognize the status of vocational education, providing students with a variety of career development opportunities [1]. The purpose of this study is to deeply explore the prejudice and discrimination of Chinese families against vocational education, and to find the causes of these problems and the impact of these problems on individuals and society by comparing with foreign vocational education. --- The Prejudice and Discrimination of Chinese Families Towards Vocational Education Family preference for academic education: The influence of traditional values on vocational education Chinese society has long held the traditional concept of attaching importance to academic education, and families generally believe that the traditional academic path provides better opportunities for their children's future development. This preference often stems from parents' expectations that their children will gain a higher social status and financial return by earning a higher degree. As a result, many families view vocational education with suspicion, viewing it as a "spare choice", believing that students who choose vocational education seem to be forgoing a more promising academic path [2]. Discrimination in non-traditional occupations: The influence of social perceptions on career choice Another factor that influences the bias of Chinese families towards vocational education is the social perception of discrimination against non-traditional professions. In traditional social concepts, certain professions are considered "decent" and "status", while others may be labeled as "inferior". This leads families to hold reserved or even negative views of some non-traditional occupations, such as beauticians, car repairmen, waiters, and so forth. Parents may object to choosing these occupations as a career direction for fear of the difficulties and social perception their children will face in these occupations. Reasons for the low status of vocational schools: misconceptions about the quality of education and career prospects The relatively low status of vocational schools in China is partly due to widespread family skepticism about the quality of education in these schools. Some parents believe that the quality of education in vocational schools is not as good as that in regular schools and worry that their children will not acquire enough knowledge and skills here. In addition, parents may misunderstand the poor job prospects of vocational school graduates and prefer their children to choose the traditional academic path [3]. --- Recognition and Development of Foreign Vocational Education --- The System of Foreign Vocational Education: Multiple Choices and Training Opportunities Compared with China, many foreign countries highly recognize and attach importance to vocational education. In these countries, vocational education is regarded as one of the important ways to cultivate all kinds of talents, and is widely respected as academic education. The diversified vocational education systems in foreign countries cover a variety of skills training in different fields, providing students with rich career options and development paths. The diversity of such systems helps meet the interests and needs of different students, enabling them to better develop their vocational skills [4]. --- The Balance Between Vocational and Academic Education: The Importance of Cultivating Diverse Talents Foreign vocational education systems usually emphasize the cultivation of vocational skills, but also pay attention to the cultivation of students' comprehensive quality and innovative ability. Compared with China, more emphasis is placed on the balance between vocational education and academic education in foreign countries, which is believed to help cultivate diversified and well-rounded talents. The idea reflects an education philosophy that places equal emphasis on "skills + qualities", enabling students to better adapt to the rapidly changing needs of society. --- The Role of Vocational Education in Economic Development: Providing Support for Social Innovation and Competition Some foreign countries have realized that the role of vocational education in promoting social innovation and economic competitiveness cannot be ignored. By cultivating qualified skilled personnel for all walks of life, foreign countries can maintain the competitiveness of their industries and enhance their overall innovation capacity. This is especially true in the fields of science and technology, engineering, medical care, and so forth. Foreign vocational education graduates have contributed a lot to the scientific and technological innovation and development of society. --- The Influence of Chinese Family Bias on Vocational Education --- Limitations of Career Development: The Conflict Between the Pursuit of Personal Interests and Family Expectations The prejudice of Chinese families towards vocational education often leads to limited personal career development. Many young people may be pressured by their families to pursue careers that are not in line with their interests and potential. Such choices that go against one's inner wishes may lead to career dissatisfaction and psychological stress, affecting one's career achievements and quality of life. At the same time, it may also lead some people to lack motivation and enthusiasm in their work, affecting their job performance and career development. --- The Talent Shortage Problem: Traditional Skills and Challenges Facing the Industry The bias of Chinese families towards vocational education has also led to some traditional skills and industries facing talent shortages. As society places more emphasis on academic education, many young people choose to pursue academic knowledge while neglecting the study of traditional skills. This has left some traditional industries and skilled positions facing a serious talent gap, affecting the development and innovation ability of related industries [5]. --- Limited Ability to Innovate: Exclusion of Non-traditional Occupations Affects Social Innovation and Change The bias of Chinese families towards vocational education may also limit social innovation and change. Innovation often comes from the intersection and collision of different fields, and some neglected non-traditional careers may be the source of innovation. However, because these occupations are subject to prejudice and discrimination, many individuals with innovative potential may be excluded from the innovation process, thus limiting the possibilities for social innovation and change. The prejudice and discrimination against vocational education in Chinese families not only restricts the career development of individuals, but also, to a certain extent, the development of the whole society. The next section will explore strategies and suggestions to solve these problems, with a view to promoting the development of vocational education and social progress in China. --- 5. --- Strategies and Suggestions for Solving Them --- The Importance of Education: The Key to Changing Family Attitudes In order to change the prejudice and discrimination of Chinese families towards vocational education, it is crucial to publicize education. The government, educational institutions as well as the media can join forces to convey the importance and value of vocational education to parents. Through publicity and education, parents can realize that vocational education is also an important way to train talents and promote social progress, so as to change their views on vocational education. --- Improve the Vocational Education System: Provide Diversified Training and Development Paths China should strive to improve its vocational education system to ensure its quality and diversity. Vocational education should cover a variety of skills training in different fields to meet the interests and needs of different students. --- A Well-rounded Education Philosophy That Encourages Personality Development and the Pursuit of Interests Fostering a well-rounded educational concept is the key to solving the problem of bias in vocational education. Education should pay attention to students' individual differences and encourage them to choose a career path according to their interests and potential. Students should be encouraged to develop their own expertise in vocational education and be aware of the importance of different professions in society. This will help reduce family and social discrimination and prejudice against specific professions [6]. Through the above strategies and suggestions, China can gradually change the problem of family prejudice and discrimination against vocational education, promote the development of vocational education, provide more career development opportunities for individuals, and provide strong support for social and economic growth and progress. --- Conclusion In this paper, the author deeply discusses the prejudice and discrimination of Chinese families to vocational education, and analyzes its comparison with foreign vocational education. Through a comparative analysis, the challenges faced by China's vocational education as well as the advantages of foreign vocational education are clearly seen. The prejudice and discrimination of Chinese families towards vocational education is rooted in traditional concepts and social culture, resulting in many young people giving up the choice of pursuing their interests and potential due to family expectations. This phenomenon not only affects the career development of individuals, but also hinders the innovation and progress of the whole society. In contrast, some foreign countries pay more attention to and recognize vocational education, providing students with diversified training and development opportunities, which is conducive to the overall growth of individuals and the development of society [7]. Through publicity and education, that can gradually change the family concept and make parents realize that vocational education is equally important. At the same time, by improving the vocational education system, we can provide students with more diversified development paths to meet different needs [8]. Finally, fostering a well-rounded education concept will help reduce discrimination against non-traditional professions and promote the development of individual interests [9]. We can gradually change the prejudice and discrimination against vocational education in families, promote the development of vocational education, and create a more favorable environment for individual career achievement and social progress [10]. This is a daunting task, but it is also a promising direction, and we are confident that we can achieve the all-round development of vocational education in China.

This paper aims to explore the prejudice and discrimination of Chinese families to vocational education. And through the comparative analysis with foreign vocational education, this research deeply explores the causes and effects of these problems. Through comprehensive literature analysis and case studies, this paper argues that the prejudice and discrimination of Chinese families to vocational education not only restricts the career development of individuals, but also restricts the development of the whole society. In contrast, the recognition and emphasis on vocational education in some foreign countries provides more career development opportunities for individuals and provides strong support for social and economic growth. In order to solve the problem of prejudice and discrimination of Chinese families towards vocational education, this paper puts forward some suggestions such as strengthening education publicity, improving the vocational education system, and cultivating the education concept of all-rounded development, aiming at promoting the development of China's vocational education and social progress.

Introduction Human immunodeficiency virus (HIV) is a member of the Retroviridae, a part of the Lentivirus family, based on its physical, biological, and genetic traits that cause HIV infection and acquired immune deficiency syndrome (AIDS). 1,2 In HIV infection, the pathological condition is started by attacking and destroying the host immune cells, such as T-helper, dendritic, and macrophage. [3][4][5] Acute HIV infection has a broad spectrum, ranging from hyperthermia to severe diseases. Thus, fever, fatigue, and myalgia are the most common symptoms of acute HIV infections. 6 The major complication of HIV infection is its progression to AIDS if left untreated, which appears if the HIV-positive patient gets a low CD4 count and becomes prone to opportunistic infections. 7,8 The most common ways to get HIV infection are through unprotected sexual contact and drug injection. [9][10][11] Also, blood transfusion can transmit HIV, but it has been eliminated in countries that apply the HIV screening system for donated blood. [12][13][14][15] Due to the availability of antiretroviral therapy (ART), which focuses on reducing the HIV viral load and maintaining the T-helper cells. Thus, normal life expectancy will be increased and progression to AIDS will be decreased by eliminating opportunistic infections. [16][17][18] HIV remains a prominent cause of mortality and disease burden in many countries, despite substantial advances in prevention, treatment, and care. 12,19 Globally, it was predicted that roughly 38 million individuals worldwide were infected with HIV. 20 This startling figure demonstrates the scope of the pandemic and its consequences for public health institutions and impacted communities. 12,20 The global HIV load is not uniformly distributed, with some places having a greater frequency of the virus than others. Sub-Saharan Africa, for example, has been disproportionately impacted by HIV, accounting for almost two-thirds of all people living with HIV worldwide. 21 In the European Region, HIV infection continues to harm the health and welfare of millions of individuals. In 2021, more than 100 thousand people will be newly diagnosed with HIV. 9,10 Saudi Arabia, with a population of more than 30 million, has been affected by the worldwide HIV pandemic. However, compared to other locations, the prevalence of HIV in Saudi Arabia has been reported to be lower than the worldwide norm. 14,22 Different countries with high HIV prevalence rates may face diverse challenges in controlling the spread of HIV, influenced by cultural norms, government policies, and community engagement in prevention and awareness efforts. 9,23 Sub-Saharan Africa, for instance, has encountered enormous obstacles in managing the pandemic due to economic restraints, limited access to free healthcare, and societal stigmas. 12,21 To curb the spread of the virus, the Saudi government, in the form of the Ministry of Health, implemented extensive preventative and control initiatives such as rigorous screening procedures for immigrants, workers, pilgrims, and visitors, as well as mandatory premarital screening program. 19,22,24 Also, as part of its HIV-fighting efforts, the Saudi government offers a proactive strategy and strives to guarantee that people living with HIV/ AIDS (PLWHA) get free access to ART and equal medical services to manage the infection properly. 22,24 Despite progress in understanding the virus and advancements in treatment choices, stigmatizing views against PLWHA have persisted since the epidemic's inception. 25 HIV-associated stigma can be defined as negative attitudes, beliefs, and actions directed toward PLWHA; it continues to be a serious issue in many areas of the globe. It stems from a lack of understanding, fear, prejudice, and societal norms. 26 The extent and character of stigmatizing views might differ among areas and populations. Some communities saw a drop in stigmatizing attitudes because of educational initiatives and advocacy work targeted at debunking myths and misunderstandings regarding HIV. 27,28 Stigma might have a serious impact on people living with HIV's quality of life; some affected subjects might get discouraged from obtaining HIV testing and treatment or not disclosing their status; therefore, social stigma may stymie attempts to manage the pandemic by deteriorating their health status, increasing the number of incidences, and increasing the mortality rate. 27,29 Medical practitioners are often the initial point of contact for people newly diagnosed with HIV; thus, they play a critical role in breaking the bad news and delivering effective care and support. However, researchers have reported that some medical practitioners might express hidden stigma toward people living with HIV. [29][30][31][32] In the Lao People's Democratic Republic, a study measured healthcare workers' discrimination toward PLWHA. Of 558 participants, 49.7% of healthcare professionals showed discriminating attitudes towards HIV/AIDS patients, while low levels of prejudice were associated with professionals having proper knowledge and experience in treating HIV subjects. 31 Also, a qualitative survey conducted in northwest England to investigate negative attitudes among dental practitioners found that one-third thought they were in jeopardy of acquiring the infection by treating patients. At the same time, one-fifth thought it was not their ethical responsibility to treat this high-risk group. 33 In the same aspect, a study done in Puerto Rico showed that most participants showed judgmental attitudes toward those patients due to the stigma that living a dissolute lifestyle is the cause of acquiring HIV infection. 29 The connection between medical professionals and PLWHA is based on trust, empathy, and efficient communication. In some conservative communities, they might be facing heavier duties to provide a safe and non-judgmental environment that helps to reduce the HIV stigma. 25 In a study done in Iran, 42.42% of the 575 healthcare providers were asked how they felt about HIV/AIDS patients. They claimed that prejudice against PLWHA was primarily a result of social stigma, religion, and ignorance of the disease's transmission. 32 Saudi Arabian society is classified as one of the largest conservative communities, where Islamic religion and cultural norms might highly affect people living with HIV. 22 Farahat et al demonstrated that higher multi-morbidity frequency is associated with increased age among PLWHA in Jeddah, western Saudi Arabia. 19 Since medical care is available for free, it is unclear whether social stigma is contributing to this increase in the rate. Medical students are the future doctors who treat HIV patients, whose treatment requires much knowledge and a positive attitude. Medical interns' stigma is underreported; thus, this study evaluates their knowledge and attitude toward people living with HIV in Jeddah, Saudi Arabia. --- Materials and Methods --- Study Design, Setting, and Study Period This cross-sectional study was conducted using a structured questionnaire to assess medical interns' knowledge and behavior toward PLWHA in Jeddah, Saudi Arabia, between January 2021 and March 2021. Graduates of five medical colleges were included in this study: College of Medicine at King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), the College of Medicine at King Abdul-Aziz University (KAU), the College of Medicine at Batterjee Medical College (BMC), the Ibn Sina National College for Medical Studies (ISNC), and the College of Medicine-Jeddah University (JU-COM). Three of the selected colleges are public, and two are private universities; all specialize in health sciences and are accredited by the Ministry of Education for various undergraduate degrees. --- Sample Size and Sampling Method This study was open to both male and female medical interns who had graduated from the schools listed and worked at hospitals in Jeddah for at least one month during the study period. So, it was decided that there are about 713 medical interns in Jeddah from five different colleges. The single proportion method will be used to figure out how many medical interns will be needed, with a figure of 50% of the population from previous research in Saudi Arabia. Cross-sectional research was used to figure out the sample size (n = the needed sample size, n = Z (/2) 2 pq / d2), and Raosoft, Inc. http://www.raosoft.com/samplesize.html was used to do the math. At the 95% confidence level (CI), the number of people who should know about HIV was estimated to be 50%, and the margin of error was set at 5%. The minimum sample size was determined to be 250. --- Instruments of the Study A structured questionnaire was developed and validated based on previously published surveys. The questionnaire has a demographics section and three themes: knowledge, attitude, and practice. It consisted of two sections: the first section included questions regarding the demographic profile; the second section had three multiple-choice questions intended to assess participants' knowledge; and the third section had two practice questions. The primary outcome variable was the knowledge score, which was taken as the total score and ranged from 0 to 16 points. In the first version of the third section, 12 (5-point Likert scale questions) were intended to measure medical interns' attitudes; three were positive, while the remaining were negative. For validation purposes, a pilot study was conducted using 89 college students. The questionnaire was validated by calculating the internal consistency and reliability test using Cronbach's alpha. In the first version, reliability testing was reported to be 0.333, considered low. After removing four negative questions, the internal reliability improved to 0.6842, which was acceptable A total of eight attitude questions related to HIV practice were represented and rated on a threepoint categorical scale as negative (8-13), neutral (14-26), and positive (27-40). The questionnaire was distributed among the intern's population using official email and was completed anonymously using the Google Forms platform. It was mandatory to complete all primary data variables to avoid missing or incomplete data in the analysis. 34 --- Statistical Analysis Qualitative data were reported as frequency and percentage, whereas continuous variables were presented as mean, standard deviation (SD), median, minimum, and maximum as appropriate. This cross-sectional study used the measure of association to identify the relationship between two or more variables. The Chi-square test of independence was performed to compare categorical data and assess the relationship between HIV knowledge and attitude scores and study variables. The Kruskal-Wallis test, a nonparametric test, was used to compare categorical or numerical variables. The significance level was set at <unk>0.05. Collected data were tabulated and statistically analyzed using JMP software (John's Macintosh Project), version 10.0 (SAS Institute Inc., Cary, NC, USA) for Windows, version 15. --- Results --- Participants' Demographic Characteristics A total of three hundred forty-six participants responded positively to the dispensed survey and were enrolled in this study. 186 (53.76%) were females, and 160 (46.24%) were males from five medical colleges in Jeddah city. Most subjects, 314 (90.75%), --- Participants' Knowledge of HIV Infection When asked how much they knew about the different body fluids that could spread HIV, they could choose more than one answer. Out of 1257 selections, more than half of the medical interns selected the correct answers. The most common fluids selected by participants were blood (328, 26.09%), semen (312, 24.82%), and vaginal secretions (302, 24.03%), but breast milk was only selected 213 (16.95%) times by medical interns, those who believed it could transmit HIV. Saliva, sweat, and tears were the least likely to be selected, as shown in Figures 1 and2. Also, the participants were asked to pick the correct answer among three statements regarding HIV transmission to evaluate their knowledge, which revealed that more than half of the medical interns, 216 (62.43%), agreed that "splashes with body fluids have a high risk for HIV transmission". However, for the last statement, "Fluid splashes to intact skin or mucous membrane", most participants (254, or 73.4%) did not agree. Also, 186 (53.76) of the medical interns agreed with the statement, "Percutaneous (needle-stick) injury has a risk of less than 1%", as shown in Table 2. Also, the participants were asked about the correct statements regarding what they would do if they experienced a needle-stick injury from a patient with unknown HIV status. 228 of the medical professionals will check the patients' --- 574 HIV status, and 220 will use post-exposure prophylaxis (PEP). In addition, the majority had selected to report the exposure to infection control, while 132 would assess the patient's HIV exposure risk. Furthermore, on-The-spot treatment for the injury and squeezing of the wound was the least to be selected, as shown in Figure 3. Hepatitis 278 was chosen by 80.35% of the people who took part in the study as the most common co-infection among people with HIV. The second one selected was tuberculosis (TB) at 45 (13.01%). The least common co-infections were Chlamydia 14 (4.05%) and Gonorrhea 4 (1.16%), as illustrated in Table 3. --- Participants' Attitude Score The attitude section of the survey was measured using a Likert scale ranging from strongly disagree to agree, as shown in Table 4. When participants were asked whether relatives or spouses should be notified regardless of the patient's consent, 157 (45.38%) of the participants strongly disagreed, while 51 (14.74%) strongly agreed. Of the 346 participants, 138 (39.88%) refused to share their utensils and toilet with HIV-positive subjects. Moreover, 176 (50.87%) said they would keep the information about HIV-positive colleagues confidential. However, more than half of the participants (58.09%) believed an PLWHA should be denied a leadership position and banned from receiving some benefits. In comparison, 66 (19.08) would not mind PLWHA having leading positions. Additionally, 244 (70.52%) did consider treating HIV-infected patients a waste of resources. Moreover, more than half of the participants, 172 (49.71%), strongly disagreed that they had the right to know the PLWHA status as a health worker. Also, 93 (26.88%) of the participants strongly agreed with the requirement of the patient's consent before testing for HIV infection. At the same time, the proportions of the strongly agreeing participants and those who were neutral about labeling the beds of PLWHA were close, at 82 (23.7%) and 104 (30.06%), respectively. The chi-square test of independence and fisher exact were used to find out if there was a link between gender and level of attitude. As outlined in Table 5, there was a statistically significant association between gender and attitude level (X 2 (2) = 8.553, p = 0.0139*). The association was strong, and we can reject the null hypothesis and accept the alternative hypothesis, women have had more negative attitude than men, as shown in Table 5. --- Discussion Despite advances in increasing awareness and lowering stigma, many areas and groups continue to face stigmatizing views concerning PLWHA. 25 Medical professionals have an important role in providing medical treatment and emotional support to help people achieve optimum health outcomes and respect their dignity and autonomy. 32,33 This study assessed medical interns' knowledge and attitudes concerning PLWHA in Jeddah, Saudi Arabia. Our findings suggest that medical interns who had relatively poor knowledge about HIV prevalence, transmission, and treatment had a lower attitude than those with better knowledge who did not show any negative attitude. Even though healthcare systems are different around the world, it has been shown that both society and healthcare providers care about PLWHA in the same way. 30 There are several variables that contribute to HIV-related stigma, including medical interns' misconceptions about how HIV is spread. 26 The study showed that the issue of transmission routes was one of the areas in which interns had relatively common knowledge. Most of the subjects correctly identified the main transmission routes, such as unprotected sex (94.57%), blood and body fluid exchange (94.19%), and sharing needles or syringes (91.47%). However, there was a clear misconception and overestimation about the role of other transmission routes. For example, they selected kissing (14.73%), mosquito bites (3.88%), coughing (0.78%), and hugging (0.78%) as routes of transmission. This can cause concern, as this lack of knowledge might lead to undue apprehension during clinical practice. Most participants were aware of which bodily fluids may transmit HIV. For instance, 94.96% of the interns knew that blood might transmit HIV, which is comparable to that reported in 2018 by Neerajaa et al, where 94% of participants believed HIV was avoidable, 89% recognized the transmission mechanism, 90% understood transmission control, and 84% had a positive attitude. 35 However, our results are less than those of Platten et al in final-year medical students (100%). 36 Aside from blood exposure, which poses the greatest risk of HIV transmission, other bodily fluids are also considered infectious, such as breast milk. Only 61.63% of the participants were aware that breast milk could transmit HIV. These results were less than those reported by Neerajaa et al in a medical intern. Saliva, tears, and sweat are not contagious unless bloody. 35 Only a few of the participants were unaware of this fact, as only 25.58%, 3.49%, and 1.94% of the interns gave incorrect answers, respectively. Upon asking about the most common co-infection among HIV patients, the results were not promising. Only 13.01% of the medical interns identified TB as the most common coinfection with HIV. On the other hand, a study done in Vietnam on final-year medical students revealed better results. (90%) subjects identified TB as the most common co-infection with HIV. 36 In our study, upon being asked what they should do if they got a needle-stick injury at work from a patient with unknown HIV status, 67.44% of interns said they would use post-exposure prophylaxis, which is comparable to the results revealed by Kasat et al in interns (68.8%). 33 Also, 43.41% of our interns said they would assess patient's HIV exposure risk, while 56.59% said they would not. This relatively low awareness regarding the proper management of HIV exposure was found among participants in this study. Other studies may be attributable to multiple reasons, such as low clinical experience, HIV prevalence in the era, and an emphasis on HIV in the curriculum. Screening for co-infection among HIV patients must be done earlier, as the CDC recommends. Hence, a lack of knowledge will compromise the patient's quality of care and lead to many complications. 37 38 As a result, the assumption that positive cases of HIV/AIDS are thought to have been acquired due to sexual misconduct is unacceptable in Islamic society. 24 This study shows that medical interns stigmatize PLWHA. This result supports a similar study to evaluate the knowledge and attitude of healthcare providers in Saudi Arabia. 24,38 The stigmatizing attitude could be attributed to the social stigma and the religious and conservative nature of the Saudi community, as HIV is associated with drug abuse and promiscuity. This appears as (31.1%) and (22.9%) strongly agree and agree, respectively, that they would feel more sympathetic toward people who get AIDS from blood transfusions compared to IV drug users (IDU). 37 This manifested well in another study, where physicians showed less sympathy toward IDU and extramarital sex. 38 The interns also showed some positive attitudes, as more than half of the sample (56.2%) would not isolate the beds of PLWHA. In contrast to another study, 78.42% of the sample asked for a different ward for HIV patients. 38 This difference could result from a lack of knowledge or from the different samples, as this study targeted medical interns and postgraduates. Lack of knowledge of transmission routes could be a reason for the stigmatizing attitudes. This manifested itself in this study in that a lot of people, collectively, would not share utensils or toilets with the PLWHA. However, the link between stigmatizing attitudes and a lack of knowledge contradicts one study while being supported by another. 39,40 It is believed that the conservative community and the association of HIV with promiscuity and IDU play major roles in the stigmatization of PLWHA. This is consistent with the findings of another study. 41 --- Conclusions Stigma continues to be a significant impediment to successful HIV prevention, treatment, and care. This study showed that medical interns have insufficient knowledge regarding the HIV mode of transmutation, the most common co-HIV infection, and the proper post-exposure prophylaxis. Moreover, the study showed that interns in Jeddah have stigmatizing behavior toward PLWHA, and a lack of knowledge about HIV can explain the stigmatizing behavior toward HIV patients. To tackle HIV-related stigma, a multimodal strategy is required, such as extensive and accurate education programs to debunk myths and misunderstandings regarding HIV transmission, treatment, and prevention. To establish a more inclusive and compassionate society for PLWHA, governments, healthcare institutions, civil society groups, and individuals must work together to address workplace stigma. Enacting and implementing legislation that safeguards the rights of PLWHA may also assist in avoiding discrimination from medical practitioners and providing equitable access to healthcare, employment, and other medical services. Based on the results, more --- Institutional Review Board Statement The Institutional Review Board (IRB) and the Research Board of King Abdullah International Medical Research Center (KAIMRC) approved this study design following the Helsinki Declaration's ethical criteria. Study Number: SP20/150/J, approval memo number: IRBC/0825/20, and the approval date is the 8th of June 2020. Before completing the anonymous survey, participants provided their informed written consent, no identifiable personal data was retained no biological specimens were collected. --- Informed Consent Statement Informed consent was obtained from all subjects involved in the study. --- Data Sharing Statement The data used to support the findings of this study have been deposited in the 4TU.ResearchData repository [10.4121/22225849]. --- Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. --- Disclosure The authors declare no conflicts of interest.

Background: Discrimination by some healthcare providers toward people living with HIV/AIDS has been documented. Differences in cultural backgrounds make it harder for future doctors, who need a lot of knowledge and a positive attitude to treat patients. In conservative countries like Saudi Arabia, not enough is known about how much medical interns know about HIV and how they feel about people living with HIV/AIDS. Methods: From April to September 2021, this cross-sectional study use non-probability random sampling and utilized a selfadministered questionnaire to collected the data from 346 medical interns who had graduated from five different medical schools. Results: Most of the subjects correctly identified the main transmission routes, such as unprotected sex (94.57%), blood and body fluid exchange (94.19%), and sharing needles or syringes (91.47%). But they did not know what the most common co-infections were for HIV patients or how to protect themselves after exposure. This paper showed that medical interns have some stigmatizing behaviors toward patients living with HIV, as 31.1% and 22.9% agreed, respectively, that they would feel more sympathetic toward people who get AIDS from blood transfusions compared to IV drug users (IDU). Conclusion: Medical interns also showed some positive attitudes, as more than half of the sample (56.2%) would not isolate beds for people living with HIV/AIDS. The study's conclusion is that HIV education and training programs should be added for medical interns, which might have a significant positive impact on their attitude.

Although the exact number of midlife and older LGBT (lesbian, gay, bisexual, and transgender) adults is unknown, LGBT older people are twice as likely to live alone, twice as likely to be single, three to four times less likely to have children than are heterosexual older adults (Services & Advocacy for GLBT Elders [SAGE], n.d.). This lack of immediate family systems can render older LGBT adults particularly vulnerable to social isolation and its consequences compared with their heterosexual counterparts. A U.S. national health survey targeting LGBT older adults reported that 59% of older LGBT adults feel that they lack companionship, 53% feel isolated from others, and 53% feel left out (Fredriksen-Goldsen et al., 2011). In comparison, according to a national survey, a little over one-third of midlife and older Americans (age 45+) were lonely (AARP, 2010). Social isolation has detrimental effects on health and well-being. For this reason, eradicating social isolation was chosen by the American Academy of Social Work & Social Welfare as one of the 12 "Grand Challenges for Social Work." As their website asserts, "Social isolation is a silent killer-as dangerous to health as smoking." This assertion is supported by House (2001), who found that the health risks of social isolation are comparable to the detrimental effects of both smoking cigarettes and obesity. Social isolation can lead to cognitive decline, coronary heart disease, and stroke (Béland, Zunzunegui, Alvarado, Otero, & Del Ser, 2005;Boden-Albala, Litwak, Elkind, Rundek, & Sacco, 2005). It can also generate negative psychological effects such as depression (Cacioppo, Hughes, Waite, Hawkley, & Thisted, 2006) and suicidal ideation (Eng, Rimm, Fitzmaurice, & Kawachi, 2002;Rapagnani 2002). In a meta-analysis on the influence of social relationship on the risk for mortality, Holt-Lunstad, Smith, and Layton (2010) found that participants with stronger social relationships had higher odds of survival. Commonly identified risk factors for social isolation within the LGBT population include individual-level factors such as living alone, childlessness, disability, poverty/ low income, mental health concerns, and stigma/discrimination (SAGE, n.d.). However, we know much less about how social environmental factors may affect midlife and older LGBT adults' perceived isolation. Although social support has often been identified as a protective factor for older adults (e.g., Krause, 1999;Tomaka, Thompson, & Palacios, 2006), the literature has inadequately identified social support factors unique to the LGBT population. We need to discover unique protective factors beyond immediate family systems for this vulnerable group because the identified protective factors of marriage and children have historically been less obtainable for this population. The current study begins filling this gap by answering the question: Does social inclusion, operationalized as aging service providers (e.g., senior centers, adult day care, transportation, employment services) who are welcoming, make a difference in midlife and older LGBT adults' perceived isolation, particularly for those who live alone? --- Living Alone and Social Isolation of LGBT Adults Despite the absence of questions about LGBT status in the U.S. census (Wang, 2017), findings from multiple studies have shown that LGBT older adults are less likely to have a partner and more likely to live alone compared with their heterosexual counterparts. Using a nation-wide sample, Fredriksen-Goldsen and colleagues (2011) found that gay and lesbian older adults were twice as likely as heterosexual elders to live alone. Wallace, Cochran, Durazo, and Ford (2011) found that half of gay/bisexual adult men (ages 50-70) in California lived alone, compared with 13.4% of heterosexual men. The proportion living alone among lesbian/bisexual adult women (more than 25%) was also higher than for heterosexual women (20%) (Wallace et al., 2011). Similar results were found in samples from New York, San Francisco Bay Area, and Los Angeles (SAGE & MAP, 2010). Transgender people may suffer even more from lack of social support, as it has been reported that medical providers often require candidates for sex reassignment surgery to divorce their spouses (SAGE & MAP, 2010). LGBT older adults are also less likely to rely on children compared with their heterosexual counterparts. A large New York study found that LGBT elders were four times less likely to have children to assist them (Brookdale Center on Aging, 1999). This is partly due to the fact that historically, LGBT people have been denied the right have children (Montero, 2014). Although same-gender couple adoption was legalized nationally in 2016, the process was long and arduous. As of 2014, only 21 states and the District of Columbia allowed same-gender couples to adopt jointly (Montero, 2014). The lack of partner and children and consequently the possible social support from the immediate family system renders this group of older adults particularly vulnerable to social isolation and its consequences. Kim and Fredriksen-Goldsen (2016) found that the level of loneliness, a subjective measure of social isolation, is much higher among LGB older adults (age 50-70) who live alone compared to those living with a partner/spouse when controlling for demographic characteristics and health conditions. Grossman, D'Augelli, and Hershberger (2000) also demonstrated that the level of loneliness for LGB older adults living alone is much higher than for those living with others. Additionally, Fokkema and Kuyper (2009) found that LGB older adults were significantly lonelier and had less social support than heterosexual older adults in the Netherlands. --- The Lens of Social Exclusion Social exclusion theory was first introduced in 1974 by René Lenoir (de Haan, 1998). It asserts that not only material resources but also engagement with and inclusion into the society are necessary for people to be integrated into the mainstream. As Duffy (1995) stated, "social exclusion is a broader concept... encompassing not only low material means but the inability to participate effectively in economic, social, political and cultural life, alienation and distance from the mainstream society" (p. 4). Social exclusion theory has been extensively used to study people in marginalized groups, based on a multidimensional perspective (Muddiman, 2000). The social exclusion perspective has been used to study both the older adult population and LGBT population but not the intersection of the two groups. For instance, Lee, Hong, and Harm (2014) used the social exclusion perspective to explain poverty among Korean immigrant older adults. They operationalized social exclusion into three dimensions-exclusion from social and civic life, exclusion from asset building, and exclusion from the labor market-and found that social exclusion contributes significantly to Korean immigrant older adults' odds of living in poverty. Within the realm of LGBT literature, discrimination, as one dimension of social exclusion, has often been singled out as a predictor of negative health outcomes (Khan, Combaz, & McAslan Fraser, 2015). For instance, Bostwick, Boyd, Hughes, West, and McCabe (2014) found that sexual orientation discrimination was associated with higher odds of mental health disorder in comparison with other types of discrimination. --- LGBT-Welcoming Aging Service Providers LGBT older adults often report feeling unwelcome at senior centers and other aging service providers (SAGE & MAP, 2010). Unwelcoming aging service providers may not be equipped to address discrimination and often fail to reach out to the aging LGBT population (Hughes, Harold, & Boyer, 2011;SAGE & MAP, 2010). In fact, whether a service provider is welcoming to its LGBT clients is a crucial component in assessing the cultural competency of these service providers (e.g., Portz et al., 2014). A culturally competent service provider would demonstrate their welcome to the LGBT population, ensure that LGBT older adults in the community are aware of their services, and provide appropriate services to this population (SAGE & MAP, 2010). Recently, there has been an increase in calls for service providers to engage in cultural competency training to serve their LGBT clients (e.g., Bradford et al., 2016;Gendron et al., 2013), which is an urgent need because multiple studies have documented that LGBT people utilize social services such as mental health services at higher rates compared with their heterosexual counterparts (e.g., Cochran, Sullivan, & Mays, 2003;Stanley & Duong, 2015). Because LGBT older adults are more likely to live alone and less likely to have social support from the immediate family system, the support they gain from the environment, the aging service providers in our study can be an important supplement to their social support system. For example, many older adults go to senior centers on a daily basis to socialize and to utilize their services. When a senior center is unwelcoming to LGBT clients, these LGBT older adults may need to hide their identity and live under daily stress and in fear of discrimination and denied service. In comparison, a welcoming senior center cannot only provide a safe environment to LGBT older adults but also facilitate educational programs for their clients, in general, to help combat homophobia in their community. A welcoming service provider is also a sign of social inclusion of marginalized populations by the mainstream society. In this study, aging service providers refer to the aging and adult service providers listed by the North Carolina Health and Human services (NC-DHHS, n.d.), such as senior centers, adult day care, home-delivered meals, in-home aide, transportation, case management, respite, and employment services (NC-DHHS, n.d.). Due to the general lack of attention of aging service providers to LGBT older adults, to our understanding, no studies have examined the contextual factors of social isolation related specifically to aging service providers or agency settings. --- The Current Study Through the lens of social inclusion, the current study begins to fill that gap. Particularly, we examine whether having experienced welcoming aging service providers (e.g., senior centers, adult day care, transportation, employment services) can serve as a protective factor for perceived social isolation. We also examine whether it can buffer the negative impact of living alone on the perceived isolation of midlife and older LGBT adults. We have two hypotheses. Hypothesis 1: Having encountered welcoming aging service providers is associated with reduced odds of feeling isolated among midlife and older LGBT adults. Hypothesis 2: Having encountered welcoming aging service providers moderates the negative impact of living alone among midlife and older LGBT adults. --- Methods --- Data and Sample The data were collected through a needs assessment survey of the aging LGBT community in North Carolina. The survey was conducted by the Center on Aging and Education at the University of North Carolina at Chapel Hill in 2013. Due to the sensitive topic and the assumption that many midlife and older LGBT adults conceal their sexuality or gender identity, a nonprobability sampling strategy was adopted. Adults age 45 and over who self-identified as lesbian, gay, bisexual, or transgender (LGBT) were recruited through e-mail announcements or fliers distributed at several formal and informal organizations, including nonprofit LGBT organizations, multipurpose senior centers, retirement communities, self-help groups, and meet-up groups for people who self-identified as LGBT. Participants were also encouraged to forward the online survey to their friends who met the study criteria. The survey yielded 240 valid responses. After excluding those to whom the welcoming service provider question did not apply, the final sample size was 222. Due to the nonprobability sampling strategy, our sample was predominantly White (93%) and non-Hispanic (96%). In comparison, 71.6% of North Carolina's population was White, 22.6% was Black/African-American, and 8.5% was of Hispanic/Latino origin (U.S. Census Bureau, 2013). The majority of our samples (67%) were from the Research Triangle area of North Carolina (Durham, Orange, and Wake County). We were aware that due to the limited resources of the Center on Aging and Education, we were not able to reach out to more areas, particularly the rural areas of North Carolina, where LGBT elderly may be particularly susceptible to social isolation due to higher poverty rates than urban areas and lack of resources in general (U.S. Department of Agriculture Economic Research Service [USDA-ERS], 2017). --- Measures --- Outcome Variable Perceived isolation was originally a three-category variable asking respondents whether they felt isolated, somewhat isolated, or not isolated. Due to a small sample size in the "felt isolated" category (N = 22), compared to "felt somewhat isolated" (N = 76) and "Not isolated" (N = 137), we recoded the outcome into a dichotomous variable, with 1 indicating feeling isolated or somewhat isolated and 0 indicating not feeling isolated. This coding scheme allowed us to draw practical clinical implications from the findings. Additionally, this coding scheme is also preceded by O'Donnell, Jabareen, and Watt (2010). --- Key Independent Variables Welcoming aging service providers was a dichotomous variable, indicating whether respondents reported that they had encountered any aging service provider in their community that they considered as welcoming to LGBT people, with yes coded as 1 and no coded as 0. Aging service providers included but were not limited to, senior centers, adult day care, home-delivered meals, in-home aide, transportation, case management, respite, and employment services. Living alone was a dichotomous variable indicating whether or not respondents were living by themselves, with yes coded as 1 and no coded as 0. --- Control Variables Follow the strategies of selecting control variables by Bernerth and Aguinis (2016), we controlled for potential confounding variables for perceived isolation and living alone. LGBT status was a categorical variable, indicating whether respondents identified themselves as gay, lesbian, bisexual, transgender, or other such as "gender non-conforming." Outness (openness about LGBT status) was a scale indicating to what extent respondents were open about their sexuality or gender identity in six areas: family of origin, friends, primary health care provider, neighbors, faith community, and the workplace. Answers about their openness to each group were coded as 0 (not open), 1 (to some degree), and 2 (open). The final outness scale derived from the mean of six scores, ranging from 0 to 2. Consulting a mental health professional was a dichotomous variable indicating whether respondents had consulted a mental health professional (i.e., a social worker, therapist, and psychologist) in the past 2 years. Hate violence was also dichotomous indicating whether respondents had experienced harassment, abuse, or violence in their lifetime, which they felt was related to their sexual orientation or gender identification. We also controlled for three demographic variables for selection bias. Employment status had three categories: employed, retired, and unemployed. Age had three categories: age 45-54, 55-64, and 65+. Income had six $20,000 intervals, ranged from under $20,000 to $100,000 and over, and was treated as a continuous variable. --- Analysis Procedure We first examined the missing values and no detectable missing patterns were found. Because we had a relatively small sample size, we deemed that multiple imputation was not appropriate and we handled missing cases (N = 23) with listwise deletion. Logistic regression was adopted to model perceived isolation. We assessed normality and multivariate linearity to ensure that both conditions were met for the independent measures. The modeling process involved three steps. First, we assessed the zero-order relationships between having experienced welcoming aging service providers and living alone with perceived isolation. Second, other control variables were added. In our final step, we examined the interaction between welcoming aging service providers and living alone. According to Williams (2010), in nonlinear models, a significant interaction term cannot be directly interpreted as evidence of differences among groups, as it can also be due to differences in residual variance. Thus, we fitted a heterogeneous model first to examine whether there was any difference between those who lived alone and those who lived with others. The results showed no heterogeneity problem with our model. The results yielded by logistic regression including the interaction term were the same as those from the heterogeneous choice model. The significant interaction term between welcoming aging service providers and living alone was added to the final model. --- Results Table 1 presents the descriptive statistics for the sample. Our sample was largely composed of gay (N = 113) and lesbian persons (N = 100), leaving a small number of bisexuals (N = 6), transgender people, and people who chose to not disclose their identity (N = 17). Table 2 presents the results of three logistic regression models. Model 1 showed that the odds of feeling isolated for those who experienced welcoming aging service providers were 61% less than for those who had not experienced welcoming aging service providers. On the other hand, for those who lived alone, the odds of feeling isolated were 2.45 times larger than the odds for those who were not living alone. In Model 2, when control variables were added, both welcoming aging service providers and living alone remained statistically significant (OR = 0.45, p <unk>.05; OR = 2.32, p <unk>.05). There was no difference between lesbians and gays regarding perceived isolation. We do not interpret the coefficients of the other sexual minority groups due to their exceptional small sample sizes. Being open about their sexual orientation/gender identity largely reduced the odds of perceived isolation for midlife and older LGBT people (OR = 0.08, p <unk>.01). Having consulted a mental health increased the odds of perceived isolation by 1.49 times (OR = 2.49, p <unk>.05). Having experienced hate violence was also positively associated with perceived isolation (OR = 2.29, p <unk>.05). Employed or retired LGBT persons had much lower odds of feeling isolated compared with the unemployed (OR = 0.07, p <unk>.05; OR = 0.05, p <unk>.05). Neither income nor age was associated with perceived isolation. In Model 3, the interaction term between welcoming aging service providers and living alone was statistically significant (OR = 0.21, p <unk>.05). For those who lived with others (living alone = 0), the effect of welcoming aging service providers was not statistically significant (OR = 0.78, p =.59). For those who lived alone, the effect of welcoming aging service providers was statistically significant. The corresponding odds ratio was calculated by multiplying the odds ratio of the interaction term (0.21) with the odds ratio of welcoming aging service providers (0.78), which was 0.16. This means that for those who lived alone, having experienced welcoming aging service providers reduced the odds of perceived isolation by 84%. In a similar fashion, we interpreted the odds ratio of living alone for two separate groups: those who had experienced welcoming aging service providers and those who had not. For those who had not experienced welcoming aging service providers, the odds of perceived isolation for those who lived alone was 5.3 times of those who were not living alone. For those who had experienced welcoming aging service providers, the odds ratio of perceived isolation for those who lived alone was largely reduced to 1.11 times (5.30 <unk> 0.21) that of those who lived with others. To graph the interaction effect, we obtained the predicted probabilities of perceived isolation for the combination of welcoming aging service providers and living alone (see Figure 1). --- Discussion The findings of this study provided fresh information about protective factors against isolation among the aging LGBT population at the environmental level, that is, beyond the circle of family and friends. Our findings suggest that for midlife and older LGBT who are living alone, having welcoming aging service providers in their areas can be very beneficial. One advantage of our study is that we were able to control for an LGBT person's outness (openness about their LGBT status). Although being open about LGBT status is often considered a positive move for reducing self-stigma (Corrigan, Kosyluk, & Rüsch, 2013), openly identified LGBT persons risk being discriminated against. Welcoming aging service providers can serve as a buffer against the negative impact of living alone and make these people less likely to feel isolated. A welcoming service provider who demonstrates their understanding of and inclusiveness toward the LGBT groups' needs may well serve as a substitute for close families or friends for people who live alone. The poignant fact that LGBT people are more likely to live alone, be single, and be childless makes our findings timely and critical for setting a new direction towards intervention development targeting socially isolated aging LGBT population. In order to reduce the perceived isolation of this population, we can focus more attention on the cultural competence training of service providers. This new intervention approach is urgently needed because existing interventions targeting social isolation have largely ignored the environmental or community level factors. The "traditional" intervention tends to focus on individual older adults, either through group-level or through individual-level activities. However, not all interventions are effective in reducing older people's social isolation. In a systematic review, Cattan, White, Bond, and Learmouth (2005) found that eight out of 18 interventions were ineffective, among which, six focused on a one-on-one support (Cattan et al., 2005). Even for interventions demonstrated to be effective in the short-term, the long-term effect remains unknown, and each intervention reaches a limited number of older adults. By contrast, cultural competence training to health and social service providers can potentially contribute to reduced isolation for substantial numbers of clients for years to come. Furthermore, multiple studies have identified that midlife and older LGBT adults have substantial barriers to accessing health care and social services. For example, King and Dabelko-Schoeny (2009) found that midlife and older LGBT adults are five times less likely to access health care and social services. Metlife Mature Market Institute, & The Lesbian and Gay Aging Issues Network of the American Society on Aging (2010) reported that almost half of lesbian and gay Baby Boomers lack the confidence to disclose their sexuality or gender identity to health care providers for fear that they would not be treated with dignity and respect. Nonetheless, health and social care have been consistently found to be one of the highest ranking needs among British midlife and older LGBT population, along with housing and other services (Addis, Davies, Greene, MacBride-Stewart, & Shepherd, 2009). Similarly, Jenkins Morales, King, Hiler, Coopwood, and Wayland (2014) found that both LGBT baby boomers and the silent generation in the greater St. Louis area report that they have barriers to health care and legal services. Because of this high level of concern about health care among the aging LGBT population, cultural competence in serving this vulnerable population is pivotal. A culturally competent service provider cannot only reduce perceived isolation for LGBT older adults but may also increase their likelihood of accessing services, therefore potentially reduce the health disparities that this group of older adults face compared with their heterosexual counterparts. Yet, in a study assessing the capacity for providing culturally competent services to LGBT older adults in an urban neighborhood in Denver, Colorado, Portz and colleagues (2014) found that only four of the agencies belonged to the category of "high competency," 12 were considered "seeking improvement," and 8 were considered "not aware." Although this only represents one community, it suggests a large untapped need for such training. In the current sociopolitical climate, service providers may show a lack of sensitivity and even overt discrimination toward LGBT people. In spring of 2017, South Dakota became the first state to pass an anti-LGBT bill, which severely limits the services that LGBT people can receive. It is predicted that a swath of state level anti-LGBT bills will follow in the near future (American Civil Liberties Union, 2017, March 10). Our findings provide timely evidence that implies the harmful impact this new wave of anti-LGBT bills will have upon older adults. As our model shows, a lack of welcoming aging service providers is associated with a much higher probability of perceived isolation for those who live alone. Considering the detrimental effect of social isolation for older adults' health and well-being, welcoming aging service providers are essential in LGBT people's lives. --- Limitations and Future Directions This study has several limitations. First, due to the sensitivity of approaching this hard-to-reach population, our sample is not representative even of North Carolina, much less the nation as a whole. Our sample is not representative of racial and ethnic minorities or people with lower income. In addition, the most isolated LGBT elders would be the most likely to be missed in a sampling strategy that depends on publicity in various group settings and/or referral by friends and acquaintances who know of potential recruits' LGBT status. As we have stated previously, most isolated LGBT older adults may live in rural areas which our survey was not able to reach due to limited resources. Future studies can target rural areas as the LGBT older adults in these areas may need social support the most. Second, we are aware of disparities in social isolation among the LGBT groups. However, due to small sample sizes of the bisexual, transgender, and "other," we were not able to examine these groups statistically. Nonetheless, to our knowledge, our study is the first to examine the relationship between perceived isolation and having experienced welcoming aging service providers. Future larger-scale research on a nationally representative LGBT sample could not only test these findings but allow a more dependable measure of whether there are racial or ethnic variations. Third, our data are cross-sectional. Therefore, no causality can be inferred from our findings. Finally, our data may contain self-report biases. Due to the lack of personality variables, we were not able to control for individual optimism/pessimism differences in terms of perceived welcome from service providers. However, we also believe that it is important and necessary to examine subjective experience of welcoming aging service providers simply because there is no standard measurement of a welcoming service provider existing yet. Future research can adopt experimental design in order to examine causality. Meanwhile, even with the limitations of these data, findings suggest that the development and testing of interventions to improve the LGBT sensitivity and competence of service providers would be an extremely valuable next step in the research agenda. --- Conclusion Using community needs assessment survey data, this study identified unique protective factors for midlife and older LGBT adults' against perceived social isolation. Logistic regression results showed that after controlling for demographics and other potential confounders, particularly an LGBT person's outness (openness about LGBT status), having encountered welcoming aging service providers is a protective factor against perceived isolation, and it also buffers the negative impact of living alone. These findings provided preliminary evidence for a new direction of intervention development targeting cultural competence training for service providers. --- Conflict of Interest The authors declare no conflict of interest.

Background and Objectives: Older lesbian, gay, bisexual, and transgender (LGBT) adults are more likely to live alone and less likely to have children compared with their heterosexual counterparts. The lack of immediate family system can render older LGBT adults particularly vulnerable to social isolation and its consequences. The current study utilizes social exclusion theory, which asserts that not only material resources but also engagement with and inclusion into the society are necessary for marginalized people to be integrated into the mainstream. The study examines whether aging service providers (e.g., senior centers, adult day care, transportation, employment services) who are perceived by older LGBT adults as welcoming to LGBT people may reduce this population's perceived isolation. Research Design and Methods: Data were collected through a needs assessment survey designed for the aging LGBT community in North Carolina. Adults aged 45 and over who self-identified as LGBT were recruited at several formal and informal groups. The survey yielded 222 valid responses. The outcome variable was perceived isolation. Key independent variables included having experienced welcoming aging service providers and living alone. Results: After controlling for potential confounders and demographics, logistic regression results showed that having experienced welcoming aging service providers was a protective factor against perceived isolation and it also buffered the negative impact of living alone. Discussion and Implications: The findings provided preliminary evidence for a new direction of intervention research-targeting LGBT cultural competence training for medical and social service providers.

Introduction The provision of optimum care for the aging population is dependent on the understanding of their views and values on healthcare issues, especially end-of-life issues. Advance care planning (ACP) is a process of reflection, discussion, and communication that enables a person to plan for their future medical treatment and other care, for a time when they are not competent to make or communicate decisions for themselves [1]. ACP generally takes one of two forms: (1) the advance directive or "living will," a mechanism that allows individuals to catalog their preferences for future healthcare; (2) the durable power of attorney for healthcare or "healthcare proxy," a document that assigns a surrogate to make medical decisions on behalf of a patient in the event of decisional incapacitation [2]. It has gained prominence internationally for perceived benefits in enhancing patient autonomy and ensuring that patients receive appropriate, high-quality end-of-life care, as well as reducing stress, anxiety, and depression in surviving family members [3]. In the United States, the Patient Self-Determination Act of 1991 aims to encourage patients to take the initiative to ensure that their values are respected at the end of their life [4]. This legislation is letting US society think over urgency of the way to deal with end-of-life care and decision-making [5]. Nowadays, up to 70% of community-dwelling older adults in the United States have completed an advance directive [6], and data from the Health and Retirement Study showed that the elderly Americans who had prepared advance directives received care that was strongly associated with their preferences [7]. China is a country with the largest population in the world and is facing a growing aging population, increased incidences of cancer, and a huge number of terminally ill patients [8]. Therefore, it is imperative to develop palliative care and implement ACP in the country. In recent years, there has been growing evidence from Hong Kong suggesting that Chinese older people residing in residential care homes are well aware of the anticipated death and welcome the opportunity to discuss issues related to end-of-life care and death preparation [9]. However, despite substantial international literature on ACP [10], there is a paucity of information on the attitudes of the elderly citizens towards ACP in mainland China. It is with this view that we conducted this study with the following objectives: (1) to identify knowledge and preferences of ACP, as well as preferences towards truth telling, healthcare autonomy, and end-of-life care of older adults living in the city of Beijing; (2) to explore the factors associated with those preferences. --- Material and Methods It was a cross-sectional survey conducted from August 6 to September 3, 2014. The study was performed at two communities in Chaoyang District, Beijing. The interviewers had training and practice sessions prior to the interviews. The face-to-face, semistructured interviews were conducted in the existing seniors clubs in the two communities. The seniors clubs were chosen because they provided the researchers with access to a large number of older persons in a short time frame. A total of 1098 people were in attendance in the seniors clubs during the period in which we visited; however, some people did not take the survey. A total of 921 participants completed the survey with a response rate of 83.9%. Of them, 21 participants were excluded from the analysis due to incompletion of the survey. A total of 900 surveys were included in the final analysis. Participants enrolled in the study were aged above 65 years, were able to understand the content of the questionnaire, and could communicate well with the interviewers. They were also willing to participate in the study. Older adults documented to have delirium, dementia, aphasia, and drowsiness were excluded from the study. The questionnaire used in the study was reviewed and approved by the hospital directors and the hospital's ethical review board. It is composed of collection of participants' sociodemographic characteristics and five questions designed to delineate participants' awareness, preferences, and attitudes towards advance care planning, truth telling, healthcare autonomy, and end-of-life care. Sociodemographic characteristics including age, gender, religion, education level, marital status, living situation, monthly income, and social support conditions were collected. Social support conditions were reflected by whether older adults can get life support from their children and the number of close friends older adults have, from whom they can get support when they have difficulties. Before asking the questions, participants were informed about the concept of ACP: "Advance care planning is defined as a communication and decision-making process that allows individuals to clarify their values and preferences for future care, and enables them to communicate their wishes to loved ones, surrogate decision makers and healthcare providers" [11]. After this introduction, they were asked the following questions: "Do you want to hear the real news regarding your own condition from the physician (Q1)?"; "Would you prefer to make your own healthcare decisions (Q2)?"; "Have you ever heard of an advanced care planning before (Q3)?"; "Do you want to document in an ACP so that your own values and preferences will be respected in case you become seriously ill (Q4)?"; "Are you willing to endure specific life-prolonging interventions (such as chronic ventilator and feeding tube) to avoid death in terms of irreversible conditions (Q5)?" Questions 1 and 2 were designed to reflect participants' attitudes towards healthcare autonomy, while questions 3 and 4 were designed to reflect participants' awareness and preferences towards ACP. Question 5 was designed to reflect participants' attitudes towards end-of-life care. These questions required a response of yes/no. If a participant did not want to answer certain questions, he/she could choose the option of "refuse to answer," which was also recorded. --- Statistical Analysis Results were presented as means <unk> standard deviations for continuous variables. Categorical variables were reported as frequencies and percentages. Participants were classified into two groups: one group was those answering "yes" for each of the above-mentioned 5 questions, while the other group was those answering "no" for the questions. A Chi-square test was used to measure differences between the participants' sociodemographic characteristics in the two groups. Groups were tested for comparability on age, gender, religion, education level, marital status, living situation, income, and social support conditions. Values of P <unk> 0.05 were considered statistically significant. All the analysis was performed using SPSS software version 16.0 for Windows (SPSS Inc., Chicago, IL, USA). --- Results Table 1 shows the sociodemographic characteristics of the participants. The mean age of participants was 74.99 <unk> 6.53 years; 53% (n = 477) of them were female. 79.8% (n = 718) of them were married, while 20.2% were single/ divorced/widowed. 11.2% (n = 101) of the respondents lived alone. 70.4% (n = 634) of them listed being high school graduate or less for their education levels. Most of them had children (97.3%) and could get life support from their children (71.3%). The vast majority of them did not have a religion (97.6%). The respondents' options for the five questions mentioned above were as follows-Table 2. Of the respondents, 80.9% wanted to hear the real news regarding their own condition from the physician, while 52.4% preferred to make their own healthcare decisions. Only a very small number of them (8.9%) preferred to endure life-prolonging interventions when faced with irreversible conditions. When introduced to the concept of ACP, the majority (78.3%) of them had not heard of it, and only 39.4% wanted to document in an ACP, whereas 41.9% did not. 18.7% of the respondents refused to answer this question. According to results of the Chi-square test, Tables 34567, survey respondents with higher education level (some college or more) had significantly higher proportion of having ever heard of an ACP, as well as preferring to document in an ACP, compared to those with education level of being high school graduate or less (23.6% versus 13.3%, P <unk> 0.01; 60.3% versus 43.5%, P <unk> 0.01, resp.). Those aged <unk>70 years had higher proportion of having ever heard of an ACP, as well as refusing life-prolonging interventions to avoid death when faced with irreversible conditions, compared to those aged <unk>70 years (21.9% versus 14.0%, P <unk> 0.05; 91.0% versus 84.2%, P <unk> 0.05). There was no difference in preferences for the questions between men and women. No statistical differences were found for gender, marital status, religion, income, having children, and support from children between those who answered "yes" and "no" for the above-mentioned questions. --- Discussion According to our knowledge, this study was the first attempt to investigate community-dwelling older persons' preferences and attitudes towards ACP, truth telling, healthcare autonomy, and end-of-life care, as well as explore the factors associated with those attitudes in mainland China. Consequently, such data will provide insight into our understanding of advance care planning and healthcare preferences of elderly community-dwellers in mainland China and serve as a foundation for later investigations on suitable ways of implementing ACP. A low level of ACP awareness and preferences was apparent; very few participants had ever heard of an ACP, which can help to achieve their autonomy. However, the favoring of truth telling and self-determination by our participants was relatively high. Most of the participants preferred to hear the real news regarding their own condition from physician, and more than half of them wanted to make their own healthcare decisions. Only a very small portion of the participants expressed that they would endure life-prolonging interventions in terms of irreversible conditions. Previous literature showed that traditional Chinese societies were strongly family centered [12,13]; healthcare decisions were often made by the family as a group, rather than by the individual, and the principle of autonomy played a lesser role in Chinese societies. In our study, encouragingly there was a significant proportion of elderly participants who favored healthcare autonomy and had negative attitudes towards lifeprolonging interventions in terms of end-of-life care. When introduced to the concept of ACP, less than forty percent of respondents expressed that they would prefer to document in it. One possible explanation is that the concept is very new to them or unfamiliar. In mainland China, end-of-life care and the concept of ACP are not widely known or taught in the medical profession, and there are even fewer public promotion activities on these topics. Over one-third of participants refused to answer the question of "Are you willing to endure specific life-prolonging interventions to avoid death in terms of irreversible conditions (Q5)?" A possible explanation may be that these participants were hesitant to talk about death and end-of-life related decisions or they considered the topic taboo or were uncomfortable discussing it. Factors influencing older people's attitudes towards ACP had been studied. We found that age and education level may be the influential factors. Survey respondents with higher education had significantly higher proportion of having heard of an ACP, as well as preferring to document in an ACP. Some previous studies also showed that higher education was associated with a greater awareness and preference for ACP [14,15]. Chinese, Filipino, and Japanese studies have shown that a higher education level and degree of acculturation are associated with more positive views towards planning and communication regarding the end of life [16,17]. From our perspective, participants with education level of being high school graduate or less may have reduced comprehension of ACP definitions and goals. Those aged <unk>70 years had higher proportion of having heard of an ACP, as well as refusing life-prolonging interventions when faced with irreversible conditions in our study. A possible explanation is that these younger respondents may be less influenced by traditional Chinese culture compared to those much older respondents; therefore they were more likely to accept the views of ACP. There have been very few studies regarding Chinese older adults' attitudes towards ACP, advance directives (AD), and palliative care. A cross-sectional study published in 2014 investigated advance directive and end-of-life care preferences among nursing home residents, which showed that most (95.3%) had never heard of AD, and only 31.5% preferred to make an AD. More than half (52.5%) would receive life-sustaining treatment if they sustained a life-threatening condition [18]. In a transnational survey, 62 patients from five hospitals of mainland China who are seriously ill were visited to study their attitudes towards end-of-life decisions. It showed that most respondents (80.3%) wanted to hear the truth directly if they were diagnosed with a terminal condition, but more than half of the respondents (55.7%) wanted continued treatment in irreversible conditions [19]. There have been even fewer studies focused on the attitudes of older Chinese people towards end-of-life decisions who migrated to and settled in Western countries. A Canadian study concluded that older Chinese people did not favor advance directives because they create negative thoughts by requiring people to contemplate their own demise [20]. Another study performed in the United States, however, showed that it is feasible to conduct a nurse-led educational seminar on ACP in a community-dwelling population of Chinese Americans; Chinese in this study were open to the topic and showed a willingness to learn about ACP [21]. A variety of factors that prevent elderly people from making advance care planning were identified in the literature, including irrelevance, reluctance to think about dying, lack of knowledge, feeling that planning is unnecessary because family knows what to do, and feeling that loved ones are unable or unwilling to discuss ACP [22,23]. In particular, there are some obstacles towards implementing ACP in mainland China for the following reasons. The first reason is associated with culture, which shapes the way people deal with illness, suffering, and death, as well as the communications and decisions related to ACP [24]. In traditional Chinese culture, Confucianism and the relative importance placed on an individual's relationship with their family and society have a deep influence on decision-making, especially at the end of life [25,26]. Family cohesion is highly valued and overrides the preferences and autonomy of an individual [27]. Moreover, the traditional Chinese superstition believed that death was a very sensitive issue, and mentioning it was sacrilegious and to be avoided [28]. Secondly, medical decision-making has been seen primarily as a duty of the family in order to protect the patient from the burden of making difficult choices about medical care [24]. It turned out to be that older adults are often excluded from decisionmaking. Despite the obstacles mentioned above, the endeavor to promote ACP and palliative care has been made in mainland China in recent years. The foundation of Beijing Living Will Promotion Association, which was approved by the Beijing Civil Affairs Bureau in June 2013, was a landmark event in this regard [29]. Since its foundation, the organization had launched a website, called "Choice and Dignity," to provide online advice on making living wills [30]. However, we should be aware that although there is a lot of evidence regarding ACP and end-of-life care discussions in Western countries, cultural attitudes towards such issues are different in Chinese societies [31]. Currently, there has been insufficient research to demonstrate the benefits of ACP to eastern Asian patients. For many Chinese older adults with the preference for making decisions as families and a relative unfamiliarity with ACP, the use of ACP may be discouraged. Moreover, patients' treatment preferences and values may change when their health changes, at the end of life and even during periods of stable health [32]. Therefore, it may be difficult and impractical to import ACP system and apply it directly in China. To implement ACP in mainland China, health professionals should conduct culturally specific advance care planning that is tailored to Chinese people's specific cultural attitudes and ethnic beliefs. Encouragingly, elderly people's favoring for self-determination as well as their negative attitudes towards life-sustaining treatment in terms of irreversible conditions in our study suggests that it is possible to initiate the topic concerning end-of-life care and ACP related issues to Chinese elderly citizens. Providing culture-sensitive knowledge, education, and communication regarding ACP is a feasible first step to promoting this health behavior in mainland China. It is important to point out the limitations of this study. First, the participants in this study were small convenience samples. Thus, the generalization of the findings requires caution. Second, associated conditions of the older adults such as multimorbidity, functional status, and prior exposure to illness were not incorporated into the study, which may also be important determinants in ACP and its perceived relevance to individuals. Finally, the questionnaire, especially the five questions in the questionnaire to delineate participants' awareness, preferences, and attitudes towards advance care planning, truth telling, healthcare autonomy, and end-of-life care in this study, was designed by us and there is no previous literature to demonstrate its validity. Moreover, we used some hypothetical clinical scenarios in our questionnaire, for which reason responses to the survey might not accurately reflect what individuals would choose in reality and validity of the questionnaire needs to be further assessed. --- Conclusion This study identified the preferences and attitudes towards ACP and healthcare autonomy of community-dwelling older adults living in Beijing. Although the majority of elderly community-dwellers in this survey appeared to have medical autonomy and preferred comfort measures in terms of irreversible conditions, a low level of "Planning ahead" awareness and preferences was apparent. Given that the concept of advance care planning and knowledge of palliative care are not well understood in China, more effort is needed to step up public education in this regard. Moreover, to implement advance care planning in mainland China, it should be tailored to individuals' needs and feasible in the Chinese healthcare system and culture. --- Conflict of Interests The authors declare that there is no conflict of interests. Submit your manuscripts at http://www.hindawi.com --- Stem Cells International Hindawi

Objectives. To investigate the preferences of ACP and healthcare autonomy in community-dwelling older Chinese adults. Methods. A community-based cross-sectional study was conducted with older adults living in the residential estate of Chaoyang District, Beijing. Results. 900 residents were enrolled. 80.9% of them wanted to hear the truth regarding their own condition from the physician; 52.4% preferred to make their own healthcare decisions. Only 8.9% of them preferred to endure life-prolonging interventions when faced with irreversible conditions. 78.3% of the respondents had not heard of an ACP; only 39.4% preferred to document in an ACP. Respondents with higher education had significantly higher proportion of having heard of an ACP, as well as preferring to document in an ACP, compared to those with lower education. Those aged <70 years had higher proportion of having heard of an ACP, as well as refusing life-prolonging interventions when faced with irreversible conditions, compared to those aged ≥70 years. Conclusions. Although the majority of community-dwelling older Chinese adults appeared to have healthcare autonomy and refuse life-prolonging interventions in terms of end-of-life care, a low level of "Planning ahead" awareness and preference was apparent. Age and education level may be the influential factors.

Since the 1990s, the Internet has been instrumental in the advancement of many social movements. An early and well-cited example of the effective use of computermediated communication (CMC) for activism and protest is the case of the Zapatista movement and the Zapatista National Liberation Army (EZLN) in their rebellion against the ongoing colonial repression of the indigenous people in Mexico (Collier, 1994;Cleaver, 1995). The Zapatista movement leveraged the Internet and CMC ''for the dissemination of information, the sharing of experience, and the facilitation of discussion and organizing... (and) for the amplification and archiving of the developing history of the struggle'' (Cleaver, 1999). Through their use of the Internet and CMC, the Zapatista movement demonstrated ''a new capacity for this and other social movements to communicate across borders and to operate at a transnational level'' and was considered a ''prototype'' for other movements (Cleaver, 1999). Related antiglobalization demonstrations inspired by the Zapatistas such as the J18 and the protests against the World Trade Organization in Seattle during 1999 similarly utilized the Internet and CMC to promote and coordinate their activism, and were also regarded as models of demonstration to be followed in subsequent years (Starr, 2000). Recognizing the importance of CMC to social movements, researchers have focused studies specifically on the use of the Internet by various movements, from grassroots activism (Tesh, 2002) to transnational mass protest movements (Clark and Themudo, 2006). Social movement activism mediated by the Internet is often referred to cyber-protest, and represents an emerging field of social movement research (Meikle, 2002;Jordan and Taylor, 2004;van de Donk et al, 2004). Cyber-protest is considered an extension of a social movement into a new media space; an articulation of the movement in the Internet (Castells, 1998). Cyber-protest is diverse, and many forms have been identified in the literature, from hactivism-electronic civil disobedience, as in the case of the 1998 blockade of the Pentagon website by the Electronic Disturbance Theater (EDT)-to culture jamming-tactical media using techniques of appropriation, collage, ironic inversion, and juxtaposition, as in the case of the ® TMark websites protesting the Bush presidential campaign of 2000 (Meikle, 2002). Although the Internet has empowered social movements to produce novel expressions of protest, they continue to rely upon the traditional tactics proven effective in the pre-Internet world adapted to a virtual environment (Meikle, 2002). For example, the EDT blockade was regarded by researchers as a virtual sit-in, a traditional and successful method for protest. Following the Zapatista model, the most abundant forms of cyber-protest leverage the Internet as a platform for communication, to disseminate information, share experiences, develop solidarity and identity, and promote, facilitate, and organize activism. The extensive usage of CMC by a social movement creates a widespread cyber-culture in the Internet, establishing many linked virtual communities for members to congregate and socialize (Meikle, 2002), forming rich and complex social systems (Jones, 1994), and representing fertile areas for conducting social research. While researchers have recognized the importance of the Internet to social movements and the value of studying virtual communities and cyber-protest to further their understanding, they have also identified challenges in conducting social research online. Manifestations of social movements in the Internet and their expressions of cyber-protest are dynamic, appearing and disappearing without warning, and distributed across multiple related websites, forums, or virtual communities, some of which may not be readily apparent to the researcher. Although they may include formal organizations as components, on the whole they are not an organization. A social movement typically lacks membership forms, statutes, chairpersons, and the like. It may expand or shrink considerably over periods of time, and exhibit phases of visibility and latency. Also, unlike political parties, social movements may have significant overlaps with other movements. Moreover, a social movement may quickly change its forms, strategy, tactics, and even some of its goals'' (van de Donk et al., 2004). The cyber-protest of social movements is ''fuzzy and fluid phenomena often without clear boundaries'', and a ''moving target, difficult to observe'' (van de Donk et al, 2004). The dynamic and open nature of social movements in the Internet, as described by cyber-protest researchers, presents unique challenges to conducting social research online. Researchers have proposed approaches to social research in the Internet, specifically attempting to address the challenges presented by the scale and complexity of the online environment. Paccagnella suggests that scholars conducting social research on virtual communities ''exploit the possibilities offered by new, powerful, and flexible tools for inexpensively collecting, organizing, and exploring digital data'' (Paccagnella, 1997). Such tools would facilitate the analysis of 'cultural cyberartifacts,' the member communications and creations representing the archived history of the virtual community. Jones (1997) presents a theoretical outline for conducting social research on virtual communities, adapting the perspectives of archaeology to CMC. Similar to Paccagnella, the approach focuses on the cultural cyber-artifacts of a virtual community, which provide an integrated framework for analysis to understand the life of the community and its members (Jones, 1997). Due to its focus on the study of cultural cyber-artifacts, the approach was called cyber-archaeology. Focusing on the cyber-artifacts of virtual communities and developing sophisticated tools for their collection and analysis, issues of scale and complexity surrounding cyber-protest may be resolved, enabling broad and longitudinal research on social movements. However, few approaches have been proposed in the literature with the technical sophistication and capabilities to assist in social research on modern virtual communities and forms of cyber-protest, and social researchers continue to be slow to react to social movements as they emerge (Edelman, 2001). In this paper, we present a cyber-archaeology framework for social movement research. We refer to the approach as cyber-archaeology, following Jones (1997), because of our focus on the cultural cyber-artifacts of cyber-protest. Our framework overcomes many of the issues of scale and complexity facing social research in the Internet, and specifically addresses many of the suggestions of Paccagnella (1997) enabling broad and longitudinal social research on virtual communities. The framework is not intended to replace the social researcher, or directly produce theoretical conclusions regarding social movements, but instead we offer a collection of powerful tools and techniques to assist in conducting social movement research in the Internet. The approach is automated and scalable, allowing a researcher to easily collect and analyze social movement cyber-artifacts across multiple linked virtual communities. The techniques for the analysis of cyber-artifacts specifically adopt perspectives of social movement theory. To demonstrate the cyber-archaeology framework for social movement research, and provide a detailed instantiation of the proposed approach for evaluation, a case study on a broad group of related social movement virtual communities is presented. --- Literature Review --- Social Movement Theory Social movements are informal networks formed through the shared beliefs and solidarity of members, which mobilize to support specific positions on social issues through various forms of protest (Della Porta and Diani, 1999). Researchers have studied social movements since the early 20th century, and theories have emerged and developed along two historical lineages: the European and the United States perspectives. The European perspective was founded in Marxism and focused on the fundamental conflict between the proletariat and the bourgeoisie, and the 'historical mission' of the proletariat to overthrow the bourgeoisie through revolution. The early U.S. perspective on social movements is referred to broadly as collective behavior (Park, 1967;Smelser, 1962), has roots in functionalist social psychology (Parsons, 1951), and considers movements as spontaneous and irrational responses to the structural strains and deprivation imposed on members by society. Following an intense period of activism in the 1960's, both the European and U.S. perspectives went through theoretical transformations during the 1970's. The European perspective on social movement theory produced the new social movement (NSM) approach. The new social movements focused on by NSM researchers included environmentalism, peace, second-wave feminism, and animal rights, among others. NSM theorists diverged from their historical Marxist roots, and adopted ''a view that contemporary western societies have outgrown the model of capitalist society suggested by Marx, rejecting the priority he affords to class struggle and to classes as agents of historical change'' (Crossley, 2002). They ''oppose economic reductionism and class reductionism: the emergence of social movements cannot be explained solely by economic changes and the position of actors in the production process'' (Fuchs, 2006a). Instead, NSM researchers stress nonclass issues in their analysis of social movements, such as gender, ethnicity, age, and environment. NSM researchers also focus as much on ''the constitutive structure and type of society in which modern movements emerge, the relation of those movements to that society, and their 'historical role' therein'' as they do the movement itself, and ''attempt to identify the central conflicts and movements (that are) definitive of the new era'' (Crossley, 2002). Theorists central to the NSM approach include Habermas (1981), Touraine (1981), andMelucci (1986). The U.S. perspective also transformed theoretically during the 1970's, and produced the resource mobilization theory (RMT). RMT developed from rational actor theory (RAT) and considered participants in social movements as actors who ''rationally calculate gains and losses that stem from certain potential actions and make choices for or against certain actions based on such calculations'' (Fuchs, 2006a). The rational actor represented a dramatic departure from collective behavior approaches that considered social movements as irrational reactions to grievances or deprivation. RM theorists viewed social movements as rational, organized manifestations of collective action and contention, with specific resources mobilized strategically to accomplish their respective objectives (Jenkins, 1983;McCarthy and Zald, 1977). Movement resources included material, human, cognitive, technical, and organizational capital and capabilities, which were utilized to coordinate activism, encourage participation, or gain political power (Klandermans, 1984). RMT also included a community of researchers who adopted a structural 'network' model of social relations and life, which allowed resources to be mobilized throughout the movement (Crossley, 2002;Diani and McAdam, 2003). The RM approach also considers other factors influencing the development of social movements such as framing (Snow and Benford, 1988;Benford and Snow, 2000;McAdam et al., 1996) and political opportunities (McAdam et al., 1996;Tarrow, 1998). Social theorists have been critical of both the NSM and RMT approaches, scrutinizing their deficiencies. We summarize the general concerns as follows. While the NSM approach focuses on macroaspects of social movements, such as the conditions from which the movement developed and the role of the movement in the history of society, it neglects to address the process by which movements achieve their social and political objectives. On the other hand, the RMT approach focuses on microaspects of social movements, such as the organizational structure and strategies for protest, while neglecting the reasons why social movements form and mobilize. Concisely, NSM explains 'why' while RMT explains 'how.'Crossley (2002) refers to the deficiencies in the NSM and RMT approaches as failures to explain aspects of social movements pertaining to agency, structure, or both. For a detailed critique of the NSM approach, see Hannigan (1985) or Pichardo (1997); for the RMT approach, see Crossley (2002). --- Social Movement Theory and the Internet Since the effective cyber-protest of the Zapatista, social movements and the Internet have developed an intimate relationship (Cleaver, 1995;Clark and Themudo, 2006), even influencing the evolution of each other (Fuchs, 2006b). The inherent structure of the Internet as a network encouraged researchers to reexamine the network model of social movements (Diani and McAdam, 2003). The network model diverged from the more traditional RMT perspective on social movements as organizations with formal, hierarchical, and bureaucratic structures. The new social movements that inspired the network model did not require membership, were decentralized, dynamic, and without formal hierarchy, and depended upon participants identifying with the perspectives and positions of the movement and its objectives (Diani, 2003). For proponents of the model, networks were significant in social movements and fulfilled many required functions. ''Networks undoubtedly facilitate mechanisms like the mobilization and allocation of resources across an organizational field, the negotiation of agreed goals, the production and circulation of information, all activities which are also essential to any type of coalition, broadly defined; at the same time, however, they also may-or may not-facilitate the circulation of meaning and mutual recognition'' (Diani, 2003). The network model of social movements was attractive to cyber-protest and social movement researchers due in part to the structural characteristics of the Internet and CMC. Social movements articulated online are fluid and dynamic networks, maintaining the linked structure of the Internet through their mobilization and expression of cyber-protest. ''In defining social movements, most observers would agree that they need some structure and are to some extent 'organized.' This structure, however, is not an organization but a network, or even a network of networks that both rests on and maintains a sense of collective identity'' (van de Donk et al, 2004). Diani was one of the early and key contributors to the network model of social movements (1990,1992), and recognized the potential to extend this model to CMC (Diani, 2000). In ( 2003), Diani proposes a comprehensive research approach to the study of social movement networks. In introducing the approach, the author reviews the literature and development of the network model of social movements, and describes how taking a network perspective of social movements can specifically improve the theories and research based on RMT, NSM, and political process theories (PPT). The author also provides a detailed description of the proposed approach. The author emphasizes two key characteristics to network structure to guide analysis: centralization and segmentation. Centralization refers to the formality of the network structure, while network segmentation refers to the ability and extent to which actors in the network can communicate without impediment. The author also identifies and describes four types of network structures: clique, policephalous, wheel/star, and segmented and decentralized. The proposed research approach to studying social movement networks consisted of four forms of network analysis: multilevel analysis, link multiplicity, time, and structural homophily. These forms of analysis represent an integrated framework to guide future research on social movement networks. Fuchs extends the network concept to describe social movements as selforganizing systems (2006a). By adopting the perspective of self-organization and viewing social movements as dynamic and complex systems on both the micro and macrolevels, the author addresses some major theoretical concerns with the RMT and NSM approaches. The author defines self-organizing systems as ''complex networks of entities that synergetically interact and produce novelty'' (Fuchs, 2006a). Through this perspective, the author connects the proposed approach to the literature on the network model of social movements. In the author's view, social movements are ''dynamic communication systems that permanently react to political and societal events with self-organized protest practices and protest communications that result in the emergence and differentiation (production and reproduction) of protest structures (events, oppositional topics, alternative values, regularized patterns of interaction and organization)'' (Fuchs, 2006a). The Internet itself serves as a dynamic communication system, and is self-organized, alluding to the tight integration between the self-organization model of social movements and cyber-protest. This relationship is described in detail in (Fuchs, 2006b). To the author, ''cyber-protest is a global structural coupling and mutual production'' of the self-organization of the Internet and the self-organization of social movements (2006b). In the case of the self-organization model, the Internet and cyber-protest have demonstrated strong influence over the development of social movement theory. --- Social Research on Virtual Communities Although we have shown that the Internet has had significant influence on social movements, and has been engaged by researchers and theorists, there are additional characteristics of cyber-protest that make conducting social movement research online appealing. The widespread use of CMC by social movement participants produces rich and complex social systems, such as virtual communities or cybersocieties (Jones, 1994). They are often vibrant with strong membership and robust communication, containing discourse that is highly revealing of the perspectives and ideologies of the social movement. There are both technical and social reasons why the study of virtual communities can benefit social research. Inherent technical features of cyber-protest and virtual communities offer aspects supporting analysis not readily found in natural settings. In a virtual community, all cultural artifacts produced by members are recorded and preserved, and can be recovered by the social scientist if and when desired. Advancement of CMC technology and the emergence of web 2.0 have provided a sufficient level of sophistication for the creation of virtual communities that can approximate real-life social interactions, producing a wide variety of artifacts with cultural and social significance. Traditional studies face the difficult task of determining the time associated with collected artifacts, but in an online environment, time is typically recorded precisely for each cyber-artifact. The accessibility of time information allows for analysis of the exact evolution of interactions and the production of cultural artifacts, which can be highly revealing of social behavior. The technical features provided by virtual communities are not the only benefits of conducting social research in the Internet. Social research focusing on virtual communities is often highly revealing of social behavior due to the phenomena described by the social identity model for deindividuation (SIDE) (Reicher et al, 1995). SIDE evolved in parallel with social movement theory, but differed in the perspectives of individual behavior within groups. Traditional models of deindividuation based on collective behavior describe factors such as the immersion in the group and lack of anonymity that lead to a loss of selfhood and control over individual behavior. However, SIDE theory shows that as deindividuation occurs in communities, individuals conform to categorical norms and organize to maximize the power of their collective identity in social movements (Reicher et al, 1995). Researchers have also studied CMC as a potential channel for social resistance, showing that due to the behavior of individuals described by SIDE, CMC was highly effective for communicating the social support necessary to resist powerful outgroups (Spears et al, 2002). These findings suggest that conducting social movement research focused on cyber-protest and virtual communities may be even more revealing of the true social behavior of participants than observation in a more natural setting. --- Cyber-archaeology Many social researchers have recognized the scale and complexity of cyber-protest, and the difficulty of conducting social research in the Internet. Paccagnella suggests that ''deep, interpretive research on virtual communities'' could be greatly assisted by ''new analytic, powerful yet flexible tools'' for ''collecting, organizing and exploring digital data'' (Paccagnella, 1997). Such tools would facilitate the analysis of 'cultural cyber-artifacts,' the member communications and creations representing the archived history of the virtual community (Paccagnella, 1997;Jones, 1997). Cultural cyberartifacts are diverse and could include websites, forum communications, e-mail, video or audio clips, documents, chat, bulletin board messages, and usage logs. The tasks performed by these tools to assist the social researcher in analyzing the cyber-artifacts of a virtual community would include data collection, preparation, and coding, storage and search, and data linking and display (Paccagnella, 1997). While the author never suggests that these tools could replace the social researcher in the analysis, since ''computers don't analyze data, people do,'' such techniques could greatly enhance the virtual community research and alleviate some of the technical challenges presented by CMC and the Internet environment. --- Cyber-Archaeology Framework for Social Movement Research We present a cyber-archaeology framework for social movement research. Our framework overcomes many of the issues of scale and complexity facing social research in the Internet, and specifically addresses many of the suggestions of Paccagnella (1997) enabling broad and longitudinal social research on virtual communities. The framework is not intended to replace the social researcher, or directly produce theoretical conclusions regarding social movements, but instead we offer a collection of powerful tools and techniques to assist in conducting social movement research in the Internet. The approach is automated and scalable, allowing a researcher to easily collect and analyze social movement cyber-artifacts across multiple linked virtual communities. The techniques for cyber-artifact analysis adopt perspectives of social movement theory, specifically the network (Diani and McAdam, 2003) and self-organization (Fuchs, 2006a) models of social movements. The proposed framework for cyber-archaeology has three phases, as presented in Figure 1. In phase one, social movement research design, social researchers identify the social movement of interest, associated virtual communities, and target cyberartifacts for collection and analysis. Phase two consists of the automated collection and classification of cyber-artifacts across one or many identified virtual communities. The third phase supports the analysis of cyber-artifacts by social researchers from the perspectives of social movement theory, including various network analysis and visualization techniques to perform multilevel, link, time, and homophily analysis, as suggested by Diani (2003). Each of these phases is discussed in detail below. The first phase of the cyber-archaeology framework consists of the social research design on the movements of interest. Researchers identify a social movement of interest and their related virtual communities, in consultation with domain experts when necessary, to develop a seed set of websites for analysis. The list of targeted virtual communities can then be expanded through link analysis of the seed set of sites, to include websites linked to and from the virtual communities already identified. Once a satisfactory list of virtual communities has been defined, the targeted cultural cyber-artifacts for research focus must be determined. A lexicon of terms pertaining to the cyber-artifacts of interest must also be developed to assist in the collection process by enabling the identification of relevant content. By limiting the search to specific cyber-artifacts of interest, the social researcher can focus their investigation, thereby alleviating issues of scale impacting social research in the Internet. The cyber-artifacts selected could be based upon, for example, influential figures in the movement, meaningful events, or core ideologies and symbolism. Focusing on specific cyber-artifacts that develop the identity of the movement and create solidarity among members deepens the social research. However, a researcher need not restrict their focus to specific cyber-artifacts; the framework supports a more comprehensive approach. Without restriction, all cyber-artifacts of the virtual communities may be collected for analysis, potentially overwhelming the researcher, but providing completeness in the research. The second phase consists of the automated collection and categorization of the targeted cyber-artifacts of interest from the virtual communities identified in the first phase. We consider the collection and classification tasks to be core to any study on social movements in the Internet. Researchers have highlighted the promise for enriching the virtual community research field by developing automated techniques to assist with these tasks (Paccagnella, 1997). Researchers have also identified opportunities to improve the understanding of social movements by broadening the research in scope, beyond a single-organization or single-issue focus (Edelman, 2001). To aid in the analysis of the collected cyber-artifacts, automated models for classification can be developed to categorize cyber-artifacts across numerous dimensions depending upon their nature and the goal of the research. Developing automated techniques for cyber-artifact collection and categorization allows social researchers to ask questions that would be infeasible to answer by traditional, manual methods, expanding the scope of research and the understanding of social movements. The outputs of the second phase, the collected and classified cyber-artifacts of the virtual communities, are then analyzed manually by the social researcher to learn about the movement. The third phase of the framework is considered optional, although these analyses are guided by social movement theory. Therefore, phases one and two of the framework could also support other forms of analysis, if desired by the social researcher. The third phase of the framework consists of the analysis of the collected and classified cyber-artifacts by social researchers from the perspectives of social movement theory. In particular, we adopt the network and self-organization models of social movements. Three forms of network analysis and visualization techniques are presented, providing insight into the patterns of self-organization exhibited by the virtual community with respect to the collected and classified cyber-artifacts. These network analyses and visualizations follow the research program proposed by Diani (2003), and cover multilevel analysis, link multiplicity, the evolution of networks over time, and structural homophily. The methods developed to support the third phase of the framework also utilize structured approaches and automation, similar to the techniques developed in the second phase, to support scalability in the social research. --- Case Study To demonstrate the efficacy of the cyber-archaeology framework, and provide a detailed instantiation of the proposed approach for evaluation, we present a case study on a broad group of related social movement virtual communities. --- Introduction Some of the most sophisticated virtual communities with strongest membership and social ties between community members belong to socially extreme movements. There are many social movements throughout the world that could be considered socially extreme or extremist, struggling for various causes based upon race, gender, sexuality, and religion. Extremist movements actively use the Internet as a quick, inexpensive, and anonymous means of communication, overcoming the limitations of traditional media, and allowing their ideological information to reach new audiences around the world. Studies have shown that these groups promote their positions through their Internet communications and publications (Glaser et al, 2002), and found evidence of resource sharing, fund raising, propaganda, recruitment, and training materials (Burris et al, 2000;Shafer, 2002;Zhou et al, 2005). Social scientists have long been interested in Middle Eastern movements supporting anti-Western positions, some of which contain participants or subgroups with socially extreme perspectives whose activism includes high-risk behavior, applying social movement theory to understand these societies and individuals (Snow and Marshall, 1984;Wiktorowicz, 2002). To be clear, we do not consider all anti-Western movements of the Middle East to be extreme, nor do we believe that all members of such movements engage in extremist behavior in their activism. However, some of the participants in these social movements hold extreme positions and engage in high-risk behavior in their activism, with grave cost. Social researchers have specifically studied the motivations and dynamics of high-risk behavior in social movements (McAdam, 1986). Communities supporting high-risk activism stress the importance of social networks, as strong social ties are required to provide the trust and solidarity among activists to encourage the high-risk behavior (Wiktorowicz, 2002). In studying these social movements with some participants whose activism includes high-risk behavior, we perform a cyber-archaeological analysis on a specific theme of cultural cyber-artifacts significant to some anti-Western Middle Eastern movements: content related to improvised explosive devices (IEDs). IEDs have social significance among some movement participants as an enabler of violent uprising and resistance, equalizer of power, and as the focus of high-risk activism and related symbolism. Examining IED-related cyber-artifacts in a global social context can provide a deeper understanding of these social movements, and the high-risk behavior of some of the participants. --- Cyber-Archaeology Framework Phase 1 In the first phase of the cyber-archaeology framework, a list of social movements of interest and related virtual communities is defined. For the case, the social movements and virtual communities were identified using the expertise of a number of government and research organizations. The seed set of social movement virtual communities was expanded through link analysis. In-links to and out-links from the virtual communities were collected and examined manually to identify additional related virtual communities to be included in the social movement research. Utilizing an automated approach to cyber-artifact collection and analysis, social movement research can extend across multiple related virtual communities. Since the social movements identified for this research communicate primarily in Arabic, a lexicon of terms relevant to IEDs was developed with assistance from domain and language experts to direct the cyber-artifact collection effort. The lexicon for IED-related cyber-artifacts consisted of more than 100 terms and their linguistic variations; a subset of the terms is shown in Table 1 (WWW resource, 2010). --- Cyber-Archaeology Framework Phase 2 The second phase of the cyber-archaeology framework consists of the automated collection and categorization of IED-related cyber-artifacts from the identified social movement virtual communities. The automated approach provides the ability to expand the research to multiple virtual communities and acquire a more complete collection of cyber-artifacts of interest. --- Cyber-Artifact Collection The collection of IED-related cyber-artifacts from social movement virtual communities entails the use of focused crawling techniques. Focused crawlers are intelligent web spiders designed to retrieve content relating to a specific subject. They ''seek, acquire, index, and maintain pages on a specific set of topics that represent a narrow segment of the web'' (Chakrabarti et al, 1999). Once the focused crawler has been properly configured, cyber-artifacts of interest can be collected from multiple virtual communities in an automated fashion with little supervision. For the purposes of the case study, these focused crawling techniques were applied in the collection of IED-related cyber-artifacts from 30 virtual communities identified in phase 1. In total, 2,541 IED-related web page cyber-artifacts were collected and qualified through manual inspection by a domain expert. Over 90% of the cyber-artifacts were gathered from a core set of 7 websites, as shown by the frequency distribution in Figure 2 (WWW resource, 2010) and collection summary in Table 2 (WWW resource, 2010). --- Cyber-Artifact Classification To assist in the analysis of the collected cyber-artifacts, automated models for classification can be developed to categorize cyber-artifacts across numerous dimensions depending upon their nature. For CMC communications, a useful categorization for analysis of cyber-artifacts is by genre of communication. Genre is a literary and rhetorical concept that describes widely recognized types of discourse (Yates and Orlikowski, 1992). Genres of communication are traditionally defined in terms of shared purpose and common conventions of substance and form (Herring and Paolillo, 2006). Shared purpose refers to the social motives and intentions behind the communication. Substance refers to the themes and topics expressed in the communication. Form refers to observable physical and linguistic features of the communication, including structural attributes, communication medium, and language such as degree of formality and use of a specialized vocabulary. Genres of communication are used within a community to establish expectations about the purpose, content, participants, form, time, and place of the interaction, describing the who, what, where, when, why, and how of the communication (Yates and Orlikowski, 2002). Traditionally, researchers have analyzed genres to understand the communication dynamics of a community or organization. Yates and Orlikowski (1992) observed that the emergence of genres within an organization is due to particular sociohistorical contexts, which are reinforced over time as the situation recurs. Researchers have also developed techniques for the automated classification of documents by genre. While genre classification is similar in nature to topic or author classification, the representative features used to characterize the document differs according to the type of classification being performed. In topical classification, features are dependent upon the topic, whereas in author and genre classification features are often considered orthogonal to topic, and greater emphasis is placed on the stylistic and structural elements of the document. Classification of collected CMC cyber-artifacts by genre of communication assists in the analysis of the social movement virtual community. Through manual analysis of the collected IED-related CMC cyber-artifacts by a domain expert, two distinctive genres of communication were identified: web pages containing descriptions on IED-related weapons, attacks, and military tactics (materials pages), and web pages with general discussion of IEDs or IED related events (discussion pages). The two genres of web pages identified are described briefly in Table 3. Sample web pages from each type are presented in Figures 34567(WWW resource, 2010). Developing automated methods for the classification of these cyber-artifacts provides the ability to enhance and expand the scope of the social research on movement virtual communities. Our genre classification approach for IED-related web pages is comprised of an extended feature representation coupled with machine learning algorithms. The extended feature representation incorporates rich structural information with stylistic and topical linguistic features, intended specifically to represent the purpose, substance, and form of the web page, indicative of genres of communication (Yates and Orlikowski, 2002). Structural features include HTML tags and information about the technical structure of the web page, which can specifically express genre information to differentiate materials and discussion pages. Stylistic features include word-and character-level lexical measures, word length distributions, vocabulary richness, function words and punctuation usage, and effectively capture the stylistic tendencies of the web page (Abbasi and Chen, 2005;Zheng et al., 2006). Topical features such as word n-grams are used to identify themes and topics in web pages. Only features exceeding a threshold number of occurrences (3) were included in the representation. The extended feature set is summarized in Table 4. The support vector machine (SVM) learning algorithm was used with the extended feature representation to automatically classify the collected IED-related web pages. The SVM is a statistical based algorithm that has been used successfully in text categorization (Joachims, 1998), and is particularly suited to learning with extended feature representations. Classification models that utilize extended feature representations often leverage methods for feature selection to identify the most relevant features for inclusion to the model, thereby eliminating unnecessary information and reducing model complexity. For the classification of the IED-related web page cyber-artifacts, we evaluate two SVM models: The first included the entire extended feature set (labeled SVM), while the second used feature selection on the extended feature set based upon the information gain heuristic, a measure of entropy (labeled SVM-IG). Both SVM classification models utilized linear kernel functions. The test bed of 2,541 collected IED-related cyber-artifact web pages was used for the development and evaluation of the automated genre classifier. As previously stated, 2,501 discussion and 40 materials web pages were identified through manual categorization by a domain expert. Due to the large number of discussion pages relative to materials pages, adjustments were made to provide equal representation of materials and discussion pages in the classification experiments. 100 iterations of the experiment were performed, and in each iteration 40 discussion pages were randomly selected from the 2,501 to complement the 40 materials web pages; a different set of 40 discussion pages were used in each iteration of the experiment. 10 fold cross validation was performed to test classification and generate performance results, measured by computing the percentage of correctly classified web pages. Overall performance values were calculated as simple averages of the 100 iterations. Overall, both the SVM and SVM-IG models performed well in genre classification of IED related materials and discussion web pages, with better than 81% accuracy. The SVM-IG model using the extended feature set and the information gain he

This paper presents a cyber-archaeology approach to social movement research. The approach overcomes many of the issues of scale and complexity facing social research in the Internet, enabling broad and longitudinal study of the virtual communities supporting social movements. Cultural cyber-artifacts of significance to the social movement are collected and classified using automated techniques, enabling analysis across multiple related virtual communities. Approaches to the analysis of cyber-artifacts are guided by perspectives of social movement theory. A case study on a broad group of related social movement virtual communities is presented to demonstrate the efficacy of the framework, and provide a detailed instantiation of the proposed approach for evaluation.

entropy (labeled SVM-IG). Both SVM classification models utilized linear kernel functions. The test bed of 2,541 collected IED-related cyber-artifact web pages was used for the development and evaluation of the automated genre classifier. As previously stated, 2,501 discussion and 40 materials web pages were identified through manual categorization by a domain expert. Due to the large number of discussion pages relative to materials pages, adjustments were made to provide equal representation of materials and discussion pages in the classification experiments. 100 iterations of the experiment were performed, and in each iteration 40 discussion pages were randomly selected from the 2,501 to complement the 40 materials web pages; a different set of 40 discussion pages were used in each iteration of the experiment. 10 fold cross validation was performed to test classification and generate performance results, measured by computing the percentage of correctly classified web pages. Overall performance values were calculated as simple averages of the 100 iterations. Overall, both the SVM and SVM-IG models performed well in genre classification of IED related materials and discussion web pages, with better than 81% accuracy. The SVM-IG model using the extended feature set and the information gain heuristic for feature selection performs particularly well with over 88% accuracy, and outperforms the standard SVM model by 7%. A t-test was performed to determine whether the performance increase gained by utilizing feature selection was significant. The performance of the SVM-IG model were a statistically significant improvement (p <unk> 0.001) over the model that used the entire feature representation without feature selection. Furthermore, the SVM-IG model was more reliable, performing with better than 80% classification accuracy in every iteration of the experiment. The information gain heuristic identified 9,268 of the 21,333 features as key and relevant to the genre classification. The results of the experiments are presented in Table 5 and Figure 8 (WWW resource, 2010). --- Cyber-Archaeology Framework Phase 3 The third phase of the cyber-archaeology framework consists of the analysis of collected IED-related cyber-artifacts through the perspectives of social movement theory. In particular, we adopt the network and self-organization models of social movements. Three forms of network analysis and visualization techniques are presented, providing insight into the patterns of self-organization exhibited by the virtual community with respect to the collected and classified cyber-artifacts. These network analyses and visualizations follow the research program proposed by Diani (2003), and cover multilevel analysis, link multiplicity, the evolution of networks over time, and structural homophily. The methods developed to support the third phase of the framework also utilize structured approaches and automation, similar to the techniques developed in the second phase, to support scalability in the social research. In the case of IED-related cyber-artifacts, learning how the highly specialized knowledge pertaining to these technologies is organized, distributed, and leveraged throughout the related social movement virtual communities is of particular research interest. Furthermore, an understanding of the communication patterns related to these cultural artifacts is significant to the understanding of the social movements, and the relations the communities have with IEDs and related high-risk activism. The third phase of the framework consists of three forms of network analysis and visualization, addressing multilevel analysis, link multiplicity, time, and homophily (Diani, 2003). The first type of network analysis, site map analysis, visualizes the self-organization of discussions and resources pertaining to the cyber-artifact of interest within virtual communities. Site map analysis addresses the multilevel analysis suggested by Diani (2003). The second type of network analysis, link analysis, visualizes the ways virtual communities interact, self-organize, and share resources. This form of network visualization addresses the link multiplicity analysis proposed by Diani (2003). The third type of network analysis visualizes the growth of virtual communities over time, addressing the temporal analysis suggested by Diani (2003). By leveraging the classification of the cyber-artifacts performed in the second phase, the three forms of network analysis and visualization can also provide insight into the structural homophily of the various classes of cyber-artifacts in the virtual communities. Comparing the classes of collected cyber-artifacts, patterns and differences in structural homophily may be revealed in the three forms of network analysis. The first type of network analysis, site map analysis, determines how IED-related cyber-artifacts are organized within a single virtual community domain, providing a form of multilevel analysis (Diani, 2003). The average site maps for websites containing IED-related cyber-artifacts were constructed to analyze the intensities of discussion and materials web pages at various levels within the domain, shown in Figures 9 and10, respectively. The site map was generated by counting the frequency of pages for each level and dividing by the total number of web sites. These site maps are indicative of the communication patterns of the participants, showing where within their virtual community these issues are being discussed and specific IED-related resources mobilized. The site maps were drawn to a 20:1 scale; thus, 5 level 1 pages on the map represents 100 actual pages. The intensities of discussion and materials web pages at various levels of the site map are represented using different color shades, with dark colors being more intense concentrations. The distribution of discussion and materials pages among the page levels in the average site maps are also presented in Table 6 (WWW resource, 2010). The majority of materials web pages occur in domain levels 2-4, while discussion content is often located in levels 2-6. The occurrence of IED-related cyber-artifacts at the middle levels of the domain may indicate that authors do not want their content to be too readily visible, but also do not want to bury their content too deep within the domain as to not be seen by community members. The concentration of discussion pages at lower levels in the domain than materials pages may be attributed to the strong identity traces in discussion content left by the message author, which the author wishes to hide. Materials pages are usually formally written manuals, often in document format, while discussion pages are casual free text with the author's stylistic tendencies evident. These observed differences between the organization of materials and discussion pages may be indicative of aspects of structural homophily (Diani, 2003). The site map analysis also identified key terms in the IED lexicon, and their occurrence distribution across site levels. Results revealed that certain terms occur more frequently at specific site levels. A selection of terms is presented in Table 7 (WWW resource, 2010). Also shown are the common levels where the terms are likely to occur, with their average occurrence per page. The second type of network analysis, link analysis, determines how multiple related virtual communities interact, share cyber-artifacts, and develop their collective identity. This visualization addresses the suggestion of Diani (2003) for analysis of link multiplicity. The link network structure, as displayed in Figure 11, was constructed to show how virtual communities containing IED-related cyber-artifacts were connected, organizing their resources across multiple related sites of the social movement. Each node of the link network represents a website, and is color coded depending upon the IED-related cyber-artifacts collected from the virtual community. The link network also includes websites not previously analyzed in the social movement study but with linkages to the investigated virtual communities. These websites, particularly those with extensive linkages to multiple virtual communities from which IED-related cyber-artifacts were already collected, are strong candidates for future related studies as their linkages may represent the presence of IED-related cyber-artifacts of interest. Three major hubs for materials related IED cyber-artifacts were identified in the network map. As shown in the link network, these websites act as major hubs for the mobilization of IED-related resources. The link network map provides insight into the relationship between virtual communities, and how the production and reproduction of cyber-artifacts occurs between related constituents (Fuchs, 2006b). To understand the evolution of the virtual communities and their self-organization over time (Fuchs, 2006b), and address the temporal dimension of network analysis proposed by Diani (2003), link network snapshots were taken over a 1-year period for two major virtual communities, presented in Figure 12. In the link network diagrams, the virtual community websites containing IED-related discussions and materials, respectively, are circled in green. Three snapshots were taken during the one year period, in January, July, and December of 2005. Both virtual communities experienced significant growth in their link networks over that time. In the case of the discussion website, total links to and from the website increased from 16 to 35. Diani suggested one of the significant aspects of social movement networks was centralization (2003). Betweenness is a measure of centrality in a link network, representing the degree to which a node provides a path to information resources to the rest of the network. The betweenness of the discussion website increased from 86.91 to 947.16, signifying an expanded role as information broker to the network. The link network of the virtual community website containing IED-related materials, expanded dramatically in comparison. Total links to and from the materials website increased from 39 to 196, and the website's betweenness measure increased from 595.14 to 31,857.12. The explosion in the materials website link network may be due to the specialized knowledge contained in IED-related materials cyber-artifacts. These resources are considered particularly valuable to the virtual communities, and are mobilized quickly and diffused throughout the movement. The differences observed in the growth of the link network supporting the virtual communities containing discussion and materials pages is revealing of structural homophily (Diani, 2003). Temporal analysis of the virtual community link networks provides specific insight into the growth and integration of the community in the larger movement, the production and reproduction of movement participation, and the emergence and distribution of its specialized resources (Fuchs, 2006b). --- Conclusion The Internet, virtual communities, and expressions of cyber-protest are significant to social movements, and offer ideal environments for conducting social research. In this paper, we present a cyber-archaeology framework for social movement research. Our framework overcomes many of the issues of scale and complexity facing social research in the Internet, and specifically addresses many of the suggestions of Paccagnella (1997) enabling broad and longitudinal social research on virtual communities. We offer a collection of powerful tools and techniques to assist in conducting social movement research in the Internet. The approach is automated and scalable, allowing a researcher to easily collect and analyze social movement cyber-artifacts across multiple linked virtual communities. The techniques for cyber-artifact analysis adopt perspectives of social movement theory, specifically the network (Diani and McAdam, 2003) and self-organization (Fuchs, 2006a) models of social movements. To demonstrate the cyber-archaeology framework for social movement research, and provide a detailed instantiation of the proposed approach for evaluation, a case study on a broad group of related social movement virtual communities is presented. Utilizing focused crawling, cyber-artifacts of interest were collected from multiple linked virtual communities. To categorize the collected cyber-artifacts by genre of communication, an extended feature representation and machine learning models were developed and tested, with nearly 90% accuracy in genre classification. Finally, the cyber-artifacts were analyzed through perspectives of self-organization and the network model of social movements, using network analysis and visualization techniques. We hope the case study has demonstrated the efficacy of the proposed framework for social movement research, and promotes its future use by researchers. Utilizing the proposed automated approaches to cyber-artifact collection and classification allows for large-scale social research across a greater number of virtual communities compared to traditional methods. A cyber-archaeology approach guided by social movement theory can produce high-quality social research and a deeper understanding of widespread movements. --- Collection and Infrastructure for Computational and Social Sciences, '' 2007-2010; and (2) NSF, ''EXP-LA: Explosives and IEDs in the Dark Web: Discovery, Categorization, andAnalysis,'' 2007-2010. --- About the Authors --- David

This paper presents a cyber-archaeology approach to social movement research. The approach overcomes many of the issues of scale and complexity facing social research in the Internet, enabling broad and longitudinal study of the virtual communities supporting social movements. Cultural cyber-artifacts of significance to the social movement are collected and classified using automated techniques, enabling analysis across multiple related virtual communities. Approaches to the analysis of cyber-artifacts are guided by perspectives of social movement theory. A case study on a broad group of related social movement virtual communities is presented to demonstrate the efficacy of the framework, and provide a detailed instantiation of the proposed approach for evaluation.

INTRODUCTION Social entrepreneurship is a contemporary approach to solving social problems that are traditionally addressed by governmental and non-profit organizations. Social entrepreneurs look for both a return on investment and a return to society. This emerging perspective is becoming very popular. The nascent literature has not yet settled on a definition of social entrepreneurship; however it does agree that social entrepreneurs have been effective when others were not. In this paper we will review the current state of social entrepreneurship. We propose a very general model that can be adapted to different settings. --- WHAT IS SOCIAL ENTREPRENEURSHIP? Social entrepreneurs look for both a return on investment and a return to society. This emerging perspective is becoming very popular worldwide. Social entrepreneurs have been effective when others were not. In many ways social entrepreneurship has been a driving force in the expansion of the social sector (Noruzi, Westover & Rahimi, 2010). Some have argued that social entrepreneurship is hundreds of years old (Svetunkov & Ponomarev, 2016), because there have always been those who make money while augmenting the efforts of governments and charities. However we are looking at the relatively new phenomenon where modern entrepreneurs seek to combine a desire for profit with altruism (Svetunkov & Ponomarev, 2016). A general example is the newly emerged field of microfinance, which serves those needing relatively low sums of money and were unable to find funding from traditional sources. A more specific example is 'Tom's Shoes' -a company that provides one pair of shoes to those in need for every one pair that they sell. The nascent literature has not yet settled on a definition of social entrepreneurship. It is in a pre-paradigmatic state, and as the field is maturing, theory is developing (Granados, Hlupic, Coakes & Mohamed, 2011). However it is still difficult to formulate an exact definition of social entrepreneurship (Abu-Saifan, 2012). Some people have used the terms 'civic entrepreneurship' (Korosec & Berman, 2006) or'social business' (Barki, Comini, Cunliffe, Hart & Rai, 2015) to describe the same phenomena. Other terms associated with social entrepreneurship such as 'philanthocapitalism', B corporations, embedded investing, impact investing and microfinance are among the Chronicle of Philanthropy's 2011 "10 Favorite Buzzwords of the Decade" (Bernholz, 2011). As noted by Jones & Donmoyer (2015), these ideas are "are not only about rhetoric and buzzwords, nor are they only discussed outside of academia". In an attempt to clarify what constitutes social entrepreneurship, Zahra, Gedajlovic, Neubaum & Shulman (2009) identified <unk>20 definitions of social entrepreneurship and go on to present a typology of social entrepreneurs which divides them into three types (see Table 1). The many definitions have not converged over time (Conway Dato-on & Kalakay, 2016). A definition must include the key concepts and apply for the wide area of activity discussed in the literature. We present the following definition: Social Entrepreneurship -Using profit making enterprises to address social, environmental and other problems that were traditionally entrusted to governmental and non-profit organizations. It is necessary to differentiate between the entrepreneur and the'social' entrepreneur (Massetti, 2008). Figure 1 shows two of the major differences. Both entrepreneurs and social entrepreneurs use both discovery and creation strategies for getting opportunities (Gawell, 2013), however the social entrepreneur is more likely to create and the traditional entrepreneur is more likely to try to discover opportunities (Korsgaard, 2011;Shaw & Carter, 2007). The more important difference is that social entrepreneurs are more interested in creating value than in capturing value, and the opposite is true for traditional entrepreneurs (Agafonow, 2014;Crisan & Borza, 2012;Santos, 2012). Social entrepreneurs also need to be differentiated from social workers, community organizers and activists. Social workers are part of the traditional governmental and non-profit landscape that attempt to address social issues. Activists and organizers bring attention to issues and situations that are not fully addressed by social workers and others. As such they help frame the opportunities and attract support for value creation by the social entrepreneur. --- WHY DO WE HAVE OR NEED SOCIAL ENTREPRENEURSHIP? Octavia Hill, Jane Addams and Mary Parker Follett questioned the self-serving view of 'capitalism' and advocated a consideration of others influences. This paved the way for SE. (Prieto & Phipps, 2014). The idea is that it is possible to "do well by doing good". (Demirdjian, 2007).Social entrepreneurship has been identified as a response to market imperfections characterized by simultaneous government and market failure (Griffiths, Gundry & Kickul, 2013;Lajovic, 2012;Santos, 2012). Government and philanthropic funding of non-profits is increasingly unsustainable (Stecker, 2014). Social entrepreneurship serves to fill the gap between the haves and have nots and societies failure to provide social services (Demirdjian, 2007). Social entrepreneurs serve as change agents (Day & Jean-Denis, 2016). It occurs in many places, but is more likely to be found in communities in decline that need both economic and social regeneration (Thompson, Alvy & Lees, 2000). Social entrepreneurship is serving the 'Bottom of the Pyramid' that Prahalad (2004) wrote about (Goyal, Sergi & Jaiswal, 2016;Pervez, Maritz & De Waal, 2013). It can be considered as providing "the essentials needed by the have-nots of society" (Demirdjian, 2007). The motivation of social entrepreneurs is "the persistency of problems in society related to poverty and persons living in marginalized communities" (Day & Jean-Denis, 2016). In this regard, perhaps all enterprises that provide social service can be categorized as social entrepreneurship (Williams & K'nife, 2012). One difference is that social entrepreneurs create sustainable public wealth (El Ebrashi, 2013) through social impact and social change (Young, 2006). Social entrepreneurs and others can create a social entrepreneurship culture allowing for many initiatives and a great positive impact (Dal Forno & Merlone, 2009). Social entrepreneurship can have its greatest impact by making itself obsolete (Trexler, 2008). Social entrepreneurship might be a part of the life cycle of an entrepreneur (Svetunkov & Ponomarev, 2016). Entrepreneur's logic changes over time (Williams & Nadin, 2011). Corporate social responsibility isn't just driven by economics, for example it could be established and sustained by a moral drive of a champion (Hemingway, 2005). When entrepreneurs are so driven they become social entrepreneurs. Advocates of social entrepreneurship believe that "social entrepreneurship is a natural expression of visionary leadership, the spiritually and ethically-based mission to seek the common good, and the virtual necessity to create sustainability for both people and planet" (Muscat & Whitty, 2009). Academic researchers, consultants and foundations are adding to and helping establish the legitimacy of social entrepreneurship (Hervieux, Gedajlovic & Value Capture Value Creation --- Social Entrepreneurs Entrepreneurs Opportunity Discovery Opportunity Creation --- Figure 1 -Opportunity, Value and Entrepreneurship Turcotte, 2010). However it is important to note that the literature has been criticized as being dominated by those who praise social entrepreneurs without critiquing them (Chell, Spence, Perrini & Harris, 2016). The few critics advocate against a'messianic script' and call for a "nonheroic practice of entrepreneurship", or.'messianism without a messiah" phrases coined by Dey & Steyaert (2010) to portray "an image of social entrepreneurship that conceives of social change without nostalgic reference to the sovereign, heroic entrepreneur". --- INTERESTING FINDINGS AND MISCONCEPTIONS Social entrepreneurship has been portrayed as a universally good thing; however it does have a dark side (Williams & K'nife, 2012). Social entrepreneurship has its limits and cannot solve social problems on a large scale (Sud, Vansandt & Baugous, 2009). It is not as scalable as government (Dees, 2007). Furthermore social entrepreneurship can undermine state sponsored development and reforms (Ganz, M., Kay, T. & Spicer, J., 2018; Nega & Schneider, 2014). It can also go the other way, government social grants might discourage SE when it goes to unintended groups (Sinyolo, S., Mudhara, M. & Wale, E., 2017). It is also important who is giving the aide and why. Mission drift and power differences can be a problem, with the social entrepreneur having power over those that are being helped and losing focus as to the original intent of the endeavor. In its extreme, establishing and maintaining power differential can be the motive for helping. For example, gang leaders provide social services as a means of control (Williams & K'nife, 2012). The local community is then indebted to the gang and tolerates or even helps them out of fear, gratitude or both. Gender is an important controversial issue in the literature. Some have found that Social entrepreneurship supports training, local networks and women's businesses (Griffiths, Gundry & Kickul, 2013). Others criticize that social entrepreneurship does not recognize the role of gender in society and entrepreneurship, and that it should have an aim of economic and social equality for women (Clark Muntean & Ozkazanc-Pan, 2016). Whereas others separate out 'emancipatory social entrepreneurship' and maintain that empowerment of women is a mediator to changing society (Haugh & Talwar, 2016). We do know that the single greatest determinant of social entrepreneurship is female participation in the workforce (Griffiths, Gundry & Kickul, 2013). There is a lack of empirical research on social entrepreneurship (Granados, Hlupic, Coakes & Mohamed, 2011) and very little empirical evidence (Cukier, Trenholm, Carl & Gekas, 2011). The literature has 'protectionists' who claim effectiveness without proof and 'opponents' who require proof (Pärenson, 2011). However social entrepreneurship effectiveness and rates of activity have been hard to measure (Lepoutre, Justo, Terjesen & Bosma, 2013). Others emphasize the impact of social entrepreneurship, but social impact has also been difficult to measure (Pärenson, 2011;Dees, 2007). It also is important to consider and measure regulatory policies that can impede or encourage social entrepreneurship (Arasti, Zarei & Didehvar, 2015) We do know some things from the existing studies. For example, places with higher rates of entrepreneurship in general have higher rates of social entrepreneurship (Lepoutre, Justo, Terjesen & Bosma, 2013). The label of social entrepreneurship alters people's perceptions and judgements as to how effective the organization is (Andersson & Self, 2015). Three quarters of cities with a population of 50,000 or more give active or moderate support to programs and whether it is supported by a city is related to perceived effectiveness (Korosec & Berman, 2006). However cities are not always where the need is greatest, and entrepreneurs that are in deprived populations and rural populations are more socially-oriented than those in relatively affluent and urban populations (Williams & Nadin, 2011). Besides the urban/rural and affluent/deprived differences, success factors and focus of entrepreneurships depends on country or region. For example -in the US social entrepreneurs focus on social injustice problems, whereas in Africa they address rural poverty (Bewayo & Portes, 2016). An interesting finding is that although the rhetoric and image of social entrepreneurship is associated with innovation, novel approaches are less likely to be successful (Renko, 2013) and although there is a perceived urgency to solve social problems, longer development time increases the likelihood of success (Renko 2013). We also know from the literature that social entrepreneurship is a world-wide phenomenon. Local dynamics and conditions often are serious influences on the prevalence and nature of social entrepreneurship activities. Studies have been conducted by scholars in many countries. A sampling of such studies can be found in Appendix 1. Perhaps it is a mistake to make a distinction between social entrepreneurs and entrepreneurs. Most in either category display both social and commercial goals. There are indications that we should consider not a dichotomy but a "continuum from purely commercial to purely social entrepreneurship" (Williams & Nadin, 2011). Research has shown the rhetoric of both types of entrepreneurs is equally economically oriented (Chandra, 2016). However there have been differences shown. Chandra (2016) also found that "the rhetoric of social entrepreneurs is more other, stakeholder engagement and justification-oriented and less self-oriented than the rhetoric of business entrepreneurs." Differences can also be found in the area of disposition. For example, agreeableness is positively related to all dimensions of social entrepreneurship, whereas openness is related to "social vision, innovation and financial returns" (Koe Hwee Nga & Shamuganathan, 2010). However social entrepreneurs prosocial motivation decreases likelihood of success (Renko, 2013). --- HOW CAN GOVERNMENT AND UNIVERSITIES HELP? The government can facilitate and support social entrepreneurship (Goyal, S., Sergi, B. S. & Jaiswal, M. P., 2016;Jung, K., Jang, H. S. & Seo, I., 2016;Griffiths, M. D., Gundry, L. K. & Kickul, J. R., 2013;Sullivan, D. M., 2007). What can government do to help? The government has insight and data to help identify and provide access to the problems that can be helped by social entrepreneurs. They can use existing organizations (Goyal, S., Sergi, B. S. & Jaiswal, M. P., 2016). For example in New Jersey, every county has a Small Business Development Center (SBDC) to facilitate entrepreneurial efforts. The SBDC work closely with the Service Corp of Retired Executives (SCORE) and the federal and local Small Business Associations (SBA). Grants and other funding can be set aside for addressing social problems (Boehm, L., 2010). Finally legislation and policies can encourage and facilitate the efforts of social entrepreneurs (Lan, H., Zhu, Y., Ness, D., Xing, K. & Schneider, K., 2014;Prakash, D., Jain, S. & Chauhan, K., 2015). Universities can also do a great deal to encourage and facilitate social entrepreneurship (Mititelu, C., Fiorani, G. & Litardi, I., 2017). Universities can use experiential learning to get students to develop a passion for social entrepreneurship (Gundlach, M. J. & Zivnuska, S., 2010). NGOs can partner with social entrepreneurs through universities (Stephenson, H. & Mace, D. L., 2009). Service learning can be used to support social entrepreneurs (Kinsella, S. & Wood, N., 2014;Peric, J. & Delic, A., 2016). Some universities already require a 'civic engagement' or'social engagement' component in their curriculum. Others have capstone experiences where students are required to solve a business problem. These courses and activities can be targeted towards social concerns. Internships at social enterprises can be beneficial to both the student and the business. Universities can put together groups of students with the appropriate skill sets to help in solving business-related or technical problems. Faculty can also be used as consultants or by using the enterprise as part of their research. --- CONCLUSION AND FURTHER RESEARCH In this exploration we proposed a definition of social entrepreneurship as 'using profit making enterprises to address social, environmental and other problems that were traditionally entrusted to governmental and non-profit organizations'. We also presented a model (Figure 1) that contrasted social entrepreneurship with traditional entrepreneurship. In the model, traditional entrepreneurs we more involved with opportunity discovery and capturing value, whereas social entrepreneurs primarily create both opportunities and value. We plan to continue to develop the model. We would like to explore the empirical support, hypothesize the reasons for the dynamics and propose empirical testing. We also would like to focus on specific issues. One issue of particular interest to us is clean water. It seems that, although clean water is a pervasive worldwide problem with many facets, social entrepreneurship has the potential to have a sizable positive impact. We would also like to identify and develop ways the government and university can help. Some current efforts are to work local SBDCs. Currently we are having teams of graduate students helping small businesses. We would like to turn some of these projects into case studies. Finally we would like to identify or possibly create data sets that capture social entrepreneurship. As identified earlier, in addition to a lack of conceptual clarity, there is very little empirical research in the area. We look forward to testing ideas and letting our empirical findings inform theory development and practice.

Social entrepreneurship is an emerging alternative to governmental and non-profit approaches to social problems. Social entrepreneurs look at the 'triple bottom line' of people, planet and profit. They aim to achieve both a return on investment and a return to society, and address social problems effectively where others have not. In this paper we review the current literature on social entrepreneurship and attempt to clarify the key concepts and dynamics. We propose a definition of Social Entrepreneurship as 'using profit making enterprises to address social, environmental and other problems that were traditionally entrusted to governmental and non-profit organizations'. We present a model of social entrepreneurship, provide suggestions for practice and provide a research agenda for scholars.

Introduction Inorganic arsenic, a form of the elements typically found in soil, bedrock, and associated groundwater is the most toxic form of arsenic to humans [1]. An estimated 220 million people worldwide are exposed to drinking water with arsenic above the World Health Organization's provisional guideline and the U.S. EPA Maximum Contaminant Level (MCL), 10 micrograms per Liter (<unk>g/L) [1][2][3]. High levels of arsenic in drinking water have been found in countries such as Argentina, Bangladesh, Chile, China, India, Mexico, and the United States [1]. Within the United States, high levels of naturally occurring arsenic have also been found in the groundwater aquifers used for drinking water supplies, particularly in the states of Utah, Nevada, Arizona, and California [4]. Health effects attributed to chronic exposure to arsenic, most commonly through ingestion of contaminated water, have been studied extensively and arsenic exposure has been associated with lung, bladder, skin and kidney, cancers along with skin lesions and impairments to the cardiovascular and nervous systems [4][5][6][7][8][9][10][11]. The International Agency for Research on Cancer has labeled arsenic as a Group 1 known human carcinogen [12], and according to a recent analysis and review, the excess mortality risk due to cancer for exposure above 10 <unk>g/L is 1 in 400 people with an increase in mortality risk for exposure above 50 <unk>g/L at 1 in 81 people [5]. From 2014-2015, it was estimated that 55,000 people across California were exposed to arsenic contamination above the legal level (10 <unk>g/L) through their local community water systems [13]. Community water systems and domestic/private wells are the primary sources of drinking water for residents of California A community water system is defined as a public water system that serves the same population year-round and must have at least 15 service connections or serve an average of at least 25 people for at least 60 days a year [14]. Community water systems can be further categorized by the number of people they serve. A small water system serves 501-3300 residents and a state small water system serves 5-14 service connections and does not serve an average of at least 25 people for at least 60 days a year [15]. Community water systems in California require domestic water supply permits which are regulated by the system operators and must maintain compliance with state and federal MCLs to ensure safe drinking water. For rural communities, small community water systems are overseen by the California State Water Resources Control Board, the Department of Environmental Health, or the California Public Utilities Commission and the state small water systems are overseen by the Department of Environmental Health at the county level [16]. For these small water systems, each county sets its own requirements for water quality testing. Domestic/private wells provide water for residential use and serve less than 4 connections and are not regulated by the California State Water Board Division of Drinking water, which makes it harder to track how many people could be at risk of drinking arsenic-contaminated drinking water. Local health officials and state regulators are concerned about ensuring safe drinking water for both community water systems and domestic/private well owners [17]. The nature of improving and regulating water quality in small community water systems or state small water systems in rural areas in the US and around the world proves to be a worthy challenge. The United Nations' Sustainable Development Goal 6 "Ensure access to water and sanitation for all" emphasizes the importance of investing in safe drinking water for community benefits including the town's continued growth and the residents' continued empowerment [18]. However, limited funding for maintaining water treatment systems and water distribution systems, remain primary challenges for small community water systems to achieve this goal. Since rural communities are often spread across many miles, it is difficult and expensive to maintain piped water to each service connection and to guarantee the delivery of high-quality water. For unincorporated communities that rely on county-level funding, it is even harder to provide infrastructure updates or routine maintenance [19]. A lack of funding often poses a challenge in meeting regulatory standards for water quality [16]. Violations of the MCL can result in increased monitoring of the system, distribution of notices to the public, and sometimes fines imposed on the services district which can take away from already minimal financial resources [20][21][22][23][24]. Our research focused on two communities in the San Joaquin Valley (SJV) of California. The SJV is located in central California with the Coast Range to the west and the Sierra Nevada Mountains to the east. This region is of particular concern because it has two of the most contaminated aquifers in the United States [25]. Due to the reliance on groundwater aquifers as the source of drinking water, the residents of the SJV are at high risk of ingesting water contaminated with arsenic above its MCL, unless the drinking water is properly treated [26]. studies have shown that the highest cumulative contaminant concentrations in community water systems in California, including contamination from arsenic, occur there [27]. In particular, it is estimated that the SJV is the region with the most people exposed to arsenic across California [28]. Furthermore, the SJV is home to a diverse population of residents with the majority of residents living in largely low-income, rural communities supporting much of California's agricultural land [29]. The SJV communities also have some of the highest rates of poverty and minority populations in the state of California [25]. According to the 2021 American Community Survey conducted by the U.S. Census Bureau, 51.4% of the population identifies as Hispanic/Latino, compared to 31.1% identifying as White alone within the eight counties (Fresno, Madera, Merced, Kern, Kings, Stanislaus, San Joaquin, and Tulare) in the San Joaquin Valley [30]. This region is also home to nine Indigenous tribes who live throughout the area, often in rural areas or unincorporated towns. Communities of color (Hispanic/Latinos and Non-Hispanic people of color compared to Non-Hispanic White) are often served by community water systems, state small water systems, small water systems, and domestic/private wells are associated with greater drinking water contamination [31]. One report estimates that among the residents who are served by community water systems that have received "out of compliance" violations, 57% identify as Hispanic/Latino while only 35.8% identify as White [16]. Additionally, studies in the SJV have shown that community water systems that serve low socio-economic status communities have higher arsenic levels and higher odds of receiving a maximum contaminant level violation [5]. Together these studies indicate the environmental injustice that exists in the SJV where low-income, Hispanic/Latino people, tribal nation members, and/or Black Americans living in rural areas are disproportionately exposed to higher levels of arsenic through their drinking water sources. These environmental health disparities in the SJV were some of the drivers of our current project where we used a health equity lens to assess the community perceptions of arsenic-contaminated drinking water and preferences for risk communication about the possible health effects. Our study represented a major community assessment component of a larger study aimed at designing an effective and acceptable arsenic remediation technology. An arsenic remediation technology of this nature would reduce current injustices regarding the availability of clean drinking water across the world. Literature has shown that solutions to arsenic contamination must adapt to the social, economic, cultural, and institutional contexts of the community it serves [32]. There are currently two major approaches to reducing arsenic exposure at the community scale: (1) the use of an alternative safer water source, and (2) reducing the arsenic concentration in the current source [32]. Across the globe, we have seen communities use alternate sources by switching to deeper wells instead of shallow wells, in hopes of avoiding arsenic-contaminated shallow aquifers [33] or by purchasing arsenic-free water from "water vendors" [34]. To reduce the arsenic concentration in a contaminated source, a few different remediation technologies have been proven to be effective including reverse osmosis, oxidation followed by adsorption and filtration, and electrolytic methods, among many others [35,36]. There are also pilot studies for different remediation technologies that would be operated at the local community water system [35,37,38]. One such technology is Electrochemical Arsenic Remediation (ECAR) invented in 2010 by Gadgil and his colleagues at the Lawrence Berkeley National Laboratory, operated by the University of California [32,38]. The ECAR technology was successfully implemented to serve 5000 people daily in a rural community of Dhapdhapi in West Bengal, India in 2016, and lowered the arsenic levels from 250 <unk>g/L to below 3 <unk>g/L [38,39]. A second communityscale installation of ECAR, this one fully powered with rooftop solar, started operation in the small town of Bahraich in North India in early 2021. The Gadgil Lab at UC Berkeley concurrently worked on a significantly improved version of the ECAR technology, called Air Cathode Assisted Iron Electrocoagulation (ACAIE) which is hundreds of times faster (and therefore more compact and less cumbersome) than the original ECAR technology. ACAIE, developed in 2019, is also more efficient and effective at removing arsenic from contaminated water [40]. ACAIE aims to treat water at less than $0.25 per liter to provide more affordable access to safe drinking water. After the successes in India, the Gadgil Lab at UC Berkeley turned its interest to solving the arsenic crisis in its home state: California. Currently, this ACAIE technology is in the field-testing phase, which began in 2022 in a community located within the SJV. Our study represented a major community assessment component of the overall ACAIE remediation technology's ongoing community-engaged field testing in SJV. Community-engaged research prioritizing participatory approaches and community perspectives has been identified in environmental health literature across the world to be critical to identifying and developing sustainable community-relevant strategies to address clean water and other environmental health issues [41][42][43][44][45]. Prior studies have shown that engaging community leaders and residents in the design and implementation of environmental health interventions has facilitated the success of these interventions. Interventions and technologies that are developed in partnership with communities and incorporate community perspectives in their design have been shown to be more readily accepted, and have longer-lasting impacts on community health. Therefore, this research centered a community-based participatory research (CBPR) approach to support three long-term goals of the larger transdisciplinary project: (1) to better develop a community education and outreach program that increases community knowledge about the dangers of arsenic-contaminated drinking water, (2) to better facilitate the acceptance of the ECAR-ACAIE arsenic mitigation strategy by trying to learn the community's relevant concerns so we might address them during future implementations, and (3) to inspire behavior change regarding water consumption by community members. This work was conducted in two rural, unincorporated communities within the SJV, California. This project was built on pre-existing partnerships with two rural SJV communities: the unincorporated, agricultural regions of Lemoore and Allensworth, both with a high percentage of residents of color including Hispanic/Latino people, tribal nation members, and/or Black Americans. Evidence-based CBPR approaches include iterative processes through building relationships with community partners and promoting co-learning and information sharing, among other practices [43,44]. In this study, we developed local partnerships and generated information-sharing strategies in order to better develop a risk communication and education program that could be implemented to increase knowledge and awareness of a topic of concern: health risks associated with arsenic-contaminated drinking water. The purpose of this community health assessment as part of the larger transdisciplinary project was to understand (1) the awareness and perceptions of at-risk rural communities about their water sources and the health impacts of arsenic contamination, and (2) the preferences for risk communication and education regarding health impacts of arsenic-contaminated drinking water. --- Materials and Methods We used a participatory approach to conduct a qualitative community assessment, including 27 community informant (CI) interviews from September 2020-June 2022. Out of the 57 potential participants, 27 (47%) completed the telephone interview. Our Cis included diverse perspectives from community leaders and residents, school district members, and public water representatives from various organizations (see Table 1). This participatory, community assessment study explored community experiences and knowledge of safe drinking water and arsenic risks and identified barriers and preferences for communitywide risk communication and education strategies. The study results will be used to design and develop meaningful and relevant community-wide environmental health education initiatives to raise community awareness about the quality of their local water systems and water safety. --- Study Area As stated in the introduction above, the San Joaquin Valley of California has been found to have high levels of water contaminants-including arsenic-with community water systems and domestic/private wells that serve populations of color (Hispanic/Latino, tribal nation members, Black Americans, etc.) more frequently receiving maximum contaminant violations compared to their White counterparts. We partnered with Allensworth and the unincorporated region of Lemoore, both communities located in the SJV who are at high risk of drinking arsenic-contaminated water (see Figure 1). Both communities are unincorporated, agricultural rural towns with about 500 residents each with a majority population of residents of color. About 95% of Allensworth residents identified as Hispanic/Latino while the residents of the unincorporated region of Lemoore largely identified as tribal nation members or other minority populations. In these agricultural communities, people often live far apart from each other with limited or no wifi/technology access. Many have low literacy and/or limited English proficiency. --- Sampling Plan and Recruitment Participants were recruited using both purposive and snowball sampling strategies. These qualitative sampling strategies are particularly effective for exploratory/formative studies such as our community assessment where little is known about these communities and for reaching hard-to-count, small rural underserved populations that do not trust researchers [46]. We purposively sampled a variety of community informants as part of the qualitative triangulation of diverse community perspectives to confirm the community issues identified by the assessment could be confirmed by multiple informant sources. At the same time, this would collectively allow for the most comprehensive, in-depth understanding of the issues of water safety and water systems in these communities. These informants included diverse community stakeholders such as local community residents and leaders (Hispanic/Latino, tribal nation, and Black), school district members (officials, teachers, and parents), community water service managers, and environmental health officers. Community informants were eligible if they were 18 or older and resided or served the greater San Joaquin Valley including Allensworth and the unincorporated region of Lemoore. Participant outreach occurred via phone, text, email, or in-person by the interview team or local community liaison/partners. We recruited a variety of community informants that lived or worked in these rural communities (community residents, community leaders, school district members, water service providers/experts) as part of the triangulation of community perspectives to ensure the community issues identified would be confirmed by multiple sources. Community informant recruitment ceased when thematic data saturation was observed in the concurrent data analyses. Triangulation and saturation of the sample strengthened the credibility of our study. --- Data Collection All the interviewers on the research team were trained to conduct interviews to be able to follow the same interview protocol in an accurate, consistent way. Community informants participated in a 30-to 60 min semi-structured open-ended telephone interview conducted by a member of the research team. Electronic or verbal informed consent was given to the interviewees via email or phone prior to proceeding with scheduling an interview. Before the beginning of each interview, the consent information was repeated verbally, and they were asked to reconfirm their consent. In the telephone interview which included 11 questions, participants were asked to elaborate on (1) their opinions and experiences with safe drinking water, (2) knowledge about arsenic risks in their communities, and (3) barriers and preferences for community-wide communication and education strategies. We asked two additional questions pertaining to the impact of the COVID-19 pandemic in 2020. All the interviews were recorded and transcribed. Interview notes were also taken during the interviews. --- Data Analysis Interview transcripts and notes were compiled for qualitative analysis. The constant comparative method was used as a technique for the qualitative thematic analysis and took into consideration key qualitative methodologies of triangulation, saturation, and credibility [47]. This method developed codes, examined relationships and interactions across descriptive and thematic codes, and compared the major themes that emerged from the coding categories. Qualitative analysis probed for parallel themes, particularly looking for knowledge and understanding of safe drinking water and associated arsenic risks, and barriers and preferences for risk communication and education to increase community awareness of arsenic-contaminated drinking water. The final codebook consisted of descriptive and thematic codes common across the 27 community informant (CI) interviews. Five research team members conducted the qualitative analysis and coding. These research team members were trained to use the same, consistent coding approach. Two to three research team members independently coded and then met together to discuss and compare each other's coding for each of the transcripts from the CI interviews and come to an agreement with the descriptive and thematic codes for the data analysis. Inter-rater agreement for the interview transcripts was determined to ensure consistency in coding. For each interview transcript, if 80% agreement in coding consistency was not reached, the researchers discussed potential issues that arose and reached a consensus about these coding issues until consistency was reached. The coding of the transcripts was an iterative process. The codebook was revised as subsequent transcripts were coded and new codes emerged. Coding consistency was recalibrated as part of this iterative coding process. Code categories were connected and grouped through thematic coding, and the researchers identified major themes from the codes. In addition, we reached data saturation mid-way through the coding of the 27 informant interview transcripts but continued completing the analysis of all 27 transcripts. We analyzed substantially more than the number of interview transcripts needed to reach data saturation or when additional interviews did not provide any new data or information and started to repeat what was previously coded. The qualitative coding used the software Dedoose Windows version is 9.0.62. --- Results --- Summary of CI Characteristics In total, 27 qualitative community informant interviews were conducted. Of the 27 community informants, 52% were female and 48% were male. 48% of participants reside or serve in the unincorporated region of Lemoore, California, 44.5% in Allensworth, California, and 7.5% from nearby areas in the San Joaquin Valley. The community informants have a range of backgrounds including community leaders, residents, and water professionals representing various organizations (Table 1). --- Community Understanding of Arsenic Risk in Drinking Water --- Community Perceptions and Experiences with Arsenic Exposure in Drinking Water Systems Community informants (CIs) expressed their awareness and perceptions of arsenic in their drinking water and shared their concerns about the failing water systems in their rural, farming communities of color. All CIs indicated they were aware of arsenic in their drinking water. Almost all participants also expressed serious concerns about the persistent risk to arsenic in their drinking water. Of those that did not express a negative perception, none of them expressed a positive perception either. One participant spoke of the ubiquitous nature of arsenic in their community stating: "Arsenic is just prevalent in this area. It's naturally occurring in our groundwater and so very likely to be present in public water systems, private wells, pretty much any groundwater source it's likely to be found." (Public Water Representative) Most CIs stated that the problem of arsenic contamination in their drinking water has been a long-term problem historically and continues today. Many CIs perceived that the water in their community was not safe for drinking and had concerns that the existing community water utility system was "archaic" and "unreliable." One participant expressed their frustration saying, "I mean, here in [town name], you never know what you're drinking, because I mean, there's so much that we don't know." (Local School District Member) Many CIs shared frustrations that community residents were being left to find their own solutions. Many CIs stated that a lot of people, including themselves, relied on bottled water for drinking and cooking due to the water contamination in the community water system or a local well, and expressed frustration at the injustice of this. Some CIs with the local schools or community organizations indicated they were able to implement temporary solutions such as point-of-use filters or free bottled water with support provided by organizations like Rural Community Assistance Corporation or Self-Help Enterprises. A few CIs discussed their hopes for a longer-term solution through digging and building a new community well that is arsenic free with funds from the state of California. However, they indicated they had been let down as no progress had been made by the state in the past ten years after funding for the community well was initially approved. A few CIs from the local schools also expressed a concern that some of the school educational funds typically allocated for teachers and students were being reallocated in order to provide the students with bottled water. Less funds for education might affect the quality of education. However, regardless, all the CIs interviewed who were associated with the local schools were concerned about ensuring both safe drinking water and a good education for the students. One community informant stated, "It [Arsenic-contaminated water] affects the student's learning capacity. I imagine what it does to someone that's 40, 50, 60, 70 years old drinking the same water. Again, it deteriorates the [student's] body from within and that has a big effect." (Tribal Nation Leader) Funding a long-term, safe drinking water solution for their communities was a hope shared by many of the CI participants. However, financial constraints of these rural communities limited their options for clean water systems. Another challenge that a few community informants identified was that water quality changed depending on where a resident lived in the community. One participant acknowledged the challenges associated with arsenic contamination at a system level and expressed frustration at what arsenic contamination did to individuals: "The mere fact that we have arsenic, any trace of arsenic in the water, is pushing a lot of very low-income people to go out and buy water [even though they cannot afford it]." (Local School District Member) 3.2.2. Lack of Understanding about Health Risks from Exposure to Arsenic and Other Water Contaminants in Drinking Water among Community Residents Most CIs shared their lack of or misunderstanding of health concerns related to the arsenic contamination in drinking water within their rural, farming communities of color. A majority of CIs expressed that they were not knowledgeable of the severity and outcomes of the health effects caused by arsenic. Only some CIs correctly stated that arsenic contamination would affect their health negatively in the long term. Even though most CIs participants believed themselves to be far more knowledgeable than the rest of the residents in their communities about the impact of arsenic-contaminated water, most of them shared that the extent of their understanding of health concerns directly related to arsenic contamination was limited. These participants also mentioned that most of the residents in their communities were not aware of arsenic or other water contaminants in their drinking water and the associated health risks. Only a few CIs specifically expressed concern that arsenic or poor water quality could be connected to cancer rates within their communities, which is a known health effect of long-term exposure to arsenic. A couple of CIs also expressed concern that some community residents did not understand the severity of long-term exposure to arseniccontaminated water. While a few CI participants noted that their water has a poor taste and odor believing those qualities to be associated with arsenic, only one participant correctly indicated that arsenic is odorless and tasteless. A couple of CIs stated that some community residents liked the taste of the water and continued to drink it, even though they knew it was contaminated with arsenic and that it was not good for them. The CI participants indicated the community residents' risk from arsenic exposure and its associated health effects were not perceived to be severe enough to deter them. Limited knowledge of the health effects of arsenic contamination in drinking was paired with misinformation about other water contaminants. With respect to additional drinking water contamination sources, a few CI participants also identified lead, bacteria, nitrate, chromium, and pesticides as other possible contaminants that altered the states of water in their communities. One CI stated that they believed another contributor to limited knowledge about arsenic was that lead was discussed a lot more than arsenic in their communities. Although water contamination was a topic often discussed among the community informants, they reported that the residents of the wider community likely had little or no understanding about arsenic and other water contaminants in their community's drinking water and the adverse effects on their health. --- Reasons for the Lack of Understanding about the Health Risks from Arsenic Contamination in the Drinking Water According to multiple CIs, arsenic contamination to the community's drinking water was often not a forefront concern for many in the community for a variety of reasons such as the lack of participation in regular community meetings about water quality, poverty, busy with their agricultural work, low literacy, and limited English proficiency. Some CIs cited one of the reasons as a lack of attendance at the monthly community action meetings where arsenic contamination in their drinking water was regularly discussed. One participant noted the importance of residents attending community action meetings where arsenic contamination was often discussed in order to stay informed about this issue by stating: "There needs to be more awareness of the water and things in the community. They [the community] don't get involved a lot, and then, they complain, and then, I tell them well, if you are going to complain, why don't you go to the meeting?" (Community Resident) Behind this lack of attendance at community action meetings, some CIs indicated there were systemic barriers that must be considered including race, poverty, and language barriers. Some of the CIs believed that the poverty level, low literacy levels, limited English proficiency, and farm worker profession of the wider rural community puts these communities at a higher risk for misunderstanding the existing water safety crisis and therefore continued to consume arsenic-contaminated drinking water. Some CIs also believed that the residents of the wider community did not have adequate income to spend on purchasing bottled water and, at the same time, did not have the literacy level or English language proficiency to fully understand community announcements about contaminated water. Additionally, some CI participants indicated most community residents were just too busy trying to survive each day to support their families and be financially self-sufficient in the farming community. Additionally, CIs discussed the water board or water utility's distribution of water district reports, statements, or flyers to the wider community. A few CIs stated that monthly water bills provided information on water contamination in English and Spanish, but stated that community residents might not take the time to read or understand these notices because of their busy work and family life in the farming community. Even though communications from the water board/utility were provided in both Spanish and English most of the time, many CIs mentioned they were worried that community residents did not read the reports due to low literacy levels in both English and Spanish. Another CI stated that they believed that a low understanding of safe drinking water was due in part to the minimal effort the public utility departments put into explaining the urgency of the issue to community residents in their reports. --- Barriers and Preferences for Risk Communication and Education --- Structural Barriers to Risk Communication and Education In the interviews, CIs reported primary barriers to risk communication about environmental health risks associated with arsenic-contaminated drinking water in these rural, farming communities of color and their challenges with reaching community residents through existing communication and outreach approaches. These barriers included a lack of community residents attending or participating in the monthly community action meetings due to other work or family priorities and limited accessibility of the existing environmental health information materials related to arsenic and other contaminants in drinking water through existing public distribution channels. Additional barriers identified were difficulties in understanding existing high health literate informational materials due to language barriers and low literacy levels and the fragmented nature of these rural communities geographically and legally for outreach (i.e., incorporated regions, unincorporated regions, tribal nations). Some CIs indicated regular community action meetings often discussed water quality and safety, but few or no community residents participated or prioritized going to them. This may have led to the lack of awareness of drinking water safety issues and was one of the primary barriers to risk communication. However, some CIs believed this lack of community participation in these meetings likely stemmed from deeper structural issues. Many CIs spoke of community residents having trouble attending meetings due to job schedules such as working long hours as farm workers or caring for their families. Succinctly put by one community informant: "The challenge is always trying to figure out what's the best sort of day and time to do a presentation. You have working families, you have people who work different times and days of the week and so it's always trying to figure out when it's going to work best for the majority of the community so that they are able to attend" (Public Water Representative). A number of CIs noted that even before the COVID-19 pandemic, attendance at community action meetings such as town halls was low and attendance worsened due to shelter-in-place during the pandemic. When meetings shifted to an online format during the COVID-19 pandemic, some CIs noted that limited internet access or limited ability to pay for internet-based access was an additional barrier that was linked to a decrease in community meeting attendance. One CI stated some of the challenges related to COVID-19 and meeting attendance: "We've tried a number of times to get people to dial in [for meetings], but that's the other part of the problem. We don't have good internet access here. It's improving, but it's not free. For somebody that has limited funds, I mean subscribing to an internet service is not viable. To try to get them to join the Zoom call, especially if they're going to pay to join on their cell phone, makes no sense to them. We've had that as a challenge so far." (Community Resident) CIs indicated that hopefully the COVID-19 pandemic will ease up and community meetings can return to in-person events which should lighten some of the barriers the community faces in attending these meetings. Other barriers that were cited by almost all CIs were (1) the lack of awareness about the severity of the issue from the public notifications about arsenic-contaminated water they received, (2) the language barriers in public communication and outreach, particularly to Spanish speakers, and (3) the high health literacy level of these public outreach materials distributed in these low literate, rural, farming communities of color. Some CIs shared that community organizations and water departments produce their own public notifications, mostly in the form of quarterly newsletters, with updates on water quality, although notification updates specific to arsenic and health were limited and only present when levels exceeded the safety standard. Some of the CIs stated that they had previously received communications from the water district about unsafe drinking water, but that these communication materials did not reflect an urgency about the arsenic contamination in their drinking water. A couple of CIs said that the water district also mailed and delivered bilingual materials to each household that notified them about their water quality. However, a couple of CIs also stated that they did not remember if these materials were in both English and Spanish or only in English. One CI stated that there was a "huge language gap" and that people with this Spanish language gap were the ones most adversely affected by the arsenic-contaminated water in their drinking water. Additionally, some of the participants stated that the community literacy level was low and that the high literacy level of the written materials presented in the community meetings or in public communications might not be the most effective way to disseminate this information. The dual language and literacy barriers were major communication and outreach issues. Another barrier unique to our study population was the fragmented nature of the towns geographically (i.e., incorporated regions, unincorporated regions, tribal nations) with people living far away from each other which is especially common in tribal nations. Additionally, due to tribal sovereignty, there are legal barriers that exist for tribal nations in these rural regions which complicate public water sharing because of the different regulations (state vs. federal) that govern water quality. --- Preferences for Education and Risk Communication about Drinking Water Safety When asked about preferred strategies to improve education and communication about safe drinking water and water contaminants, CIs responded with several suggestions including the importance of educating the school children and having them bring this information back to their families and community residents. Additional suggestions were ensuring easy-to-understand bilingual materials are provided in English, Spanish, and indigenous languages, building trust with these communities for communication and outreach, and a few other public health education and promotion strategies. The most discussed strategy to improve risk communication about drinking water safety and health within these rural, farming communities of color was to provide educational materials about safe drinking water to the students at the local elementary and middle schools. Some CIs believed that by providing education about safe drinking water to the students they would then be able to take that drinking water safety and health information back to their parents and community residents, a practice known as language brokering. A CI that works directly with students affirmed that they found their elementary and middle school students often shared what they learned at school with their parents or guardians at home. One participant stated this philosophy simply: "Educate the young people, they'll educate the community, and then we can move forward in a positive way that will benefit everybody." (Rural Community Leader) Another CI acknowledged that this communication strategy was already being implemented through their community's annual project-based learning summer workshops for local high school students and described its success by stating: "The kids basically have a little seminar, and then [afterwards] they talk to their parents at least somewhat about it. Again, it depends on whether their parents are interested in or concerned about those kinds of issues or not and whether or not they'll hear them out. From what I've noticed, it seems to be the case that the information usually at least gets to their parents." (Rural Community Leader) One CI also suggested developing and holding annual water quality and environmental health training for elementary and middle school teachers, so they would be better prepared to discuss water topics and answer any questions that their students might have. Many CIs explained the importance of having in-language educational materials that the students could bring back to their parents in English and Spanish since most of the students were bilingual and many parents were monolingual, Spanish speakers. By having bilingual materials, parents who were literate in Spanish and/or English could engage with their children's school materials and for parents who were not literate in either language, these students could elect to read the materials to their parents in their preferred spoken languages. A few CIs cited this bilingual communication and outreach approach to have been successful thus far through their annual project-based learning summer workshops for high school students. They believed that it could work similarly if adapted to the school year curriculum and implemented in their local elementary and middle schools to improve community education about drinking water safety and health. A majority of CIs suggested that relevant community educational materials developed by external community educators or partners about safe drinking water in English should also be translated in-language (e.g., Spanish) to have a larger impact on outreach to the broader community. A participant said that the community action meetings they have attended have been "only in English" which has led to a divide in the knowledge level between English speakers and non-English speakers. Many CIs stated that in-language communication and outreach should be applied to flyers or written materials that were sent out or distributed at these community action meetings or other local events so that these educational materials could reach as many people as possible across the community. Many CIs described their towns as "

Due to chronic exposure to elevated levels of arsenic in drinking water, thousands of Californians have increased risk for internal cancers and other adverse health effects. The mortality risk of cancer is 1 in 400 people exposed to above 10 µg/L of arsenic in their drinking water. The purpose of this community assessment was to understand the perceptions and awareness of the residents and public water representatives in rural, unincorporated farming communities of color in San Joaquin Valley, California. In our research, we asked 27 community informants about their (1) available water sources, (2) knowledge about the health impacts of arsenic, and (3) preferences for risk communication and education regarding the health impacts of arsenic-contaminated drinking water. Through our qualitative coding and analysis, we found that most community informants indicated that there was limited community awareness about the health effects of drinking water with elevated arsenic levels. Preferences for risk communication included using in-language, culturally relevant, and health literate health promotion strategies and teaching these topics through the local K-8 schools' science curriculum with a language brokerage approach to transfer student knowledge to family members. Key recommendations include implementing these communication preferences to increase community-wide knowledge about safe drinking water.

's school materials and for parents who were not literate in either language, these students could elect to read the materials to their parents in their preferred spoken languages. A few CIs cited this bilingual communication and outreach approach to have been successful thus far through their annual project-based learning summer workshops for high school students. They believed that it could work similarly if adapted to the school year curriculum and implemented in their local elementary and middle schools to improve community education about drinking water safety and health. A majority of CIs suggested that relevant community educational materials developed by external community educators or partners about safe drinking water in English should also be translated in-language (e.g., Spanish) to have a larger impact on outreach to the broader community. A participant said that the community action meetings they have attended have been "only in English" which has led to a divide in the knowledge level between English speakers and non-English speakers. Many CIs stated that in-language communication and outreach should be applied to flyers or written materials that were sent out or distributed at these community action meetings or other local events so that these educational materials could reach as many people as possible across the community. Many CIs described their towns as "close-knit communities" and ones that took care of one another. CIs also described their rural, farming communities of color as not trusting or receptive to outside assistance, even if the potential partner organizations were there to provide a solution for the arsenic crisis. A suggestion made by some CIs to external research partners was the importance of working closely with their community through cultivating relationships over time with trusted community residents who could advocate and support a research group's presence in their communities. A few other strategies to improve communication and outreach about arsenic-contaminated drinking water were presented by CIs. Some CIs agreed that a community action meeting with a visual aid or presentation would be the most effective educational format for meeting participants to increase their awareness about safe drinking water and water contaminants. Some participants also encouraged external community educators or part-ners to take part in their regular community or town events throughout the year to be able to reach and educate an even broader community. One participant stated: "If you're able to set up basically an event within the [community] event that's already there, you'd be guaranteed a pretty good amount of community exposure, which then would allow you to hopefully get a good amount of community participation." (Rural Community Leader) A few CIs indicated that these community action meetings should also include the incentive of free food. Previous town meetings that included free food were more widely attended by community residents compared to meetings that did not include free food. Another CI emphasized the importance of word-of-mouth as a primary communication in these rural, farming communities of color stating that, "Word of mouth could go a long way. If you get somebody, like if my Tia (aunt) or my comadre (godmother) would tell me, then I would trust her, right? Because I feel like she's a trustworthy person. I think if you could just talk to as many people as you can and hopefully the word of mouth gets around." (Community Resident) Finally, another CI said that using personal stories about arsenic in drinking water or other contaminated water from local town residents could go a long way because people related more to these personal stories from town residents. --- Discussion In this study, we learned from community informants in rural, farming communities of color about the little or no community awareness of health risks associated with arseniccontaminated drinking water and the longstanding problems of their water systems in these unincorporated rural regions of San Joaquin Valley (SJV). We also learned about existing barriers to risk communication and education due to low literacy and limited English proficiency. Community informants made a number of suggestions for strategies to increase the risk communication and community education about the health risks of arsenic in drinking water including developing an innovative Science, Technology, Engineering, and Math (STEM) education curriculum about drinking water safety and health starting with educating local elementary school and middle school students in these communities. They mentioned that the knowledge these students learned in school or workshops could then be brought home and transferred to their families and community residents. Another suggestion was to incorporate linguistically inclusive risk communication messaging in Spanish and indigenous languages through community gatherings and trusted messengers. Additionally, some of the limitations of this study including sampling methods, interview techniques, and the impact of the COVID-19 pandemic are addressed. --- Community Awareness and Barriers to Knowledge Acquisition Our results identified the lack of awareness among rural community residents about health risks associated with arsenic, the predominant concerns of community residents, and the existing barriers to gaining knowledge about these risks. Many CIs stated their belief that their communities were at high risk for arsenic contamination in their groundwater, which is supported by the recent literature that showed that the highest cumulative contaminant concentrations (Arsenic, Nitrate, etc.) occurred in the San Joaquin Valley [27]. Despite expressing high general awareness about the risk for arsenic contamination in drinking water, the CIs expressed limited knowledge about the specific health risks associated with arsenic (cancer, skin lesions), similar to what was found in one 2004 study where knowledge of arsenic as a risk to health was high, but knowledge of specific health risks associated with consumption of arsenic contaminated water remained low [48]. Additionally, CIs expressed concerns that the outdated infrastructure of their community water system or state small water system was perpetuating poor water quality in their unincorporated rural, farming communities. Existing literature supports these concerns and shows that community water systems serving lower socioeconomic-status populations, such as the residents in Lemoore and Allensworth, are more likely to receive water contamination violations due to a lack of resources within the community to address these issues [5,20,21]. When these violations occur, those who are responsible for local water quality are required by law to let the community know. However, CIs also reported that these public violation notifications were often confusing or not understandable due to their high literate reading level and lack of clarity about the severity of reported water quality. One recurring barrier the CIs reported was the low literacy level of their community and the associated high literate-risk communication materials, stating that this barrier could be the reason that community residents did not know about the health risks associated with arsenic-contaminated drinking water and why they did not understand the public notices provided by the water service district. Research conducted in 2017 by the National Center for Education Statistics (NCES) for the Program for the International Assessment of Adult Competencies (ages showed that 52% of people living in the United States had a literacy score of level 2 or below indicating intermediate or low levels of literacy [49]. According to the same study, 67% of non-native-born residents compared to 49% of nativeborn residents in the United States had a literacy score of level 2 or below [49]. A 2003 NCES study examined health literacy for adults and found that 88% of adults had intermediate or low levels of health literacy, and Hispanic/Latino adults had lower average health literacy across all racial/ethnic groups [50]. Health communication experts stated the importance of creating simple, direct messages to better reach diverse audiences, especially for those with low literacy levels [51][52][53]. When working with communities, health educators need to listen to the risk communication barriers expressed by community members and work with the community to identify and implement the relevant community strategies needed to raise awareness about the risk of water contaminants and sources of safe drinking water. This community-defined health promotion approach can be a pivotal point in tackling the lack of awareness about the arsenic crisis in groundwater across San Joaquin Valley. --- Community Suggestions for Health Education Strategies Our results identified novel approaches to raising community awareness about water contaminant risks by providing Science, Technology, Engineering, and Math (STEM) education about drinking water safety and health to elementary school and middle school students and for the students to transfer their knowledge home. These results also identified health promotion and education approaches to community adults of color by creating linguistically inclusive materials for health communications in Spanish and indigenous languages that affirm the existing literature. The primary approach that the CIs suggested as the most effective way to inform more rural, farming community residents about the health risks associated with arsenic-contaminated drinking water, was to educate the primarily Hispanic/Latino and tribal nation farming communities' children and adolescents first as a step toward educating their family members and other community adults. Since these STEM education courses would be designed for children, this approach would use "language brokering" between children and parents as a strategy to educate adults with limited English proficiency and/or low literacy about the risks of arsenic-contaminated drinking water. Minimal research exists examining the approach proposed by the community informants where the youth would take a STEM class about drinking water safety and health and pass knowledge along to their parents or other adults in the community. One study conducted between 2008 and 2010 in Bangladesh showed the success of a community education intervention with school children ages 8 to 11 years old [54]. This study demonstrated an increase in knowledge for students across 14 elementary schools in rural villages in Bangladesh. Among students who received the education intervention in comparison to those who did not receive the education intervention, the intervention group reported a higher rate of behavioral change in using cleaner drinking water wells [54]. Although the cited study in Bangladesh was not a community-defined education approach, we have hope that in taking the CI recommendation about educating the children as a means to educate family members and adults in the community, we could see similar results in rural, unincorporated towns in San Joaquin Valley, California when we couple a STEM education program with clean water sources. We were not surprised that most CIs overwhelmingly stated the importance of providing community health information in both English, Spanish, and Indigenous languages to be inclusive of all community residents. The COVID-19 pandemic has made clear the importance of culturally relevant and linguistically available public health messaging to overcome linguistic barriers contributing to health disparities across communities of color where the primary language is not English, as emphasized in previous literature [32]. CIs also discussed the phenomenon of language brokering where residents who have limited English proficiency rely on others (often children or adolescent youth) to translate presentations and communication materials, including health information, from English into the minority language [55][56][57]. By developing linguistically inclusive community health information, the need for adolescent language brokering in everyday instances could be reduced. --- Limitations Through the use of purposive sampling aided by our partnership with community leaders, our CIs may have been largely representative of the more involved community residents and community leaders rather than those who are less involved in the community which could have introduced selection bias. To address this type of bias, we asked all CIs if they could recommend other potential participants in line with snowball sampling. We also sought out additional CIs at a community event to recruit residents of the wider community. Since we worked closely with rural and unincorporated communities of color, our results might not be generalizable to populations in larger towns or in communities that do not reflect our study's population demographics. Interviews were not anonymously conducted between the interviewer and the interviewee, which may have led to social desirability bias in the responses that were collected. All responses were coded and de-identified before sharing any results outside of the research team. However, due to the nature of the interview, the researcher knew the name of the participant they were speaking with. As such, responses may have been given in a manner that seemed to be a "correct answer" instead of one that was true to the participant out of fear that answers given would be shared with community leaders or other community residents. CI interview guides were structured in such a way as to prevent leading questions that might increase the likelihood of this type of bias; however, full anonymity could not be provided during the interviews and thus social desirability bias is always a concern. Due to the COVID-19 pandemic, all interviews were conducted over the phone or through a virtual meeting platform which might have hindered our relationship-building capacity with community informants. This also led to limitations in data collection in that we could not view body language or have interactions outside of scheduled phone calls. Additionally, the pandemic has limited our in-person interactions with our community partners which could have hindered some of the trust we were hoping to build with the community and could have skewed interview participants' opinions about the study's community engagement approaches. --- Conclusions Our community assessment was designed to understand (1) the awareness and perceptions of rural, unincorporated communities of color about their water sources and the health risks of arsenic contamination from them, and (2) the barriers and preferences for risk communication and community education about drinking water safety and health. A major lesson learned for conducting research in hard-to-reach communities that we identified, and is affirmed by prior environmental health research studies [41], was that community-engaged research that prioritizes participatory action approaches, community input, and community feedback in all phases of the study are critical to identifying and developing sustainable and community relevant strategies to address clean water and other environmental health issues. The results from this study emphasized the overwhelming need for more communitywide education about the health risks associated with arsenic contamination of drinking water. Our results provided community-defined recommendations for communication strategies including the need for health-literate, linguistically inclusive messages and developing an elementary and middle school bilingual Science, Technology, Engineering, and Math (STEM) education curriculum to increase youth, family, and community knowledge about the health risks of arsenic exposure from local drinking water. As a next step, we plan to design and translate risk communication/educational strategies identified from our results for our environmental health community education initiative about water safety and health in these communities for K-8 students, high school students, and adult community residents. It is important to incorporate our community assessment findings into future public health intervention programs for rural, unincorporated regions that continue to rely on groundwater drinking water sources with exposure to water contaminants including, but not limited to arsenic. Additionally, it is important to take community perspectives into account for future public health interventions when there are acute water crises, such as water supply interruptions as described in a previous study [58], in addition to chronic water crises, such as the arsenic crisis in the SJV. We recommend that any public communications related to contaminated drinking water should be (1) linguistically and culturally inclusive for the predominant language and racial/ethnic communities in a specific geographic area and (2) written in easy-to-understand language for residents who have low literacy levels in English, Spanish, or another language. For the towns we partnered with, we recommend directly implementing our study findings and beginning a language brokering approach via STEM education curriculum in the local schools to increase student, family, and community knowledge about arsenic contamination. Using a language brokering approach for risk communication that brings home the in-language information about the safety of community water systems and arsenic risk in local drinking water to the limited-English proficient and/or low-literate parents and family members from their educated children who learn about these issues of water contaminants and water safety through their STEM curriculum in the local schools. Children or students can bring home this knowledge to adults and children in their homes and improve parent, family, and community knowledge about arsenic contamination and safe drinking water. In addition, children who learn about these issues early in life will have the potential to one day transform their communities into safe water havens and sustain the knowledge and skills needed to ensure water safety in their communities into the future. For similar rural, unincorporated towns who are at risk for drinking contaminated water, we recommend speaking with community residents and leaders to understand their perspectives and preferences for risk communication and education which may also include an educational intervention for school-aged children. Incorporating questions about residents' primary source of drinking water (domestic wells, small community water systems, etc.), and conducting baseline knowledge testing about water contamination, might also help researchers structure appropriate intervention programs. If our future community outreach and education program approach proves effective, it could be adapted to increase awareness of additional environmental hazards among other rural, unincorporated towns, particularly those in the San Joaquin Valley of California who are at risk of drinking from contaminated water sources. --- Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available for the audio recorded interviews due to confidentiality. However, de-identified transcripts can be provided upon request. --- Conflicts of Interest: The authors declare no conflict of interest.

Due to chronic exposure to elevated levels of arsenic in drinking water, thousands of Californians have increased risk for internal cancers and other adverse health effects. The mortality risk of cancer is 1 in 400 people exposed to above 10 µg/L of arsenic in their drinking water. The purpose of this community assessment was to understand the perceptions and awareness of the residents and public water representatives in rural, unincorporated farming communities of color in San Joaquin Valley, California. In our research, we asked 27 community informants about their (1) available water sources, (2) knowledge about the health impacts of arsenic, and (3) preferences for risk communication and education regarding the health impacts of arsenic-contaminated drinking water. Through our qualitative coding and analysis, we found that most community informants indicated that there was limited community awareness about the health effects of drinking water with elevated arsenic levels. Preferences for risk communication included using in-language, culturally relevant, and health literate health promotion strategies and teaching these topics through the local K-8 schools' science curriculum with a language brokerage approach to transfer student knowledge to family members. Key recommendations include implementing these communication preferences to increase community-wide knowledge about safe drinking water.

levels of defending behavior. In contrast to these studies and studies on initiating bullying, no prior studies have examined under which conditions children are more or less likely to participate in bullying behavior more generally (initiating, assisting, and reinforcing bullying). To bridge this gap in the literature, the current study will focus on individual-and classroom-level predictors of these behaviors, and more specifically on the cross-level interactions between the predictors. The longitudinal approach in the current study also adds to the extant literature on bullying which often is cross-sectional in design. --- Individual Characteristics and Bullying Behavior Bullying takes place on a large scale despite the fact that children generally hold a negative attitude toward bullying (Andreou et al. 2005;Boulton et al. 1999;Menesini et al. 1997;Rigby and Slee 1991). Indeed, individual bullying-related attitudes are only weak or at best moderate predictors of bullying behavior (e.g., Boulton et al. 2002;Rigby 2004;Salmivalli and Voeten 2004). Why then do children engage in bullying? It has been suggested that bullies have the primary aim to obtain or maintain a dominant position in the peer group (Adler and Adler 1998;Pellegrini and Long 2002). In line with this it has been found that bullies strive for respect and dominance (Ojanen et al. 2005;Sijtsema et al. 2009). In addition, Olthof et al. (2011) showed that bullying behaviors are associated with both social dominance and the desire to be dominant, with the ringleader bullies (initiators) scoring the highest on these measures. It seems that bullies get what they aim for, as research has shown that both dominance (power) and status (popularity) are key to bullying (Prinstein and Cillessen 2003;Vaillancourt et al. 2003). Importantly, researchers have studied the construct of popularity as distinct from likeability, because children who are perceived as popular by their peers are not by definition liked (Adler and Adler 1998;Cillessen and Rose 2005). While bullying is related to dominance and status ("perceived popularity"), most studies report that bullies are at the same time more rejected by their peers than non-involved children (Boulton and Smith 1994). Salmivalli and colleagues (1996) have also found that among girls, bullying is related to both peer rejection and peer acceptance, which might indicate a controversial status in the peer group. Moreover, the study of Olthof and Goossens (2008) showed that boys who engage in bullying behaviors desired particularly the acceptance of other boys who engage in similar behaviors, although they did not actually receive this desired acceptance. To summarize, although bullies are often perceived by their peers as popular or respected, they are rarely well-liked. Moreover, although popularity and likeability refer to an individual's relative social standing in the peer group, it is the group that assigns status or acceptance to its members. Thus, children depend on their peer group (e.g., classroom) in the realization of their goals related to affection or status (Caravita and Cillessen 2012;Salmivalli 2010). This opens the door for grouprelated influences with regard to social standing (popularity and likeability) among peers and involvement in bullying behaviors. --- Group Characteristics and Bullying Behavior Next to individual attributes, classroom characteristics are also likely to contribute to bullying behavior. Classrooms are social settings of which children are involuntary members and where they spend most of their time at school. Classroom differences in bullying frequencies can be regarded as classroom norms; they refer to codes of conduct that prescribe behaviors that members of a group can enact (Bendor and Swistak 2001). Descriptive norms refer to how widespread certain behavior is within a group. The greater the prevalence of behavior, the more likely individuals believe that engaging in that behavior is normative and legitimate (Ang et al. 2010;Rimal and Real 2003). One way to assess descriptive classroom norms is to aggregate bullying behaviors to the classroom level, that is, the extent to which bullying behaviors are on average displayed by children in a classroom (cf. Chang 2004). Several studies have shown that such descriptive norms indeed influence individual behaviors, over and above individual-level correlates. This was shown, for example, for aggression, (Mercer et al. 2009;Thomas et al. 2011), defending (Pozzoli et al. 2012), prosocial behavior (Chang 2004), and also for bullying (Sentse et al. 2007). Hence, descriptive norms on bullying might explain why a child is more likely to bully in some classrooms than in others. At the same time, we have to account for the injunctive norm of the group, that is, the attitudes toward bullying that are on average held by children in the classroom. That is, next to prevalence of behavior, collective attitudes towards such behavior are another source of how normative and legitimate behavior is for individuals in the group and it can be assumed that children are more inclined to bully when their context (classroom climate) is permissive with regards to such behavior as compared to when their context is less permissive. Some researchers have argued that the injunctive norm is more important than (i.e., overruling effects of) the descriptive norm (e.g., Henry et al. 2000, for aggression, andPozzoli et al. 2012, for defending) whereas others have shown the opposite, with descriptive norms being more influential than injunctive norms (Scholte et al. 2010, for bullying). The latter study showed that the injunctive norm with regards to bullying is known to be quite disapproving of bullying but, just like individual attitudes, only weakly to moderately related to actual (individual) involvement in bullying (Scholte et al. 2010). The reason for why classroom norms cause children to engage in bullying behavior may be because it is rewarding to do so. In a cross-sectional study it has been found that in classrooms where bullying occurred at high levels (i.e., was normative), bullying was less likely to be associated with peer rejection and more likely to be associated with peer acceptance (Sentse et al. 2007). Hence, one reason to act in correspondence with the classroom norm is the fear to be rejected by classmates if one's behavior deviates from what is observed to be the norm. More specifically, it can be hypothesized that (1) over time, rejected or low accepted children will engage in bullying behavior especially in classrooms with a "pro-bullying" norm, i.e., the motivation to act in line with the classroom norm out of fear to be (increasingly) rejected by their classmates if their behavior deviates the classroom norm. This "pro-bullying" norm means that either the descriptive norm is high (i.e., high levels of bullying behavior in the classroom) or the injunctive norm is low (i.e., low levels of anti-bullying attitudes in the classroom). Secondly, as bullying is related to dominance and status (popularity), children may engage in bullying behavior to obtain, maintain, or even increase their dominant position in the group (Cillessen and Borch 2006;Prinstein and Cillessen 2003). This reasoning, however, results in two contrasting hypotheses with respect to norms. First, as reviewed above, popular children are not necessarily liked. Thus, if popular children want to obtain or maintain their status in the group without losing affection (likeability), it can be expected that their engagement in bullying behavior dependents on the classroom norm. Over time (2a), popular children can be hypothesized to engage in bullying behavior, especially in classrooms with a "pro-bullying" norm, i.e., the motivation to act in line with the classroom norm to obtain or maintain a dominant position in the classroom without the risk of losing affection. However, as popularity is often associated with (negative) behaviors that distinguish these children from the rest of the group (e.g., Dijkstra et al. 2009), it could also mean that (2b) the motivation of popular children to obtain or maintain their status in the group leads them to engage in bullying behavior especially in classrooms with an "anti-bullying" norm. The latter means that either the descriptive norm is low (i.e., low levels of bullying behavior in the classroom) or the injunctive norm is high (i.e., high levels of anti-bullying attitudes in the classroom). --- The Present Study The present longitudinal study will investigate the extent to which individual characteristics and classroom norms are predictive of bullying behavior during 1 year in grades 4 to 6. Individual level factors include anti-bullying attitudes and indices of social standing in the classroom (acceptance, rejection, and popularity). Based on the arguments and previous studies described above, rejection and popularity are expected to be positively associated with bullying behavior, whereas anti-bullying attitudes and acceptance are expected to be negatively associated with bullying behavior. Gender will be controlled as boys are more likely than girls to engage in bullying behaviors (Nansel et al. 2001;Salmivalli et al. 1996), but we had no reasons to assume that associations between the study variables would be different for boys and girls. At the classroom level the focus is on the predictive role of classroom norms, that is, the descriptive norm (bullying behavior of the classroom) and the injunctive norm (antibullying attitudes of the classroom). The central aim of this study is to examine the moderating role of these classroom norms in the associations between individual social status (likeability, rejection, and popularity) and individual probullying behavior over time. --- Method --- Participants We used the three waves of data collected for the KiVa anti-bullying program evaluation (see Kärnä et al. 2011 for recruitment and intervention program details). The data were collected in May 2007 (pretest; grades 3 to 5), December 2007 and May 2008 in grades 4 to 6 in 78 schools, representing all five provinces in mainland Finland. Most students were native Finns (i.e., Caucasian), the proportion of immigrants being 2.5 %. Not all classrooms/schools participated at all three waves; some schools only participated at first wave (pretest) or at wave 3 (posttest) (total of 2,493 students, 32 %), leaving the number of students participating in all three waves at 5,270 (64 %), of which 2,151 students attended control schools and 3,119 attended intervention schools. To increase the reliability of peer-reports, we excluded data from students who were in classrooms where fewer than 10 students had filled out the questionnaires (n=100 students). Because the KiVa intervention would (theoretically) fundamentally change the associations between our study variables, we focused on students from the control schools only. Thus, our final sample included 2,051 fourth-to sixth-grade students (49 % boys; M age =11.16 years;) from 130 stable classrooms (M class size was 21.86 students). Important to note here is that students remain in the same classroom all day, irrespective of the subject to be taught. --- Procedure The data were collected through internet-based questionnaires, after parents had given their active consent (91.7 % of the target sample). Testing sessions were held during regular school hours in computer labs under the supervision of teachers. Each school had access to computers. Teachers were given detailed instructions concerning the procedure two weeks prior to the data collection. If teachers had any questions or concerns, they could obtain support via phone or email. The order of the questionnaires as well as the order of the items within questionnaires were randomized. At the beginning of the testing session, the term bullying was defined for the students. The definition included the three main components of bullying: intent to harm, repeated nature, and imbalance of power (see e.g., Olweus 1999). Teachers read the definition out loud and students were then asked to read the same definition from their computer screens. Additionally, a shortened version of the definition (i.e., "It is bullying, when a person is repeatedly hurt on purpose") always appeared on the upper part of the computer screen when students responded to bullying related questions (i.e., Participant Role Questionnaire). The students were assured that their answers remain strictly confidential and will not be revealed to teachers or to parents. --- Measures Bullying Behavior (Waves 1, 2, and 3) Bullying behaviors were measured with the Participant Role Questionnaire (PRQ) (Salmivalli and Voeten 2004). The Bullying subscale included three items describing the initiation of bullying (i.e., "Starts bullying"; "Makes the others join in the bullying"; "Always finds new ways of harassing the victim"). The Assisting subscale consisted of three items describing active participation in the bullying that is not initiated by that person (i.e., "Assists the bully"; "Joins in the bullying, when someone else has started it"; "Helps the bully, maybe by catching the victim"). The Reinforcing subscale consisted of three items describing behaviors that reflect reinforcement of the bullying (i.e., "Comes around to see the situation"; "Laughs"; "Incites the bully by shouting or saying: show him/her!"). Participants were given a class roster and asked to nominate an unlimited number of classmates who fit the description in an item. For each participant, received nominations were summed per item and divided by the number of possible nominators. The final proportion scores, ranging from 0 to 1, were created by averaging across the 9 items. Internal consistency of these bullying behaviors was good at all three time points (Cronbach's alpha's were 0.94). Moreover, we checked the validity of this combined scale by looking at associations with self-reported bullying. These associations with self-reported bullying were similar for the separate participant role scales and the combined bullying scale at wave 1 (bullying/initiating: r=0.28; assisting: r=0.27; reinforcing: r=0.25; combined bullying scale (r=0.28) and wave 3 (bullying/initiating: r= 0.23; assisting: r=0.25; reinforcing: r=0.24; combined bullying scale r=0.25). Anti-Bullying Attitudes (Wave 2) Children responded to items from the Provictim Scale (Rigby and Slee 1991), which measures their attitudes toward bullying, victimization, and defending. The original 20-item Provictim scale, which has been found to distinguish between students who indicate that they support the intervention of teachers and peers to stop bullying and those who believe that bullying should be ignored (thus showing discriminant validity; see Rigby and Slee 1991), was modified into a 10-item version to fit the present context better. In other words, we chose the items reflecting attitudes to bullying behavior rather than perceptions of the victim or perceived responsibility to intervene in bullying. Answers on items like "I feel bad seeing a child bullied" could range from 0 (completely disagree) to 4 (completely agree). Answers were averaged across the items to create an Antibullying attitudes scale (Cronbach's alpha=0.82). Higher scores are indicative of higher anti-bullying attitudes, that is, disapproval of bullying. Social Standing (Wave 2) Acceptance and rejection were assessed by asking students to nominate an unlimited number of classmates they liked most (acceptance) and liked least (rejection). The average number of nominations given was 5.61 for acceptance and 5.00 for rejection. To assess popularity, participants nominated up to three classmates they perceived as most popular (i.e., "Who are the most popular [students] in your class?"). For each student, the received nominations per item were summed and divided by the number of nominators (i.e., participating classmates) to account for differences in class size. Scores could vary from 0 to1 (proportions). Classroom Norms (Wave 2) A classroom-level indicator of bullying (descriptive group norm) was created by averaging the individual Bullying subscale proportion scores (incoming peer nominations from wave 2) for each classroom. Similarly, a classroom-level indicator of anti-bullying attitudes (injunctive group norm) was created by averaging the individual Anti-bullying attitude scores for each classroom. This approach of calculating classroom norms is similar to other studies using classroom norms (see e.g., Pozzoli et al. 2012;Pöyhönen et al. 2013;Sentse et al. 2007). --- Strategy of Analysis To account for the nested structure of our data we constructed a multilevel random intercept and -slope model in Mplus 7 (Muthén andMuthén 1998-2012) for bullying behavior (measured at wave 3). We defined two levels in our data, that is, an individual (student) level and a classroom level. School was not used as a second or third level, because we defined no variables on the school level and because students were restricted in their peer nominations to their own classroom. The independent variables (all measured at wave 2) at the individual level were rejection, acceptance, popularity, and anti-bullying attitudes. Gender (0=girl, 1=boy) and grade (4, 5, 6) were added as covariates in the analyses. Moreover, we controlled baseline (wave 1) bullying behavior and controlled for the effects of bullying behavior (wave 1) on subsequent social standing (wave 2). The independent variables at the classroom level were classroom levels of bullying (descriptive group norm) and classroom levels of anti-bullying attitudes (injunctive group norm). Because the descriptive classroom norm is based on the classroom averages of wave 2 bullying behavior, we chose wave 1 bullying as baseline behavior when predicting wave 3 bullying, in so that there was no overlap between predictors, and between predictors and dependent variable. We also controlled for the effects of individual bullying on classroom bullying, and individual anti-bullying attitudes in classroom anti-bullying attitudes. To facilitate interpretation of the effects we centered all our independent variables around their grand mean before they entered the multilevel prediction, i.e., for every participant the grand mean was subtracted from the raw scores. Cross-level interactions were modeled by examining between-classroom variability in the associations between individual social standing and individual bullying behavior (i.e., random slopes) and predicting this variability (random slopes) by descriptive and injunctive classroom norms. This approach and its interpretation is similar to creating cross-level interaction terms in other statistical packages and will be reported as such in the tables. See Fig. 1 for a schematic overview of the multilevel random intercept and -slope model. As some variables were skewed, the maximum likelihood robust (MLR) estimation of Mplus 7 was used to generate robust standard errors and full information maximum likelihood parameter estimates (Muthén andMuthén 1998-2012). The latter means that all available pieces of information were used and hence all participants were included in the analyses, irrespective of whether they had some missing values. --- Results --- Descriptive Statistics Means and standard deviations of the study variables are reported in Table 1, as well as the mean differences between boys and girls. On average, boys scored higher than girls on measures of bullying behavior and rejection. In contrast, girls scored higher than boys on anti-bullying attitudes (i.e., disapproval of bullying) and acceptance. There were no significant gender differences in popularity. In addition, we have ran ANOVA's to test for mean differences between the three grades. There were significant grade differences in mean baseline pro-bullying behavior (F=5.02, p<unk>0.01), anti-bullying attitudes (F=6.80, p<unk>0.01), and rejection (F=13.55, p<unk>0.01). We followed up by post-hoc (Turkey HSD) comparisons which revealed that the mean of probullying behavior was significantly higher in grade 6 (M=0.10) as compared to grade 4 (M=0.08) but not grade 5 (M=0.09), and that with higher grades, the means were significantly lower of antibullying attitudes (Ms=3.20, 3.11, 3.06 respectively) and rejection (Ms=0.18, 0.16, 0.14 respectively). Within-and between level correlations between the study variables are reported in Table 2. Correlations were computed separately for boys and girls but since they were all in the same direction we discuss them together. Wave 1 and wave 3 bullying behavior were highly correlated, indicating quite high levels of stability in this behavior over time. Comparably small to modest correlations were found between bullying behavior and the other study variables; bullying behavior was negatively associated with anti-bullying attitudes and acceptance, and positively associated with rejection and popularity. Notably, grade (which also serves as an indication of age) was only weakly or not significantly related to any of the study variables. Classroom bullying behavior (descriptive norm) was positively correlated with individual bullying behavior and negatively with classroom antibullying attitudes (injunctive norm). The latter was only weakly correlated with individual wave 3 bullying behavior. In order to be sure that gender does not moderate the results (although we had no theoretical reasons to assume different associations for boys and girls), we additionally tested the difference in correlations for boys and girls using the Fisher r to z test. The correlations did not differ significantly with the exception of the correlation between popularity and bullying behavior, which was slightly stronger for boys. Hence, gender entered the multilevel analysis as covariate and not as moderator. --- Multilevel Analyses Unconditional Model We first estimated an empty random intercept multilevel model (i.e., unconditional model) for bullying behavior (wave 3) to see how much variance existed at the individual and at the classroom level. The Intraclass Correlation Coefficient (ICC) indicated that there was a significant amount of 11 % of the total variance to be explained at the classroom level (p<unk>0.01), which justified the use of multilevel modeling with predictors at both the individual and classroom level. Individual Level Model Our first predictive model only included level 1 predictors with fixed (across classrooms) slopes; see first column of Table 3. This one-level model was built as follows: first, the covariates gender and grade were entered; second, baseline bullying was entered; third, the predictors anti-bullying attitudes and social standing indices were entered. As none of the effects changed in direction or significance during the three steps, we will only discuss the final step of this one-level model. While controlling for the effect of baseline bullying behavior on later bullying behavior and its effect on subsequent rejection, acceptance, and popularity, being a boy and higher levels of popularity or rejection were associated with more bullying behavior, whereas higher anti-bullying attitudes were associated with less involvement in bullying behavior. Grade and acceptance did not significantly add to the prediction of bullying behavior. Random Slopes Model Before turning to our two level model including classroom level predictors, we examined whether there was significant classroom (level 2) variability in the associations between social standing and bullying behavior by allowing the slopes of bullying behavior on the three social standing indices to be random. The between (classroom) level variances of these slopes were all significant except for the slope of acceptance which was only marginally significant (popularity: Est. = 0.006, p<unk>0.05; acceptance: Est. = 0.009, p<unk>0.10; rejection: Est. = 0.033, p<unk>0.001). Now we will turn to the two-level model that includes predictors at the classroom level to see whether these moderated the effects of social standing on bullying (i.e., explain the classroom-level variability of these slopes). --- Descriptive Norms Model In the second predictive model (see second column of Table 3), the classroom level of bullying behavior (descriptive norm) was added to the previous model. We saw a strong, positive effect of the descriptive norm on bullying behavior. Additionally, we regressed the random slopes on the descriptive classroom norm which showed that the descriptive norm moderated effects of social standing indices on bullying. Simple slope analysis for low (-1 SD) and high (+1 SD) levels of the descriptive classroom norm revealed that: (1) in classrooms with high descriptive norms, higher levels of rejection were associated with increases in bullying behavior over the course of the school year, whereas the bullying behavior in classrooms with low descriptive norms did not vary as a function of rejection, see Fig. 2; (2) low acceptance was associated with increases in bullying to a greater extent in classrooms with high descriptive norms as compared to classrooms with low descriptive norms, see Fig. 3; and (3) higher levels of popularity were associated with increases in bullying behavior in classrooms with high descriptive norms, but the bullying behavior in classrooms with low descriptive norms did not vary as a function of popularity, see Fig. 4. Injunctive Norms (Full) Model The third and final column of Table 3 shows that the injunctive norm did not significantly add to the prediction of bullying, and it only moderated the association between popularity and bullying behavior, showing that (4) higher levels of popularity were associated with increases in bullying behavior in classrooms with high injunctive (anti-bullying attitude) norms, but in classrooms with low injunctive norms the bullying behavior did not vary as a function of popularity, see Fig. 5. We evaluated the model fit of these multilevel models using the Satorra-Bentler difference test, which is used in a similar fashion as a standard <unk> 2 difference test but accounts for MLR estimation (see last row of Table 3). With every step, the model significantly improved fit as is shown by the significant TRd values. In addition, Pseudo R squared was calculated as indicator of the proportional reduction in residual variance as compared to the unconditional model. For the one-level model, Pseudo R 2 =0.64 and for the (full) two-level model, Pseudo R 2 =0.73. --- Discussion This study was set out to investigate under which conditions, with regard to individual characteristics and classroom norms, children are more or less likely to participate in bullying behavior (initiating, assisting, and reinforcing) over the course of one year. By doing so our study extended the existent literature on bullying, which is mainly based on cross-sectional studies and focused on (precursors and consequences of) bullying and victimization rather than broader defined participant 0 0.05 0.1 --- low high --- Bullying behavior --- Rejection Low descriptive classroom norm --- High descriptive classroom norm Fig. 2 Cross-level interaction between rejection and the descriptive classroom norm predicting bullying behavior. Note: Low and high correspond to -1 SD and +1 SD, respectively. Descriptive classroom norm refers to level of bullying behavior in the classroom. Simple slopes for low descriptive norm b=0.04, p=0.12 and high descriptive norm b=0.17, p<unk>0.01 role behaviors. Overall, while controlling for baseline behavior and its effect on subsequent social standing, our multilevel analyses revealed that both individual and classroom level factors are important for explaining bullying behavior within one year. Moreover, crosslevel interactions revealed that the relations between social standing and bullying behavior were dependent on the descriptive classroom norm, but not on the injunctive norm (with the exception of popularity). --- Individual Characteristics and Bullying Behavior In line with previous studies, we found that boys were more likely than girls to engage in bullying behavior (Nansel et al. 2001;Salmivalli et al. 1996). In addition, our study showed that social standing in the classroom at time 2 is related to subsequent bullying behavior at time 3 even while controlling for the reverse pattern; over time, children high on rejection were more likely to engage in bullying behavior while accounting for the prior level of bullying behavior and the prior effect of bullying behavior on rejection. Previous studies on associations between engagement in bullying and likeability (Boulton and Smith 1994;Salmivalli et al. 1996) were all cross-sectional in nature. Our study shows that not only bullying behavior is linked with subsequent higher rejection, higher popularity, and lower likeability, but also that social standing is related to subsequent bullying behavior. (high on rejection or low on acceptance) being more likely to engage in negatively evaluated behaviors such as bullying, which might, in turn, explain and maintain their lower likeability among peers. However, there is also evidence that bullies (and their assistants and reinforcers), just like other children, have friends and are usually part of friendship cliques of which the members show similarity in bullying behavior (Espelage et al. 2003;Huitsing and Veenstra 2012;Salmivalli et al. 1997;Witvliet et al. 2010). This indicates that, although in general these children score low on likeability, they may receive acceptance from specific children when differentiating between genders or behavioral similarity (see Olthof and Goossens 2008;Veenstra et al. 2010). Future research examining acceptance and rejection by boys versus girls, or by bullies versus non-bullies (in contrast to classmates in general) and bullying behavior is needed to get more insight in these processes. In a similar vein we found that popularity at time 2 was positively associated with later bullying behavior at time 3, again after controlling for the effect of bullying behavior on popularity. This finding supports the view that bullying behavior is not only a way to obtain, but also to maintain dominance and status among peers. Consistent with this, studies have found that popularity is more strongly linked to perceptions of power than to social preference (Vaillancourt and Hymel 2006) and that adolescents in peer groups that are perceived as popular use bullying as a means to maintain their position in the hierarchy (Adler and Adler 1998). Finally, as expected, we found that children's generally negative attitudes toward bullying had a small, negative effect on bullying behavior at the individual level, whereas at the classroom level, there was no effect of the injunctive norm (i.e., anti-bullying attitude typical of the classroom) on engagement in bullying behavior (cf. Boulton et al. 2002;Rigby 2004;Salmivalli and Voeten 2004;Scholte et al. 2010). One likely explanation for this discrepancy is "pluralistic ignorance" (Juvonen and Galván 2008), which refers to group members (classmates) who individually have anti-bullying attitudes but still think that others accept bullying behavior because they witness this behavior in their group. Especially if children do not vocalize their negative attitudes toward bullying, this might lead to a false belief that the majority of the group (classroom) approves of bullying behavior. If this is the case, the discrepancy between high anti-bullying attitudes and involvement in bullying is not that paradoxical anymore. This again highlights the necessity to look at (other) group-related factors and their interactions with individual characteristics when studying bullying behavior. --- Classroom Norms and Bullying Behavior The reason for choosing the classroom as the context in the present study is because previous research revealed that bullying takes place mainly within classrooms (Smith and Brain 2000) and because children are typically restricted to their classmates in their peer nominations for social standing. In line with our expectations and the results of previous research (Espelage et al. 2003;Scholte et al. 2010), we found that the descriptive classroom norm of bullying at time 2 is positively associated with individual bullying behavior at time 3, above and beyond the effect of the injunctive norm. This finding implies that children, when it comes to bullying, are strongly affected by their classmates and the bullying climate in the class. In contrast, there was no effect of the injunctive norm on individual bullying, which is in line with a previous study that 0 0.05 showed that the injunctive norm effect disappeared as soon as the descriptive norm was included in the prediction (Scholte et al. 2010). This might be because behavior is more salient, and hence more influential, than (non-vocalized) attitudes. The results also revealed that the descriptive norm moderated the effects of individual social standing on individual bullying behavior. In classrooms characterized by a high descriptive bullying norm, higher rejection, lower acceptance, or higher popularity was associated to increases in bullying behavior whereas these associations were less pronounced or absent in classrooms characterized by a low descriptive bullying norm. The pro-bullying classroom norm seems to make it more likely that children who score relatively high on popularity or rejection (or low on acceptance) will perceive bullying as legitimate behavior to receive positive feedback or attain social prestige and a respected reputation as compared to others. This is in line with the social misfit hypothesis, which postulates that children will be rejected by their peers when they display behaviors that deviate from the group norm (see Wright et al. 1986;Sentse et al. 2007). Looked at it this way, over time, highly rejected or low accepted children might hope to improve their social standing in the classroom by 'fitting in', that is, by engaging in bullying behaviors especially when they are in a classroom where bullying appears to be the norm (Garandeau and Cillessen 2006;Sentse et al. 2007). In the case of popular children, their increased engagement in bullying behavior in classrooms characterized by a high descriptive norm might result from an effort to maintain their dominance status in the classroom without the risk of losing affection from classmates. However, a contrasting finding resulted from the interaction with the injunctive classroom norm. It showed that in classrooms with a high anti-bullying attitude, higher popularity was associated with increases in bullying. This finding would be more in line with the assumption that popular children want to set themselves apart from the rest, act as leaders and not as followers, in order to gain and maintain visibility and social dominance which are part of the definition of (perceived) popularity (Cillessen and Rose 2005). Both, seemingly contrasting, findings might point to a bidirectional relation between popularity, classroom norms, and engagement in bullying behaviors. Dijkstra et al. (2008) showed that among adolescents, bullying behavior displayed by especially the popular children in the classroom (i.e., the popularity norm) determines whether bullying is socially accepted. That study, however, was cross-sectional in nature and thus it is difficult to conclude whether it is a 'popularity norm' that explains acceptance over time of the children who engage in bullying, or whether being accepted or perceived as popular while engaging in bullying leads to a 'popularity norm' over time. Future studies may want to look into this specific bi-directional relationship in more detail. --- Strengths and Limitations The longitudinal design, utilizing three waves of data collected in a large population-based sample, the multilevel perspective including predictors from both the individual and the classroom level, and the focus on a broader definition of bullying behavior set our study apart from previous research into bullying. Aside these strengths, also some limitations must be acknowledged. Firstly, the current study did not tap into the network nature of the processes under study. That is, the study of bullying behavior and social standing in the classroom could have been more informative if we had dyadic measures available, indicating by whom children are accepted or rejected, and by whom they are perceived as popular for the network (e.g., classroom) as a whole. Related to this point, the peer nominations in the current sample were restricted to classrooms only and hence we had no information on possible relationships with children from other classrooms or even outside the school while this could have been informative. Secondly, we should be aware of possible false inferences when assumptions about aggregated scores are made based on the aggregation of individual level data. However, next to the aggregated group level we also accounted for the associations at the individual level in so that false inferences on associations are unlikely. Thirdly, although the current study was specifically focused on bullying related behavior, other participant role behaviors including outsiders are worthwhile investigating in order to fully grasp the dynamics of the bullying process. Lastly, our sample was representative for mainland Finland, but it is not clear whether the findings can be generalized to other countries.

This three-wave longitudinal study was set out to examine the interplay between individual characteristics (social standing in the classroom) and descriptive and injunctive classroom norms (behavior and attitudes, respectively) in explaining subsequent bullying behavior, defined as initiating, assisting, or reinforcing bullying. The target sample contained fourth-to sixth-grade students (n=2,051) who attended the control schools in the Finnish evaluation of the KiVa antibullying program. Random slope multilevel analyses revealed that, over time, higher popularity or rejection, or lower acceptance were associated with increases in bullying behaviors, especially in classrooms with a high descriptive bullying norm. In contrast, the injunctive norm did not moderate the associations between social standing and engagement in bullying, except for children high on popularity. Theoretical and practical implications of the results are discussed.

study did not tap into the network nature of the processes under study. That is, the study of bullying behavior and social standing in the classroom could have been more informative if we had dyadic measures available, indicating by whom children are accepted or rejected, and by whom they are perceived as popular for the network (e.g., classroom) as a whole. Related to this point, the peer nominations in the current sample were restricted to classrooms only and hence we had no information on possible relationships with children from other classrooms or even outside the school while this could have been informative. Secondly, we should be aware of possible false inferences when assumptions about aggregated scores are made based on the aggregation of individual level data. However, next to the aggregated group level we also accounted for the associations at the individual level in so that false inferences on associations are unlikely. Thirdly, although the current study was specifically focused on bullying related behavior, other participant role behaviors including outsiders are worthwhile investigating in order to fully grasp the dynamics of the bullying process. Lastly, our sample was representative for mainland Finland, but it is not clear whether the findings can be generalized to other countries. Replication of the results in other countries is warranted, as well as studying the associations among varying age groups. Despite these limitations, our study extended previous research in showing that social status in the group is not only a consequence but also a motivation to engage in bullying. Moreover, it showed that relations between social standing in the group and engagement in bullying behavior depend on the descriptive classroom norm, but not on the injunctive norm (except for popularity). These findings support the view of bullying as a group processes which must be approached as such in tackling bullying in intervention and prevention programs. --- Conflict of Interest The author declares that she has no conflict of interest.

This three-wave longitudinal study was set out to examine the interplay between individual characteristics (social standing in the classroom) and descriptive and injunctive classroom norms (behavior and attitudes, respectively) in explaining subsequent bullying behavior, defined as initiating, assisting, or reinforcing bullying. The target sample contained fourth-to sixth-grade students (n=2,051) who attended the control schools in the Finnish evaluation of the KiVa antibullying program. Random slope multilevel analyses revealed that, over time, higher popularity or rejection, or lower acceptance were associated with increases in bullying behaviors, especially in classrooms with a high descriptive bullying norm. In contrast, the injunctive norm did not moderate the associations between social standing and engagement in bullying, except for children high on popularity. Theoretical and practical implications of the results are discussed.

INTRODUCTION As consummate niche constructors on multiple scales, humans have the ability, and tendency, to make even the most unfamiliar familiar. This process requires work and energy, but human plasticity, flexibility, and creativity enables us to acclimatize, or even adapt, to novel spaces and experiences. Most would agree that understanding these processes is core to bioanthropological work on human variation and adaptation, but few recognize how relevant these perspectives are to understanding our own research experiences, especially that of going to and returning from the field. The geospatial movement, the cross-cultural shifts, and the biosocial trials and tribulations that accompany the entire fieldwork process are aptly condensed by the subtitle of The Hobbit "there and back again." In this essay we offer personal and professional insights into the "back again" aspect of fieldwork arguing that we need to both 1) be more cognizant of its often problematic and difficult nature across all career stages and 2) fold such awareness into the training and preparation of PhD students across the broad range of work involving human biology and fieldwork. Going to the field, wherever that might be, is often central to the practice of anthropological work. An anthropologist's role is to immerse oneself in another society, observe, measure, and assess individuals "elsewhere" to construct a better understanding of the surrounding world and the human role(s) in it. While "elsewhere" can be in a place physically/culturally distant from an individual's lived experience or can be a place personally familiar like an ancestral homeland or current community but still requires adopting new roles and perspectives, such tasks require one to step away from home, along with its schedules, structures, and familiarities. This process of change is understood to be difficult; thus, preparing to go to the field is approached with rigor. There are seminars, credited coursework, and whole books (see MacClancy & Fuentes, 2013, Hewlett, 2019, or Ice, Dufour, & Stevens, 2015), dedicated to preparing novice anthropologists with the initial discomforts, culture shock, and mental shifts needed to approach fieldwork in a productive and healthy way. However, while the stress of going to the field is often anticipated and acknowledged, what happens when coming back is frequently not. Comparatively, there is very little professional, social, and emotional preparation for the eventual end of fieldwork, the "back again." In this phase, the changes that were experienced when going to the field are now, in essence, happening in reverse, accompanied by the compounded knowledge, customs and habits acquired during the fieldwork itself. As a result, the researcher who returns is not identical to the individual who left. The home they left has also not remained frozen in a state of cryogenic familiarity. Life goes on. Here, we acknowledge the complexity, diversity, and difficulty of the "back again" and share individual experiences of returning from the field. We also include some of our collective thoughts on modes for better training future fieldworkers. Our goal is to normalize the difficulty and shed light on what the process has been like for a range of scholars at different stages of their careers when they returned from dissertation field sites around the world. We hope that this will be used as a tool for future students and scholars to prepare to come home as successfully and consciously as possible. --- CONTEXT Changes in activity patterns, sleep schedules, social interactions, and diets are known to have substantial physiological and psychological impacts on the human body and mind (Lieberman, 2013). These transitions are generally anticipated and prepared for when adjusting to the fieldwork environment, but it seems less so on return. In addition, some of the changes that were made to adapt to fieldwork conditions, from speaking different languages to culturally specific practices of eating, drinking, physical activity (or lack thereof), smoking, or ameliorative prescription drug use (i.e., sleeping pills, digestives, painkillers) may not translate effectively to everyday life back home and can create challenges. Fieldwork stints may vary in duration; but no matter the length, field seasons can disrupt home life, and often require committed engagement to invest substantively, socially, and emotionally with new communities. Logistical (e.g., time zone differences, hours needed for participant observation) and technological restrictions (e.g., lack of access to internet or electricity) can hamper even the best of intentions to maintain sufficient social ties with home. Changed or broken ties with family and friends are also common, a reality often not anticipated by those new to fieldwork. Cumulatively, these factors may include'reverse' culture shock when a researcher returns home, forcing researchers to reinterpret their lives and adjust to new ways of being. For those who work intensively with communities and engage in "deep hanging out" (Rosaldo, 1994) as part of their field experience, coming back to lab work and article writing-activities done frequently in solitude, or in the formal confines of the academic landscape-can be isolating and depressing. Without the proper time and capacity to process the fieldwork experience, the actual production of knowledge, such as transforming field data to an academic article (not to mention a dissertation!) can be exceedingly difficult. Further, returning to the pace and patterns of academic and scholarly life (e.g., conferences, programmatic requirements, productivity/publishing expectations, grant writing, social media immersion, departmental politics, stress regarding next projects/positions, etc.) can induce negative responses, including panic and despair. Some of these psychological states arising from such a transition can be long-lasting. The lack of acknowledgement, attention, and care given to these processes further exacerbates these issues and can lead to a range of adverse outcomes, ranging from discontent and ennui to more serious anxiety and depression. The issues with coming "back again" are serious and deserve both scholarly attention and social care. Here, we offer examples of departing to the field and returning to the United States/United Kingdom. This provides a degree of connectivity across the some of the readership, but we recognize that a serious engagement with this issue will require a much broader discussion across multiple voices and home locations. We hope that this article will at least initiate broader conversations, both within and between institutions and professional organizations. Here are the authors' stories of "back again." --- STORIES Mallika Sarma (Rocky Mountains, USA). I have always strived to be a Highly Productive Person. Always with at least two different checklists in hand and comfortable with the epithets of being "overprepared" and "intense", I found that my energy was a perfect fit for doing really good fieldwork. I loved the speed and intensity of being in the field, trying to get as much, and as thorough, data as possible while engaging and connecting with my field assistants and participants. My "fieldwork life" became my world while my family and friends back home, always supportive though slightly confused, yet nonetheless impressed and proud of my seemingly glamorous occupation, faded into the back of my priorities. In the field, I learned to loosen. I unclenched and took in, moved with the flow, made cozy in what made me most uncomfortable, because it was the only way to be. Returning, however, was a calcifying process, stiffening to the expectations of the academic and research world, of life back home, and of my own ambition. Reconciling my field-self with my returned-self was hard. As a human biologist studying adaptations to extreme environments, I have been fortunate to do fieldwork in some of the most remote and gorgeous places on Earth. The beauty of my different field sites made returning to post-industrial midwestern America particularly bleak. During graduate school, I lived alone with my cat (who was happily boarded and spoiled rotten by my parents during my field seasons), which was far from the usual intensive social communities I formed while in the field. Though the physical distance wasn't much, when I returned home from dissertation fieldwork, I felt like I had come back to a different world. I was not prepared for the intense loneliness and blanket despair I felt. Compared to my other fieldwork experiences, dissertation fieldwork and eventual return felt like it should have been easy. I worked in Central Wyoming, still in the United States, speaking English, with access to internet (mostly), and within relatively close range of my immediate family. With the exception of some race-related vigilance (after all, I was a brown woman alone in a notoriously red state during the Trump era) and watching out for blizzards or an odd grizzly bear, my attention could be singularly focused on data collection. Without the need of a sat phone or anything else equally cumbersome, every now and then my committee checked in on me and friends kept me in the loop with the petty drama and happenings back home, but never enough to disturb me from my immersion into my new community. But after the high of fieldwork, that Highly Productive Person had collapsed into a Barely Functioning Person, wrapped in blankets on the divan consuming Netflix with the same purpose and rigor I brought to the field. Somehow the conspicuous whiteness and wealth of my then home institution was more overwhelming and suffocating than what I experienced in the American West. My fieldwork experience existed in my head, but I could only partially share it with my support system and their consistent, but distant, support suddenly felt inadequate. Everything felt more tenuous getting back into the data when I was so far away from the source. What if all my data were bad? Or worse, what if my work was inconsequential? There were other, less-existential issues. As I think is common, I struggled with food a lot on return. When in the field it was "eat what you can get," which was necessary and sometimes glorious. However, on return, without the social cues that come from eating in a community, I would just not. Or I would fall back on "eat what you can get" to my own detriment. What made me an excellent field anthropologist felt like my undoing on returning: an all-encompassing deep investment in my work and community and the subsequent loss thereof. I was forced to realize that re-formation of one's home community requires targeted investment, just as much as in the field. I found that planning time off with my people in the transition between field seasons and everyday life allowed me to reconnect with them as well as properly mark the end of one momentous thing and the beginning of another. I needed both official and unofficial gatherings of other fieldwork-returning friends to talk about the transitions back. I relied on conferences and other professional meetings where I could just be around other people who would get it. I joined a dissertation support group where we could talk through the struggles of writing and the feelings of never getting enough done. I also threw myself back into my hobbies and athletic training and began paying attention to everyday life again. I recognized that at home, particularly immediately after coming back from the field, I required a strict schedule and diet, including scheduled writing, work-out, and relaxation time and even automated reminders to eat and drink water (that I keep to this day), to stay healthy and sane. But perhaps the most important of all was re-engaging in friendships and relationships back home with gusto. Three years after completing dissertation fieldwork, I am only now slowly recovering to what I understand is my new pace. As I embark on a new area of research, I see these transition steps when returning from the field as critical to other major transitions in my life, including getting married, moving cross-country, and starting a new job. If anything, I have learned that making home again is the only real constant. --- Sheina Lew-Levy (Central Congo Basin). "I think you should come home." Five months into what was meant to be a six-month field trip in the Republic of Congo, I called my supervisor. I had started fieldwork in a village while assisting another researcher, sharing a teary goodbye before heading into the forest to stay in a BaYaka forager forest camp. Now, the camp was planning to move to a remote fishing pond. My interpreter had to go back to his job as a schoolteacher. I had to decide whether I would follow my participants on my own or go home. Today, five years and four field seasons later, the field has become a second home. I text my interpreters constantly, asking if they've received any news from the community I work with. I dream of running and singing in torrential downpours with the teenage girls. I light candles and mourn when I learn that friends I've made in the field have passed on. I crave a meal of forest tubers and drinking palm wine by the fire. When I submitted my PhD, it was returning to the field after-the nighttime laughter, the quiet evenings, games with children-that helped me feel whole again. But during that first field season, I struggled. I felt disconnected from my interpreter, who seemed eager for work to end each day so he could ditch me and go talk to his new friends. A usually sedentary person, I walked miles and miles daily, in the rain, hip-deep in ponds, following children as they wove through the forest to forage and play. The rice and beans I had purchased went moldy. The food my interlocutors shared with me, though delicious, seemed to offer little in the form of nutrition, likely because I had not yet developed the gut bacteria necessary to digest it (Schnorr et al. 2014). I remember eating a can of sardines in my tent when everyone had gone to bed at night, crying (I hate sardines). Having assumed a tropical rainforest would be, well, tropical, I had brought a flimsy sleeping bag liner in which I shivered throughout the night. I turned my sat phone on every night, desperate for text messages from home. I found respite with the women and the children. Any free moment I had, I sat near the women, listening to their conversations, trying to decipher their words and meanings. Sometimes the women called me over to ask me questions and tell me stories. Other times they seemed annoyed I was there. The children, on the other hand, seemed thrilled to have me around. They taught me to speak Yaka, play games, and understand social norms by correcting me when I transgressed them. They started my fire every morning. They seemed bemused when it became clear I couldn't identify koko (Gnetum africanum) a generic-looking, but highly nutritious leafy green. I remember sitting by my fire at night, my eyes closed, listening to the chatter and laughter and occasional whines of babies as they settled. I remember laughing from my tent, along with everyone else, when one camp member was reprimanded by his wife for snoring loudly in the night. These social bonds kept me going. But, by the time I called my supervisor, I felt that my body was wearing out. I can't remember now whether I suggested I come home, or whether he could hear the exhaustion in my voice. What I do remember is feeling a deep sense of relief when I agreed that coming home was probably best. The camp was relieved too-the families I stayed with were always very concerned about my safety, and I think they could sense I was drained. Professionally, my transition back to university life seemed successful. There was data to enter and clean, and a new field season to plan. Personally, I was a mess. I was so thin I could see my ribs and spine poking out through my skin-my partner said I looked like a deflated balloon. At Christmas, I snuck into my partner's parents' kitchen at midnight to eat microwaved cheese on crackers. I was constantly worried about the well-being of my interlocutors, who had no access to medical services in the bush. I called my mom crying. "You've always found transitions difficult" she would say. I picked fights with my partner constantly, finding it difficult to settle back into coupled life. So, for better or worse, I leaned into my work. Specifically, I leaned into data entry. So mind-numbingly boring, I could watch TV while I plugged away at my Excel file. I entered focal follows to Buffy the Vampire Slayer, interviews to Gilmore Girls, field notes to The Office. I felt effective even as I felt lazy. These shows followed me to bed, where I spent any of the time I wasn't working. When my partner complained that I was polluting our soundscape, I turned the volume down-but only slightly. The familiar storylines (I've watched these shows since I was a teenager) and simple characters kept me company, helped me fall asleep, kept my buzzing mind still. It took two months for my weight and mood to regulate and for me to feel more comfortable with silence, just in time for the next field season. With more years of experience under my belt, I've learned how to 'do' fieldwork. I now know what gear works at my field site. I pack a duffle bag of Clif bars and instant oats. I speak Yaka well enough to share jokes with the women and tease the children. My body has become more resilient too: it takes less time for my sleep, digestion, and muscles to adjust to the physical demands of my work. But returning from the field continues to be a struggle. I still pick fights, I still worry, and I still eat cheese at midnight. I still watch too much TV when I get home. Over the years, I've learned a few things the hard way: until my metabolism regulates, I should be careful with what I eat and how much I drink; my partner is incredibly forgiving; transitions are indeed hard, and they probably always will be. --- Theresa Gildner (Southeastern Ecuador). After four previous field seasons, I na<unk>vely believed I was fully prepared for my transition from fieldwork to final data analysis and dissertation writing. I had spent months in Amazonian Ecuador as a member of the Shuar Health and Life History Project collecting data examining lifestyle patterns, hormone levels, and parasitic infection. This last season marked the end of data collection; however, as I transitioned from the field, I was simultaneously juggling several life changes (e.g., my recent marriage, move to a new city, and impending transition out of grad school). In addition, the field season took place in the fall rather than summer, truncating my timeline to finish laboratory analyses and write my dissertation before defending in the spring, adding stress during the post-fieldwork transition period. Because I was applying for jobs while in Ecuador, I had to periodically travel back into town to access the internet. While there were benefits to this arrangement, including restaurant food and hot showers, these trips resulted in a series of "mini-transitions" between data collection and my life back home, which made it difficult in many ways to maintain the mentality I needed to focus on data collection. Trips to town were disruptive in other ways. Several important news stories broke while I was in the field that year -including the Las Vegas mass shooting -and it was very difficult to read about these events so far from my family and friends, an experience not unique to this particular field season. Throughout my fieldwork, the news dump experienced upon regaining access to the internet was been one of the most difficult aspects of transitioning. It can be overwhelming to learn about distressing world events you have missed while off the grid, especially when loved ones back home have had time to process the news. Emotional support offered by study teammates also learning about these news stories for the first time can be extremely helpful. My teammates were often open to discussing these world events at length as we all came to terms with the news, while loved ones back home were sometimes less willing to dwell on these stories if they had already spent time grappling with their implications. I found that the lifestyle and dietary changes associated with prolonged fieldwork consistently result in unexpected lingering physical effects. These changes include an inability to easily consume many typical American foods for weeks after returning from the field. This was especially difficult during my final field season because I returned the week of Thanksgiving. While glad to be home in time to celebrate with my family, I was unable to eat most of my favorite holiday food. I learned that, while it may be tempting to enjoy all your favorite treats immediately upon returning home, you need to be patient and allow your body time to readjust back to your usual diet and activity patterns. A final complication at the end of my dissertation data collection was a long delay in international sample shipment from Ecuador to the U.S. I called shipping representatives in Ecuador and the U.S. daily, but it still took weeks for the samples to finally arrive safely in the lab, leading to considerable anxiety about whether I would be able to analyze my data and finish my dissertation on time. As someone who has never dealt well with deviations from set plans, one of the most challenging aspects of fieldwork for me has been accepting that things rarely go exactly as expected. My experience with sample shipment highlights how even the best laid data collection and analysis plans may be affected by factors outside of your control. In these instances, the ability to pivot and restructure is key. By shifting my writing schedule while I waited for samples, I was still able to finish my degree the following spring as planned. Throughout this difficult transition period, the support of my dissertation committee, friends, and family was critically important in helping me adjust and finish my PhD. My fellow graduate students and research teammates played a huge role, as our shared experiences enabled them to empathize with my concerns and offer valuable advice. I have found that building social time into my schedule to maintain these supportive relationships is invaluable. For example, I participated in weekly trivia nights with other graduate students, research teammates, and friends outside of academia. This friend group has been an incredibly important source of support, providing a space to share my frustrations and ask for input. However, while the encouragement of others may ease the transition out of the field, I also found that it is essential to allow for personal time, ideally taking a few days once to you leave the field to decompress and get a little distance from your project. This space can help you transition from field-related stressors and routines, while also providing needed time to process your experiences. For me, this meant spending time in activities that did not require much critical thought, such as catching up on TV shows or movies I missed while in the field or pleasure reading. Additionally, allowing yourself some time to adjust (both mentally and physically) may help your study in the long term, leaving you recharged and ready to hit the ground running when you jump back into data cleaning and analysis. --- Ben Trumble (Lowland Bolivia). When conducting field research, you realize that there are many things you were never taught in grad school. For example, I never took a course that prepared me for animal husbandry (what do you look for when buying a cow? what is a fair price? how many people will it feed?), nor did my hominid paleo-anatomy courses teach me the best way to butcher a peccary (hint: do not puncture the bladder). Do not get me wrong, I had excellent mentors who helped shape my research and taught me much, from how to hold a pipette, to how to conduct interviews, but they couldn't prepare me for everything. My days conducting research following hunters and collecting saliva samples were long, starting with a pre-dawn hike to a hunter's house, followed by hunting for twelve hours or more over rough terrain. Nighttime came quickly, so by the time I got to my tent, I only had an hour or two to eat and enter data before I was sleeping in preparation for the next day. Meals were simple with little choice-they were dictated by the fish I could catch, the meat hunters gifted me, or the canned foods I brought with me. Oddly, the thing I was least prepared for was getting back to grad school and all the sudden, free time and personal choice. Sure, I had long days in the lab analyzing samples and fulfilling my duties as a research assistant, but my day started later and ended earlier than in the field. Instead of being excited about free time, it provided stress. No longer did I have a singular task, but instead many; finishing my dissertation, finding a postdoc, catching up with friends and family, getting back to the gym, seeing all of the movies, television, and internet memes I had missed. Instead of the river deciding which fish I ate for dinner, or a hunter's skill determining if peccary was on the menu, I was overwhelmed with choice at the supermarket. Not only did I have the option of buying meat, but I could also get a different dozen cuts from a dozen different animals all in the same refrigerated display-the hundreds of daily choices honestly felt paralyzing. The bolus of time meant that I was not spending every waking moment on my all-consuming goal of increasing knowledge (it was a long time ago, and I was still young and na<unk>ve with delusions of grandeur). All of those options meant time away from conducting research, and that was why I wanted to be scientist, so I began to feel guilt whenever I was not working. If I was not going to be singularly focused on research, shouldn't I have just gotten a job after college instead of spending more than half a decade in grad school? Even after being home for a month, I still did not feel I had caught up with all of my friends, which made me feel like I had let them down, but at the same time, catching up with them was also preventing me from finishing my dissertation. Without a singular goal, I felt like I was being pulled in many directions. In retrospect, I should have talked to friends, family, or a therapist about these feelings. I eventually realized that so many aspects of fieldwork are uncertain (weather, vehicular issues, injury, illness), and I was able to cope with that uncertainty because I had a clear goal. When I got back, I no longer had to face flash floods or trucks stuck in the mud, but instead a different kind of uncertainty about how I should be focusing my time and energy. Much like butchering a peccary, it was not something that could be taught in graduate school, but something I had to experience and learn with my own hands. Over time I began to feel less guilty, found the focus I needed, and became better at multi-tasking (as I write this at night, over a week past the deadline...let's just say it's a work in progress). --- Michaela Howells (Indonesia and American Samoa). Leaving home was easy. Why was coming back so hard? My undergraduate study abroad in Indonesia was transformative, exciting, and exhausting. It challenged me in ways I never knew possible and laid the groundwork for becoming a successful graduate student and professional. It was a gentle but intense introduction to fieldwork. I pushed through my limits and in some cases established them (turns out car sickness is not improved in a foreign country in the back of a van winding through curvy roads). My world view expanded. I met the love of my life. I took my first steps towards becoming a field anthropologist. Returning home was so much harder than I had ever expected. Entering back into my life felt constrictive, hollow, and lonely. As a first-generation college student, my family had no framework to draw upon. My family and friends were happy to see me, but they didn't seem interested in my experience. They didn't want to hear stories or look at pictures. I had the distinct feeling that I was boring people. My life changing adventure was less interesting to them then catching me up on their daily lives. There were stinging comments about my "vacation". I went from living in a very social group setting in Bali, attending my first professional conference in China (International Primate Society), and traveling through Hong Kong, Singapore, and Malaysia with my newfound partner -to being unceremoniously back in my childhood bedroom. My life had totally changed. I had left a different person then when I returned and was shocked by this discordance. Simultaneously, I had the distressing feeling that it never happened. That I had imagined the biggest experience of my life to date, that everything I had grown into would dissipate or wither away. I knew how I wanted to spend my career and was scared it would never happen. Years later I realized that I was terrified that this would be the highlight of my life, that it would be the last chance I would have to experience fieldwork, and that my fledgling relationship would wither with distance. Thankfully I was wrong on all fronts. Returning to school helped my mental health substantially. Faculty were genuinely excited for my experiences, and I was surrounded by others who had returned from their own field schools. The realization that I felt more at home at school then my parent's home was simultaneously jarring, confusing, and comforting. Ten years later I was preparing to come home from my two-year dissertation fieldwork in American Samoa. The idea that I would get funding and complete my dissertation research was so fantastical to me that I hadn't imagined my return. I had spent my whole academic life preparing for my dissertation fieldwork but gave little thought to re-entry. All of my belongings were packed away. My office space gone to make room for other students. My home-space vacated to free up space for my husband to take in a roommate to defray our costs, and for easy moving if necessary while I was overseas. But it also made for easier divestment if our relationship ended while I was abroad -ours wouldn't have been the first marriage to end during fieldwork. I created a comfortable life for myself in American Samoa. I felt useful, purpose driven, and smart. I worked hard and faced challenges but felt the most fit and relaxed I had been in years. It was the first time in my life I had lived alone. I enjoyed the community I was a part of and was heart-broken to leave. I returned with more questions than I left with. My re-entry was made possible by some serious mental gymnastics. I had to convince myself that I would come back to the mainland US long enough to finish my PhD and then would move closer to the ocean and warm weather. Colorado's climate seemed antithetical to American Samoa's and provided an icy highlight of the differences between my lives. I was depressed and felt like I was finished in academia. As raw as this time felt, it was made easier through staged re-entry, patience from my friends and loved ones, and the addition of our Samoan dog Uli. Uli provided consistency during re-entry between my two homes. She helped stave off the feeling of disconnection and discordance and eased the process. I had set my schedule so that I wouldn't need to be at my desk writing until a month after I returned. This gave me the space to experience my intertwined feelings of loss, pride, relief, and despair without immediately being expected to have an intellectual engagement with my experience. I exercised, learned how to cook, and listened to music. I felt sorry for myself and missed the life I stepped away from. In retrospect, I am thankful my loved ones and advisors were able to provide the space for me to grieve before moving onto the next stage of my life. Now, as a tenured faculty member, field work is a more predictable occurrence. Re-entry from the field is easier for me now due to improved technology and contact, social media, collaborative work with those who I can reenter with, increased career and financial security, strong family support, and personal behavioral changes that make the process easier. I have discovered that it is critical that I schedule buffer time into my return schedule. When possible, I schedule a few days alone on my travels home. These days create a liminal space for me to readjust to re-entry. This is not vacation time, but instead time to be alone, type up my field notes, and enter data. It provides crucial mental space from the conflicting pressures of fieldwork and homelife, and time for radical self-care. This practice provides important mental space before I fully re-enter, while forcing me to slow down and process my experiences before refocusing on reconnecting. --- Agust<unk>n Fuentes (Mentawai islands, Indonesia). Unlike most folks doing their dissertation fieldwork today, when I embarked thirty years ago for my project there was a more limited range of communication options. In fact, as the field site was on the island of Pagai Utara, Mentawai, West Sumatra, Indonesia, I had no option for connecting to my family, advisor, and friends (platonic and romantic) aside from a two to three-day journey from the field site to Padang, Sumatra, that involved three boats, <unk>4 hours of hiking and <unk>18 hours of sea travel, in each direction and was completely dependent on the weather. I note this to lay clear that aside from once every 6-8 weeks or so I was not in contact with "my" world at all (except for some BBC on shortwave on Sundays). And the field experience, while in some ways the best experience of my life, was really rough and simultaneously one of the worst. In late 1992, before my original end date but very much just in time, I left the field site, made my way to Jakarta and boarded a plane for the USA. A lot had transpired, and I was mentally and physically wiped out: I'd lost 25 pounds (<unk>16% of my pre-project body weight), my girlfriend of the time, and any sense of what I was "doing" or why I was doing it. I was on a plane to Los Angeles that stopped over in Hawaii. At 40,000 feet, somewhere over the Pacific, I decided that I'd had enough and that this whole academic thing was not for me. I argued my way off the plane at Hawaii, picked up my backpack and spent the next few weeks trying to find myself in a new place (I'd never been there outside of the airport before), with only one contact (the aforementioned ex-girlfriend). I set myself up, got a little work and a space to set my stuff, and called my Advisor to let her know I was out of the PhD thing and my Mom to tell her where I was. Both women, in hindsight, were right on target in their approach with me. They were kind and calm, talked me through the moment, convinced me to just breathe and take some time. The world would not stop even if I stepped out for a spell. Time passed without thinking about anything related to my project, my "career" or my future aside from a lot of internal dialogue about existentialism, loneliness, William Burroughs (a long story), and what being a haole was like. In retrospect, I recognize I was steeped in anthropology, in what I love about the discipline, but did not realize it at the time. Eventually, and I guess expectedly (at least by my Advisor and my Mom), I realized I needed to head back home, to the San Francisco Bay Area where both my Mother's house and my PhD program were and figure out what I was going to do. On the plane back to the continent it occurred to me that maybe I had not completely failed. Maybe I had indeed done something all that time in the rainforest, with the Mentawains, the langurs, gibbons, snakes and acid spitting ants. I just didn't yet know what that was. Coming back to campus (UC Berkeley) took a while (of me living on my Mom's couch and feeling sorry for myself). But once I did, my Advisor, my friends and my family recognized the trauma (both externally experienced and self-induced) and they offered a soft landing, during which I began to remember why I'd gone to graduate school in the first place. My fellow graduate students formed a bubble of friendship and facilitation, I re-enrolled, and walked back into TA'ing. Slowly but surely over the course of <unk>6 months, I began to pull out my notes, my data, my drawings, and developed more than 40 rolls of film. At that point, seeing what I had done, recognizing the shortcomings and faults, but realizing that

Leaving "home" to pursue fieldwork is a necessity but also a rite of passage for many biological anthropology/human biology scholars. Field-based scientists prepare for the potential changes to activity patterns, sleep schedules, social interactions, and more that come with going to the field. However, returning from extended fieldwork and the reverse-culture shock, discomforts, and mental shifts that are part of the return process can be jarring, sometimes traumatic experiences. A failure to acknowledge and address such experiences can compromise the health and wellbeing of those returning. Here, we offer personal stories of "coming back" and give professional insights on how to best ready students and scholars for returning from fieldwork. We argue for an engaged awareness of the difficult nature of returning from the field and offer suggestions for individuals and programs to better train and prepare PhD students pursuing fieldwork. By bringing forward and normalizing the difficulty of the fieldwork-return process, we hope that this reflection acts as a tool for future scholars to prepare to come home as successfully and consciously as possible.

's house and my PhD program were and figure out what I was going to do. On the plane back to the continent it occurred to me that maybe I had not completely failed. Maybe I had indeed done something all that time in the rainforest, with the Mentawains, the langurs, gibbons, snakes and acid spitting ants. I just didn't yet know what that was. Coming back to campus (UC Berkeley) took a while (of me living on my Mom's couch and feeling sorry for myself). But once I did, my Advisor, my friends and my family recognized the trauma (both externally experienced and self-induced) and they offered a soft landing, during which I began to remember why I'd gone to graduate school in the first place. My fellow graduate students formed a bubble of friendship and facilitation, I re-enrolled, and walked back into TA'ing. Slowly but surely over the course of <unk>6 months, I began to pull out my notes, my data, my drawings, and developed more than 40 rolls of film. At that point, seeing what I had done, recognizing the shortcomings and faults, but realizing that something of value had occurred. Principal of which, was substantive growth in my ability to think with and about ecology, biology, behavior and what the world is like. That is when my path as a scholar genuinely started. --- DISCUSSION It is important to remember that coming back is part of the journey. In rare cases, students leave for the field and choose not to return. However, while accurate statistics are hard to find, there is the sense that more commonly a non-insignificant percentage of graduate students do return from dissertation fieldwork but then decide to no longer continue their PhD. This may be in part due to the trauma of "back again." Though not everyone who leaves academia does so because of this specifically, we hope that the vignettes here -from scholars with different experiences -help illuminate the black box of returning from the field. One goal of sharing these experiences is to dissipate stigma surrounding the struggle to return, while also informing efforts to support students through the transition back home. While the process of returning from the field is unique for each individual, we have identified five common themes connecting the vignettes shared here that are likely factors shaping the re-entry process (Figure 1). First, many researchers struggle with the transition from singularly focused work to the numerous, sometimes overwhelming, responsibilities and expectations at home. Second, there is the need for time to "zone out" and create buffer time between fieldwork and continued research. These vignettes demonstrate that this buffer time can manifest as planning an end-of-fieldwork celebration, watching television, or even taking an extended pit-stop in Hawaii. They also demonstrate how this buffer time is necessary for both mental and physical recovery. Third, it is apparent that the stress of returning from the field is often heightened by the stress that comes with anticipating and preparing for the transition from a PhD program to a postdoc or faculty position. A fourth common theme is the central importance and tenacity of social relationships; there is the fear or actuality of romantic endings, ignoring or leaning into friends and family, and the challenges from transitioning from one community to another. Lastly, individuals tend to struggle with expectations of productivity and/or data analysis, where work (or lack thereof) can be used as a crutch to avoid dealing with the challenges posed by returning from the field. It is also important to note that these common concerns do not occur in isolation, but are often experienced simultaneously, further compounding the stress experienced during the transition "back again". Open and more transparent discussion about the struggles of returning from the field should be encouraged and normalized. Particularly in light of an ongoing changing fieldwork environment, whether it be due to travel restrictions or greater reliance on virtual data collection, we must value and create a space to discuss what it means to return. We conclude this reflection with a few suggestions for researchers, and for programs, to make the "back again" experience part and parcel of the training and expectations of fieldwork (see Fig 2). Below, we suggest specific action items that individuals and programs can implement to ameliorate the re-entry crises and facilitate a more successful anthropological experience. --- Action items for Researchers 1) Prepare for the return.-Although there is tremendous diversity between field experiences and re-entry situations, there are ways that individuals can prepare for the return. Being honest with yourself about your experiences before you are back can help you recognize what you will need when you return. It can also help you untangle the commonly experienced jumble of emotions. This kind of frank self-reflection can take many forms; for example, journaling or freewriting gives you space to be brutally honest and vulnerable in a way you do not have to share with others. It can help you process your re-entry experiences and provide a crucial pause for breath. These writings can also help you determine how to talk to your support system in a meaningful way. You can also write up a more formalized fieldwork report, where you denote what went well and what went poorly. While this documentation can be kept to yourself, it can help with eventual grant reporting and manuscript methods. Additionally, having this information pre-written may relieve some of the reticence experienced when having to write up these details later. In addition, preparing to return can include finding small rituals that you can sustain from location to location. Perhaps it's listening to certain songs or re-reading a book, continuing an exercise regime or engaging in some other kind of regular physical activity. You can also bring meaningful rituals home with you; for instance, a colleague still sweeps her house every day like she was taught by her host family in the Central African Republic. These behaviors play an important role in processing experiences. Some may feel that processing these emotions and experiences in these ways is somehow weakening or indicates their lack of suitability in the field. However, we encourage you to recognize this process as helping to transition into a healthy work-nonwork life integration and can help clear your mind. 2) Create a support system.-Family and friends outside of academia may have no framework to understand your experience or the challenges you may face upon re-entry. Talking to them prior to leaving and re-entry may help reduce this gulf of information. Share a friend's experiences to help prepare them for yours. Give them a copy of this article. Also, seek out other people who recently returned from the field to confide in who can empathize with your experiences and share their own. Walk together, share meals, or binge watch a show that everyone else watched while you were away. Support systems can and should include faculty members in your department who can act as allies and help provide authority in protecting your time and re-entry. While this can be your primary advisor, it is better to have multiple faculty members that look out for you. Even with the most supportive primary advisor, multiple mentors, particularly those in positions of authority in the department, can help ease the transition. 3) Give yourself space and time as needed.-The expectation that graduate students and professionals should be able to seamlessly transition between field sites and academic homes is not only misleading, but does a disservice to our field. Elaborated in the vignettes above, in academia, many feel the pressure to continuously work, resulting in feelings of guilt when taking personal time and setting work-life boundaries. However, it's necessary to recognize that being kind to yourself upon re-entry is important for both physical and mental health. This includes creating an interstitial space for yourself that enables you to process this transition. This could be establishing literal space, for instance days or weeks taken in a neutral location. It can also be metaphorical and emotional space, clarifying to friends and family that you may need some time and distance before reconnecting. It may even be helpful to enlist close friends or family members to act as your buffer between you and more insistent loved ones. Be honest with yourself and your advisors about the time you will need to transition back, including dedicated time for non-work events and the expected lag in pace and productivity that will likely occur upon return. This transition back is physical and mental. As shown in the vignettes, your physical body and its capacity for work, exercise, digestion, and more, will have likely shifted in the field. This may manifest in changes in diet, body composition, or even physiological function. It may take time to come back to a familiar state, and sometimes instead will transform into a new normal. Mental changes occur in conjunction with changed physicality. Arriving home after time "elsewhere" means a sudden mental shift not just about your changed physical state but also in adjusting back to cultural or environmental norms. This shift includes processing how your feelings about these norms may have changed. This can include dissonance, guilt, and helplessness in reconciling the abundance, wealth, and wastefulness of a North American university town compared to the scarcity and thriftiness of a remote field site. Or there may be dissonance in the beauty and splendor of the physical environment of your field site compared to the starkness of returning to an academic institution. Further, it is important to understand that the time and space you needed while in the field is probably very different from what you will need upon return. Planning these realities into your schedule will result in a healthier re-entry and a more productive professional life. Training would be improved with honest discussions by faculty and recently returned students about their re-entry experiences. Sharing these experiences can be uncomfortable; however, it should be noted that not sharing can have detrimental effects on student wellbeing. Faculty advising fieldwork students should be encouraged to take special training courses on how to support these students and how to identify red flags. Students can also be matched with non-supervising faculty or more senior students that can be a mentor and guide them in what to expect both in the field and upon return. --- Action items for 2) Post-fieldwork: Expect a diversity of re-entry strategies and offer flexible timelines for meeting program benchmarks.-For mentors, having an open and frank discussion with students throughout their fieldwork and re-entry process will enable you both to set realistic and healthy goals and strategies. This communication reduces the risk of students disappearing or becoming non-communicative after completing their fieldwork and potentially slowing their progress in the program. Students and early professionals need to hear that planning for re-entry is as important as all of the other components of research and can help them be successful in their other goals. This can be facilitated through a field return orientation, where the next set of expected benchmarks can be explained to all returning students, and the relative flexibility of each of these benchmarks. During this time, departments should remind students of resources available to them upon return, from help finding housing to where to find mental health services. Programs should also normalize and encourage time off/vacation between field seasons and resuming academic activities. Given the different requirements and experiences of students, it is likely that there will be a need for flexibility. By introducing this flexibility as a norm, students will be better situated on an achievable path to success, one relevant and contextual to their own experience, rather than failing to meet generalized, but unrealistic, expectations. 3) Post-fieldwork: Facilitate social gatherings, workshops, and provide resources for post-field cohorts/individuals with financial and infrastructure support from the program.-In addition to improving the mental health of students, these events can help promote healthy ways forward after fieldwork. Departments should mark students' return in some way to ensure that students feel welcomed back and that they can reintegrate, such as through an email, or return party, or department story sharing. They can structurally and financially facilitate formal dissertation writing groups as well as more casual social events (e.g., group meals or movie nights). In addition, departments can create materials like "so your friend/child/spouse returned from the field" pamphlets or infographics, to help support networks better understand the challenges of fieldwork and fieldwork return. Some faculty and students may be resistant to these kinds of events and identify them as a waste of time or taking time from their writing (or more realistically their procrastination). This can be addressed through a normalization and celebration of these workshops and gatherings as an optional but important and highly encouraged part of the re-entry experience, while also emphasizing re-entry challenges as part of the fabric of academic life. Caveats: Privilege and a changing fieldwork landscape.: As emphasized throughout, the process of returning differs for each individual. An intersectional approach is essential. An individual's entire academic experience, including their fieldwork, is critically shaped by factors such as race, gender, socioeconomic status, research funding, sexuality, parents' prior education, parental status, and generational wealth. While research has shown that anthropologists generally report a significant lack of family-career balance and high stress associated with the profession, this distinction can further be intensified by lower socioeconomic status and being a woman (Lynn, Howells, & Stein, 2018). Further, in periods of global uncertainty and precarity, such as the COVID-19 pandemic, these inequities have only been further exacerbated (Gewin, 2021). The re-entry experience in particular is heavily impacted by an individual's access to social and financial support. Having a closely engaged personal support unit that can understand what research travel entails and can absorb ongoing responsibilities at home, is not an experience shared by all. Yet, these factors significantly impact a field researcher's ability to let non-research concerns fade into the background. Many first-generation students have no framework upon which to base their experience of personally transformative academic work, or even extensive travel, abroad. Just as important, their personal support networks, such as their close family and friends, likely also have no framework to understand unique fieldwork experiences. Having a supportive (both behaviorally and financially) partner/spouse or even parents, a stable academic job, a self-set schedule, and childlessness/access to childcare provides flexibility others may not have. Through programmatic choices, institutions can help facilitate and bolster all students, but particularly those that may require additional support. Diversity, equity, and inclusion are imperative for the success of biological anthropology as a field (Anton et al. 2018, Bolnick et al. 2019) and thus, programmatic structures recognizing these differences are requisite to help mitigate these inequities. In addition, the research landscape is rapidly changing due to factors such as technological advances, increasing climate-related natural disasters, and global pandemics (Ocobock et al., 2021). What we understand as anthropological fieldwork is consequently changing. "Going to the field" may instead be (re)connecting with communities remotely online or relying heavily on local collaborators to collect the majority of the in-person data. Further, with greater access to online networking options, from messaging apps like WhatsApp to social media sites like Facebook and Twitter, the deep and active engagement with a community can persist in ways that even 50 years ago it could not. However, the sometimes-dramatic change in mindset between "fieldwork" and "field return" is likely to be consistent and normalizing this transition remains crucial. Our aim here to share our stories of returning from the field and providing individual and programmatic suggestions to improve the return experience begins the conversation that we hope will continue. While the history of field-based academic research is long, it has often been built for individuals with considerable existing power and privilege, thus softening many of the negative effects of the return process. If there is to be more equitable access and flourishing of field-based science, it is required to at least acknowledge how difficult returning can be for everyone, since it likely impacts marginalized and vulnerable groups most. Embracing these perspectives and successfully implementing structural support will likely take time, however destigmatizing the struggle of return and setting those coming back up for success, will make for happier scholars and better scholarship. In the end, we must recognize that our work does not conclude with just going "there", but in fact, is only starting when we come back again. We have identified five of the most common factors that likely impact individuals as they go through the re-entry process. These factors are connected and some or all can contribute to the additive and compounded stressors of returning home. We suggest specific action items that individual researchers and institutions/programs can implement to begin to ameliorate the re-entry crises and facilitate a more successful research experience. --- DATA AVAILABILITY STATEMENT Data sharing not applicable -no new data generated, or the article describes entirely theoretical research.

Leaving "home" to pursue fieldwork is a necessity but also a rite of passage for many biological anthropology/human biology scholars. Field-based scientists prepare for the potential changes to activity patterns, sleep schedules, social interactions, and more that come with going to the field. However, returning from extended fieldwork and the reverse-culture shock, discomforts, and mental shifts that are part of the return process can be jarring, sometimes traumatic experiences. A failure to acknowledge and address such experiences can compromise the health and wellbeing of those returning. Here, we offer personal stories of "coming back" and give professional insights on how to best ready students and scholars for returning from fieldwork. We argue for an engaged awareness of the difficult nature of returning from the field and offer suggestions for individuals and programs to better train and prepare PhD students pursuing fieldwork. By bringing forward and normalizing the difficulty of the fieldwork-return process, we hope that this reflection acts as a tool for future scholars to prepare to come home as successfully and consciously as possible.

Introduction An important part of the moral challenges of genetic engineering originates from misunderstandings fostered by the use of metaphors, asserts philosopher Peter Janich (Janich 2001). What Janich writes with regard to the genetic information metaphor holds true for synthetic biology's machine metaphors as well. The machine metaphor is a central part of a specific understanding of what synthetic biology is. It is both an expression of this understanding of what synthetic biology is and it shapes synthetic biology research accordingly, as well (Hellsten andNerlich 2011, Vincent 2016). For one thing, metaphors contain ways of thinking about the ontology of one's object of research and about the nature of one's research activity itself. For example, with regard to synthetic biology whether or not the set of ontological characteristics suggested by the machine metaphor captures adequately what an organism, as a matter of fact, is, and how it can be described and explained is the subject of recurring dispute (Boudry and Pigliucci 2013, Nicholson 2013, Rosen 1993). For another thing, ontological assumptions of this kind have ethical implications regarding one's perception of safety issues, regarding what one thinks of as justified ways to act and react to the object, and regarding what one assumes a technology to be capable of. In what is to follow, some of the ontological assumptions of the machine metaphor will be identified and analyzed. Whether or not these assumptions adequately capture what an organism is, will not be at the center of the argument, though. Instead, the argument will focus on the latter ethical implications that a machine metaphor inspired understanding of synthetic biology and its research objects may have. As a result, it will not argued here that the machine metaphor should be rejected tout court. The machine metaphor may for example help to look for and identify controllable causes of molecular intracellular processes, and engineering these regularities may help to develop single cell organisms with valuable novel traits. It will be argued, though, that one needs to keep in mind that the machine metaphor is a metaphor and as such captures at best parts of the characteristics of organisms. One should be aware that this mismatch can influence what one expects organisms to be able to do, and how one behaves towards them, in ethically relevant ways. --- The machine metaphor and its implications In synthetic biology, to name but a few examples, single cell organisms are described as "genetically engineered machines" (iGEM Foundation n.d.), intracellular molecular processes are "genetic circuits" (Weiss et al. 2003), and genetically engineered organisms are referred to as "platform organisms" (Roelants et al. 2013). Obviously, synthetic biology makes use of the metaphor of the machine and its variants, most notably its information technology equivalent, the computer, as a central conceptual lens to describe, explain and modify molecular intra-and intercellular processes. Applying machine metaphors in order to describe and understand natural objects and processes is not a new phenomenon. In the seventeenth century, the mechanical clock was used to this end, in the eighteenth century the balance was an influential metaphor, in the nineteenth century the steam engine metaphor appeared, and in the twentieth century and up to day the computer, the information processing machine, replaced these older machine metaphors (Lunteren 2016, Zaccaria, Dedrick, andMomeni 2017). All of these machines are artefacts designed by humans, and their structural and material make-up serves as a means to attain a predefined state of affairs and thus fulfills a specific purpose for humans. Given the fact that living beings are not designed by humans, and their make-up, according to evolutionary theory, cannot be explained by reference to a specific purpose it is meant to fulfill, the widespread use of machine metaphors in modern science might come as a surprise. At the same time, if one succeeded in explaining organisms or parts of organisms as if they were designed to fulfill a specific purpose, this is to say that one succeeded in identifying causal relations in organisms that allows for their purposeful redesign and reengineering. In this sense, machine metaphors fit neatly into the modern concept of the natural sciences according to which if one understands an object, one should, in principle, be able to build it, and successfully building an object counts as an experiment that supports the theoretical hypotheses that went into its design (Köchy 2012). Against this backdrop, the widespread use of machine metaphors in synthetic biology can be seen as an expression of the vigor with which synthetic biology introduces rational design and building ideals into biology. The engineering methodology figures prominently in synthetic biology's multi-disciplinary background of research approaches. Biology, chemistry, physics, IT and engineering all are part of this background, with engineering approaches occupying center stage (Heinemann andPanke 2006, Schyfter 2013). Engineering principles and approaches, such as designing, modularizing, and standardizing suggest using the machine metaphor for its objects of intervention. In addition, the machine metaphor fits neatly into a larger story of what synthetic biology is and what it is aiming at. Following this story, synthetic biology constitutes the latest step along the line of developing scientific bottom-up explanations of macro-objects and their behavior (Church and Regis 2012). At the lowest level, physics analyzes the movement and structure of atoms by identifying and analyzing subatomic structures and parts. At the next level, chemistry analyzes complex molecules by scrutinizing the simpler molecules and atoms of which complex structures consist. At yet another level, living molecules, which is to say organisms, become the objects of analysis. This is the realm of "analytic" molecular biology research, which traces the behavior of organisms back to their inner molecular genetic structures. At each level, analytic knowledge allows one to intervene technologically, alter the objects in question and devise novel objects. This is the "synthetic" side of each analytic science. In chemistry, for example, naturally occurring compounds such as sugar or vitamins can be produced synthetically and, in addition, compounds not known from nature, such as plastics, can be synthesized. Synthetic biology, it is expected, will lead to similar developments with regard to organisms. Synthetic biologists will be able to rebuild naturally occurring organisms and to create novel organisms by means of DNA synthesis (Kastenhofer 2013). This straightforward account of scientific progress, and the place of synthetic biology within it, is closely connected to an ontological assumption. If explaining complex molecules amounts to knowing the simple molecules contained within the complex structure, and if explaining the function of an organism is tantamount to identifying and analyzing the genetic structure of this organism, then it is obviously assumed that the functioning of a complex object is the result of the laws and regularities governing the behavior of this object's parts. As a correlate, if one knows the parts of an object and their functions, one can reliably predict the overall behavior of the object in question. This ontological assumption is the link to the machine metaphor. With regard to machines, the very same principles of part-whole relations and predictability are at work in reverse order. When one designs and assembles a machine, the idea is to combine parts with reliable and predictable functions in order to come up with a complex result that, in turn, and due to the reliability of its parts, fulfills specific functions and purposes in reliable and predictable ways. Consequentially, if what one does in synthetic biology is correctly captured by the account given above, the objects of one's research must be thought to share important ontological aspects with those entities that one calls "machines". If one transfers the relevant characteristics of the machine paradigm to the realm of synthetic biology and single cell organisms, this, then, is the resulting list: Firstly, the behavior of a single cell organism is to be explained by reference to its molecular, genetic parts. Thus, single cell organisms can be designed and can be built part by part. Secondly, designing and building an organism entrenches a specific function in the organism. This function fulfills, from the perspective of the designer, a specific purpose. Thirdly, the resulting organism's behavior can be predicted reliably given sufficient knowledge of its parts (Boldt 2013). --- Envisaging potential environmental adverse effects When one intends to build a reliable object with specific functions, and when one assumes that the reliability and predictability of the object is a result of the reliability and predictability of the behavior of its parts, a natural point to start will be to analyze these parts. As soon as one can assume to have sufficient knowledge about the parts, determining the behavior of the object must appear to be, in principle, a matter of computation. In turn, if the object behaves in unpredicted ways, leading to unintended side effects, the cause of the failure must be assumed to be located in the inner make-up of the object. Now, on the one hand, if one imagines this object to be, for example, an electronic device, this perspective covers adequately the demands of building safe machines and of dealing with side effects, one may assume. On the other hand, if one imagines the object to be an organism, there are good reasons to doubt the adequacy of this outlook. For one thing, organisms evolve. Evolution is a process involving chance mutations within a genome. Chance mutation is, by definition, an unpredictable process. This counteracts any attempt to design a stable, reliable organism that preserves its design function from generation to generation. Evolutionary change thus is a main obstacle to building organisms the behavior of which can be predicted over long periods of time. For example, it has been proposed to insert toxin producing genetic circuits into genetically modified organisms. Under normal conditions the genetic circuits are repressed and do not express the toxin. If one observes that the organism proliferates in unintended ways, or leaves its assigned host environment, one can activate the gene circuit by adding a metabolite, for example. Then, toxin is produced and the microbes die. However, as experience shows, these "suicide genes" are prone to undergo mutation over time, which compromises their functionality (Wright, Stan, and Ellis 2013). For another thing, organisms interact in manifold ways with each other and their environment. In microbial multispecies communities, bacteria influence each other via direct physical contact, metabolic interdependencies and coordinative signaling systems, thus maintaining ecosystem equilibrium (Guo, He, and Shi 2014). With regard to microbial communities, this modelling requires close ongoing collaboration between theorists and experimentalists (Zaccaria, Dedrick, and Momeni 2017). Modelling a multispecies ecosystem not restricted to microbes obviously would be an even greater challenge. On the one hand, this is not much of a surprise for molecular biology research practice. On the other hand, if one confines oneself to the machine paradigm, these multi-level interactions must come unexpectedly. In order to build a reliable machine one would restrict the number of external factors that can influence the workings of the machine to those that are necessary for the machine to fulfill its design purpose. If organisms are viewed as machines, thus, one would not expect them to interact in these multiple ways with the environment. This may lead to underestimating possible adverse effects of genetically altered organisms in the environment, since on the one hand they are products of synthetic biology and described as machines, on the other hand, they are still natural living beings. As an aside, in addition to chance mutation and ecosystems interactions, reproduction and growth challenge ideals of the machine perspective as well. Besides predictability and reliability, machines are supposed to be entities that efficiently fulfill their functions. Up to a certain point, growth and reproduction are prerequisites for microbes to produce a specific valuable metabolite, for example. Above this limit, though, growth and reproduction take up energy that otherwise could be used in the metabolic process to produce larger amounts of the required substance. Life's tendency to preserve itself both with respect to individuals and with respect to generations thus at a certain point counteracts attempts to utilize organisms as efficient production facilities. Therefore, optimizing organisms in this respect would be tantamount to devising non-living biological molecular networks, such as genetic cell-free production systems. Turning back to safety related characteristics of natural organisms that are difficult to incorporate into the machine paradigm, it is worth noting that evolutionary genetic changes are small, stepwise changes. Mutagenesis is a selective alteration of the genome, and it builds upon the design as it has evolved up to that point, regardless of whether a novel design may fit an ecological niche better than existing organisms. Now, at first sight the machine metaphor does not seem to prompt engineering radically genetically altered organisms rather than sticking more closely to known ones. After all, if natural organisms can be explained in terms of the functioning of a machine, building a machine might just as well consist of rebuilding a slightly altered natural template. On a wider understanding, though, taking into account the content of the machine metaphor regarding parts-whole relations and concepts of scientific progress, confining oneself to re-designing natural organisms must appear to be a gratuitous restriction of one's scientific and technological abilities. Rebuilding natural organisms may be a first step to underpin the claim that one has accurately analyzed and explained the functioning of organisms. The ultimate proof of this claim, though, is designing and engineering organisms that are not known in nature. Similarly, to forego exploring potential benefits that only novel organisms may possess would appear to be an imprudent confinement to the well-trodden paths of nature, a confinement that one cannot justify scientifically or technologically from the machine point of view, once sufficient knowledge of genetic parts is available. Thus, this striving towards novelty is fueled by epistemological reasons and by reasons pertaining to exploring uncharted terrains of organismic functions and their possible benefits. Reflections on safety are no inherent part of this perspective, but must be added from the outside, as it were. Now, taken to the extreme, designing novel genetic organisms may include organisms with DNA made from alternative amino-acid systems. Given these xenobiological organisms' reduced abilities to interact with natural organisms, they may in some cases offer safety advantages (Schmidt 2010). Nonetheless, as a general rule, the more radical an organisms differs genetically from natural organisms, the less one can predict its environmental effects, including adverse effects. As the example of invasive species, novel organisms to the invaded habitat, shows, these risks do exist. To sum up, chance mutation and evolutionary change, and, in addition, organism-environment interactions are characteristics of organisms that challenge attempts to predict their behavior reliably. Mutations modify the internal genetic structure of an organism, potentially leading to altered functions, and they occur, qua being chance mutations, unpredictably. Organism-environment interactions involve high numbers of variables and values rendering attempts to model the development of multispecies ecosystems over time challenging. From the machine point of view, characteristics of organisms such as mutagenesis and multi-level environmental interactions come unexpectedly. In order to build a reliable machine one would want to avoid any chance changes to the design that have an effect on the function of the machine, and one would want to restrict the number of external factors that can influence the workings of the machine to those that are necessary for the machine to fulfill its purpose. Thus, if one restricts one's perspective to the machine paradigm, one may overlook potential risks and side effects when a genetically altered organism, viewed as a machine, is engineered for an application in the environment. Finally, the ontological assumptions inherent to the machine metaphor render engineering genetically radically altered, and in this sense novel organisms the ultimate proof of success with regard to explaining organisms. By its own logic, this striving does not anticipate environmental safety risks. Claiming that synthetic biology views organisms through the lens of the machine paradigm, and arguing that the machine paradigm tends to downplay evolutionary change and interaction within ecosystems does not necessarily entail calling for a replacement of the machine metaphor. The machine paradigm may be helpful in many respects when one analyzes genetic structures and develops novel gene networks. It does entail though that when it comes to applications in the environment, supplementary expertise is needed to assess environmental safety risks. This expertise needs to include a focus on evolutionary change and ecosystems interactions, which is to say that it needs to include expertise in ecosystems biology, evolutionary biology, and developmental biology. This is, of course, no news for existing safety assessments of genetically engineered organisms for environmental use. These and other multidisciplinary safety assessment procedures already do exist. Being aware of the possible impact of the machine metaphor on safety perception backs up and reinforces these practices. One special issue deserves mentioning here. Mutagenesis and novelty differ with regard to safety assessment challenges. To start with, mutagenesis renders the safety assessment of any engineered organism difficult. Nonetheless, in the case of engineered organisms that are slightly modified versions of a natural template, risk assessment can rely on experience with this template. In contrast, releasing a novel synthetic organism into the environment raises safety concerns because the effects of this organism as it is, prior to any genetic mutation, on existing ecosystems is hard to assess. After all, in this case there is no single natural template, which could serve as a basis for risk assessment. This second challenge may actually render plausible a demand to refrain from releasing synthetic organisms at all. This demand, though, is warranted only in the case of radically altered organisms. One could imagine that introducing novel synthetic organisms stepwise, starting from a non-radical version of the synthetic organism via ever more radical designs, could help to deal with this problem adequately, if this procedure allows enough time to get acquainted with each version of the organism and its environmental effects. --- Machines and moral agents As long as synthetic biology is confined to genetically engineering single-cell organisms, discussing issues of moral status may appear superficial. What is more, scenarios of synthetic biology modifications of mammals and humans in the future mention, for example, reconstructing the mammoth by way of modifying elephant germ cells, and genetically engineering the human germline in order to improve the immune system (Church and Regis 2012). Since modifications of this kind leave all those capacities of animals and humans intact that are essential for their moral status, there would be no reason to suppose that ascriptions of moral status to the mammoth, or to the genetically altered human, would differ from ascriptions of moral status to elephants or non-genetically-altered humans, it seems. What would change, though, when synthetic biology turns to multi-cell organisms, mammals, and humans, is that, following the machine metaphor, an understanding of animal and human behavior is imported that may stand in tension with and thus may weaken understandings of these behaviors that lend credibility to ascriptions of moral status. Many ethical theories, among them utilitarianism and deontolgy, agree that a necessary condition for moral agency is the capacity to distance oneself from one's interests, and to reflect upon these motivations from an evaluative point of view that takes into account well-being, interests, or rights of others. In utilitarianism, for example, this reflective distancing enables an evaluation of possible actions in terms of the effects of the consequences of these actions on the interests of others that are affected by the actions. In Kantian deontology, the distancing comprises evaluating whether the maxim of one's envisaged action (i.e., the statement that one wants to perform an action of a certain type in a certain situation in order to attain a specific end) could be upheld in a hypothetical world in which all moral agents would follow this maxim. The ability to distance oneself from the interests that would otherwise lead one to act in a certain way is not easily reconcilable with the machine paradigm. To begin with, this ability appears to include the ability to prove any prediction of one's future behavior wrong simply by acting differently. If one is confronted with a prediction of one's behavior or if one has a suspicion that someone wants to take advantage of a certain regularity of one's past behavior that he reckons will continue in the future, one can change one's behavior and act contrary to these expectations, if one has an interest in not being predictable. What is more, suspending one's actions and reflecting upon them appear to be best understood as an interpretive activity. Determining interests or maxims involves determining actions that someone, following his interests or maxims, respectively, is prone to perform. As Alisdair MacIntyre, among many others, has argued, actions are not entities that can be perceived immediately as what they are. Their identity depends on the ends they are supposed to serve. These ends can be manifold and link up to larger narratives about who the agent is and what kind of life she leads. For example, whether what I am doing is mowing the lawn, exercising, doing a favor to my wife, or irritating my neighbor cannot be discerned by looking at my movements. Understanding my action presupposes knowledge about me, my relations, my habits, and my plans (MacIntyre 2013, 237-263). In turn, such knowledge presupposes knowledge about what it is like to experience emotions, to be in relation with others, and potentially to make sense of and find meaning in life as a whole. When attempting to discern what an action is, one thus needs to interpret the events in question in terms of different narratives that make sense of them and picking the one that appears to fit best to this specific and other of my actions. Looking backward at past actions, these narratives help to understand what one actually did and why. At the same time, looking forward, these narratives shape how one will act in the future. They render certain options more meaningful than others, and thus lead one to act in certain ways. Now, let us suppose one puts together a robot equipped with artificial intelligence of a kind that allows the robot to, one, change its behavior when it is confronted with predictions of it. Two, the robot continually engages in reflecting about narratives that make sense of its actions, and applies these narratives to shape its future behavior. Three, these narratives are based on knowledge of what it is like to experience emotions, to be in relation with others, and to make sense of life as a whole. As it seems, in this case there would be no reason not to ascribe the status of a moral agent to this robot, unless, and I am following a realist supposition on this issue here, it turns out that the robot does not really possess these properties but somehow mimics them (Torrance 2014). At the same time, though, would there be good reason to call this robot a machine, that is to say, an object the behavior of which can be explained in terms of parts that are effectively arranged in order to attain specific behavioral ends? Given that this robot can stop doing what it is designed to do when it is told about its design ends, and given that it continually shapes and reshapes its ends and its future by interpreting its actions in terms of larger narratives about what is important and lends meaning to its life, this does not appear plausible. While at the micro-level of bit processing and electronic limb control, this robot may well be described as a machine, at the macro-level this description would not fit any more. In much the same way, while isolated processes at the level of molecular interactions in the human body may be comparable to machines in meaningful ways, humans as acting and reflecting whole organisms capable of moral agency are not. Applying the machine metaphor to humans would amount to calling into question the reality of the ability to distance oneself from one's interests, to prove wrong predictions of one's behavior, and to shape one's future according to larger human life narratives. Since these are conditions for moral agency, applying the machine metaphor to humans would amount to calling into question the reality of the human ability for moral agency. One may compare this line of thought to the genetic determinism assumption. According to this assumption, DNA is a blueprint that determines phenotype traits and behavior. It was argued further that since phenotype human traits and behavior are determined by genetic make-up, the genotype is the actual agent behind human behavior, rendering beliefs in human moral agency an illusion (Dar-Nimrod and Heine 2011). The machine metaphor is well suited for similar claims since its explanatory scheme of parts reliably causing behavioral effects of the entity as a whole does not leave room for those explanatory schemes that sustain ascriptions of moral agency. --- Machines and moral patients Engineering human DNA by way of synthetic biology circuit design is still science fiction, even though research approaches to alter genetic circuits in mammalian cells to prevent disease are already explored (Black, Perez-Pinera, and Gersbach 2017). Envisaging synthetic biology multi-cell organisms, including insects, is much closer to reality, though (Markson and Elowitz 2014). Now, since animals do not possess the status of moral agents, applying the machine metaphor to them cannot undermine their moral status in this respect. Both according to everyday attitudes and according to many ethical theories, though, some animals have moral status in the sense that protecting their well-being is a moral obligation, and inflicting harm on these animals constitutes a prima facie moral wrong that needs to be justified by other, overriding ethical obligations in order to be acceptable. That is to say, some animals can be said to possess the status of moral patients. The question then is whether applying the machine metaphor to animals may undermine practices of ascribing the status of being a moral patient to them. In animal ethics, different sets of traits are identified as necessary conditions for being a moral patient. Being able to experience suffering and enjoyment, for example, is an intuitively appealing first and basic candidate. These experiences are not neutral sensations but involve an element of evaluation. Experiencing pain and joy is bound up with needs, desires, interests, or, as an overarching term, attitudes (Basl 2014). All of these concepts suggest the idea that animal behavior of the relevant kind must be understood as being stretched out between a perceived actual state of being and an anticipated future state of being. Metaphorically speaking, unlike causal descriptions, descriptions in terms of desires, interests, and attitudes suggest that chains of behavior are not pushed forward by starting conditions according to laws of animal behavior, but are due to pulling forces, namely the attractiveness or unattractiveness of different possible future state of affairs. Weighing these courses of action according to what matters for the animal introduces an element of human-like agency that is captured in Tom Regan's concept of being "the subject of a life" (Regan 1983). If this is a correct understanding of moral patiency, being a moral patient and being a moral agent can be interpreted as ascriptions to organisms along a line of an increasing ability to take a stance towards one's motivational states. Avoiding pain, seeking pleasure, having needs, having desires, following interests, and shaping one's life according to narratives may be understood as steps along this line of an increasing ability to suspend immediate impulses to act and to shape one's future in accordance with evaluative reflection (Jonas 2001). Again, then, the capacity to be the subject of a life stands in tension to the machine paradigms assumption that the behavior of an object is to be explained by internal parts that effectively, reliably and predictably cause this behavior. Applying the machine metaphor to animals that possess the status of moral patients undermines these moral status ascriptions since it calls into question the assumption that the behavior of an animal can be a basic form of agency of a subject of a life. As a consequence, while applying the machine metaphor to genetic processes in and between human and animal cells may help to accumulate technologically useful knowledge about these processes, applying it to animals and humans as whole organisms does not do justice to the characteristics of animal behavior and human agency and renders ascriptions of moral status unintelligible. --- Spearhead science and technology The machine metaphor invokes an account of scientific progress according to which the behavior of any complex object can be explained by reference to its parts. A complete explanation of reality thus starts from its most basic parts, moves on via intermediate ones, and ends with explaining highly developed organisms and human beings. In principle, the interactions of objects at each level are supposed to be contained in theses explanations, since they can be seen as a subset of an object's individual behavior. If one strictly adheres to this paradigm, explanations in terms of, for example, teleological concepts, or folk psychology, or narrative accounts of historical events, are valid only if they can be shown to conform to physiological, chemical and physical explanations. The latter explanations, then, appear as "real" explanations, whereas the former ones at best are shortcuts for those real ones. From the point of view of the machine paradigm, the preference for explanations in terms of parts is a preference for explanations that refer to causal laws or law-like regularities, since only regular behavior allows for reliable predictions, which is what one needs if one wants to build a machine. Teleological and narrative accounts of events, by contrast, are accompanied by ambiguity and uncertainty. A person's intention, for example, cannot be deduced directly from her action as a physical, observable event, but depends on the larger framework of the person's long-term plans, professional aims, private aspirations, situational inclinations, and many other attitudes, character traits and objectives. Any one of these features can play a role in grasping why the person acted in a certain way und thus in grasping her intention. Now, even if the person herself picks out one possible end and declares that her intention was to reach this rather than another end, doubt necessarily remains as to whether she is ultimately correct, and other convincing stories might be offered that shed new light on her behavior. Academic disciplines such as history and social and political science inevitably have to get along with these ambiguities and uncertainties. By their own standards, this does not undermine claims to knowledge and expertise. From the machine point of view, though, these claims to knowledge must remain dubious. What is more, following the machine paradigm and its implications for scientific progress, synthetic biology can be singled out among the natural sciences with regard to its state of development. While physics has already reached a level of analytic accuracy that enables, among many other things, the production of novel synthetic elements, and while chemistry has already analyzed chemical reactions up to a point where it becomes possible to synthesize novel compounds such as plastics, biology now has advanced to precisely the threshold where purposeful designing, arranging and rearranging of genetic parts becomes possible to such a degree that the resulting genetically engineered organisms can be regarded as "novel". This implies that right now synthetic biology is to be regarded as the spearhead of scientific and technological progress. From its own perspective, it is a discipline that is about to develop the most accurate knowledge about organisms up to date, and it is about to turn this knowledge into reliable technological means to modify and control behavior (Bensaude Vincent 2013). Now, many of the serious and unresolved problems societies grapple with involve organisms. To name but a few, food has to be provided for a growing population. Crop plants are unable to grow in large areas of the earth due to drought. Oil spill threatens marine ecosystems. All of these problems are well known, and obviously so far bringing to bear the available technological tools on these problems has not helped to solve them, nor have social and political measures had decisive effect. If one is to look for new solutions, following the machine paradigm one will in this situation first of all place one's hopes on technology rather than social and political measures, since the latter cannot, as it is assumed, be informed and guided by truly reliable knowledge. What is more, since available technologies have been utilized to no avail, it will seem natural to turn to the latest significant development in this field. Given the machine paradigms background story, synthetic biology thus can come to be regarded as not merely one among other, but rather the foremost candidate for solving societal and environmental problems. Critics of what has come to be known as "technological fixes" to societal problems have long argued that these approaches suffer from a variety of drawbacks (ETC 2007). For instance, while for-profit companies in rich parts of the world may develop these technologies, the site of application and the end users may be located in low-income countries. Technological solutions thus may contribute to global inequality and entrench economic dependency. What is more, high-tech applications may fail to provide sustainable, long-term solutions, since local knowledge and resources may not suffice to repair the application and keep it functional. Consequently, opting for low-tech solutions, or alternatives to technical solutions, or changing societal conditions, for example regarding education and training, may often promise better outcome than high-tech applications. Owing to the machine metaphor and its wider implications, one may tend to overestimate the share synthetic biology applications may contribute to solving societal problems. Again, this is not an argument for abandoning the machine metaphor, nor is it an argument to renounce developing synthetic biology applications and assessing their potential benefits. Rather, it is an argument for staying realistic with regard to synthetic biology hopes and promises, for keeping track of the whole field of possible technological and social solutions to societal problems, and for embedding synthetic biology applications in a social context that allows long-term safe and just use. --- Conclusion The machine metaphor in synthetic biology is a powerful conceptual lens. It can fruitfully lead research to analyze the relation of genetic parts to the functions of the organism as a whole and to determine ways to engineer organisms. At the same time, firstly, the machine metaphor systematically fades down characteristics of organisms such as evolutionary development and ecosystems interactions. This may lead to underestimating potential environmental side effects of synthetic organisms. Moreover, from the point of view of the machine metaphor, confining oneself to building slightly altered natural organisms falls short of exploiting the full potential of synthetic biology, which from this point of view lies in its ability to engineer novel organisms, such as organisms containing synthetic DNA parts taken from a variety of natural templates, and organisms containing DNA based on alternative base pairs. How to assess possible side effects of novel synthetic organisms on the environment is an open question, though. Secondly, from the perspective of the machine metaphor, viewing an entity as a moral agent or patient, respectively, becomes dubious. From this perspective, the characteristics of living beings that warrant ascriptions of moral status must be reinterpreted as illusionary phenomena since they stand in tension to explanations in terms of parts that reliably bring about specific, predefined effects. Thirdly, the machine metaphor contains a concept of scientific progress and synthetic biology's place within this progress. This concept stresses some aspects of the nature and potential of science and technology in general and synthetic biology in particular, and hides others. Inherent in this perspective is a tendency to overestimate the potential contribution of synthetic biology applications to solving societal and environmental problems. Being aware of the implications of the machine paradigm can help to include supplements to this perspective that overcome its ethical limitations and to reinforce practices that already mitigate these effects. With regard to environmental safety, awareness of the force of the machine metaphor thus backs up practices of multi-disciplinary safety assessments. Building novel organisms stepwise may be another measure to mitigate risks. With regard to the potential moral status of organisms, awareness of the explanatory scheme implicit in the machine metaphor helps to understand that ascriptions of moral status hinge on characteristics of animal and human behavior that are out of the reach of the machine metaphor. Thirdly, and finally, with regard to assessing means to solve societal problems, awareness of the internal logic of the machine metaphor and its epistemic background story may safeguard against overestimating the capacities of synthetic biology to successfully deal with these challenges. --- Authors' contributions All contributions to the text stem from the single indicated author, JB. --- Ethics approval and consent to participate Not applicable. Competing interests I declare that I do not have any financial and non-financial competing interests. --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The extent to which machine metaphors are used in synthetic biology is striking. These metaphors contain a specific perspective on organisms as well as on scientific and technological progress. Expressions such as "genetically engineered machine", "genetic circuit", and "platform organism", taken from the realms of electronic engineering, car manufacturing, and information technology, highlight specific aspects of the functioning of living beings while at the same time hiding others, such as evolutionary change and interdependencies in ecosystems. Since these latter aspects are relevant for, for example, risk evaluation of uncontained uses of synthetic organisms, it is ethically imperative to resist the thrust of machine metaphors in this respect. In addition, from the perspective of the machine metaphor viewing an entity as a moral agent or patient becomes dubious. If one were to regard living beings, including humans, as machines, it becomes difficult to justify ascriptions of moral status. Finally, the machine metaphor reinforces beliefs in the potential of synthetic biology to play a decisive role in solving societal problems, and downplays the role of alternative technological, and social and political measures.

. With regard to the potential moral status of organisms, awareness of the explanatory scheme implicit in the machine metaphor helps to understand that ascriptions of moral status hinge on characteristics of animal and human behavior that are out of the reach of the machine metaphor. Thirdly, and finally, with regard to assessing means to solve societal problems, awareness of the internal logic of the machine metaphor and its epistemic background story may safeguard against overestimating the capacities of synthetic biology to successfully deal with these challenges. --- Authors' contributions All contributions to the text stem from the single indicated author, JB. --- Ethics approval and consent to participate Not applicable. Competing interests I declare that I do not have any financial and non-financial competing interests. --- Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The extent to which machine metaphors are used in synthetic biology is striking. These metaphors contain a specific perspective on organisms as well as on scientific and technological progress. Expressions such as "genetically engineered machine", "genetic circuit", and "platform organism", taken from the realms of electronic engineering, car manufacturing, and information technology, highlight specific aspects of the functioning of living beings while at the same time hiding others, such as evolutionary change and interdependencies in ecosystems. Since these latter aspects are relevant for, for example, risk evaluation of uncontained uses of synthetic organisms, it is ethically imperative to resist the thrust of machine metaphors in this respect. In addition, from the perspective of the machine metaphor viewing an entity as a moral agent or patient becomes dubious. If one were to regard living beings, including humans, as machines, it becomes difficult to justify ascriptions of moral status. Finally, the machine metaphor reinforces beliefs in the potential of synthetic biology to play a decisive role in solving societal problems, and downplays the role of alternative technological, and social and political measures.

INTRODUCTION Across contemporary computing disciplines there is an abundance of systems with a social dimension, either designed for the primary purpose of being used between people (such as Facebook), or those designed with respect to the wider social context, such as control room interfaces. In this paper we focus on the former: technologies whose usefulness is largely conditional on its use by multiple people in some kind of broadly cooperative fashion. The trend of social technologies is enabled by the increasingly widespread availability of sophisticated, connected infrastructure and systems, for example smartphones, wired and wireless broadband internet, internet-enabled digital picture frames, streaming radio and various XML-based formats and protocols. Systems often have very little sophistication on the client side, with user experience and functionality determined server-side, or increasingly, 'in the cloud', and thus highly amenable to rapid change. Moreover, these technologies have for many people become commonplace, mundane and thoroughly integrated into everyday life. Along with gains in technology, the HCI field's perspective on the largely inescapable social dimension of technology has gained in sophistication. It is now accepted that technology is not used in a vacuum and that to understand use and improve design, we must pay heed to the multifaceted environment in which it is used. If we accept Lave's [23] and Suchman's [35] notion that action is intertwined and inseparable from the social environment then it follows that to understand action, we should favour genuine use in everyday settings rather than formal, contrived laboratory studies [5,7]. Ethnographic fieldwork has been widely used for developing insights about people and cultures that can inform the design of technology. An ethnomethodological analytic stance that examines the ways in which people make sense of their world, display this understanding to others and produce the mutually shared social order in which they live could be particularly instructive for understanding the social use of new technologies. Repurposed as it is from social science, ethnography however sits awkwardly with the design of new technologies, as its primary remit is to investigate and describe a culture, rather than proffer suggestions of a possible future. Moreover, trends in technology such as ubiquitous and mobile computing hamper observational fieldwork's effectiveness as significant elements of interaction are outside of the purview of the ethnographer [11]. While useful, ethnography is not a complete approach to investigating use and needs to form part of a wider exploratory, iterative research programme. Use is not an inherent property of a system or artefact, rather something that is developed and maintained over --- Use Once the initial prototype is deployed and participants begin using, misusing or under-using the system, there is opportunity to begin data collection and analysis. Quantitative data can come relatively 'free' by logging system usage, with qualitative methods useful for explaining or further exploring discovered phenomena. Largely, evaluation in RAID takes place through use and design. We draw from ethnomethodology in our emphasis on situated use and willingness to be led by discovery. Observational fieldwork, often employed in ethnomethodologically-oriented studies, was used in the investigation of the prototypes. Prototypes can and should be instrumented to collect as much data as possible given privacy and ethical concerns. Rather than simply logging successful interaction transactions, it can be useful to also capture attempted, failed or aborted transactions. Distributed logging scenarios present some logistical difficulties with the storage and transport of data [9]. A question log is maintained for open questions regarding the use of the system. These arise throughout the RAID process, and the log helps the team to prioritise and focus data analysis efforts. The log can support a kind of unit testing or validation when designing alterations or new features. Questions are added at design time to be later resolved when the changes are in actual use. --- Reflection The reflection stage is concerned with digesting available information from the design and use stages to produce considered design and methodological responses. It is through this reflective design step whereby the researcher opens new areas of inquiry. Schön describes this as part of a "move-testing" experiment, in which theory, developed through reflection can be confirmed or negated in the course of action [33]. Understanding what to respond to and how is critical for the next iteration to progress instead of regress. A continual reflective journal is kept to record observations, thoughts and ideas during the design process, as well as data extracts and notes on the prevailing social environment. The journal, together with its associated artefacts, permit a dialogue to take place between the design and the designer as well as the research programme itself. Ephemeral observations gained as a result of being an embedded researcher can be recorded, for example the 'back story' behind events, or ad-hoc remarks made by participants. Along with notes, we included screenshots, photographs, extracts of interviews and user-created artefacts, charts of usage data and so on into entries made on a weekly basis. The journal was a working document, rough and incomplete, yet served as a valuable aid not only for on-going reflection and design, but for when exhaustive data analysis was carried out much later. In deploying changes, we change the world to some degree: we change the design, change ourselves and change others' behaviour. Unexpected events and phenomena abound. The reflection stage is where pause is taken to consider these outcomes with respect to the research and design goals and analysed usage, before devising responses. The design motto can be useful to guide and moderate response, as not all observations can be reasonably acted upon. Responses can also be a reflection on the research programme: perhaps new methods or tools need to be used or existing techniques refined. The emergent experience of the system -for example how Rhub transcended being a discrete prototype with discrete technical features and became an everyday, natural tool for coordination -can easily escape analysis. The reflection stage is an excellent opportunity, with assembled observations of usage, analysis and notes to explore notions of higher-order experience. --- RELATED WORK Iterative design and development has long been accepted as an important means to mitigate some of the inherent risk of design: we never fully know what we are designing until we begin. Even under tight time constraints [14] or limited cycles [30], iterative design is considered beneficial. RAID has a similar workflow to various iterative development methodologies, such as Rapid Evolutionary Development [1] but differs in that RAID has a greater emphasis on the reflective process and is focused on exploring emerging use and appropriation rather than striving for software quality and correctness. There are many cases in the literature where a'real-world' prototype has been used. Motivations vary, but commonly include: testing and capturing usage in realistic scenarios; deploying in non-public contexts such as the home; probing to see how the prototype is appropriated; and how and what kinds of use develop [25,34,37]. Few, however, deploy prototypes for long continuous periods, thus restricting opportunity for appropriation or non-novelty usage. Moreover, there is little articulation of the methodology used or reflection of its implications. Some notable systems that were deployed for a long period and used by a number of people, such as the SPAM and Hermes systems [9] only had limited iterative development during the deployment. Crabtree's treatment [10] of "technomethodology" [6] argues the case for real-world deployments and provides an outline for how it might proceed in practice. Technomethodology is positioned as an approach for bridging ethnomethodology and system design. Prototypes are deployed in the world as breaching experiments and use studied with an ethnomethodological perspective. New design solutions are devised based on findings and then redeployed, forming an iterative cycle. RAID however emphasises long-term use of a single prototype and far more intertwined relationship between design and analysis. A commitment to in-situ, longitudinal design has been explored in the context of co-designing a museum guide with curators for museum patrons [16]. A key difference is that our approach focuses on in-situ use by target users of a continuously usable prototype and emphasises exploring and fostering long-term usage. Issues relating to the wider use of a system are difficult to anticipate, due to their complexity, uncertainty and emergent nature [29]. Nathan et al. [29] present four criteria to help envision an understanding of these issues early in the design process, however we suggest that it is through design and use that questions which arise in this process can begin to be resolved. The interpretivist perspective of RAID, to learn through use in an exploratory, iterative fashion is inspired by Action Research [24]. The four stages of Action Researchplanning, action, observation and evaluation -are simplified in the RAID framework and tailored for technology-oriented design. Here, ethnomethodological studies take place around a deployed prototype, with observations and analysis informing the next prototype's design. Appropriation has been described as the process through which artefacts are adopted for use in everyday practice [13]. Appropriation work takes place in a number of different dimensions, such as social, technical and organisational [2]. As in the RAID framework, Carroll [8] argues appropriation is a valuable resource for design, and should not be seen as something that takes place once the design process is complete. --- EXPLORATORY PROTOTYPE In its original formulation [21], technology probes were primarily designed for open-ended use with minimal functionality 2. They were introduced early in the design process and not altered during deployment, actively logging user activity while subjects experimented with them. Probes were presented as being different from prototypes which are usually designed for a particular purpose, have rich functionality, evolved iteratively, introduced late in the design process and primarily utilised for gathering usability feedback. In our work, we weren't satisfied with dichotomy between probe and prototype, thus preferring the term 'exploratory prototype' as some form of middle way. The exploratory prototype is a usable and useful system, deployed over a long-term basis in real, natural settings. Like probes, they seek to be the tools or toys which participants can pick up, use and integrate into their everyday life as they choose. It is the analysis of people's activity with these tools that informs design. Like prototypes however, exploratory prototypes are iteratively developed and have rich functionality. By presenting a single evolving prototype rather than a series of different prototypes, we aim to smooth over logistical difficulties often encountered when deploying prototypes, such as delivery, configuration and training. Moreover, participants are able to build trust in the prototype over time, which might be manifested by using it for more personal activity, recommending it to others or even coming to depend on it. Difficulties in deployment can severely limit the number of participants researchers are able to recruit and maintain, which in turn hampers design feedback. Today's commonly available internetconnected devices mean it is simple to offer a technically sophisticated exploratory prototype which can be iteratively developed with zero or minimal impact on participants. --- DISCUSSION In this section we dissect elements of the RAID framework, relating them to the literature and our experiences with the two cases of Rhub and Nnub. --- Feedback through use The necessity to "reconcile the fragments" [11] of data from disparate sources was for us, largely mediated through the reflective journal. In the journal we brought together extracts from our other information sources and analysed them as a whole. With time stamping across data and artefacts such as the change log, we could consider usage with respect to the state of the design in time, or in relation to changes that occurred before or afterward. Logged usage data needs to be significantly pre-processed to be useful, such as through time-series trending and activity clustering. In some cases, we created instrumentation to provide the design team with live usage metrics of a feature in parallel to creating the feature itself. For example, as Rhub grew, it became necessary to offer enhanced functionality to limit its push messaging. One design response was a text message command which would stop all Rhub messages from being forwarded to that person's phone for a specified time period, or by default, several hours. While designing and implementing this feature, we were curious as to whether it would be used, for what reasons, and if time periods would be specified. These questions were noted in the question log for follow-up in interviews with participants after the feature had been deployed for some time, as well as for the creation of simple data visualisations depicting live usage of the feature with regard to received message quantity and so forth. Participants' interactions leave their mark in the data differently, perhaps only appearing as faint traces, necessitating follow-up interviews or workshops to form complete accounts. In Nnub, sometimes a particular posted notice would be accessed more than expected, leading us to examine the pattern and chronology of access to discern whether the pattern was attributable to the notice design, aspects of search or interaction, or other contextual factors. For example, while text posts were more frequently posted, it was picture posts that were viewed the most, so we looked for ways to simplify their creation, such as integration with Flickr and a drawing mode. We observed five major patterns of participants' usage that were helpful in the design process: deluge, accretion, drought, erosion and missteps. Designers can use these patterns for pre-emptive evaluation and consideration during design (for example, how will sudden surges of activity be handled?) and also for suggestions of analytical perspectives on usage. --- Deluge A rapid influx of usage can expose scalability issues with usability and technical architecture. For example, one group of Rhub users treated it like instant messaging, sending a rapid, large number of short messages, overloading the system and resulting in messages being delivered out of order or not at all. This incident occurred after many months of stable usage by a much larger group, the difference, however, was they used it akin to text messaging, sending longer, but less frequent messages. In Nnub the introduction of scribbles led to a large number of scribbles per day, displacing notices (which usually take more time and effort to construct) from view. This led us to redesign the interface to give appropriate visibility to both forms of content. --- Accretion In systems that allow people to create artefacts (such as photos, groups, annotations and so on) or have other byproducts of use, gradual accumulation of these may reveal design deficiencies. As the number of artefacts rise, attention needs to be paid to their usability and management. It can also be useful to examine what kinds of artefacts are being created and for what purpose, perhaps with a view to better supporting this usage scenario. For example, we noticed people creating locations in Rhub for their home address, so they could set their location as 'home', and for this to be visible to friends. After observing the accretion of home locations, which aren't really useful for others, we decided to special-case this scenario and the creation of 'home bases' which are represented differently and do not appear in public location listings. --- Drought Under-use or absent use of features can hamper the viability of other dependant features or create a disinviting, barren user-experience. Use can be encouraged by improving usability and utility, while objective review of the feature might lead to its redesign or removal. In the case of Rhub, we were curious about the use of location-based services, however too few participants were setting their location for these higher order services to offer value. As a result, we made location-setting successively easier and more rewarding which in turn led to greater use of dependent features. Non-use of Nnub by some community group leaders was followed-up by ethnographic study, proving informative. Both systems had a majority of passive users who preferred to observe rather than participate. Although acknowledging the benefits of a digital noticeboard and the need for outreach beyond their group, some group leaders only used email for group announcements, due to lack of time. We explored this use context further, leading to design interventions aiming to better support or potentially transcend existing email practices, for example allowing people to post content with email. --- Wearing in Erosion or wear can reveal repeated use in physical artefacts. For example, patterns of worn paint on a mobile phone might reveal how it is usually held. Software-based systems do not physically wear, however user activity can be logged and then later analysed for trends and established usage patterns. Like the accretion pattern, it may reveal activity deserving of further analytical focus or design response. Basic usability can be improved for example by shortening navigation trails and surfacing information to more accessible locations. More substantial changes can also be made. Over time it became apparent that Rhub was used mostly for group communication, and in particular, for ad-hoc group coordination, such as planning a night out. Our users had a loose, dynamic perspective of groups. We thus successively altered the design to support a more fluid group structure and additional support for coordination, allowing people to 'tune in and out' of event organisation as interested dictated. --- Missteps Users' missteps -trying to do something the system doesn't support, or causing an error, for example -are valuable for design. Missteps can surface basic usability flaws as well as mismatches between the design and participant's expectations and intuitions of how it should work. Because deployment and use is outside the lab, missteps are not as easily captured as in traditional usability testing. Extensive logging and exception handling is critical to respond to missteps and judge their importance, and serves as a useful input to qualitative processes to understand the nature of the misstep. In the case of Rhub, we significantly improved its text messaging interface through observation of attempted commands sent by users. Similarly with Nnub, we provided support for creating notices based on content from other websites after we observed the practice being performed using crude copy and pasting. In this way, features are co-evolved with users' expectations of what they can be useful for and how to use them. --- Feedback over time Exposing the prototype to use over a long period better positions it to capture intermittent and periodic events which might otherwise fall outside of a short term study period. For example, consider how activity within a household changes throughout the day, week and year. During weekdays, there might be a regular pattern of early morning activity, a lull during the day and burst in afternoon which tails off into the night. When examined on a weekly basis, weekend patterns emerge. When examined on a yearly basis, events such as holidays and birthday celebrations become apparent. In our deployment of Nnub, clear usage patterns emerged that corresponded to community activity, such as surges of use when the nearby school was starting or finishing. Identifying and analysing such patterns can be informative: correlation of usage and important community events led us to explore how events were publicised, discovered and communicated within the community. Over time, the system is exposed to more varied use contexts, not only because people themselves vary their activity, but because new people begin using the system. Alternative perspectives on existing practices become evident, or in some cases entirely new use contexts emerge. For example support for PDF notices in Nnub was not implemented until a government agency began using the system and had exacting requirements on how notices were to be presented. This requirement did not emerge until Nnub was deployed for three months and now that it is implemented, sates one particular user group and opens the possibility for new and interesting practices by others. --- Continuously available The RAID framework suggests that the exploratory prototype be continuously available and usable, so that participants can build trust and come to rely on it. With the knowledge the prototype will not be taken away after a short period, participants are able to view their usage and conscious appropriation-work with a long term outlook [7] and invest in appropriation. Work such as tailoring and negotiation with others to establish mutually beneficial usage patterns can be amortised over a long period, rather than be neglected for short term gain. Care can and should be taken when analysing the in situ use of short-deployment prototypes, where there is considerable novelty usage and minimal appropriation. Fundamental differences exist between novelty usage and longer-term'sustainable' usage. In our experience, novelty usage is characterised by playful experimentation, which stabilises as users -with regard to others and the environment -establish how to use the system in a beneficial manner. Analysis of novelty usage is still useful for design, however any such analysis should be considered and presented in this frame and not necessarily representative of normal use. If a prototype is not able to sustain long-term use, more fundamental questions arise regarding its actual usefulness and desirability. Keeping the prototype stable and working during iterative development with minimal resources is a core challenge. This can be partly mitigated by managing participants' expectations of the level of functionality and reliability of different parts of the system, for example clearly signposting areas which are new and not completely tested. In our experience, we found that participants had very high expectations of the prototype and did not fully appreciate the difficulty in delivery of desired functionality. However, this pressure is lessened by the privilege of having people use a system, and serves in a variety of ways to remind designers of priorities from a use perspective. In any research project, participants are a valuable asset, and care must be taken that they are not lost due to frustration with system. In the case of Rhub, alarmingly soon after its introduction, people depended on it for communication and coordination. If a Rhub message did not reach the intended recipients due to a system fault, people would naturally be disappointed and reduce their usage, or use it only for inconsequential messages. In one case, a leader of a student group assignment established a Rhub group and invited the other members. A number of messages were successfully sent to the group via Rhub, however on the one occasion that he attempted to use Rhub to organise a group meeting, Rhub failed to deliver the message due to a fault. After this, the group effectively abandoned Rhub and thus we lost four participants and the exploration of a new usage scenario for the system. --- Contingency and context Use is contingent. How we use something and for what purpose cannot be dictated in advance by the designer: he or she can only make some ways easier than others. There is a place for formal laboratory-based studies of interaction; however, it is in the artefact's natural environment that it is exposed to realistic contingency. For new technologies, studying everyday in-situ use of prototypes can be necessary as people are poor at anticipating their need or potential use before they are actually using it and have integrated it into their normal routine [31]. Following on from Heidegger, it is by way of everyday use that a tool transcends being present-at-hand to become readyto-hand at which stage activity can take place through the artefact rather than with the artefact. The appropriation process might realise the artefact's intended uses or perhaps, by using it in ways the designers never intended, transform the artefact entirely. Use does not take place in a vacuum: we observe others' use, others observe us and we operate within semi-rigid cultural and societal structures. Thus how someone uses an artefact and what that artefact means to them is contingent on properties far beyond the material and functional attributes of the artefact itself. While we acknowledge that all action takes place with regard to context, some design problems involve action that is particularly sensitive to the context in which it is usually carried out. If a reasonable facsimile of the context with its salient features cannot be recreated, testing the action outside of the context can offer only limited insight. For example, exploring action in a waste water treatment plantwhich provides a rich phenomenological experiencedemands grounding to the actual plant [7]. Social, personal or intimate contexts like the home may also reveal their contingencies best through in-situ use without researchers present [33]. With social technologies the context of action is enormously variable and difficult to effectively recreate or represent in a laboratory environment. --- Utility Everyday use can be realistic use. Where the exploratory prototype offers enough value or utility, people will use it not because they are paid or asked to follow a synthetic scenario but because they have a genuine want or need. In addition, they can use it when the desire arises, rather than be limited by a short deployment period. Evaluation of contrived use can give a distorted impression of the system [3]. Artefacts offer utility or value in different ways. Some might enable an activity that was previously impossible or difficult to accomplish. Other artefacts might offer value because they are enjoyable to use, or support selfexpression and creativity, such as a musical instrument. Where the utility of an artefact is compelling enough, people are willing to sacrifice a high level of usability [12]. For example, text messaging was not dismissed in its infancy due to limited text input capabilities or 160character message limits. Text messaging's utility -along with factors such as an affordable unit-based pricing model -led to its long-term regular use by millions of people. In turn, this has provided opportunity for researchers to explore the way it is used and the socio-cultural practices that have emerged around it (for example, [17,22,26]). Findings from this research has led to the development of a variety of technologies which aim to better support practices evident in text messaging usage, our prototype Rhub being but one example. Although original text messaging implementations suffered from numerous shortcomings in their usability, widespread, rapid growth in usage demonstrated its utility and provided the means for improving design by studying actual usage. We have previously suggested [20], along with others [15], the importance of utility, and its relatively minor role in HCI discourse. Longitudinal design processes such as RAID help to ground design and hold its actual usefulness to account, as people will cease to use an artefact if it does not provide enduring value. --- Rapid evolution: opportunity and risks Traditionally, once a prototype is deployed and being used the designer is relatively powerless to influence how it is appropriated. Means external to the artefact are available, such as training or marketing, but little change can be made to the artefact itself to encourage or discourage particular ways of use. Prototypes need to be recalled or replaced with an updated version and there is accompanying difficulty and effort in this transition, inhibiting iterative design. Connected systems and devices, however, can be iteratively changed with minimal inconvenience to users. This new trend of continuous or online iterative design enables it to be an activity that takes place over the life of the artefact, rather than taking place once or at sparse intervals. By responding to contingency and emergent use more quickly, exploration of a design space can be more expansive and thorough. Change, and the communication thereof, must be carefully managed so that in the pursuit of exploring unfolding use, participants are not alienated by a changing or unstable system. If the prototype is continually evolving, a clear concern is how to update participants with new, removed or changed functionality. During the early deployment of Rhub, bugs were fixed hours after they occurred and new functionality introduced on a daily basis. If new functionality is silently introduced, it may be some time before it is stumbled upon, and thus not serve its purpose for evaluation through use. Non-use through non-discovery of new functionality is quite different from non-use through disinterest or dislike. Too much change risks lessening the benefits of the long-term prototype, as participants will not be able to bond with a kaleidoscopic system. To keep users informed, we sent periodic emails to users of the respective systems. The newsletters were a digest of entries from a frequently-updated development blog, which early adopter users monitored. Early adopter users are a useful resource for idea generation [38]. By making usage of features externally observable where possible, early adopter use could be noticed by others and was an effective way of spreading knowledge of new features. --- The embedded researcher For both systems, we were active participants in not only the use of the system, but the wider social context in which they were initially employed. As in Action Research [24], from which we draw inspiration, we were thoroughly embedded in the context, playing an active role in encouraging and facilitating use. Accordingly, we were in an excellent position to observe use of the systems in practice and how they changed the social context. Because of our existing personal ties to the user community -which we wished to maintain -we were easily sensitised to concerns such as privacy and spamming. Moreover, with both case studies, we hoped to strengthen ties between participants, including ourselves. As with many highly-situated investigative approaches, questions can be raised regarding generalisability and validity, especially as we are actively shaping use through iterative design. We would suggest it is a necessity to maintain an open and honest stance, to be led by the unfolding use and understanding. Generalisability and validity can be improved through triangulation of data gathered with varied analytic perspectives as well as deployments in multiple contexts [28]. A native's knowledge and ease of access to the social context was highly beneficial for fieldwork and initial design and deployment, and eased some of the difficulties associated with the observational study of mobile and ubiquitous technologies [11]. However, as the user population grew and new sub-groups arose, our privileged position was diminished. How well the RAID framework adapts to deployments in alien or relatively inaccessible contexts is an open question. --- Confined design and re-design One of the concerns rightfully levelled at the use of functioning long-term prototypes is that once deployed, the potential for the design to evolve is limited: the premium placed on use and user feedback may come at the expense of design flexibility. There is a balance to be struck between providing utility in the prototype, yet for the design being sufficiently 'open', to be led by participants and the exploratory process. Too much "closure" [27] and users will either use it exactly as intended, or not at all. Too much openness and perhaps utility is diminished. In functioning software prototypes, openness seems to equate with providing power and preventing excessive structure and detail. In contrast, non-functioning lo-fidelity prototypes don't provide functional power but can furnish an array of different designs with minimal effort particularly with respect to physical affordances and evoking imagination about use. However, these nonfunctioning prototypes do not yield the same form and depth of data in context and over time. Functional prototypes can be narrowly scoped at first to explore particular core aspects of functionality over time. The art of RAID is deciding how to develop the prototype and being judicious about when to discard the prototype and begin anew, paying particular attention to the relationship established with user communities and responsibilities in relation to user generated content. Clearly there is a place for both approaches, for example, conducting early workshops with numerous design alternatives, followed by a course of long-term use. With Rhub, we were able to completely redesign some aspects of functionality whilst still within the overarching design paradigm, and as far as the participants were concerned, same system. Data from long-term use can also feed into fresh, experimental designs which can be explored tangentially. --- CONCLUSIONS In this paper we have argued for a reflective, agile, iterative approach to socio-digital systems, emphasising longitudinal, everyday use of an evolving, continuously available exploratory prototype. This is particularly important for social technologies, which tend to be used in an ad-hoc manner highly contingent on context. In examining the use and appropriation of technology with this perspective, we are able to proactively explore a design space, evolving the prototype as participants' activities, needs and understanding evolve. The RAID framework suggests iterative cycles of design, use and reflection around an exploratory prototype, assisted with the use of artefacts such as the change log, question log and reflective journal. These artefacts help to document ongoing change in relation to observations and questions regarding use. Time is a good test of a design intervention's usefulness and relevance in any given context. Over time, we can observe use in time frames and contexts that suit peoples' lives rather than rapid research or product development cycles. We can see the ways in which people bridge existing practices with the new system, "making it at home" in their personal and social environments. Everyday use exposes the prototype to natural, everyday contingencies, providing reason and opportunity to use the system in new ways. New possibilities emerge from use in related or unexpected contexts. With a continuously available, long-term prototype, we aim to foster participants' investment in appropriation and use, potentially resulting in richer, more realistic observations and testing the enduring validity of research claims. Analysis of how people use the exploratory prototype is core to the framework. We identified five usage patterns which are particularly informative for design: deluge, drought, accretion, wearing in and missteps. Aspects of appropriation and configuration activity are revealed through these patterns of use, non-use and difficulty of use. Wider availability of online systems taking advantage of pervasive mobile devices and connectivity improves the feasibility of rapid development and continual data collection across a wide variety of technological artefacts. RAID exploits this trend in order to rapidly co-evolve design with use and understanding. As the prospect of long-term use and real uptake are the major questions that surround many designed systems, the approach is particularly promising. However it requires an artful consideration of appropriate design response to both user and research needs. This paper serves as a reflection on our experience with long-term evolving prototypes but raises many further questions that are beyond the scope of this paper such as the role of the designer and how to disentangle emergent use from designer-directed use.

One of the major challenges in the design of social technologies is the evaluation of their qualities of use and how they are appropriated over time. While the field of HCI abounds in short-term exploratory design and studies of use, relatively little attention has focused on the continuous development of prototypes longitudinally and studies of their emergent use. We ground the exploration and analysis of use in the everyday world, embracing contingency and open-ended use, through the use of a continuously-available exploratory prototype. Through examining use longitudinally, clearer insight can be gained of realistic, non-novelty usage and appropriation into everyday use. This paper sketches out a framework for design that puts a premium on immediate use and evolving the design in response to use and user feedback. While such design practices with continuously developing systems are common in the design of social technologies, they are little documented. We describe our approach and reflect upon its key characteristics, based on our experiences from two case studies. We also present five major patterns of long-term usage which we found useful for design.

Introduction Critical investigations into the functions, relevance, and values are considered crucial for reevaluating concepts and theories within the realm of social science. These inquiries offer the opportunity to examine social concepts within specific societies and regions. Some argue that a theory that proves functional in particular contexts may lose its significance when applied to other places. For instance, Weber's theory of capitalism appears relevant when applied to modern European countries, but its degree of relevance diminishes when applied to certain areas in China, India, and Southeast Asia. 1 When assessing the applicability of a theory, scholars should emphasize not only its theoretical formation (epistemology) but also engage in critical-philosophical reflection. This reflective approach, often referred to as the "reflexive turn," according to Salvatore, aids in discovering a functional path and direction for contextualizing the theory. 2 The "turn" implies the need for elaborating genealogical inquiries as well as engaging in critical and comparative studies. 3 This elaborate project of the "reflexive turn" is necessary to establish the origins of concepts and their ideological values. 4 However, this reflection is not without its challenges. Modern social sciences have predominantly developed under the influence of European scholarship, which has often enjoyed unrestricted access to the invention of concepts and theories. 5 While these concepts and theories may possess universal validity in a general and abstract sense, their historical and concrete manifestations are conditioned by temporal, spatial, and cultural frameworks. 6 Historically, they have been integral components of Western civilization, at times claimed as "the global" perspective. 7 Consequently, the Western social sciences have exerted significant dominance, shaping how social concepts and theories are perceived and interpreted in other regions such as Asia, Africa, and the Middle East. 8 This influence stems from the disparities between the academic traditions of the West and other regions, each distinctively characterized by their own differences and complexities in political, economic, military, and cultural contexts. 9 Additionally, the profound impacts of colonialism, which have shifted from the center to the periphery, have further contributed to this reality. 10 This historical fact, indeed, warrants critical examination. However, it is essential to avoid solely focusing on the issue of civilizational domination and colonialism. While acknowledging their significance, it is equally important to consider the emergence of resistant thought against Western scholarship (Occidentalism). These factors are not relatively unimportant and should be given due consideration. 11 Yet, it is crucial to give due consideration to the universal virtues of the West, particularly concerning the development of social science. This perspective acknowledges that Malay-Indonesian scholars can learn from the West, as its global influence is not an exclusive representation of the entire globe but rather a part of 5 Edward Said, Orientalism (London: Penguin Books, 2003), 325. 6 Syed Hussein Alatas, "The Autonomous, the Universal and the Future of Sociology," Current Sociology 54, no. 1 (2006) Edward Said, Culture and Imperialism (London: Vintage, 1993), 377; Syed Farid Alatas, "Academic Dependency and the Global Division of Labour in the Social Sciences," Current Sociology 51, no. 6 (2003), 599. https://doi.org/10.1177/00113 921030516003. it. Moreover, it is important to recognize that "the global" in this context does not imply a singular meaning but encompasses diverse civilizations, allowing for the existence of multiple modernities. 12 For instance, values such as independence, freedom of conscience and thought, autonomy, equality, and justice are universal values not limited to the West but have evolved over time in various regions across the globe. While acknowledging the significance of dichotomizing between the West and other regions or the center and the periphery, it is crucial to critically assess its relevance within the scope of this examination. Undoubtedly, the exploration of traumatic memory residues that ideologically emphasize the narrative of "Eastern victims of Western oppression" will elicit critical responses from intellectual perspectives influenced by post-colonialist modes of thought. 13 It is important to emphasize that acknowledging the problems and influences of colonization does not imply undermining their significance. It is imperative to address and mitigate these problems and influences accordingly. In conjunction with these efforts, a reconsideration of diverse conceptions of cosmopolitanism within the framework of multiple modernities may pave the way for a potential trajectory in the development of social science, encompassing various locations, particularities, vernaculars, temporalities, intellectual backgrounds, and notably, traditions as the most influential elements. 14 In this context, the term 'traditions' refers to the fundamental element that underlies perspectives in social science. However, it is important to note that these traditions are neither immutable nor static; they are dynamic entities with the potential to generate creative insights. 15 According to Salvatore, "...Tradition should not be understood as non-reflexive, primordial culture, but more dynamically, as the ensemble of practices and arguments that secure the social bond and provide cohesiveness to human communities of 12 Arnason, "East and West"; Salvatore, The Sociology of Islam,23. 13 Ibid., 232. 14 Mannheim, Essay on the Sociology, 100; Tariq Ramadan, The Quest for Meaning: Developing a Philosophy of Pluralism (London: Allen Lane, 2010), 22-25. 15 Shaharuddin Maaruf, "Some Theoretical Problems Concerning Tradition and Modernization among the Malays of Southeast Asia," in Yong Mun Cheong (ed.), Asian Tradition and Modernization Perspectives from Singapore (Singapore: Time Academic Press, 1992), 242-244. varying scale." 16 So, practically, tradition has facilitated communicative actions, transfer of knowledge, and socio-political contestations by social agencies. When the social process of tradition has occurred, some assert that it is "understood as a creative link, as the triumph of memory over the power of time." 17 In the context of this inquiry, traditions of the Malay-Indonesian world are encouraged to creatively elaborate on concrete social situations and problems since they "can only have a positive value if they are based on the creative tendencies in society." 18 At the same level, social science understanding must consider the essentials of tradition. 19 Based on this point, therefore, it is important to offer examinations and reflections on two major alternatives of engagement of social science and tradition in the Malay-Indonesian context. This article provides an intellectual reflection on the significance of integrating social science and tradition within the framework of Malay-Indonesian scholarship. Its objective is to shed light on the thought process employed by Muslim intellectuals in the region. Consequently, it aims to pave the way for the advancement of social science by exploring potential avenues for contributing to its overall progress and development on a broader scale. 20 This article contends that the Malay-Indonesian tradition offers an alternative perspective within the realm of social science, which exhibits a distinct inclination toward perceiving and formulating certain concepts. This inclination is shaped by unique qualities, alternative values, intellectual stances, and thinking styles among scholars. As a result of these factors, this alternative perspective tends to be autonomous, problem-oriented, and transformative, as it actively engages with social change and transformation. In examining this perspective, the article considers two conceptual approaches: autonomous social science and prophetic social science. It is essential to discuss the relevance of both approaches in the overall dynamics of social theory development. --- Autonomous Social Science The concept of autonomous social science was originally proposed by Syed Hussein Alatas (b. 1928 -d. 2007), a renowned sociologist, Muslim intellectual, and social activist from Malaysia. Although his concept has not gained the same level of recognition as Karl Marx's critique of capitalism in Das Kapital (1867), it holds significant importance and is found throughout Alatas' extensive body of work. Notably, two of his papers, "The Development of an Autonomous Social Science Tradition in Asia: Problems and Prospects" (2002) and "The Autonomous, the Universal, and the Future of Sociology" (2006), prominently emphasize this perspective. 21 The application of this concept can be traced through his books, "Intellectuals in Developing Societies" and "The Myth of the Lazy Native: A Study of the Image of the Malays, Filipinos, and Javanese from the 16 th to the 20 th Century and Its Functions in the Ideology of Capitalism," both published in 1977. 22 Alatas defines autonomous social science as a conceptual viewpoint encompassing "the linking of social science research and thinking Asian problems." 23 It is essential to note that the mention of Asia does not imply an endorsement of cultural nativism. Rather, it signifies that Asian scholars have the opportunity to leverage their traditions to make valuable contributions to addressing the issues faced by their own societies or communities. 24 This concept necessitates a profound moral commitment and unwavering intellectual conviction. It primarily functions at the level of intellectual reflection, where its optimum role is realized. 25 The foundation of the concept of autonomous social science lies in the universal values that emerge from religion, cultural traditions, local wisdom, and the ethical principles governing social life and behavior. These values serve as a collective set of moral guidelines, illuminating the intellectual perspective of scholars as they engage in the examination, evaluation, consideration, and formulation of specific concepts. This intellectual endeavor is of paramount importance, as it enables these values to actively shape the inclination of social science thinking. When these values, such as freedom, independence, autonomy, equality, and justice, are reflected and incorporated into intellectual works, they contribute to a reflective and theoretical approach to problem-solving within society. This endeavor stands as a pivotal aspect in shaping the conceptual framework of autonomous social science. 26 Hence, as emphasized by Karl Mannheim, the concept of social science should be both diagnostic and reconstructionist, as social scientists bear social responsibilities that urge them to actively engage in the social process and establish an organic connection with society. 27 In the application of the concept of autonomous social science, the cultivation of creative and critical thinking plays a crucial role. Such thinking becomes particularly important in addressing two key challenges: the tendency of intellectual thinking to become stagnant when social scientists confront difficult and complex problems, and the earnest effort required to critically question prevailing ideas that are often accepted uncritically, without adequate justification. 28 This creative and critical thinking is also important because it can (a) give injection of the development of idea; (b) ensure that the false thing should be fixed; (c) emphasize clarity than ambivalence; (d) uncover the ideological deviation; (e) give new meaning persistently and dynamically to the social concepts that have been applied. 29 In the word of Erich Fromm, "critical thinking is...an approach to the world... [which] stands in the service of life, in the serving of removing obstacles to life individually and socially which paralyze us."30 Without adopting this stance, scholars often encounter challenges in critically understanding both complex problems and the prevailing ideas that may be obscured by specific socio-political and cultural biases. For instance, excessive reliance on consuming concepts or theories without effectively applying them to achieve their intended goals, and a lack of attention to the significance of connecting scholarly work with society, inevitably restricts the full potential of scholars. Highlighting the significance of creativity and critical thinking, it is essential to recognize that social scientists must avoid becoming hindered by excessive reliance on a singular perspective. Such dependence can undermine essential virtues such as freedom, independence, and autonomy. Alatas further suggests that Asian scholars can benefit from studying and learning from the Western experience. He highlights this further, "In the Western world, the autonomous tradition is decisive and vigorous, and the demarcation line between general universal sociology and the autonomous studies of subjects peculiar to specific Western countries is clearly observed." 31 The invention process of social science in the West is argued to be free from any kind of domination by external influence. In the context of pre-colonial period, in the Muslim hemisphere, Alatas mentions that Ibn Khaldun's social science is also autonomous. He eloquently argues that "It was sociology born out of a historical setting unimpeded by the domination of a hegemonic external intellectual tradition from a previous colonial power." 32 --- Prophetic Social Science The concept of prophetic social science was introduced by Kuntowijoyo (b. 1943 -d. 2005), an esteemed litterateur, Muslim intellectual, social activist, and professor of social science at Universitas Gadjah Mada (UGM) in Yogyakarta, Indonesia. Kuntowijoyo's ideas on prophetic social science can be explored through his notable works, namely "Muslim without Masjid" (Muslim Tanpa Masjid) (2001) and "Paradigm of Islam: Interpretation for Action" (Paradigma Islam: Interpretasi untuk Aksi) (2001; new edition 2008). 33 In contrast to Alatas, Kuntowijoyo's concise exposition of prophetic social science can be discovered in his most recent book, "Islam as Science: Epistemology, Methodology, and Ethics" (Islam sebagai Ilmu: Epistemologi, Metodologi dan Etika) (2008). 34 However, it is worth noting that the practical implementation of Kuntowijoyo's concept is often absent, despite his publication of numerous literary works encompassing novels, short stories, poetry, and critical literary essays that reflect his intellectual inclination. 35 The concept of prophetic social science refers to a form of social science that draws inspiration from prophetic values, particularly within the Islamic tradition. 36 This concept emphasizes the significance of social change and transformation, as Kuntowijoyo asserts that Islamic teachings encourage Muslims to actively participate in endeavors aimed at social development. 37 Furthermore, this concept does not solely focus on the normative aspects of Islamic teachings but also takes into consideration the empirical, historical, and temporal dimensions of Islam. 38 In developing his concept, Kuntowijoyo also recognizes the value of Western historical works. Scholars such as Snouck Hurgronje, Schrieke, Pijper, and Clifford Geertz provide him with valuable insights regarding empirical methods of social research, even though he is cognizant that their scholarly works are not politically value-free. 39 This is similar to transformative social science, which involves the exploration of religious teachings and theology to formulate social concepts or theories. This approach has been popularized by theologians such as 33 Kuntowijoyo, Muslim tanpa Masjid (Bandung: Mizan, 2001); Kuntowijoyo, Paradigma Islam: Interpretasi untuk Aksi (Bandung: Mizan, 2008). 34 Kuntowijoyo, Islam sebagai Ilmu: Epistemologi, Metodologi dan Etika (Yogyakarta: Tiara Wacana, 2006). 35 Alatas, The Myth. 36 Heddy-Shri Ahimsa, Paradigma Profetik: Mungkinkan? Perlukah? (Yogyakarta: Universitas Gadjah Mada, 2011), 6. 37 Kuntowijoyo, Muslim Tanpa Masjid, 357-376. 38 Kuntowijoyo, Islam sebagai Ilmu, 85. 39 Kuntowijoyo, Muslim Tanpa Masjid, 103-105. Gustavo Gutierrez, John Sobrino, and Cornel West. 40 Or social scientists like Joe R. Feagin, Hernán Vera, and Kimberley Ducey. 41 While the former focuses on elaborating critical social sciences through the reformulation of theological discourse, aiming to translate the religious spirit into a catalyst for social and political change, the latter endeavors to develop alternative social sciences that draw upon universal religious values. The purpose of these alternative social sciences is to actively apply these values in order to stimulate social and political transformations. This concept comprises two fundamental pillars of ideas. The first pillar involves a modern reconstruction of religious thought, serving as the philosophical foundation of the concept. 42 This reconstruction is enriched through the reflections of two prominent figures in philosophy, namely Muhammad Iqbal and Roger Garaudy. 43 Both emphasize that the prophetic vision is oriented toward serving humanity, particularly marginalized groups (mustad<unk> 'af<unk>n), and finds manifestation through socio-historical actions. 44 It requires a heightened consciousness of prophetic ideals, which should be distinguished from Sufistic consciousness. The prophetic consciousness underscores the importance of manifesting love for God in daily social life, social development, and social transformation. In contrast, Sufistic consciousness primarily emphasizes the individual worship of God. The second is some selected Western social concepts, namely "Thomas S. Kuhn's 'paradigm', Levi's Strauss''structuralism', Peter L. Berger's 'objectification' and Antonio Gramsci's'relative autonomy'." 45 There are some points in elaborating on these concepts: (1) Kuntowijoyo reflects that this concept is not a "normal science", 46 but dynamic science which is ready to be evaluated and examined critically to find many possibilities of creating breakthroughs or something fresh and new; 47 (2) To invent the "prophetic weltanschauung" scholars should select the certain relevant resources since the resources strongly determine the quality of the paradigm; 48 (3) The resources of the paradigmatic thought of the prophetic social science is constituted from not only the Muslim society and relevant academic works but also scriptural resources, especially the Qur'<unk>n. 49 Consequently, Islam here is not viewed as a religion per se but as knowledge or science that is very valuable to enrich the paradigm. Indeed, it does not mean that Kuntowijoyo affirms the constitution of pseudo-scientific work since he is neither ideological nor narrow-minded when considering it; (4) Kuntowijoyo agrees with the insightful notion of Peter Berger and Thomas Luckman in defining social reality as useful resources, since "reality is socially constructed and that the sociology of knowledge must analyze the process in which this occurs"; 50 (5) This concept has a constructive and pragmatic purpose that attempt to generate the ideals of social transformation. In this point, the idea of "relative autonomy" has its role, mainly to "understand the traits and functioning of any religion within a definite society." 51 Through this intellectual elaboration, furthermore, Kuntowijoyo develops three concepts of social science (civil society, historical activism, and the importance of prophetic consciousness) and three prophetic principles (humanization, liberation, and transcendence). 52 These three concepts are derived from the Qur'<unk>n Surah '<unk>l 'Imr<unk>n (verse 110), "You are the best ummah produced for mankind. You enjoin what is right and forbid what is wrong and believe in Allah." Following this verse, the concepts: (1) the scriptural identification on the best ummah (societies, communities, peoples, tribes, and nations) means this is imperative to attempt to develop the civil society based on the prophetic consciousness; 53 (2) developing civil society should be understood as creative efforts of creating history which could be achieved through social activism; 54 (3) the prophetic consciousness is the critical consciousness on the ethical calling of religion which encourages and engages human and society to be more active in serving humanity; 55 (4) "enjoin what is right" means how to accomplish the mission of humanization or "humanizing human" in this contemporary time and situation; 56 (5) "forbid what is wrong" means how to join to the efforts of liberation and at the same time to fight against the dehumanistic, dominant and hegemonic structures in terms of social, economy, politics and culture; 57 (6) "believe in Allah" which means the transcendence of the spiritual awareness of human on what religion names as the God. 58 They are the ideas that creatively formulate the concept of prophetic social science. --- The Development of Alternative Social Science in the Malay-Indonesian Context The autonomous social science and prophetic social science possess distinctive orientations that contribute to the dynamic development of social science. This contribution operates on two levels: intellectual reflection and conceptual formation. As a result, certain characteristics, such as being problem-based, autonomous, and engaged in social change and transformation efforts, have emerged. These characteristics significantly shape the intellectual stance, clarity, mode of thought, and applicability of concepts. Consequently, the orientations, forms, and perspectives in this context differ to a significant degree from social sciences rooted in Western traditions. The concepts of autonomous social science and prophetic social science have the potential to introduce "alternative paths" in the development of social science, which has long been dominated by Western scholarship. These concepts possess characteristics such as being problem-based, autonomous, and attentive to social change and transformation. These alternative paths aim to break free from the influence of dominant Western perspectives, as they acknowledge the diverse societal issues, traditions, ideals of modernity, and scholarly directions present in different contexts. It is important to critically examine the problem of domination and the distinctions between Western and non-Western social concepts, although there may be some similarities. 59 However, it is crucial to deepen this critical evaluation and expand the boundaries of critique. Understanding the problems at hand and offering viable solutions, for instance, are valuable endeavors. The fundamental issue regarding the domination of scholarship lies in the lack of capacity and enthusiasm for critical thinking. In the case of Malays and Indonesians, scholars often lack the reflective ability to consider alternative perspectives. They frequently find themselves trapped in the application of Western theories, unaware of their applicability, functionality, and most importantly, their disconnection from the realities of their own societies. Numerous works in anthropology and sociology, as highlighted in "Social Science and Power in Indonesia" edited by Vedi R. Hadiz and Daniel Dhakidae (2005), 60 serve as prime examples of uncritical inclinations within social science. These works were predominantly produced to fulfill the orders of rulers, dating back to the colonial era until the 1990s. During the New Order era in Indonesia, under the firm grip of the Suharto regime, social science research was directed toward legitimizing the state development project. The roles of economists, 59 Said, Culture and Imperialism. 60 Vedi R. Hadiz and Daniel Dhakidae (eds.), Social Science and Power in Indonesia (Singapore: Institute of Southeast Asian Studies, 2005). particularly those who graduated from the University of California, Berkeley, were significant in this regard. 61 A similar situation can be observed in Malaysia, where several academic works are primarily generated to serve the political interests of those in power. Unfortunately, these works overly rely on irrelevant theories that fail to explain the social problems of society. 62 In response to this concerning trend, Alatas emphasizes that "the phenomena of servility and intellectual bondage are not the same as genuine creative assimilation from abroad." 63 The attribute of "alternative" in the concepts of Alatas and Kuntowijoyo emphasizes functionality, diagnostic capability, rootedness in society, and an orientation toward solving societal problems. 64 This stands in contrast to dominant views in social science that often prove to be inapplicable, irrelevant, and dysfunctional. Furthermore, the "alternative" aspect implies being rooted in tradition, such as Islam in the case of Kuntowijoyo, which enables independent and diagnostic thinking to identify the central problems within society that require immediate resolution. Moreover, it carries an orientation to actively transform culture, tradition, and even society toward improved conditions and situations. 65 Furthermore, it is essential to address the central questions: "What is the main challenge?" and "What are the obstacles or problems we face in confronting this challenge?" It is evident that these alternative concepts did not emerge out of thin air. Throughout the colonial era in the Malay-Indonesian world, spanning Singapore, Malaysia, and Indonesia, and continuing into the present era of globalization, Western dominance has prevailed in the domain of scholarship. Whether in the form of Orientalism, Eurocentrism, or other manifestations, the influence of the West has been pervasive. During the colonial era, social science works were deliberately 61 Ibid., 2-30. 62 Alatas, Intellectuals; Alatas, The Myth. 63 Syed Hussein Alatas, "Academic Imperialism," in Syed Farid Alatas (ed.) Reflections on Alternative Discourses from Southeast Asia (Singapore: Centre for Advanced Studies and Pagesetters Services, 2001), 36. 64 Syed Farid Alatas, Alternative Discourse in Asian Social Sciences (New Delhi and London: Thousand Oaks and SAGE Publications, 2006), 82; Pradana Boy ZTF, "Prophetic Social Sciences toward an Islamic-Based Transformative Social Sciences," Indonesian Journal of Islam and Muslim Societies 1, no. 1 (2011): 115. 65 Soedjatmoko, The Primacy of Freedom in Development (Lanham, MD: University Press of America, 1985). produced to serve colonial interests, while in contemporary times, they have been shaped by globalism, nativism, and even Islamism. Unfortunately, misguided responses to Western domination have fostered resentment, and the underlying problems remain unresolved. 66 One notable example is the scholarship within the postcolonialism framework, where the primary focus often revolves around disproving the detrimental effects of "Orientalism" on colonized subjects. The efforts of prominent post-colonialists predominantly involve presenting "Occidentalism" as a reactionary response, aiming to counter the project and discourse of Orientalism. However, a concerning trend has emerged, characterized by the promotion of nativism, cultural chauvinism, and Islamism in academic circles. Some scholars within this trend tend to prioritize the development of concepts or knowledge rooted solely in their own traditions while outright rejecting any influence from the West. 67 The worst is emerging the symptom of intellectual slavishness. This inclination brings into the academic trend of academism, objectivism or neutralism, positivism, 68 and definitely, Eurocentrism. In this case, everything should be totally considered in the frameworks and the ideals of the West. 69 In simple terms, we encounter two distinct problems: internal and external. The internal problem stems from our own intellectualism and mentality, while the external problem has been shaped by the historical legacy of colonialism. In this complex context, Syed Hussein Alatas criticizes scholars who fail to cultivate a critical awareness, referring to them as having fallen into the trap of captive minds. According to him, the traits of a captive mind include: (1) "it is alienated from major issues in society"; (2) "it is fragmented in outlook"; (3) "uncreative and incapable of raising original problems"; (4) "a way of thinking dominated by Western thought in an imitative and uncritical manner"; and (5) "incapable of devising an analytical method independent of current stereotypes." 70 Captive minds are not limited to just Malay scholarship but are also prevalent in Indonesia. To address this issue, a critical and diagnostic perspective is crucial in order to unravel and overcome the constraints of such thinking. In this way, it is important to consider Mannheim's assertion to break this "block of thinking". Actually, scholars must have "(1) the ability to identify and being affected by a particular problem of issue confronting the society; (2) the ability to determine and define the problems that emerged; (3) the ability to analyze the problems from various dimensions; (4) the ability to offer viable solutions to solve the problems; and (5) the courage to challenge the existing ideas that have caused the aggravation of the problems confronted." 71 However, the contextualization of the notion of alternative social science requires a careful approach. This can be achieved through three essential steps: Firstly, it is necessary to understand and analyze the most significant historical episodes within the Malay-Indonesian society that have successfully fostered social civility, cohesiveness, and order. By examining these episodes, valuable insights can be gained for the construction of a social epistemology. Secondly, the historical depictions should be evaluated through the lens of autonomous social science (prophetic social science) to determine their suitability as primary sources of constructing social epistemology. This evaluation will help ascertain whether these depictions can serve as reliable data for the development of alternative social science. Lastly, it is imperative to foster critical awareness among scholars and intellectuals, encouraging their active participation in the independent development of social science. This entails breaking away from the dominance of grand theories originating from Western social science and instead focusing on the practical application of local knowledge to effectively address societal problems in the region. By doing so, alternative social science has the potential to inspire and drive meaningful social change. Yet, in order to effectively energize the transformation aimed by alternative social science, it is crucial to move beyond mere theoretical patterns and focus on the practical realization of its contextualization. This requires a clearer depiction of how alternative social science can be implemented in the context of Malay-Indonesian society. By providing a practical project, it becomes possible to activate social engineering that is guided by a deep understanding of the evolution of autonomous social science. Such a depiction can be unearthed from a specific episode of the region's historical epoch, one that has been widely recognized and embraced by society. In the modern era, an example of this is the practice of Islam Nusantara by Muslims in the region. Islam Nusantara represents a collection of arguments and practices of Islam adopted by Muslims in the Nusantara peninsula, which have shaped patterns of social relations and cohesiveness through communicative actions throughout a specific historical epoch in the region. It is an expression of Islam that has profoundly influenced the development of society and culture. In essence, Islam Nusantara serves as a crucial element in the production of Nusantara social science. By acknowledging and integrating the principles of Islam Nusantara, alternative social science can be enriched and take into account the unique socio-cultural context of the Malay-Indonesian society. This integration fosters a deeper understanding of the society and allows for the formulation of more relevant and effective strategies for social change and transformation. In this context, "Islam" is not viewed as a transcendental religion disconnected from the social history of a particular society. Instead, it is understood as a historical and empirical set of arguments and practices that have influenced the societal dimensions of the region. From an anthropological perspective, Islam Nusantara has emerged as a result of the historical reality and socio-intellectual dialectics between Muslims and the scripture (revelation). Muslims in this region have studied and embraced Prophetic values from the scripture and other Islamic sources such as hadith and traditions. When the social life of Muslims is guided and developed based on the principles of Islam Nusantara, and when this social development fosters social order and contributes to the emergence of Nusantara civilization, it can be argued that the Muslim society in the region is practicing the Islamic values of prophetism. Through the lens of prophetic social science, it is recognized that a society practicing the values of prophetism aims to strengthen social civility within its own social fabric. This strengthening project, inspired by prophetic consciousness and Islamic teachings (Islam Nusantara), can only be achieved through concrete social activism. In their pursuit of this activism, Nusantara Muslims actively participate in the agenda of humanization and the fight against dehumanization in various aspects of life, including social, political, and economic domains. Moreover, the practices of Islam Nusantara exhibit distinct characteristics compared to other practices found in Muslim-majority countries, particularly in the Middle East. While Islam Nusantara generally embraces principles of peace, tolerance, pluralism, multiculturalism, and harmony, the realities of Islam and Muslims in Middle Eastern countries are often marked by tensions, conflicts, and inhumane conditions. Therefore, the practices of Islam Nusantara can be viewed as a valuable source of knowledge production. The Nusantara society and its societal patterns offer insights that can autonomously contribute to the production of a unique social epistemology. By recognizing and studying the dynamics of Nusantara society, alternative social science can develop a special body of knowledge that is rooted in the local context and can effectively address the challenges and complexities of the Malay-Indonesian society. Indeed, despite the potential for contextualizing the notion of alternative social science and drawing from the tradition of Islam Nusantara as a primary scientific ingredient, there is a lack of awareness among scholars in the region regarding the essentials of producing alternative social epistemology and, consequently, the development of alternative social science. To address this, there is a need to launch a campaign to promote the notion and garner broader support from the public. This campaign should also aim to encourage local scholars and intellectuals to actively participate in the constructive process of producing social science that aligns with the principles of alternative social epistemology. Furthermore, it is crucial to foster critical and intellectual discussions surrounding the notion of alternative social science. These discussions should delve into paradigmatic issues, scientific approaches, methodologies for scientific analysis, considerations for data collection, classification, and selection, as well as conceptual and theoretical debates. By engaging in these discussions, scholars and intellectuals can deepen their understanding of the notion and refine their approaches to producing alternative social science. Through comprehensive and rigorous intellectual engagement, a stronger foundation can be built for the development of alternative social science in the Malay-Indonesian context. This will not only enhance the scholarly discourse but also contribute to the formulation of effective strategies for addressing societal challenges and fostering meaningful social changes. Indeed, the trajectory of engaging social science and tradition, drawing intellectual insights from Muslim scholars in the Malay-Indonesian world, and attempting to contextualize alternative social science (autonomous and prophetic social science) is open and accessible for any social scientist who is committed and dedicated to addressing these issues. The concepts of autonomous and prophetic social science provide valuable inspiration for those seeking to develop "alternative" approaches to social science that break free from the domination of Western social science and traditions. While the realization of this project requires further sharing, elaboration, and contestation, it is encouraging to recognize that the groundwork has been laid. Scholars and intellectuals who are passionate about this endeavor can build upon the existing concepts and frameworks, contribute their own insights, and engage in rigorous intellectual discussions to advance the development of alternative social science in the region. By embracing the principles of autonomous and prophetic social science, social scientists can explore new avenues of knowledge production, challenge existing paradigms, and offer fresh perspectives on societal issues. This intellectual journey holds the potential to foster a more inclusive, contextually relevant, and transformative social science that addresses the specific needs and challenges of the Malay-Indonesian context. --- Concluding Remarks Autonomous social science and prophetic social science hold great significance in the context of Malay-Indonesian scholarship as they embody the engagement between social science and tradition. These concepts emerged as responses to the dominance of Western social science, which was historically shaped by colonialism and Orientalism. They aim to challenge this domination and address the complex issues that have arisen within scholarship, such as globalism, nativism, Islamism, chauvinism, academism, objectivism, positivism, and Eurocentrism. In this context, two fundamental problems exist, the general social problems of society and the problems within scholarship itself. It becomes clear that these problems cannot be effectively

This article aims to examine the interaction between social science and tradition in Malay-Indonesian scholarship, with a particular focus on the perspectives of Muslim intellectuals. This interaction is represented by two approaches: autonomous social science and prophetic social science. The article argues that both approaches have their distinct inclinations in perceiving and formulating theoretical foundations. These inclinations are shaped by their unique qualities, alternative values, intellectual stances, and thinking styles. Moreover, during the process of development, they are influenced by various factors that give them their distinct characteristics, such as autonomy, a problem-based approach, and a strong commitment to rethinking society and driving change and transformation. The social concept of Islam Nusantara represents this development. It is a collection of arguments and practices of Islam adopted by Muslims in Nusantara, which have shaped patterns of social relations and cohesiveness through communicative actions throughout a specific historical epoch in the region. The article also highlights two essential factors that significantly shape these approaches: the dominant schools of thought in social science and the peculiar Malay-Indonesian traditions. While Western social sciences have played a central role in understanding certain concepts, Malay-Indonesian traditions have contributed significantly to shaping the paradigm of local scholars.

to advance the development of alternative social science in the region. By embracing the principles of autonomous and prophetic social science, social scientists can explore new avenues of knowledge production, challenge existing paradigms, and offer fresh perspectives on societal issues. This intellectual journey holds the potential to foster a more inclusive, contextually relevant, and transformative social science that addresses the specific needs and challenges of the Malay-Indonesian context. --- Concluding Remarks Autonomous social science and prophetic social science hold great significance in the context of Malay-Indonesian scholarship as they embody the engagement between social science and tradition. These concepts emerged as responses to the dominance of Western social science, which was historically shaped by colonialism and Orientalism. They aim to challenge this domination and address the complex issues that have arisen within scholarship, such as globalism, nativism, Islamism, chauvinism, academism, objectivism, positivism, and Eurocentrism. In this context, two fundamental problems exist, the general social problems of society and the problems within scholarship itself. It becomes clear that these problems cannot be effectively solved without critical intellectual reflection. The concepts of autonomous social science and prophetic social science offer three key characteristics that enable the questioning, problematization, and potential resolution of these issues: autonomy, a problem-based approach, and a strong commitment to rethinking society and driving change and transformation. These characteristics reflect the inclination of alternative social science rooted in the Malay-Indonesian tradition. This invention holds significant potential to inspire social scientists worldwide to establish independent, autonomous, and cosmopolitan intellectual perspectives. However, the implementation of these concepts and inventions faces challenges. The successful realization of these ideas as a scientific project relies on critical awareness, creative initiatives, and strong efforts of local scholars and intellectuals. The way in which they address these challenges remains an open question, and their actions will determine the extent to which the contextualization of these concepts can be effectively carried out.

This article aims to examine the interaction between social science and tradition in Malay-Indonesian scholarship, with a particular focus on the perspectives of Muslim intellectuals. This interaction is represented by two approaches: autonomous social science and prophetic social science. The article argues that both approaches have their distinct inclinations in perceiving and formulating theoretical foundations. These inclinations are shaped by their unique qualities, alternative values, intellectual stances, and thinking styles. Moreover, during the process of development, they are influenced by various factors that give them their distinct characteristics, such as autonomy, a problem-based approach, and a strong commitment to rethinking society and driving change and transformation. The social concept of Islam Nusantara represents this development. It is a collection of arguments and practices of Islam adopted by Muslims in Nusantara, which have shaped patterns of social relations and cohesiveness through communicative actions throughout a specific historical epoch in the region. The article also highlights two essential factors that significantly shape these approaches: the dominant schools of thought in social science and the peculiar Malay-Indonesian traditions. While Western social sciences have played a central role in understanding certain concepts, Malay-Indonesian traditions have contributed significantly to shaping the paradigm of local scholars.

The growing cancer mortality, poor access to palliative care, treatment hesitancy, poor survivorship, and limited oncology healthcare infrastructure have been investigated in Sub-Saharan Africa. In Nigeria, the orthodox treatment and management of cancer is still greatly influenced by superstitious belief and spiritual connotations. Yet, the sociospiritual and culturally sensitive meanings attached to cancer and treatment are still lacking in the literature. This study explored the socio-spiritual strategies of older adults living with cancer in Nigeria. The study adopted a qualitative approach and data were collected through a semi-structured interview with 15 older adults from the Oncology unit, University of Nigeria Teaching Hospital (UNTH) Enugu State, Nigeria. The data were managed and inductively coded using NVivo 12 and analyzed thematically. The findings of our study show that cancer is traditionally conceptualized as the aftermath of a disconnection with supernatural/ancestral forces. By implication, such misconceptions affect the level of dispositions and steps taken toward diagnosis and treatment. To this end, our study argues that blending spiritual needs, social support, and culturally sensitive cancer education could improve the quality of life and coping capacity of older adults living with cancer. The study recommends the infusion of sociospiritual interventions to cancer treatment for older adults in Nigeria. Abstract citation ID: igad104.3338 --- EXPLORING THE STRESS-RELATED GROWTH AMONG ELDERLY KOREAN IMMIGRANTS IN THE CONTEXT OF COVID-19 PANDEMIC. Jooah Lee 1, Yongseop Kim 2, Hyejin Park 3, and Junhyoung Kim 3, 1. Henry M. Gunn High School, Palo Alto, California, United States, 2. Indiana University, Bloomington, Indiana, United States, 3. Texas A&M University, College Station, Texas, United States While some studies investigated stress-related growth (SRG) during the COVID-19 pandemic, little research has been pursued to explore positive psychological changes associated with COVID-19 among older Asian immigrants. Thus, we aimed to find positive changes resulting from the COVID-19 pandemic based on the SRG framework. Using a purposeful criterion sampling strategy, semi-structured in-depth interviews were conducted with 11 participants. Content mapping and content mining questions were used. Data were qualitatively analyzed using the constant comparative method. Challenges and stressors associated with the pandemic such as fears of physical or verbal attack, COVID-19 infection, strict COVID-19 regulations and mandates, and limited opportunities to interact with others were identified based on the statements of participants. We identified three salient themes that were characterized as components of SRG: (a) increasing leisure-time physical activity participation, (b) developing a closer relationship with others, and (c) improving resilience. The findings of the present study present qualitative evidence that older Korean immigrants experienced positive changes associated with the pandemic such as being physically active, developing closer relationships, and improving resilience. These changes are associated with the main elements of SRG, and it appears that the pandemic did lead to older Korean immigrants experiencing SRG.

neighborhood safety trajectories that may relate to healthy aging. This qualitative study used data from a sample of participants in adulthood and older adulthood (aged 35-89, Mage= 64 years) from two predominantly black, low-income neighborhoods that have had different histories with regard to disinvestment and systemic racist policies. We used 60 in-depth interviews to explore perceptions around neighborhood safety trajectories and their perceived impact on aging outcomes. The results highlight several dimensions of perceived changes in neighborhood safety over time and how these are perceived to relate to aging outcomes, e.g., psychological stress. The findings also show a few differences between the two neighborhoods concerning policing strategies, individual behaviors, and resource availability, all of which may contribute to perceptions of safety. Implications of this work highlight the importance of developing culturally relevant interventions that target upstream social determinants of healthy aging and health disparities.

Introduction Rural communities within Wyoming [USA] are becoming increasingly ethnically and racially diverse, and Wyoming's primarily White rural school and district administrators find the increasing diversity challenging. The sources of this challenge include the presence of cultural homogeneity; poorly received culturally proficient professional development by those with deficit and racist mindsets; and school, district, and community cultures that are exclusionary and isolated spaces rich in the ignorance of issues surrounding social class, race, and culture that are common across other states that are experiencing an increase in rural diversity within their student populations (Howley et al., 2014). Duncan and Stock (2010), talking about Wyoming schools and principals, stated: Schools are seeing increased diversity in their student populations and now cater to transient students, immigrants, and students from widely differing socioeconomic classes, which creates a vastly different context "for teachers and principals who were socialized in homogenous settings with students of the same race, ethnicity, socioeconomic status, and language" (Crow, 2006, p. 315). Principals must be sensitive to cultural differences and the various instructional approaches required. Not all principals feel equipped to lead in such complex, changing environments. (p. 296) Additionally, in a study conducted in the Rocky Mountains, Hesbol et al. (2020) found a lack of pre-and in-service training and education in the areas of diversity, social justice, and race for school and district administrators; this constrains the ability of school and district administrators to work with issues of diversity and social justice across the region. Further, not all rural administrators recognize, welcome, or come prepared for the "browning" of their communities and increasing diversity within their states, particularly with regards to English language learners (Reyes, 2006). The term "browning" comes from Dr. James Johnson, an African-American demographer from the University of North Carolina (USA) who discusses both the Browning and Graying of America as two powerful demographic processes driving dramatic change in American society (Columbus Metropolitan Club, 2017). Specifically, the Browning of America is the rapid increase in immigration-driven population change coming from Latin America, Asia, and the Middle East that outpaces the growth of the White population due to increases in marriage and birth rates. See Table 1 for data on the Browning of Wyoming communities (93% of Wyoming's population growth comes from Non-white/ Hispanic) and Table 2 for the Browning of Wyoming schools (White less than 2% vs. Hispanic over 18%, for example). Wyoming appears to be one of the states that does not consistently recognize, welcome, or engage the increasingly diverse population within the state, based on the statements and actions of Wyoming principals and superintendents (Castaneda et al., 2006). This qualitative study was designed to investigate Wyoming educational administrators' views on diversity and the intersection between race and rurality among central figures in the success of rural districts and schools-district superintendents and school principals. Therefore, the study's research question focuses on how a new, emergent framework called Cowboy Epistemology in Wyoming affects school principals' and district superintendents' efforts to promote awareness and change around diversity and social justice issues. --- The Wyoming context Wyoming is a top-ten state in K-12 education, ranked number seven nationally in 2018 [scoring highest among all the western states, first nationally in school finance, eighteenth nationally in student achievement, and ninth nationally in addressing the achievement gap between low-income students and their more affluent peers] (Quality Counts, 2018). As a state, Wyoming increased its diverse population from 2010 to 2018 according to U.S. Census data as shown in (Han et al., 2015), and operating in an ultra-conservative and individualistic political culture (Fowler, 2013) that avoids conversations regarding diversity and racism. Further, Wyoming has been slow, and at times resistant, to embrace diversity and culturally responsive teaching stances (Han et al., 2015) despite Educational Leadership Constituent Council (ELCC) educational leadership standards adopted at preservice instruction and state licensing certification to support culturally responsive instruction at the school and district levels (NPBEA, 2011a(NPBEA,, 2011b)). Rural principalships in Wyoming, and many Western rural locations, are often filled by local educators supported by a "grow your own" mentality that aligns with academic and cultural homophily practices in which rural "hiring committees and superintendents often select applicants that match their own characteristics and experiences rather than optimum criteria" (Pendola & Fuller, 2018, p. 3). In doing so, superintendents respond to community pressures for familiar faces and families that "fit in" to the local social community context, perpetuating local cultural and racial practices and stereotypes. Additionally, Pendola and Fuller (2018) noted that rural principals are leaders in the local community and must engage the community in a manner consistent with the community's values and expectations to maintain trust, connection, and effectiveness with community members. Woodson (2015) noted that cultural homophily is an attraction between likeminded people based on shared cultural traits and, as a result, helps individuals interact and feel at ease with one another. Congruent with the demographic divide positionality described by Han et al. (2015) and the effects of cultural homophily upon the rural principalship noted by Pendola and Fuller (2018) is the issue of Whiteness. The positionality of Whiteness in this manuscript and Wyoming is one of White supremacy and White superiority. White supremacy, in this case, is not associated with hate groups such as the Ku Klux Klan but is best defined by Ansley (1997) as: A political, economic, and cultural system in which whites overwhelmingly control • An over-reliance by political and business leaders on the fossil fuel industries of oil and coal despite a strong presence and potential for future growth of wind energy and other alternatives, and evidence of climate change for educational and other governmental funding. • A rural and isolated way of life that is characterized by ranching and agricultural jobs and lifestyles. • A territory and state settled as a result of Manifest Destiny primarily by White Anglo-Saxon settlers who believed they were anointed as a "special" race and superior to those they encountered (Native Americans and Hispanics) in order to spread the Christian faith and democracy and to do so by force if needed. School and district administrators are positioned within the emergent Cowboy Epistemological reality to varying degrees based upon the following: The key to understanding cultural proficiency begins with recognizing of different worldviews. Differing cultural worldviews become problematic when one's worldview causes one to judge a culture as "inferior" or "less than" rather than "different." Cultural worldviews range from seeking to eliminate other cultures to seeking to interact with other cultures in a way that builds on the best of both worlds. This range of worldviews is reflected in the points along the cultural proficient continuum. (p. 57) The cultural proficiency continuum gives leaders and stakeholders a working vocabulary to discuss and explore unhealthy and healthy values and behaviours of individuals and policies and practices of organizations as well as a way to assess the limitations of current practices and behaviours along with a method to make recommendations for future growth and development (Lindsey et al., 2009). The cultural proficiency continuum ranges from the negative perspectives of cultural destructiveness, cultural incapacity, and cultural blindness to the positive perspectives of cultural precompetence, cultural competence, and cultural proficiency. Thus, with cultural proficiency, school or district administrators can "either reinforce existing structures or promote change through dialogue and collaboration" (Lindsey et al., 2009, p. 54) within the local variations of Cowboy Epistemology. --- Methodology The following research question guided this study: 2012) noted that, in the twenty-first century, studies with 20% response rates can be just as accurate as studies with 60% response rates and provide as consistent results. Demographically, 20 (42%) superintendents and 62 (20%) principals statewide responded. There were 59 (72%) male and 23 (28%) female respondents. Regarding district size, according to the number of students, 39 (47%) of the sample indicated "Less than 750," 14 (17%) mentioned "751 to 1,500," and 29 (35%) said "1,501 or more" students. --- Data analysis A qualitative methodological approach was used to analyze the open-ended responses (Corbin & Strauss, 2008). Initially, the researcher used the margins to capture coding (Salda<unk>a & Omasta, 2016, p. 48) based on the conceptual framework. Then open coding was used to categorize data. Based on this analysis, similar codes were grouped into themes that led to establishing "a structure of thematic ideas in relation to it" (Gibbs, 2009, p.60). To ensure the research was conducted rigorously and methodically "to yield meaningful and useful results" (Nowell et al., 2017, p. 1), the researcher used several validation strategies. The techniques included describing researcher bias, searching for rival evidence, providing detailed descriptions, and maintaining an audit trail (Creswell & Poth, 2016, pp. 259-263). --- Positionality statement The author is a former school and district administrator from a large urban school district rich in --- Findings The findings of this study were centered around the research question and reported by theme. --- Research question The research question was: how does a new, emergent framework called Cowboy Epistemology in Wyoming affect school principals' and district superintendents' efforts to promote awareness and change around diversity and social justice issues? The themes that emerged through the categorization of codes (Gibbs, 2009) were in the following areas (1) diversity, (2) political, (3) instructional, and (4) community/socioeconomic. The author has chosen to provide samples of Wyoming administrator quotes/responses to illuminate each theme, provide context, and uncover the intersection between issues of race (diversity and social justice) and rurality. --- Diversity The area of diversity was a major theme that emerged from coding. There is not a lot of racial, political, or religious diversity in our community. This is both a blessing and a curse for grown-ups and students. We get very comfortable with sameness, and this can, at times, breed misunderstanding of others. That being said, I think our students and staff, as well as the community as a whole, are very tolerant of others who look different. --- Political A second central theme that emerged from coding was political in nature. The majority of responses by Wyoming educational leaders in this area were located in "the first three points along the [culturally proficient] continuum [which] are comprised of unhealthy values, behaviours, policies, and practices that emerge from barriers to cultural proficiency" (Lindsey et al., 2009, p. 6). A Wyoming principal provided an example of cultural destructiveness, who stated, "I feel that the national emphasis on it [diversity] could well spill over into our state. Taking valuable time and resources away from the classroom to pursue non-academic hot-button issues." Multiple administrators noted examples of cultural incapacity and the resulting conflict when discussing diversity as part of their micropolitical role. For example, "helping board, community, and some staff understand the needs of diverse students gives conflicting priorities over the state and nation." A Wyoming principal echoed this, citing, "[The] overarching reach of political party ideology that drowns out individual constituency discussion." A second political theme concerned resource allocation. Specifically, a Wyoming principal stated, "In the area of social justice, I feel budget cuts have impacted students with diverse needs." This was echoed by a Wyoming principal who noted: After working in two school districts, I feel the accountability for students considered "others" are not addressed at the level that they could be-especially students that are English Learners, low-SES, American Indian/Alaska Native, etc. There is a lack of exposure to diversity for educators and legislators-some are isolated and do not know what challenges educators face. Therefore, Wyoming school and district administrators must advocate and speak up for those most in need of help and resources in this area. --- Instruction The area of instruction emerged as a significant theme during coding. The majority of responses by Wyoming educational leaders in this area were consistent with cultural precompetence, as noted on Lindsey et al.'s (2009) culturally proficient leadership continuum. Specifically, Wyoming administrators were concerned with Wyoming students' lack of exposure to diverse students, experiences, cultures, and ideas within the instruction space. A Wyoming principal stated: I see the major issue being the ability of our students to work with a diverse group. Given the lack of diversity in many of our communities, our students simply don't have the exposure that would aid in this process. Finding a way to support the soft skills our students will need to work in a team with people from diverse backgrounds is important. A second Wyoming principal echoed, "I think the time invested in teaching/standards/assessment and being in such a remote, vanilla location drowns out working on issues of diversity and social justice." Another Wyoming principal asserted: I think the challenge is to open students' minds to the outside world when we live in such a rural and generally homogeneous state. I don't like the term social justice, but I do think we need to constantly work to expose students to different ideas and the world that exists outside Wyoming borders. What is concerning about instruction and cultural precompetence is that it is inconsistently progressing, as noted by the following comments. A Wyoming principal noted: I really don't see any major issues. However, I would comment that we have a lack of diversity in Wyoming schools. In our area, students are either White or Hispanic. We love having foreign exchange students in our school to expose students to different cultures. A different Wyoming principal commented, "I am an elementary principal. Our focus is reading, writing, math, and science. I do not perceive diversity or social justice as a major issue for elementary education." An outlier in this area was a Wyoming principal who acknowledged the shortcomings in this space: "At this point, the issue is awareness for students in terms of diversity and social justice. Many of our schools do not educate students in these areas so that they have a clear sense of acceptance and tolerance." The lack of instruction in the areas of diversity and social justice perpetuates the cycle of Whiteness, cultural homophily, and demographic divide and reinforces the mindset of Cowboy Epistemology within the state of Wyoming. --- Community/Socio-economic The final area that emerged as a major theme during coding was community, particularly in socioeconomics and tolerance. The responses by Wyoming educational leaders in socio-economics sat across the culturally proficient leadership continuum. For example, in terms of cultural proficiency, a Wyoming principal remarked, "I feel lucky to be a part of a district that has focused on social justice, especially in our dual immersion school. I feel competent at blending social justice and appreciation for diversity into our daily academics." Two Wyoming principals in the space of cultural precompetence noted, "Wyoming lacks racial diversity, but it does have ample economic diversity," and "Wyoming is slow to change and is not bringing in enough people to diversify our economy. This leads to a lack of diversity which is harmful to our students not only academically but socially." However, in regard to cultural incapacity and cultural destructiveness, Wyoming administrators acknowledged the following troubling issues in community and tolerance. A Wyoming principal stated: The lack of diversity in our schools and the ultraconservative nature of some of our adult patrons is worrisome. I worry that there is a growing lack of tolerance in our state and that that may result in the marginalizing of some sectors of our population. A Wyoming superintendent echoed, "I see the biggest issue is people's lack of acceptance of people with differences. Civility is eroding, not just in schools, but in society as a whole." Finally, a Wyoming superintendent noted, "At this point, the issue is awareness for students in terms of diversity and social justice. Many of our schools do not educate students in these areas so that they have a clear sense of acceptance tolerance." --- Discussion The discussion is centered around the research question and reported by theme. --- Diversity The major finding in diversity was the non-recognition of diversity as an issue despite the double and triple-digit growth of students of color (see Table 2) and educational leadership standards promoting diversity [ELCC building and district level standards] within the state of Wyoming. This cultural blindness by Wyoming educational administrators reinforces the existing structures of demographic divide, cultural homophily, and Whiteness in Wyoming and are evidence that the foundational frameworks of Cowboy Epistemology have taken root in the offices of school and district leaders in Wyoming. This finding suggests that much work needs to be done. A second political finding compounds the initial political finding that resources, accountability, and expectations differ for diverse Wyoming students. While Wyoming administrators felt that budget cuts, accountability, and expectations were not at the levels they should be, they also did not advocate for resources, budgets, accountability, and expectations at or above the levels they should be for diverse students. Additionally, Wyoming administrators indicated that those in political power, such as legislators, were not diverse or educated about the needs of diverse students within the state, which supports the Holmes (2021) finding that Wyoming superintendents preferred to work with internal stakeholders before engaging with external stakeholders such as legislators and political officials. --- Instruction As instructional leaders, educational administrators in Wyoming find themselves aligned with what Ladson-Billings (2014) identified as a fixed mindset in the area of culturally relevant instruction due to a lack of depth and relevance in multicultural classroom instruction and avoidance of instruction in the areas of diversity and high student interest such as politics, global warming, and "real" American history. The comment by the Wyoming elementary principal that there is no room for diversity and multicultural instruction in the elementary curriculum is wrong and uninformed, and when coupled with the comments regarding the lack of student exposure to issues of diversity and social justice, creates students with the potential to perpetuate racist, bigoted, hate-filled attitudes in settings beyond the K-12 classroom. An example of this is reflected by multiple students who, on the 2020 University of Wyoming Climate Survey, identified their gender as Apache Attack Helicopters (Swinton & Van Brunt, 2018). Additionally, due to the lack of diversity within --- Community While Wyoming has communities that are focused on and engaged in the work of diversity, i.e., dual-language schools, these communities are rare. The majority of communities in Wyoming experience a lack of diversity and a lack of tolerance for those who are different from them in ways that go beyond race, color, and gender. There is a general lack of acceptance of differences, an eroding of civility within communities across the state, and an absence of discourse on a wide variety of opposing views. In large part, Wyoming administrators choose to accept these community values, mores, and norms of behaviour, and channel their energies and efforts into other areas. --- Implications, limitations, and next steps The diversity and social justice implications for educational administrators in Wyoming are significant in these times of multiple pandemics. In order to move from blind and adverse stances on diversity, politics, instruction, and community, educational administrators across the state of Wyoming must adopt more competent, asset-oriented, and culturally aware/proficient leadership positions that stop "ignoring issues of social class, race, and culture" (Howley et al., 2014, p, 627). This requires educational administrators to engage in culturally relevant and sustaining leadership professional development and education, lead their communities in critical conversations, and develop long-term advocacy-centered relationships with critical internal and external stakeholders. In terms of limitations, this study is confined to the boundaries of the state of Wyoming and the principals and superintendents within; thus, while there is a great deal to be learned from this study, the study has limited generalizability. Additionally, the study is limited in data collection as the participants completed an online questionnaire focused on a wide variety of topics in which diversity and social justice were only one question out of eleven. Further, as the study is based on participants' self-reporting, the data must be taken at face value. Finally, the population surveyed had been subjected to frequent requests for research participation and had indicated through feedback via professional organization leadership that research fatigue had set in upon many in the population (membership); therefore, the researcher, in order to improve the response rate, asked for limited demographic information to provide for greater participant anonymity and confidence when asking potentially controversial subjects. Just as Cowboy Joe is the universal symbol for the state of Wyoming, so is the motto of --- Data Cowboy epistemology: Rural school and district leadership for diversity and social justice --- Author William (Toby) Holmes EdD is an asistant professor of educational leadership in the Department of Teacher Education and Educational Leadership and Policy at the University of New Mexico. He is a former principal of a five-star elementary school in Las Vegas, Nevada, USA. Dr. Holmes is a member of the Choctaw Nation of Oklahoma, primarily focused on research and teaching in leadership communication, motivating language theory, culturally sustaining instructional leadership, and issues focused on rural leadership. Email: [email protected] ORCID: 0000-0001-6846-722X

This qualitative study focuses on the intersectionality of race and rurality by looking at the responses of Wyoming principals and superintendents to the issues of diversity and social justice within Wyoming. The responses are presented and analyzed through a new framework called Cowboy Epistemology, and the Cultural Competency Continuum (Lindsey et al., 2009). It appears that despite double-digit increases in diversity between 2010 and 2018, some Wyoming school and district administrators continue to demonstrate actions and practices congruent with the demographic divide, cultural homophily, and Whiteness along with cultural worldviews that suggest a failure to: (1) value diversity, (2) engage political organizations and individuals in a manner that advocates for the needs of diverse students, (3) implement multicultural instruction beyond superficial means, and (4) engage the community in tolerance for others who are different from the traditional White Wyoming ranching, conservative, materials extraction, isolationist way of life. While outliers and standards for social justice and diversity exist in Wyoming among and for administrators, more needs to be done to prepare and train administrators to engage in culturally proficient and sustaining instructional leadership so that administrators can serve all students, engage in community leadership, and resist the negative influences of Cowboy Epistemology, demographic divide, cultural homophily, and Whiteness. Chief among the more needs to be done for Wyoming administrators is the adoption of culturally responsive school and culturally sustaining instructional leadership practices and training on the culturally proficient continuum.

Introduction Emerging debate on the general level of vaccinations' acceptance is widespread across developed countries. A large movement against vaccination is becoming greater and greater through traditional media channels, social media platforms, and online communities [1,2]. As such distrust grows, countries are required to face a new social challenge, whose negative consequences may lead, progressively and dangerously, to an erosion of the general level of populations' protection and to the reappearance of "old fashioned" pandemic diseases [3]. In response to this trend, many initiatives have been promoted by decision makers to reinforce the obligation to vaccinate [4,5]. The World Health Organization (WHO) Strategic Advisory Group of Experts (SAGE) on immunization uses the term "vaccine hesitancy" to indicate the individual's propensity to refuse some or all vaccines as established by clinical guidelines and schedules [6,7]. Motivations underlying the scarce propensity to receive vaccination have relevant behavioral foundations. Prior studies documented that individual attributes, such as age, gender, ethnicity, health status, behaviors, and perceptions directly impact vaccination attitude in large communities [8,9], as well as specific categories of individuals, such as adolescents [7,10], adults and the elderly [11,12], health workers [13], and homosexuals [14]. Behavioral determinants of a scarce propensity to receive immunization are grounded in the general idea of invulnerability in individuals, who, at least for some infectious diseases, feel themselves healthy and strong enough to successfully solve their health problems without any immunization protection. Moreover, the high level of distrust toward the side effects of vaccines and the perceived economic interests of pharmaceutical companies in creating new medical needs are perceived as strong reasons for refusal. Other factors that likely affect vaccine hesitancy and influence behavior are (i) confidence, i.e., trust in the effectiveness and safety of vaccines as well as in the health care system and in the motivations of policy-makers involved; (ii) complacency that indicates the perception of individuals about the risks of vaccine-preventable diseases along with the relative need of vaccination; and (iii) convenience, which concerns all factors influencing the access to vaccines such as availability, affordability, accessibility, health literacy, and the appeal of immunization services [6]. Social and psychological circumstances may need to be considered to implement successful preventive strategies. Particularly, research on adolescents' behavior and development is crucial to describe the different reactions to immunization programs. Such relevance is motivated by twofold reasons, both connected to the extraordinary changes they experience during this phase. First, during the early adolescence (11-14 years old), they undergo many vaccinations against infectious diseases, such as the HPV immunization and recalls to tetanus, diphtheria, and pertussis. Second, adolescence represents the time span when consciousness, responsibility, and awareness toward healthy behaviors and health prevention emerge. Recent studies highlighted the role of parents' networks in addressing children vaccination choices, arguing that the influential power of networks is a greater predictor than any other individual characteristic [15]. In other words, parents play a key role in vaccine acceptance, but the immunization choices for their children are not taken individually. Rather, they are a product of an influential path where health care providers, family members, and friends act as sources of information, direction, and advice [15]. Although it is undeniable that parents and family concur to build a personal identity, particularly in a period when the young individual is not fully developed, children and their way of living have greatly changed over time. Such changes have affected, in turn, their degree of freedom in making decisions. This is even truer in the case of health choices, often displaced in an intimate and personal sphere, such as sexuality [16]. Schools are primarily involved as social structures where young individuals establish and develop network ties and, given their educational function, they are eligible settings for health promotion interventions [17]. Given the mutual dependency between social interactions and health behaviors, friendships are identified as an important element for adolescents' health [18]. Prior studies documented the importance of adolescents' network in the promotion of healthy choices, such as the propensity to physical activity [19] or the tendency to obesity [20]. Scholars found an association between the similarity (homophily) in certain health statuses and friendship networks, confirming the occurrence of contagion phenomena among young individuals [21][22][23][24]. These homophilous behaviors can be explained by the process of friendships' selection [23,24] or, conversely, by the sharing of habits among friends [25]. Other studies analyzing the role of friendship networks in adult obesity or smoking cessation [26,27] have pointed out the necessity to grow preventive strategies, addressing health behaviors, that consider the relational context of individuals [28]. However, few studies analyzed the role of social networks in health promotion topics, such as immunization behavior, and no exhaustive conclusions are available about vaccination hesitancy. According to the available knowledge, social networks allow a greater social exposure to medical information, which, in turn, raises people's perceptions of the benefits of immunization [29]. Moreover, imitation (contagion) effects among peers and community members may lead to similar decisions about immunization [30]. Although the role of social communities in shaping vaccination preferences and attitudes is already recognized, less is known about where such preferences and attitudes frequently emerge. The aim of this study is to understand factors affecting the vaccination behavior of adolescents. Specifically, we aim to explore the relationship between pupils' positive responsiveness to vaccination and their friendship networks maintained during school hours (school network) and outside school hours (after-school network). --- Materials and Methods --- Research Setting This study is part of the larger Italian research project "VacciniAmo le Scuole" aimed at encouraging the promotion of vaccination and primary prevention across the Italian educational system [31]. In Italy, routine immunization programs, including poliovirus and diphtheria-tetanus have been mandatory since the early 1960s. The Hepatitis B vaccine was added only in 1991, with the introduction of a universal vaccination for newborn and children up to 12 years. As the reduction and eradication of vaccine-preventable diseases have become a paramount priority in the public health agenda, other vaccinations, even if not mandatory, are free of charge for all the individuals in the country, strongly recommended by the Italian Ministry of Health and included in the health benefit packages (LEAs). According to the process of decentralization started in the late 1990s in Italy, regions are required to set up their vaccination strategies, which must be consistent with the National Vaccination Plan [32]. Public facilities are in charge of providing vaccination to children, which are usually administered within specific departments of the Local Health Authorities (LHAs). According to this organizational setting, pediatricians and General Practitioners (GPs) run counselling activity to their communities and families about vaccines [33]. Moreover, a prominent role in health promotion activity is played by schools. This justifies our study framing and the research setting chosen to run data collection [17]. --- Data Collection and Participants According to the project's objectives, a 90-min training program about vaccination was carried out in all four classrooms of a secondary school located in the central Italy. During the training, background information about vaccines was provided by public health physicians, and a role-playing activity allowed participants to experience practical situations about immunization. The training gave the opportunity to administer a structured paper questionnaire to the attending students. Before the training, pupils' parents received (at home) a similar structured paper questionnaire together with the request for the informed consent to allow the data collection. These questionnaires represent the primary data source for this study. Questionnaires submitted to students and their parents (one respondent per each pupil) slightly differed and were structured in three different sections. The first section collected demographic characteristics of the respondents and, only for pupils, specific questions about the school and after-school friendship networks. Particularly, students were asked to list up to 8 school mates-both within and outside their own class-with which they interact the most both during school and after-school hours. The second part of the questionnaire attempted to appreciate the level of knowledge and awareness about immunization. The third part of the questionnaire was aimed at collecting the attitude and feelings toward vaccinations in both the categories of respondents. In order to measure the pupils' responsiveness toward vaccination, data on students' vaccination profile were collected by the Local Health Authorities to which the school administratively belonged after one month since the delivery of the training program at the school. --- Data Analysis As our goal is understanding how friendship ties may affect the responsiveness toward vaccination, our data are relational [34]. Social network analysis (SNA) techniques were applied to understand the role of friendship networks in determining pupils' vaccination positive response behavior. SNA comprises a set of tools and methods to investigate the relational data [34], supporting researchers in understanding the role of social relations in individual or organizational choices. In particular, we applied multiple regression quadratic assignment procedures (MRQAPs), a statistical technique widely used in previous social network research to explore factors associated with the formation of network ties [35,36] as well as the adoption of similar behavior [37]. MRQAP tests aim to offer an alternative to traditional statistics for inference. MRQAPs are permutation tests for multiple linear regression model coefficients when data are organized in square matrices of relatedness among n objects and when the unit of analysis is the dyad. Such a data structure is typical in social network studies, where variables indicate some type of relationship between a given set of actors [38]. The MRQAP test, well-known as "double semi-partialing" or DSP, has been proved to show robustness under all multicollinearity conditions and thus solve the problem of the structural autocorrelation of network data [39]. One important implication of this methodology is that all covariates have to be transformed into pseudo-network data. In our case, all the covariates represented individual attributes and preferences of pupils as measures of differences/similarities among each pair (dyad). Practically, continuous covariates (i.e., parent's age) entered the models as absolute differences between pupil i (sender) and pupil j (receiver). In this case, the smaller the difference, the greater the similarity (homophily) in the dyad. On the contrary, categorical covariates (i.e., pupils' attending class) or binary covariates (i.e., gender) were transformed as binary variables describing the "match" between the two, assuming a value of "1" if both pupils in the dyad were in the same category or a value of "0" otherwise. In this paper, we used the logistic regression quadratic assignment procedure (LRQAP), a specification of the multiple regression quadratic assignment procedures family [35,36], which is best suited for binary networks and variables. Given the binary nature of our dependent variable, namely, the similarity in positive responsiveness, LRQAP was adopted to produce estimates of our regression models (in order to test the robustness of our findings, we conducted additional MRQAP tests and the results of this supplementary analysis, consistent with those present in this paper, are available from the authors upon request). All analyses were performed using the UCINET 6.598 software package [35]. The school and after-school friendship network data collected were described in one-mode sociomatrices aimed at appreciating the inter-personal ties among pupils. --- Dependent Variables "Positive Responsiveness". The dependent variable in our study describes the similarity in the responsiveness to the training program pupils have been involved in and that occur in each dyad. In particular, the variable describes the presence of a positive reaction in terms of new vaccinations observed in the month that followed the training intervention. This information was collected through the pupils' vaccination profile observed at the relevant Local Health Authority and was measured as a binary variable assuming the value of "1" when the pupil positively reacted to the training intervention and accepted being vaccinated in the month after and a value of "0" otherwise. Then, we built a squared matrix in which pupils are reported in rows and columns, and interception cells indicate the occurrence of similarity in their positive reaction to the training intervention ("1" if it was recorded, "0" otherwise). --- Explanatory Variables Network Measures and Social Mechanisms "School Friendship Network". In the questionnaire, pupils were asked to name the classmates with whom they talk the most and usually spend time with during school hours. Using this information, an adjacency matrix representing these interpersonal relations among children was built up. The matrix indicated both in rows and columns the pupils enrolled in the study and in the cells, the value of "1" when a friendship relation occurred, with a value of "0" otherwise. "After-School Friendship Network". Students enrolled in the study were asked to list the classmates with whom they talk the most and they usually spend time with after school hours. As in the case of the school friendship networks and according to the information collected, we built an adjacency binary matrix representing the friendship among children outside school. In the matrix, the pupils enrolled in the study were observed both in rows and columns, and the presence of their relation was reported using binary values in the cells. "Social Mechanisms." In order to investigate the social processes underlying the configuration of the two friendship networks observed, three different social mechanisms have been taken into consideration in our models. First, given the positive emotions generally involved in a friendship, we considered the reciprocity of pupils' relational ties. Particularly, reciprocity indicates the likelihood of individuals to reciprocate relations and thus to be mutual in the social exchange process. Second, we considered the networks' closure, which refers to the natural tendency of individuals to triangulate in their networks and, eventually, to connect and cooperate within small groups. Third, a measure of preferential attachment was used in our models. Specifically, the preferential attachment indicates the likelihood of actors to link to others depending on their level of popularity. In practical terms, the preferential attachment indicates the extent to which an actor is more likely to become progressively more popular as an effect of his/her centrality [40]. We specified and included these three social mechanisms for the two friendship networks tested in our models. --- Control Variables "Pupils' Ethnicity". Prior studies documented that some forms of homophilous behavior may be predicted by the homogeneity in the racial ethnicity [41]. Thus, we considered a binary variable indicating the pupils' racial ethnicity. This information allowed us to derive a similarity matrix assuming in both rows and columns the pupils and in the cells, the value of "1" whether pupils in the dyad were found to be similar in terms of this attribute and a value of "0" otherwise. "Pupils' Gender". Since we were interested in investigating any form of gender homophily in our sample, we considered in our models pupils' similarity in the gender attribute. Considering the related binary variable, we built up a matrix assuming the value of "1" whether the pair of actors in the dyad belonged to the same gender and a value of "0" otherwise. "Pupils' Attending Class". We considered in our models the class attended by pupils within the school. A categorical variable aimed at appreciating this attribution information allowed us to derive a squared matrix representing in both columns and rows the pupils and in the cells the value of "1" whether they belonged to the same class, and a value of "0" otherwise. "Pupils' Perceived Relevance of Vaccination". In the questionnaire, children were asked to declare their perception about the relevance of vaccines in the prevention of infectious diseases. This perception was operationalized by a 7-point Likert scale ranging from "1" (not relevant at all) to "7" (very relevant). Because of the categorical nature of this variable, we built up a matrix describing the absolute difference in the perceived relevance between the pair of actors in each dyad. "Parents' Perceived Relevance of Vaccination". Prior studies documented that parents' individual characteristics may predict the responsiveness to vaccination [7]. As we collected information from parents about their perceived relevance of immunization in the prevention of infectious diseases, we included such information in our models by a matrix describing the absolute difference in the perceived relevance between the pair of parents in each dyad. "Parent's Age". Similarly, we chose to consider parents' similarity in terms of age as a potential predictor of the similarity in the responsiveness to vaccination. Since this variable is continuous, we calculated the absolute difference in age for responding parents' pupils in the dyad. "Pupils' Medical Visits". Similar behaviors in the access to the NHS system may predict positive responsiveness to the vaccination training program. Under this hypothesis, we considered in our models the average number of medical visits that pupils underwent every year, as self-reported by their respondent parent. In order to appreciate this variable, we described the degree of similarity in each dyad using a matrix, reporting both in rows and columns the name of pupils and in the cells the absolute difference in the number of visits occurring in the dyad. --- Results Out of 83 students attending the different classes, 49 pupils and their parents agreed to participate in the present study. All data pertaining to these students were collected and considered in our analysis (response rate was 60%). Table 1 shows the demographics of sampled pupils. We represented graphically the 49 pupils enrolled in the models by using the Netdraw software package [35]. Figure 1 describe the sociograms of the pupils' school friendship network and after-school friendship network, both based on different attributes included in the models tested. In particular, the size of each node (pupil) indicates the perceived relevance of immunization in the prevention of infectious diseases as declared by pupils and its color indicates the pupil's gender. Red nodes represent females, while blue nodes indicate males. Table 2 shows descriptive statistics of the variables enrolled in our study as well as the quadratic assignment procedure (QAP) correlation results. "School friendship network" is positively and significantly correlated with social mechanisms underlying social ties, such as the "Reciprocity" (r = 0.607, p <unk> 0.000(0)1) and the Closure (r = 0.633, p <unk> 0.000(0)1). Similarly, "Closure" and "Reciprocity" during school hours are moderately correlated with each other (r = 0.527, p <unk> 0.000(0)1). "After-school friendship network" is, in turn, associated with its own social mechanisms, such as "Reciprocity" (r = 0.586, p = 0.000(0)1). "School friendship network" is associated positively and significantly with reciprocal ties occurring after school hours (r = 0.462, p <unk> 0.000(0)1). In addition, "after-school friendship network" is associated with the "school friendship network" (r = 0.473, p <unk> 0.000(0)1) and the reciprocal ties during school hours (r = 0.462, p <unk> 0.000(0)1). Similarly, the two measures of "Reciprocity" (both during and after school hours) are associated positively and significantly (r = 0.473, p <unk> 0.000(0)1). The social mechanisms of "Preferential Attachment" both during and after school are moderately associated (r = 0.563, p <unk> 0.000(0)1). The similarity in the attendance of pupils' class is associated with "School friendship network" (r = 0.472, p <unk> 0.000(0)1), "Reciprocity" (r = 0.472, p <unk> 0.000(0)1), and "Closure" (r = 0.67 p <unk> 0.000(0)1). Three different models were tested in order to appreciate the statistical association among variables. Model 1 includes variables that represent social relationships that students maintain during school day as well as control variables. Model 2 includes after-school relationships and related social mechanisms as well as the other possible factors explaining vaccination behavior similarities Finally, Model 3 aims to test only the effect of control variables. Table 3 shows LRQAP findings. No statistical significance was found among the dependent variable and the school friendship networks and social mechanisms variables. Nevertheless, after-school friendship networks seem to be positively and significantly associated with the similarity in vaccination responsiveness (OR = 1.473), together with the reciprocity of pupils' ties (OR = 1.837) and preferential attachment (OR = 1.341). These findings suggest that pupils are more likely to share information and knowledge about health behaviors through social relationships maintained after school hours rather than through those established during the school day. According to the explanatory variables enrolled in our models, our results show that the similarity in racial ethnicity (OR = 6.130 in model 1; OR = 5.391 in model 2; OR = 5.983 in model 3) is positively and significantly associated with the similarity in vaccination behavior. Similarly, belonging to the same class seems to positively act as an antecedent of the similarity in vaccination responsiveness (OR = 1.836 in model 1; OR = 1.681 in model 2; OR = 1.820 in model 3). Finally, a negative and significant association was found between the similarity in pupils' gender and the dependent variable (OR = 0.865 in model 1; OR = 0.836 in model 2; OR = 0.860 in model 3). Thus, vaccination behaviors are more similar for those students having the same ethnicity or diverse gender as well as for those belonging to the same class. Finally, no statistically significant association was found between the dependent variable and the similarity in the perceived relevance of being vaccinated expressed both by the pupil and his or her parent, in the frequency of his or her access to the general practitioners, and in parents' age. --- Discussion In a historical period of economic, political, and social pressures, the debate on vaccination, as well as on the legitimate role of welfare states in the promotion and prevention of health [42], is extraordinarily hot. Behavioral mechanisms of trust regarding vaccination procedures underwent a gradual erosion. At the same time, a strong concern of the heterogeneity in the distribution of vaccination coverage is observed from country to country [43] and, in Italy, even from region to region. In a recent review of the vaccination regional plans (RPPs), Rosso et al. (2015) showed how such variability is not only in the outcomes, but also in the appraisal activity of regional plans, and the initiatives against unequal access to vaccine immunization are still far from an effective implementation [44]. To uncover this concern, it is necessary to consider the contextual factors affecting the policy and management decisions [45], such as laws and regulation and the level of integration between professionals and public authorities [46]. In this sense, previous studies have shown that financial incentives of doctors decisively encourage the immunization level in the reference population [46]. Finally, especially at this point, the importance of the media and the public in promoting a certain perception of vaccines may not be ignored [47]. Individuals and families are the preferential recipients of this communication flow, according to which they adjust their level of trust/mistrust toward vaccines and their consideration of the state as the principal health promoter [42]. The role of social networks in determining health choices has been robustly recognized in different context and communities. Assuming a micro-level perspective of analysis, our paper aims to disentangle the role of social networks, and, more specifically, of friendship ties, in the formation of vaccination behaviors among young adolescent attending secondary school in Italy. Our findings show that social ties between students are positively associated with their vaccination behaviors. More specifically, friendship ties occurring after school are more likely to create similarity in vaccination positive responsive behavior. Although schools are recognized as ideal social contexts for health promotion activities, our results seem to be consistent with prior studies suggesting that more effective interventions are needed to foster the interaction between schools and the surrounding environment [17,47] and to translate planning objectives into genuine benefits for the target population [47]. In Italy, we may consider that schools have experienced profound changes. Several reforms have characterized the organizational life of these settings. Public policy interventions addressed in schools have been numerous and turbulent. Although health promotion has been always considered as a priority in the educational system, it is necessary to take into account the numerous challenges school managers have had to face in the recent past and present, such as job insecurity of teachers, lack of financial resources, and infrastructure inadequacy. Some critical points have to be addressed to effectively manage prevention programs, which should regard capacity building activities for project management, commitment, and leadership, as well as the availability of human, financial, material, and temporal resources [47]. Moreover, the WHO SAGE group recommends integration between health and non-health services through joint initiatives in order to address complacency and convenience hesitancy factors [6]. Our results also document the association between such similarity in positive responsiveness and certain individual attributes of pupils, such as their racial ethnicity and the class attended. The level of homophily in these characteristics seems to affect the attitude toward vaccination. Moreover, a diversity tendency (heterophily) has been observed in regard to gender. In other words, pupils are more likely to be similar in terms of vaccination responsiveness when they are different in terms of gender. This last result surely deserves more attention in future research. Understanding the network ties that classmates establish and maintain after school is of interest to implement effective vaccination strategies at organizational, regional, and national levels. We believe that our findings are of interest from at least three different perspectives. First, our paper contributes to clarify the role of social networks in determining choices related to health issues and behaviors. Although several empirical studies demonstrated a strong association between the two, our study may represent a new original piece of evidence in the broader research topic of social networks and health care. Second, our results may have relevant methodological implication. The use of social network analysis may be of interest for those who are interested in exploring the mechanisms underlying the social structures of any community and setting. Third, our results may be helpful for managers at different level of decision making. The investigation of social structures could be useful to understand the spread of a disease within a population [48]. Moreover, policy makers and schools' managers could project targeted vaccination campaigns using social network data [49]. Notwithstanding the important potential contributions that the paper may provide researchers and decision makers with, the study is not without limitations. Firstly, we observe that the present study is cross-sectional precluding the possibility to infer causal relationships among variables. Future longitudinal studies will be helpful to shed more light on how social selection and influential processes play a role in this context, explaining more widely how social networks affect the propensity of pupils toward vaccinations. Secondly, our study may have not considered other variables that can potentially help to unfold pupils' attitude toward vaccination. Thirdly, the empirical context of this research is represented by one school in the Italian NHS. The limited research scale of our setting may significantly constrain the generalizability of our findings. Furthermore, we note that several factors that affected the intervention delivery when our survey was administered can help to explain the low response rate achieved. The most significant is probably the extensive spreading of the no-vax movement in the regional area where we conducted our research, which seemed to have influenced pupils enrolled for the study along with their parents, who significantly limited the consent given to their children for participation in the survey. Despite the above limitations, we believe that our analytical approach, once replicated at a larger scale in other health systems, could helpfully support policymakers' interventions aimed at understanding vaccination problems. --- Conclusions The present study informs policymakers about the importance that school and after-school friendship networks have for the vaccination behaviors of adolescents. Adolescents' attitudes and behaviors likely affect their health conditions and propensities toward the health system over time. Social networks created with classmates seem to be important in this vein because they help explain how individuals' opinions regarding certain educational and health interventions take shape in scholastic contexts. Only the joint effort of policymakers in these two fields, education and healthcare, will improve our ability to tackle individual vaccination hesitancy as well as to implement more coordinated and focused health interventions. Moreover, understanding the role of social networks in shaping health behaviors may lead to effective health promotion campaigns that could be delivered in institutional or recreational places taking into account social ties. We can consider the present research as a first attempt to incorporate social network effects in the study of vaccine hesitancy and its determinants. If replicated at a broader scale, this study could better inform policymakers to promote strategies aimed at tackling vaccination hesitancy.

Psychological and social characteristics of individuals are important determinants of their health choices and behaviors. Social networks represent "pipes" through which information and opinions circulate and spread out in the social circle surrounding individuals, influencing their propensity toward important health care interventions. This paper aims to explore the relationship between students' vaccination health choices and their social networks. We administered a questionnaire to students to collect data on individual students' demographics, knowledge, and attitudes about vaccinations, as well as their social networks. Forty-nine pupils belonging to 4 classrooms in an Italian secondary school were enrolled in the study. We applied a logistic regression quadratic assignment procedure (LR-QAP) by regressing students' positive responsive behavior similarity as a dependent variable. LRQAP findings indicate that students' vaccination behavior similarity is significantly associated with after-school social ties and related social mechanisms, suggesting that pupils are more likely to share information and knowledge about health behaviors through social relationships maintained after school hours rather than through those established during the school day. Moreover, we found that vaccination behaviors are more similar for those students having the same ethnicity as well as for those belonging to the same class. Our findings may help policymakers in implementing effective vaccination strategies.

Background Globally, drug and substance abuse (DSA) is a leading cause of years lived with disability (YLDs) and was responsible for 183.9 million disability-adjusted life years (DALYs) [1]. The widespread and growing DSA epidemic has caused a devastating impact on adolescents and young people with an increased likelihood of developing substance use disorders (SUDs) [2,3]. Substance use disorder is defined in this study as the recurrent use of alcohol and drugs that leads to clinically and functionally significant impairment [4]. Despite the high global burden of SUDs, there is a considerable treatment and support gap, particularly in low-and middle-income countries where between 76 and 85% of the people with the condition do not receive treatment [5]. The United Nations' 2030 Sustainable Development Goals are pushing for the prioritization of action in improving mental health and substance use treatment and prevention [5,6]. In addition to public and self-stigma due to negative stereotypes and prejudice [7], individuals who misuse alcohol and illicit substances are often marginalized and underserved by healthcare services due to providerrelated stigma caused by the erroneous belief that addiction is caused by weak character and poor choices [8]. Thus, patients suffering from SUDs might be considered a lesser priority or less likely to be treated with dignity and compassion when compared with patients with other conditions. Though scarce in practice, interventions such as using social support services tailor-made for individuals who have committed to recovering from SUDs have been proven effective [9]. Besides the lack of facilities providing rehabilitation services for these individuals, most illicit drug users are unaware that they need support and treatment because they consider drugs a solution and not a problem [10]. Even after acknowledging that they are overwhelmed by the problem, most of them do not seek treatment because they: (i) may not be ready to stop, (ii) do not have health insurance, (iii) fear the effects of withdrawal symptoms on their job performance or (iv) do not know where to get treatment [11]. Zimbabwe adopted the World Health Organization (WHO) Special Initiative for Mental Health: Universal Coverage for Mental Health, launched in 2019 [12]. The aim was to address the SUDs and increase access to equitable mental health services. Drug and substance use in Zimbabwe has increased significantly during the past decade and is now rated number 8 in the country's top ten risk factors accounting for DALYs [13]. Many young people take illicit alcohol, methamphetamine, codeinecontaining cough syrups, marijuana/cannabis, and sedating opiates [14]. These illicit drugs are either clandestinely manufactured within the local communities or smuggled from neighboring countries through porous entry points and well-organized syndicates at the official border posts [15]. The drugs are then sold to the young people in many places where unlawful transactions take place sometimes by women who masquerade as innocent fruit vendors [16]. The DSA problem in Zimbabwe can be attributed to increased poverty and dwindling economic opportunities that have left many young people idle and depressed, only to find solace in illicit substance use. Similar to global reports, DSA is a significant challenge among Zimbabwean youths [17,18]. According to the Zimbabwe Civil Liberties and Drugs Network (2019) report, the proportion of young people involved in DSA increased from 43% to 2017 to 57% in 2019 [19]. In 2019 about 45% of admitted mental health patients were youth abusing drugs [20]. As a result of the massive brain drain caused by the mass immigration of skilled mental health providers, there is a massive treatment gap in mental health and SUDs treatment in Zimbabwe [21]. There is a severe shortage of human resources for mental health, and the number of psychiatrists is estimated at 18 (0.1 per 100,000 population), 17 based in Harare [22]. Moreover, mental health services are chronically underfunded, with limited specialist treatment facilities [12]. Spiritual factors play a major role in mental health treatment in Zimbabwe. Many patients consult faith healers for managing psychosomatic and mental health disorders [23]. Spiritual powers, informal counseling, and rituals for healing are conducted as treatment by traditional and Christian faith healers [24]. --- Role of religion in SDA Religious institutions like churches serve as a proxy for information dissemination hubs and social connections and significantly influence individual behavior. The church and other religious-affiliated organizations have promising yet underexplored potential to provide social support services for treating SUDs in communities where substance use treatment services are limited [25]. The synchronicity of DSA among youths and the increase in churches within the local communities highlights the underutilized potential avenue to counter the scourge of substance use. However, stigma related to DSA is common among believers who are expected by society to live a venerated lifestyle. Thus, the substance abuser is confronted with the anticipated conflict between social stigma and the desire to live a reverent life. Anticipated stigma leads to failure to disclose one's addiction problem and reduced likelihood of seeking treatment. The anticipated stigma associated with social rejection can further exacerbate the abuse leading to SUDs. The available scientific evidence shows that social support in the form of emotional comfort and knowledge about treatment options from significant others facilitate recovery from SUDs over time [26]. Practitioners often reserve their decision to discharge their clients because they feel that nonspecialized providers, such as self-help groups and religious organizations may be ill-prepared to address SUDs given their generalist training [27]. In addition, individuals providing these services in nonspecialty settings may have limited resources to effectively amplify treatment outcomes. Despite the controversy among addiction experts in clinical practice, research evidence reveals that partnering with faith-based organizations in the fight against DSA is a protective factor, particularly during recovery [28]. In Zimbabwe, some churches have been marred with abuse for political [29] and personal gains, as well as extortion and moral misconduct. Nevertheless, many religious organizations have been actively involved in mediating and ameliorating conflicts from an individual to national levels and established community projects such as social safety nets, education, and health facilities. Religious involvement can mitigate indulgence in illicit substance use and its adverse outcomes [28,30]. To bridge the support and treatment gap in resourceslimited contexts like Zimbabwe, we hypothesized that faith-based DSA recovery support and treatment could positively impact community prevention and treatment of substance use disorders and reduce relapses postrecovery. To assist church-going adolescents and young people recovering from DSA, the Africa University Clinical Research Centre established peer-support groups in which participants shared their experiences and solutions to the challenges they face. This study aimed to (i) establish the barriers and facilitators to accessing psychosocial support in a church context, (ii) determine the role of the church in DSA recovery, and (iii) explore strategies that the church can implement to promote access to psychosocial support to adolescents and young people recovering from DSA. --- Methods --- Study design This was a cross-sectional study that employed qualitative methods and was conducted over seven weeks from January to March 2022. The researchers chose to use the qualitative research method because it highlights the process and delivers an in-depth understanding of the study populations' subjective, perceived meanings, interpretations, and behaviors [31]. As Agar (1996) explained, qualitative research 'humanizes stereotypes' like SUD patients [32]. This study was part of the Pillars & Plants: Prevention, Support & Empowerment for Youth (PAPSE) project, which aimed to establish a chronic disease self-management clinical and e-therapy platform in a clinical trial and offer support group activities for adolescents and young people recovering from DSA. Africa University Clinical Research Centre, United Methodist Church Youth Pastors, and the Pastoral Care and Counselling Services provided the project support. The project was launched in 2019 and initially recruited and offered support to university students recovering from DSA. In 2019, young people recovering from DSA from the United Methodist Church circuits were recruited for weekly support group discussions with church youth pastors and mental health experts. The first cohort conducted face-to-face group sessions; however, the virtual platform was adopted for this second cohort due to COVID-19 pandemic restrictions. All participants were provided with data support to enable them to attend the weekly support sessions. The interviews of this study were conducted using young people recovering from DSA who had participated in the program's second cohort (2020-2021). --- Study setting The city of Mutare is in Manicaland Province, which borders Zimbabwe and Mozambique. After a long diamond rush discovered in 2006 in Chiadzwa communal areas near Mutare, many unemployed youths from all over the country and neighboring countries flocked to the city seeking to exploit the newly discovered economic opportunities. The city became a hub of illicit business, which attracted an influx of cheap alcohol and spirits from Mozambique. According to the 2015 Zimbabwe Demographic Health Survey, 86% of Zimbabweans were Christians [33]. A local study revealed that about 75% of the Zimbabwean population consult spiritual and biomedical care providers for mental health [34]. Regarding mental health matters, some people shun the formal health system for being culturally insensitive [35]. --- Study participants and recruitment For the purpose of this study, the term youth was defined as any individual aged between 16 and 25 years of age. The study focused on a diverse group of church-going youths between 16 and 25 years recruited from the PAPSE DSA treatment program for young people recovering and DSA. The program recruited these young people from 28 United Methodist Church circuits in Mutare City to be part of support groups designed to offer them psychosocial support during their recovery. Eligible participants were youths in the support groups, and youth pastors who represented the church were recruited as key informants. Invitations to participate in the study were sent to participants via the two WhatsApp groups, consisting of twenty DSA recovery program support group participants. Eighteen young people and three youth pastors responded to the invitation, and these were purposively selected for the interviews. The informed consent forms for those under 18 were signed by their parent/guardian. All the DSA recovery participants were black Africans, and 12 were males. Of the three-youth pastor interviewed, two were male, and their age range was 26 to 35. The key informants were purposively recruited from the pool of youth pastors who were actively involved in the weekly support group discussions. These key informants were interviewed on barriers and facilitators, as well as the role of the church in accessing psychosocial support among young people recovering from DSA. --- Data collection The study participants received an informed consent form which they signed and returned to the researcher coordinator either in person or as a scanned document. The researcher then called the study participants to schedule the interview date and time. In-depth interviews were conducted by FFM and CM using semi-structured questionnaires for the youths recovering from DSA and the church pastors. Both interviewers were Zimbabwean citizens who understood all the languages used by the interviewees. Data collection was the first point of contact between the data collectors and study participants. Although the interviews were conducted in English, participants were encouraged to express themselves in the local language-Shona freely. The interviews were conducted using recorded telephone calls, and the files were saved on the cloud soon after the interview. A password-protected recording device was used as a backup, while phones were placed on 'do not disturb mode during the interviews. The researchers developed the questionnaires guided by variables associated with access to psychosocial support emerging from similar previous studies. The data collection tool was categorized according to the characteristics influencing psychosocial support for DSA recovery. First, participants' sociodemographic characteristics (age, sex, marital status, and occupation) and experiences with substance abuse (duration of abuse, turning point, and the decision to seek help). Second, the barriers and facilitators for access to psychosocial support in the church context (disclosure; confidentiality; support from church leadership; stigma from church members, family, and peers) and the role of the church in DSA recovery (existence of psychosocial support structures, discussions around DSA among youths). Lastly, questions were asked regarding possible strategies to promote access to psychosocial support. The questionnaires were pretested using participants from the Africa University campus church circuit. The authors monitored the interviews for the saturation of themes. After interviewing the eighteen young people recovering from DSA and three youth pastors, the authors agreed that data saturation had been achieved, taking into consideration participant diversity and interview quality. --- Data analysis The study followed the stages for the analyses in phenomenology. Data from the recorded interviews were transcribed verbatim for analysis by FFM and CM. The researchers translated the interview scripts into English, and nonverbal cues, e.g., sighs, pauses, and laughter, were included in the data transcription on a Microsoft Word document. To improve accuracy in the interpretation of the interview transcripts, three research team members participated in the transcription of all the interview scripts. The research coordinator, FFM, led the analysis and had regular debriefing meetings with CM and FMM. An iterative qualitative research approach was applied to all the transcripts, which were read through multiple times to identify themes emanating from the interviews. Thematic network analysis of the qualitative data was conducted using techniques proposed by Attride-Stirling (2001) [36]. To facilitate decision-making and to problem-solve in an intelligible form while disentangling arguments, a thematic network enables the extraction of i) basic themes (lower-order premises in text), organizing themes (basic themes grouped), and global themes (overall themes capturing the principal metaphors in the text [36]. The thematic network used the hermeneutic approach to break up the text to find the embedded explicit rationalizations and implicit significance. --- Trustworthiness of the study To enhance the trustworthiness of the study, all researchers underwent a mandatory Good Clinical Practice on Social and Behavioral Research Best Practices for Clinical Research training two weeks before data collection. The questionnaires were pretested using four young people and two youth leaders, and modifications were made before the interviews. Upholding data integrity and treatment fidelity was done by developing standard operating procedures for the data collection activities. To track fidelity and minimize protocol deviations, debriefing meetings were conducted on the morning of every day of data collection. These meetings were held throughout data collection and analysis. Discussions were used to resolve disagreements, and this was conducted until all parties involved reached a consensus. The researchers discussed with third parties who organized the PAPSE program and were responsible for the day-to-day running of the program to obtain feedback on the study results. We used quotes as a way of providing a vivid illustration of the participants' responses on the questions asked. The Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines for reporting on qualitative research were used in reporting the study findings. --- Ethical considerations The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All procedures involving human subjects/patients were approved by the Africa University Research Ethics Committee (Approval number: AUREC 2460/22). Written informed consent was obtained from all participants, and permission to conduct the study was obtained from the Director of the Africa University Clinical Research Centre. For possible psychological adverse events during and after the interviews, the participants were referred to university psychotherapists who were available for free counseling via phone or email 24/7. --- Results Table 1 displays the sociodemographic characteristics of the young people recovering from DSA who participated in this study. The primary outcome of interest, which was access to psychosocial support, was the global theme. The organizing themes were facilitators and barriers and the church's role in accessing psychosocial support among young people recovering from DSA. The basic themes around facilitating factors were acceptance that DSA was a health problem that requires treatment, trust in the confidante, support from peers and parents, and a wellorganized support program. The basic themes around barriers were having negative role models, radical religious beliefs that castigated drug-abusing youths, denial of one's addiction problem, and societal stigma and discrimination of illicit drugs. The church may improve access to psychological support through rehabilitative programs and offer counseling services to young who have committed to quitting drugs. The initial response reflecting on the study participants' experiences revealed that substance use began during their years in school. The obstinate use of illicit drugs and substances led to dependance and SUDs. They talked about how they regret missing life opportunities because of substance abuse. "I started using drugs in high school. I was 15, and the senior boys I hung out with were doing it, and it looked cool. I failed to stop after that and dropped out of school. I am 24 now and have no hope for a future unless I stop this and accept that I need help. " (24,Male). Seven participants reflected on how they lost connection with family and friends. The subsequent loneliness and isolation intensified their drug-abusing habit. --- Facilitators --- Individual acceptance The study participants reported that acknowledgment of one's drug addiction problem was integral to the recovery process. Understanding that using drugs was a perilous path that conflicts with Christian values and that there is a better and sober life led them to seek psychosocial support. "It was only when I accepted my drinking problem and that it had gotten out of control that I realized that I needed help. " (21,Female). "I think I had a revelation. I started feeling empty, lonely... I needed God. " (22,Male). After realizing their mistakes, they noted how bad the situation had deteriorated and hence were willing to commit to new beginnings; as one participant described: "Self-reflection and being honest with myself is the only thing that made me seek for help. I thought I was headed for danger, but the truth was that I was already in danger. " (18,Male). Some participants had to witness the consequences of DSA for them to decide to quit. "I saw a dirty man lying face down in a roadside ditch. He looked dead. That was an awakening moment for me. I figured myself in his shoes, and I knew I had to stop!" (20,Male). --- Confidentiality The availability of someone the participant would trust emerged as a critical element in disclosing one's drug addiction and ultimately seeking support. As one participant described: "The bond I once had with one of the elders' wives allowed me to be open up to her. Surprisingly, she did not judge me, and this took a huge load off my shoulders. " (22,Female). The young people also felt that access to support from outside the church prompted them to seek counseling. "You can't trust these 'holy church youths' (laughs); they gossip a lot. I was so relieved when I talked to a church pastor from our church who was one of the counselors for the PAPSE Program. She was such a good listener, I trusted her, and I am glad she was good at keeping secrets. " (20,Female). --- Availability of an organized support program Given the sensitivity of drug abuse issues in the community and the associated stereotypes, the interviewees wanted assurance of confidentiality. This was likely going to be possible if the support group was organized. One of the leaders stated: --- "We barely have efficient structures. It feels like we fail the youth who need us focusing on other issues....we tend to neglect them. " (Male Pastor). One of the youths supported this: "Church pastors speak against substance use and encourage abstinence, but formal structures to support don't exist. I believe drug recovery counseling for the church could be more effective if the pastors adapt the organized approach introduced by the PAPSE program. " (22,Female). --- Peer and parental support Peer pressure was identified as the leading cause of drug uptake by the young people who participated in this study. However, peer support had a pivotal role in recovery. Young people can openly counsel each other, and peers who understand the dangers posed by DSA can be the compass to seeking help. --- "It was because of my friend that I felt I should seek help from professionals and God because I knew he had not forsaken me. " (21, Female). Despite having all their needs provided for at home, some young people still fall for DSA. The family's involvement in the decision to seek support lightened the burden of solitude. "I had a very supportive family; they loved me and gave me attention and affection. I was the one who got lost and started doing drugs with the other children in my hood. I kept justifying my (drug) habit, but I am glad my parents never gave up on me until I had that turning point. I started attending church again...I still feel like a prodigal son from the Bible. " (25,Male). --- Barriers --- Stigma and social rejection When asked about barriers to seeking psychosocial support, the most central obstruction mentioned by the study participants was the stigma associated with DSA. The notion that some'sanctified' youths who abstained from drugs were unwilling to associate with them made it difficult for the affected youths to seek support from within the church (from ten respondents). "They portrayed perfection in my eyes as good youths who did no wrong, and I felt they would not accept me and would judge me harshly. " (20,Female). The feeling that they were being discriminated against bore the assumption that they did not belong to the religious institution. --- "In their (abstaining church youths) eyes, I was that bad apple, an outcast with a bad influence on other youths. I did not see them accepting me as one of them. " (19, Female). --- Radical religious beliefs As part of psychosocial counseling, some church elders prescribed seeking answers from the. Bible, praying and fasting as a way of getting inspiration to quit drugs. Such generalized guidance did not align with the youths' expected path to recovery. --- "I stopped seeking help from her (church elder) because she emphasized that I should study the Bible, believe it, and pray. That is a lot to read; she never pointed out where I should read from for the formula to succeed. " (25,Male). Some relatives and the community may perceive DSA as a mystic malady caused by external forces like demons, witchcraft, or evil spirits, which causes indirect discrimination. "I heard people saying someone in my family had bewitched me, and they were making a fortune out of my misery... One day, I overheard my mom telling one of the church elders to pray for me so that the evil spirit in me would go away...I understand she wanted to help, but I was infuriated by her walking around saying these things about me. " (18,Male). Another major barrier to seeking psychosocial support was a stigma related to a lack of mental health literacy. Some appointed youth leaders were perceived as too old to understand the mental health problems faced by the current generation of youths. --- "The age gap (between the young people and some youth pastors) makes everything hard. They would not understand what exactly is going on... their main thrust and the ultimate solution would be praying for you to cast out demons when I need a compassionate ear. " (24, Male). --- Denial Some parents dissent from the news that their child abuses drugs, especially when the child. goes to church. One of the youth pastors mentioned this: "No parent wants to believe that their child is doing drugs... that can be a sign of parenting failure. " (Female Pastor). Parents or guardians always want the best for their children and tend to look away from the dark side of the child's behavior thus missing an opportunity to help. --- "I told my aunt about my addiction problem, and she was adamant that we would pray and fast, and the problem would go away... It didn't!" (16, Male). --- Negative role models One of the barriers to seeking psychosocial support was negative peer pressure and the associated temporary cushion it brings. --- "I started drinking with this group of boys I looked up to. They accepted me and gave me more attention than my parents did. I needed that honestly; unfortunately, the affection came with a terrible addiction. " (18, Male). In some circumstances, the youth could not see the point of seeking psychosocial support because those who were supposed to help were also in a similar predicament. "Seeking help from youth leaders who are substance abusers is like a blind man leading another blind man. " (20,Female). Seeking support can be difficult in situations where the family is also involved in substance abuse. --- "Since my mother died, my father has turned to intoxicating alcohol to drown his sorrows. I couldn't rely on him for my recovery. " (18, Male). --- The role of the church The participants' narratives identified the church as a potential protective environment to provide counseling services for young people recovering from DSA. The church can implement this effectively through the various opportunities for prosocial activities against defiant behavior like substance use. --- "As leaders, we ought to identify the needs of the youths and exert our attention in that direction. We have to speak to all youths and not discriminate. We should create a safe and nurturing environment for those struggling with an open mind. " (Female Pastor). Developing innovative ways of keeping the youths engaged in productive tasks to deter them from risky activities such as DSA. --- "As a church, we must meet the socioeconomic needs of the young people facing life challenges before seeking to restore their faith. " (Male Pastor). When asked to elaborate on how these needs can be met, the pastor added: "We have groups such as choir, sports teams, and outings that are currently inactive. We should stimulate the youths to become active and join. Some problems need professional help... more than just prayer, and we should refer these to those trained to do the work. " As a source of Supreme Power through individual belief, the church can be a fortress that protects young people from DSA. One participant echoed this: "I knew my faith in God could help me sober up. I grew up in a Christian household and stopped attending church due to drugs. Restoring my faith in God would help me sustain my resolution to stop drugs. " (20,Female). One shared feeling reported as the church's role was: "An affiliation with other progressive youths gave me solace... that dedication to serve the Lord has kept me from drugs. " (17,Male). --- Strategies to promote access to psychosocial support Findings from the interviews revealed the need for increased awareness of current youth challenges. The campaigns should be able to reach out to youths within and outside the church since the church is a safety net for the vulnerable. --- "As Christians, we should have a welcoming attitude even to those who abuse substances and are not part of our church... interact with them in the community, listen to their side of the story and intervene on a case-by-case basis than judging from a distance or imposing our interventions on them. " (Male Pastor). To overcome idleness, the church should create opportunities for youths, especially those out of school. "The church should encourage and support youths to start projects through seed grants or promote sporting and entertainment activities to identify talent... some of the addiction is a result of unemployment... giving them (youths) some responsibility prevents indolence. " (24,Female). Figure 1 shows the thematic networks for accessing psychosocial support among church-going youths. --- Discussion Under the facilitating factors organizing theme, the researchers noted the following basic themes acceptance, confidentiality, peer and parental support, and organized support program. Negative role models, radical religious beliefs, denial, stigma, and discrimination were identified under the barrier factors organizing theme. The church was seen as the arbitrator between the barriers and accessing psychosocial support (global theme) among church-going youths. The church ensures that young people get involved in projects and programs that avert idleness. --- Facilitators Individual acceptance of the addiction problem facilitated access to treatment and support. Concerning mental health issues, acceptance is pivotal for coping and recovery [37]. Acceptance creates room for a positive mindset for behavior change associated with positive and clinically relevant outcomes [38,39]. Without selfacceptance, the mental health problem will persist, and beneficial interventions will be less helpful [40]. The selfconscious recognition that one needs treatment should be emphasized while counseling young people recovering Fig. 1 The thematic networks of access to psychosocial support for church-going young people recovering from DSA from DSA because it has a bearing on the retention of patients in treatment. Confidentiality was one of the facilitators of accessing psychosocial support among the church-going youths who participated in this study. Trusting and confidential relationship issues are sensitive among young people's engagement with mental health services providers [41]. Thus, facilities and individuals providing counseling services should be discreet and youth-friendly by offering open engagement, truthful responses to identified problems, and ensuring retention to care [42]. Training the youth pastors tasked with counseling duties in related ethical issues and providing a private environment for the counseling sessions is pivotal to attracting young people recovering from DSA in a church context. A similar study also echoed the need to train church leaders to handle SUD victims [43]. Peer and parental support were common facilitators to accessing counseling among the study participants. Openly disclosing one's addiction challenges is facilitated by an established trusting relationship [41]. With support from peers and family members, the youth facing addiction challenges can build a buffer against social stigma and thus repel isolation and the associated anxieties [44]. Family and friends can be trustworthy, confident, and treatment buddies throughout recovery. Peer support groups encourage those experiencing SUD recovery to express their feelings, thoughts, and personal concerns in an environment that does not make them feel discriminated against. The youths interviewed in this study applauded the development of the PAPSE program and how it incorporated the church leaders in an organized way that met their counseling needs. Organized and goal-directed programs motivate the youths to seek treatment because they provide systematic and explicit guidance [45]. A well-coordinated program reduces ambiguity and conflict, while treatment milestones can be measurable through mutually set targets. In contrast, the lack of professionalism and unplanned environments where therapists may fail to uphold ethical standards can deter young people from seeking mental health treatment [46]. Standardized structures devoted to psychosocial support of young people recovering from DSA in places of worship in Zimbabwe are limited and we recommend that the religious organizations institute these to improve access to support for this vulnerable population. --- Barriers We noted that stigma and social rejection were significant barriers to accessing psychosocial support among young people recovering from DSA. This is consistent with other studies in which fear of rejection was a concern among individuals struggling with addiction which would be perceived as weak by their peers and significant others [47][48][49]. The internalized belief based on misconceptions about self significantly influences well-being, ultimately affecting access and utilization of health services [50]. While drug addiction is a source of stigma, the thought of social rejection after disclosing one's problem can exacerbate mental health challenges. A transparent channel and non-judgmental approaches can encourage more young people to disclose, seek and access SUD treatment early. Some church counselors emphasized reading the Bible, fasting, and prayer, a strategy noted as ineffective by the study participants. One of the effective strategies to heal mental health problems is genuine, empathetic listening [51]. Receptivity and good listening skills by the counselor allow the victim to freely express themself without feeling that they are being judged. These can be achieved using motivational interviewing techniques guided by four principles: resisting the righting reflex; understanding the patient's motivations; listening with empathy; and empowering the patient [52]. The counselor should allow the patient to confide without rushing to prescribe solutions since individuals with addiction challenges often feel misunderstood. A counselor's patience permits understanding the problem from the patient's perspective and provides a specific solution. Instances of negative role models were noted as a barrier to seeking support among church-going youths facing drug addiction challenges. Trusted primary sources of attachment, like family members, can be a source of a refugee during the recovery. However, the knowledge that the individual who should be part of the solution is also part of the problem can induce negative thinking and resentment. Parents who have SUDs are likely to have unhealthy family functioning styles and lower care scores [53]. Living or constantly associating with someone still using substances can trigger cravings that may lead to relapse [54]. Thus, to improve treatment outcomes, DSA counselors need to have an upstream mentality and understand the conditions in which their clients reside. Similarly, church leadership for the youth and those who appoint addiction counselors should do so after a careful assessment that includes background checks. The treatment process can be impeded if the parents/guardian of the child is in denial. The parent often feels responsible when their child is a DSA leading to denial and experiences such as stress, self-accusations, anger, sadness, and pressure to assist their child to overcome the addiction problem [55,56]. The counselor must understand the patients' social support base to provide comprehensive therapy. --- Role of the church and strategies to promote access to psychosocial support Religious organizations are critical in offering recovery support services, coping ability, and rehabilitation of DSA patients [51]. Youth involved in faith-based activities exhibited better resilience and the ability to cope without using drugs when compared to their counterparts [28]. Introducing social activities to reduce idleness or financial gains can positively impact the recovery program. Volunteering for charity work for congregational substance abuse recovery programs is associated with positive health outcomes [57]. Volunteerism help patients reconstruct their new identity towards recovery, and besides a good feeling of self-worthiness, helping others improve self-esteem and optimism while reducing depression and helplessness. Rendering service to the community in the substance treatment field can also sustain one's mental health recovery and maximize multiple aspects of life functioning [58]. Young people expect to be financially independent, and the alarming unemployment rates in Zimbabwe can be a barrier to alcohol and DSA recovery, even with good counseling. This was also noted in a study by Stokes and colleagues (2018) [39]. Employment is deemed one of the critical indices of accessing psychosocial support among drug addicts, as work provides opportunities for socialization and healthy relationships with non-substance abusers [59]. Thus, strengthening commitment to recovery [60]. Recovery programs within and outside the church should forecast improving the employability and source training for the youth to prevent relapse. In addition, the church can also advocate for the recovering youths in a way that will improve a positive employer attitude. To succeed in its mandate, Christians must reflect on self-concept and reinforce the culture of love and respect. --- Strengths and limitations This study brought novel knowledge of the challenges faced by church-going youths, and according to the authors' knowledge, it is the first of its kind in Zimbabwe. However, our study had some limitations as it was conducted using young people from one Christian denomination. Thus, the findings cannot be generalized to other churches and contexts. All the participants were selfselected from the youth involved in the PAPSE program leaving out those who had the addiction but had not enrolled for treatment. Using self-reports may be a source of bias as participants can give socially desirable responses. Therefore, a more robust quantitative study could be conducted to enhance reliability. --- Conclusions Acceptance of one's addiction problem is critical to initiate seeking psychosocial support. Confidentiality with disclosed information, support from trustworthy relationships, and the availability of a well-coordinated recovery program enable young people to seek support. An organized church-mediated drug abuse support program has the potential for high-success treatment outcomes and retention. By providing innovative projects, inclusive programs, and social support, the church can bridge the psychosocial support gap between the support services and the barriers, mostly interactive relationship based. Formal training of church-based counselors or pastors can have profound treatment outcomes by introducing the ethical aspects of psychotherapy and reducing the preconceived social stigma associated with SUDs. To assist youth in prompt access to mental health support services, Christians willing to help must approach this sensitive age group with an altruistic lens. --- Author contributions FFM, FMM, and CM conceptualized and designed the study. FFM and CM collected the data, and all authors contributed to the data

Background The church and other religious-affiliated organizations have promising yet underexplored potential to provide social support services for young people recovering from substance abuse in communities where drug and substance rehabilitation services are limited. This study aimed to establish the barriers and facilitators of accessing psychosocial support, the role of the church, and strategies to promote access to psychosocial support for youths recovering from drug and substance abuse. Methods This was a qualitative cross-sectional study, and semi-structured interviews of 18 church-going youths and three youth pastors were conducted in eastern Zimbabwe. Data were collected using recorded telephone interviews. Data were transcribed and analyzed using the thematic network analysis technique of producing basic themes, which build into organizing themes. Organizing themes produces one overarching global theme. The Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines for reporting on qualitative research were used in reporting the study findings.The interviews produced the following basic themes under organizing theme barriers: stigma and discrimination, parental/guardian denial, radical religious beliefs, and negative role models. Under the organizing theme facilitators, the basic themes were acceptance, confidentiality, peer and parental support, and an organized support program. The church acted as the bridge between the barriers to access to services and support seeking through innovative, inclusive projects and activities, as well as a pillar of social support. Conclusions Acceptance of one's addiction problem is critical to initiate seeking psychosocial support. Confidentiality, support from trustworthy relationships, and the availability of a well-coordinated recovery program enable young people to seek support. We recommend formal training church-based counselors in the ethical aspects of psychotherapy to reduce the preconceived social stigma associated with drug and substance abuse.

Using self-reports may be a source of bias as participants can give socially desirable responses. Therefore, a more robust quantitative study could be conducted to enhance reliability. --- Conclusions Acceptance of one's addiction problem is critical to initiate seeking psychosocial support. Confidentiality with disclosed information, support from trustworthy relationships, and the availability of a well-coordinated recovery program enable young people to seek support. An organized church-mediated drug abuse support program has the potential for high-success treatment outcomes and retention. By providing innovative projects, inclusive programs, and social support, the church can bridge the psychosocial support gap between the support services and the barriers, mostly interactive relationship based. Formal training of church-based counselors or pastors can have profound treatment outcomes by introducing the ethical aspects of psychotherapy and reducing the preconceived social stigma associated with SUDs. To assist youth in prompt access to mental health support services, Christians willing to help must approach this sensitive age group with an altruistic lens. --- Author contributions FFM, FMM, and CM conceptualized and designed the study. FFM and CM collected the data, and all authors contributed to the data analysis and interpretation of the study results. PTM wrote the first and final manuscript, and all authors reviewed the different manuscript drafts. --- Funding None. --- Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. --- Declarations Ethics approval and consent to participate The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All procedures involving human subjects/patients were approved by the Africa University Research Ethics Committee (Approval number: AUREC 2460/22). Written informed consent was obtained from all participants, and permission to conduct the study was obtained from the Director of the Africa University Clinical Research Centre. --- Consent for publication None. --- Competing interests The authors declare no competing interests. --- Authors' information The researchers hold the following credentials: FFM (BSc Psychology), PTM (BSN, MPH, Ph.D. Student), CM (M. Ed in Educational Psychology, BSc Psychology), and FMM (MPH, MD). During the study period, FFM was an undergraduate student, PTM was a Research Support Officer, CM was a senior lecturer, and FFM was the Director of the Africa University Clinical Research Center. All researchers were female except PTM. FFM and CM had one and ten --- Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Background The church and other religious-affiliated organizations have promising yet underexplored potential to provide social support services for young people recovering from substance abuse in communities where drug and substance rehabilitation services are limited. This study aimed to establish the barriers and facilitators of accessing psychosocial support, the role of the church, and strategies to promote access to psychosocial support for youths recovering from drug and substance abuse. Methods This was a qualitative cross-sectional study, and semi-structured interviews of 18 church-going youths and three youth pastors were conducted in eastern Zimbabwe. Data were collected using recorded telephone interviews. Data were transcribed and analyzed using the thematic network analysis technique of producing basic themes, which build into organizing themes. Organizing themes produces one overarching global theme. The Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines for reporting on qualitative research were used in reporting the study findings.The interviews produced the following basic themes under organizing theme barriers: stigma and discrimination, parental/guardian denial, radical religious beliefs, and negative role models. Under the organizing theme facilitators, the basic themes were acceptance, confidentiality, peer and parental support, and an organized support program. The church acted as the bridge between the barriers to access to services and support seeking through innovative, inclusive projects and activities, as well as a pillar of social support. Conclusions Acceptance of one's addiction problem is critical to initiate seeking psychosocial support. Confidentiality, support from trustworthy relationships, and the availability of a well-coordinated recovery program enable young people to seek support. We recommend formal training church-based counselors in the ethical aspects of psychotherapy to reduce the preconceived social stigma associated with drug and substance abuse.

Introduction The number of conceptions and births follows a seasonal, cyclical, but usually fluctuating pattern over time, influencing even the characteristics of later life (Didikoglu et al. 2020). It is a global phenomenon observed both in contemporary and historical populations (see Quételet 1826;Lam and Miron 1991), including the natural fertility ones, like the Old Order Amish (Greksa 2004) or the Hutterites. In the latter case, the conceptions increased from December to June and declined from July to November (Surbey et al. 1986). Overall, the seasonality's diversity is substantial all over the world. In the United States, there was a peak in births during September and another one during April-May (Lam and Miron 1991). This general pattern depended on the latitude: the northern states exhibited a spring/summer peak while the southern states an autumn peak (Martinez-Bakker et al. 2014). Some differences existed in England and Canada, concerning mainly spring months (Lam and Miron 1991). According to the Lam and Miron (1991) classification, a "European pattern" of seasonality of births emerged in the past, corresponding mainly to Western European countries, consisting of "a global spring peak, a local September peak, and a significant trough during the late fall and early winter". In central and southern Europe, where seasonality was less pronounced, the local September peak persisted, and the decrease in the number of births was observed from October to January. In India, the diversity was high, while in Japan, the seasonality had more affinities with the North American and European populations. More details can be found in Lam and Miron (1994). In more recent studies, the number of births is higher from February to April in northern and South Australia, but, in the southeast, the peak comes in September-October (Wilson et al. 2020). Singh et al. (2020) discovered two general patterns (1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) for India; the first corresponds to a conception excess from October to February (Uttar Pradesh), while in Kerala, a peak exists in July-August. In Sweden, the high number of births during spring was followed by low numbers in the last quarter of the year (Dahlberg and Andersson 2018). Seasonality patterns do not remain stable over time. In Poland, for example, the pattern of a spring peak and a local excess in September remained unchanged for a long period. In 1990-2009, however, a new pattern emerged, with birth peaks in July and September (Cypryja<unk>ski 2019). In Malta, patterns altered following the introduction of contraception in the 1960s (Grech et al. 2003). In Sweden, the seasonal variations of birth rates declined significantly during the 21 st century (Dahlberg and Andersson 2018). In France, the historical September birth spike, corresponding to a peak of conceptions in New Near's Eve, is the only one that still stands out among the low-amplitude seasonal variations (Régnier-Loilier 2010). Overall, as Roenneberg (2004) suggests, human seasonality has declined in the recent era. The human reproductive cycles are either an adaptation to the seasonal cycles or subject to cultural factors (Wood et al. 2017). Many environmental, social, and biological agents affect them (see Doblhammer-Reiter et al. 2000), including several sociodemographic variables, like maternal age, marital status, education and birth order (Bobak and Gjonca 2001; see also Lummaa et al. 1998;Surbey et al. 1986;Cypryja<unk>ski 2019;Yang 2021). Among the essential sociocultural determinants are the religious festivities, holidays and sexual activity cycles, which, along with fecundity, vary in a year. In the USA, an increase of conceptions occurs around Christmas, but not in the summer (Lam and Miron 1991). By contrast, in England and Wales, sexual activity increases during summer, besides the Christmas effect on conceptions (Wellings et al. 1999;Hennink et al. 2000). In Croatia, a similar positive effect of Christmas on conception rates is found (Pola<unk>ek et al. 2005). However, the "Christmas effect" has disappeared in Sweden after the 1990s (Dahlberg and Andersson 2018). Other religious obligations may restrict human sexuality in a year. In Spain, for example, it was found that conceptions decreased significantly during Lent (Simó-Noguera et al. 2020). In Romania, the fasting periods and the religious affiliation are essential factors regulating the date of conception (Herteliu et al. 2015). In Malta, the late summer peak is related to contraceptive planning, which in its turn is affected by the Roman Catholic ethos and social pressures (Grech et al. 2003). Thanksgiving day increased coital frequency in the Old Order Amish (Greksa 2004). Therefore, these events may either boost or depress sexual activity and conceptions in a year. There is evidence that several ecological/climatic factors may affect the frequency of sexual intercourse. As Darrow et al. (2009) suggest, the hot weather in the summer may reduce conceptions through reduced coital frequency or decreased fecundability (e.g., decreased sperm quality; see also Yang 2021;Greksa 2004). Seasonal variations in fecundability have also been found elsewhere, with a peak in the late autumn and early winter (Wesselink et al. 2020). In Australia, the hot and humid summers in the north, and the cold winters in the southernmost parts of the country reduce the number of conceptions below their regional averages for the year (Wilson et al. 2020; see also Cornelisse et al. 2016). In other studies, the effect of photoperiod or light intensity is emphasised (Cummings 2007;Darrow et al. 2009). Indeed, other factors contributing to the annual birth pattern -besides photoperiod-can lead to changes in the coital frequency in a multivariate scheme, including, as said before, holidays, seasonal patterns in fetal loss, and seasonal preferences in pregnancy planning (Darrow et al. 2009). Seasonal variation of workload must be also included among them. In Italy, the intensity of the agricultural workload varies significantly in a year and becomes more intense in the summer, primarily when grain was harvested without sufficient nutrition (Ruiu and Breschi 2019). Roenneberg (2004) recognises a combined effect of industrialisation, microclimate and human behaviour on human conception rhythms and seasonality, because after the "shifting work from outside to inside, both the daily and the annual changes of light and temperature are drastically buffered." Any relevant publications are scarce in Greece. Krassas et al. (2007) recognise that in the general population of Greece, the number of births increases during summer, which gives a rise of conceptions during the autumn and the first month of winter. Gavalas (2009) found that in two villages of Paros (Naoussa and Kostos), conceptions were high in the summer and September, reduced towards November, and increased until February. Zafeiris (2012), found a high but changing seasonality of births and conceptions among the populations under investigation. More specifically, in the Vlachic population of Karitsa until 1979, the number of conceptions increased in May and remained above average until July. Afterwards, seasonality declined. In the Pomaks of Rhodopi, there was an excess in conceptions lasting from June to September. These temporal trends are connected with the annual religious calendar, cycles of agriculture and other factors. Thus, the main research question in this paper concerns the seasonality of conceptions (and thus births) in 5 mainland populations. The scope is to identify general patterns of sexual behaviour and fecundability throughout a year and to examine the existing heterogeneity, i.e. possible convergences and divergences among these populations. Further, as we approach the modern era, the second topic of interest is to observe changes in the seasonally of conceptions (births) that occurred over time. The main characteristics of the populations under study are described in Table 1. Five populations with significant similarities and differences are considered. Three of these are mountainous. One of these, the population of Metsovo, is of Vlachic origin, and is located in the Pindus Mountains. Metsovo is one of the metropolises of Vlachs in Greece, respresnting a well developed region in the past compared to the other four populations (see Zafeiris et al. 2015). The other two mountainous populations are residing in Organi and Kehros and come from the Rhodopi mountain in the homonymous Department of Eastern Macedonia and Thrace (Zafeiris 2020). These are Pomak populations. The fourth population is a lowland population residing in Dion, in the Department of Pieria, in the Region of Eastern Macedonia. It consists of a mixture of locals and newcomers from the Mounts Olympus and Pindus (Vlachs and Sarakatsans), as well as of Refugees from Asia Minor and Euxinus Pontus (see Zafeiris and Kaklamani 2019). These populations have undergone significant social, economic and cultural changes over time (see Table 1). Finally, the General Sample represents an assortment of populations scattered in ecological niches of continental Greece and will be used for comparative purposes in relation to the geographically localised populations. --- Data and Methods For the purpose of the study we use data derived from the Vital Registration System of the municipalities of Metsovo and Dion and of the former communities of Organi and Kehros, which were verified and completed during fieldwork. The date of child conception to the mother was estimated as the recorded date of birth minus 260 days (Jones 1997). Estimating the date of conception in this way takes into account only registered live births; foetal loss and misscarriages are excluded from our estimates, as relevant data are unavailable. Hence, to obtain comparable results, the assumption that the rate of foetal loss is constant across seasons and across the periods under research is implicit. Such an assumption, however, may not be fulfilled, especially concerning the different periods under investigation. It may be more realistic to assume that overtime, as socioeconomic circumstances and prenatal care improve, foetal loss would decline. That could affect comparability between diferent periods, but it is unlikely to affect the estimates of the extent of seasonality within a specific period.per period of conception is unknown for these populations. In any case, the number of stillbirths is very small and this further suggests that the assumption adopted does not quantitatively affect the result. A special epidemiological study is needed for this,though even in this case any results must be questionable due to the long time gap between the time the events took place and the time the data were collected. Instead in this paper only the recorded births were taken into consideration, in the absence of any other solution in order to include foetal losses. Overtime, socio-economic conditions, cultural factors and reproductive practices change, and tend to differentiate the observed seasonal patterns of conceptions and livebirths. Hence, the analysis refers to four discrete periods 1951-1964, 1965-1980, 1981-1992 and 1993-2002 which essentially correspond to the fertility transition in these populations (Zafeiris et al. 2015;Zafeiris and Kaklamani 2019;Zafeiris 2020). The fertility transition in all of these populations led to the decrease of the total fertility rate, with various rhythms. Specifically, fertility was very high in the Pomaks until 1964, with more than 4 children per woman (Zafeiris 2020). In Metsovo, where fertility transition started earlier, period TFRs fluctuated between 2.8 and 2.6 children per woman in the period 1951-1964 (Zafeiris et al 2015). In Dion, fertility remained high for the cohorts reproducing during this era (Zafeiris and Kaklamani 2019). After 1964 and until 1980, fertility declined rapidly in all the populations studied, and a new demographic regime of low fertility appeared in the 1980s. As a result, women give birth to fewer children over time, a fact that is portrayed in table 2 of this paper. Due to the heterogeneity of the five populations with respect to religion, socio-economic milieu, employment activities and certain traditions, the statistical analysis considers them separately. Their characteristics are presented in detail in Table 1 (Metsovo, Dion, Organi, Kehros, General Sample). The analysis is carried out as follows. The micro-statistical information on date of birth and the consequently estimated conception date was converted to aggregate data, referring to numbers of conceptions by month and year of observation. However, one must take into consideration the unequal length of months in a year, which obviously distorts the real seasonality patterns. In order to confront this problem the observed aggregated monthly data were adjusted to numbers of conceptions of months of equal number of days (30 days). The statistical analysis was performed using the adjusted numbers of conceptions by month and period of study shownin Table 2. We employ the following procedure to portray seasonal variations of conceptions. Using the adjusted monthly numbers of conceptions, we construct seasonal indices applying a variant ratio to moving averages method. We calculate centered moving averages which are much smoother than the original data, so that they reflect a clearer picture of the long term trend of conceptions (Kvanli et al. 1996). The ratios of the observed data to the centered moving averages reveal, in relative terms, the seasonality of the phenomenon. These ratios are considered as dependent variable. Subsequently, we express each month in terms of binary dummy variables, and we introduce these dummy variables in the models as regressors. Finally, we estimate multiple linear regressions without a constant term (Brooks 2008); the estimated regression coefficients correspond to twelve seasonal indices, showing the percentage excess or deficit of monthly conceptions for a specified period. Finally, we also calculat the coefficient of variation as an overall measure of the seasonal fluctuations. The coefficient of variation (CV) is calculated as the ratio of the standard deviation to the mean, indicating the extent of variability in relation to the mean of the population (see Kirkwood, 1979). --- Results Conceptions show intense seasonal fluctuations and a long-term downward trend over time. Figures 123456present the observed number of conceptions by month as well as the 12-month moving average for each population separately and for the whole sample, for the overall period of the study. The total number of conceptions (Figure 1) follows a downward trend, which appeared initially in 1968 and was established more clearly in 1981; seasonality is evident, but the range of seasonal variation gets smaller over time. Several differences, however, are observed across samples and periods. Note that the number of births for the period since 1993 does not include the Pomaks of Organi and Kehros, due to a lack of relevant data. Therefore, total numbers of births for this period (Figure 1) are not comparable to the other periods included in the analysis. --- Analysis by period During the first period (1951)(1952)(1953)(1954)(1955)(1956)(1957)(1958)(1959)(1960)(1961)(1962)(1963)(1964), individual reproductive behaviour presents many similarities across populations, as the number of conceptions tends to be higher in late spring and some summer months (Figure 7; see also Table 3). Nevertheless, there is still high diversity. First, a Christmas or a New Year's Eve effect is not easily detected. Conceptions in January remain either very low, like in Kehros, or close but below the value of 100, which denotes an equal distribution of conceptions between the months of a year (dotted black line in Figure 7). The only exception is observed for Organi in January, but people there are Muslims, and those days they did not celebrate the arrival of the New Year, as fieldwork evidence suggests. Note that in the previous month, i.e., December, the number of conceptions was very low in all populations. Therefore, at least for the Christian populations of the analysis, the advent of Christmas and New Year holidays may have slightly increased sexual activity and, consequently, conceptions in January compared to December. In the following months, conceptions decrease in all populations. For the Christians, this could be an effect of the fast of Lent, during which the more religious population groups avoided any sexual activity. However, this is also observed among Muslims. The conceptions remained low since late autumn and throughout the winter, irrespective of religion, culture, domestic economy, social status or geographic location; evidence for limited sexual activities during the colder months of a year because of hormonal or other physiological reasons are discussed in the introductory section of this paper (this limitation besides humans has been found in other biological species; see Retana-Marquez et al., 2003;Demir et al., 2016). Thus, while religious obligations seem to be essential initially, the observed similarity with the practices of the Muslim populations denotes the multivariate nature of the phenomenon and weakens but does not completely negate the effect of religious restrictions on it. As a matter of fact, after the Greek Orthodox Easter, usually in April, the number of conceptions increases rapidly in Dion, this being direct evidence of the religious festivities on sexuality. People of the general samplefollow with a one month delay, as do the other populations. However, the early increase in child conceptions in Dion highlights the agricultural orientation of the region where, as a lowland area, the growing season begins earlier compared to the mountainous areas of Greece. The conceptions reach a peak there during May and, afterwards, they decrease rapidly until almost October. Similarly, in the General sample whose members are scattered in various regions of Northern Greece mainly, the delayed increase of conceptions in May is accompanied by a decrease towards July and a relative stability until October. In the mountainous populations, conceptions are high in late spring. After a summer plateau, a second peak is observed in August-September in Rhodopi and in October in Metsovo. The coefficient of variation within each population and period may indicate the observed heterogeneity among the populations (Figure 8). In the first period (1951)(1952)(1953)(1954)(1955)(1956)(1957)(1958)(1959)(1960)(1961)(1962)(1963)(1964), the seasonality is intense in all populations but less so in the general sample. In the following period (1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976)(1977)(1978)(1979)(1980)(1981)(1982)(1983)(1984), the seasonality of conceptions declines considerably except maybe for Dion and, to a lesser degree, for Metsovo. The situation is somewhat similar to the previous period (1951)(1952)(1953)(1954)(1955)(1956)(1957)(1958)(1959)(1960)(1961)(1962)(1963)(1964); conceptions are relatively high during the late spring/summer months (May to August). In the following period (1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992), seasonal fluctuations become milder, and the pattern of seasonality becomes less clear in most populations (Figures 7 & 8). However, June is still the month exhibiting high seasonal indices of conceptions for all populations; the broadest range in seasonal indices is observed in Metsovo and Kehros. On the other hand, February and October show a relative drop in conceptions in all populations except for the region of Dion. In the last period (1993 onwards), there were insufficient or no data on conceptions for Organi and Kehros; only for the General sample and the populations of Metsovo and Dion it was possible to estimate seasonal indices. In Metsovo, which exhibits the largest range of seasonal fluctuations, an elevated number of conceptions is observed in September and during winter (November and December). In the General sample, a high number of conceptions is observed during the autumn months of September and October, while in Dion, only in May a modest rise of conceptions can be observed. --- Analysis across periods Examining the seasonal indices over time, we see several differentiations both between populations and between periods. In general, however, it appears that there is a tendency for an increase in the intensity of conceptions among the mountainous populations during the late spring/summer season, irrespective of people's religion. Looking at the monthly data on conceptions, it is noticeable that the range of seasonal fluctuations over time tend to decrease in all samples (Figures 7 and8). The last period (1993 onwards) is perhaps an exception for the General Sample and the region of Metsovo, but this may be a feature of this period or may be due to the realisation of a smaller number of conceptions compared to previous periods. In fact, the small number of births during that period may have affected the observed seasonality patterns, but to an unknown extent. The estimated seasonal indices, which portray the relative excess or deficit of the conceptions by month, also depict similar patterns of variation of the events across different periods. --- Discussion and conclusions The present study uses vital registration data regarding five distinct population subgroups of mainland Greece (residing in Metsovo, Dion, Organi, Kehros, and a General Sample) to explore the seasonality of conceptions for the period 1951-2002. These populations exhibit different characteristics concerning the area of residence (Metsovo, Organi and Kehros are mountainous areas whereas Dion is situated in a plain), religion, socio-economic status, employment activities and certain traditions. The analysis aims to add evidence on this sparcely researched topic for Greece. The findings indicate that there is seasonality of conceptions in all populations under study, more prominent in the period 1951-64. Seasonality decreases over time, especially from 1981 onwards, though at a different pace for the different populations. The extent of seasonality differentiates across populations; it is less pronounced in the general sample throughout the whole period under investigation. Conceptions are at low levels during winter and early spring in all populations. Among mountainous populations, there is a tendency of an increased number of conceptions during late spring and early summer, irrespectively of religion or socio-economic status. By contrast, in lowland areas (Dion) the increase in conceptions is observed slightly earlier (in April), reflecting both religious practices (greater sexual activity following Lent) as well as an earlier onset of agricultural activities in plains compared to mountainous areas. Qualitative fieldwork evidence suggests that the older women of the christian populations considered sexual relations during Lent a supreme sin and avoided them. Of course, even in the strictest environments, people can often act and go beyond the rules. a fact which permits several conceptions to occur during that time. As religiosity declined over time, younger generations were not that keen on following cultural and religious norms. In Muslim populations, any effort to identify religious effects on seasonality is dubious because religious holidays have a mobile character following the Gregorian calendar. For example, the Ramadan festivities, a period of daily fasting, may occur in different months during subsequent years. Hence, it seems that people with different religions and ecological landscapes have significant similarities in their sexual and reproductive activities. Therefore, the Greek Orthodox Easter could partially contribute as an inhibiting factor of procreation for the Christians, but other factors seem equally essential, indicating the multivariate pattern of human existence. The case of Metsovo is indicative of this complexity. While in the first period of study Metsovo follows Dion, though with a significant time lag in the Lent period, in the second period under study (1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976)(1977)(1978)(1979)(1980), conceptions increase in Metsovo from April to May, but remain below average. A brief discussion of such factors follows. The first such factor reflects the agricultural production cycles, which differ between the mountains and the plains: the intense agricultural work starts earlier in the lowlands. However, is this a matter of time or climate, as found for several populations worldwide (see introduction)? During those times, most people in Dion were growers of tobacco or cereals and several vegetables, while a few were cattle breeders, having abandoned their previous livestock and mountainous life (Zafeiris and Kaklami, 2019). By contrast, Metsovo retained its character as a Vlachic Metropolis (Zafeiris et al., 2015). A variety of occupations characterised its population. Besides the limited cultivations, people were still stockbreeders, woodcarvers, conductors, and merchants. Pomaks, being in the worst economic situation compared with the others because of their geographical and cultural isolation, were small farmers, stockbreeders and loggers. They were trying to survive in an adverse environment without actively participating in the market economy (Zafeiris, 2020;Zafeiris and Koukli, 2022). Besides this heterogeneity, all the mountainous populations have several analogies in their procreational activities. Thus, climate and several other ecological factors (like the type of ecological habitat) are mainly responsible for the monthly distribution of conceptions irrespectively of their socio-economic characteristics. As ethnographic evidence collected during fieldwork suggests, another factor must be considered: the mechanization of production first in the Christian populations and later in the Muslim ones. However, the lack of solid evidence does not allow an estimation of its effects on seasonality. Additionally, the harsh winters in the mountains may have affected seasonality patterns. The working cycles differ significantly between the study's mountainous and plain populations. A typical example is that the mountain goats and sheep are born much later each year than in the lowlands. Thus, as ethnographic evidence suggests, the working cycles are intensified later in the mountains. However, the workload in the fields permits the couple to find time and space for sexual intercourse, freed from the confines of his household. Many women expressed their despair at the lack of freedom and the total embrace of their family during the winter, when all members of a household lived together within the extended family. After all, the extended family is the typical family formation in all populations, except perhaps for the General sample. In this system, the couple, their male progenies with their wives and grandchildren, often share the same room under the same roof, which is divided into individual spaces usually using rugs. If the family's economic situation allows, new adjusted rooms can be built, in which the two spouses live together with their children, but separately from the other couples of the extended family, yet still under a common or neighbouring roof (see Zafeiris et al., 2015;Zafeiris and Kaklamani, 2019;Zafeiris, 2020;Zafeiris and Koukli, 2022). Obviously, in this limited space, considering that the bad weather inhibits life outside home, the possibilities for sexual relations are significantly limited. The couple feels freer when working in the countryside, and thus its sexual activity increases. Moreover, fieldwork evidence suggests that couples need to plan for the "arrival" of a new child. This is more beneficial when there is no demanding farm work as, on the one hand, the child will enjoy his mother's attention to a greater extent, and, on the other hand, it will have grown to some extent when his mother works in the fields a few months after its birth. Thus, people recognized this necessity, and they acted accordingly. Additionally, a mother should not be in advanced pregnancy when she needs to work in the fields. After all, family planning seems necessary in these populations, regardless if it is socially accepted or recognized. One could suggest that these populations should have been in a natural fertility state where family planning was absent. However, data not cited here revealed that though a couple had their first child nine months after marriage or soon thereafter, the couple could regulate the length of their reproductive life, stopping it when the desired number of progenies was achieved while also practising birth spacing (Zafeiris et al., 2015;Zafeiris and Kaklamani, 2019;Zafeiris, 2020;Zafeiris and Koukli, 2022). Fieldwork evidence suggests that coitus interruptus and abstinence from sexual intercourse were widespread methods. Therefore, the timing of a birth depended on various physiological, ecological, economic and cultural characteristics, not to mention the estimated workload by the time of the birth. Nevertheless, in more recent times, as economic development and modernization took hold, traditional patterns became gradually less important. The dependence of these populations on agricultural and livestock activities for their survival decreased, as did their cultural isolation and traditionalism. Consequently, seasonality of conceptions declined substantially, as the relevant circumstances changed.

The paper studies seasonality of conceptions among five distinct population subgroups of mainland Greece for the period 1951-2002. The populations explored include those residing in Metsovo, Dion, Organi, Kehros, as well as a "General" Sample consisting of persons located in various areas of continental Greece. The populations under investigation present diverse characteristics regarding religion, cultural background, socioeconomic status etc. Records of births were derived from the Vital Registration System of the respective municipalities and communities of the populations under research were constructed. The date of child conception was estimated as the recorded date of birth minus 260 days. The analysis focuses, among others, on the construction of seasonal indices, applying a variant ratio to moving averages method which reveal, in relative terms, the seasonality of the phenomenon. Subsequently, these ratios are considered as the dependent variable in regression models while months, expressed in terms of dummy variables, are introduced as predictors. Four main sub-periods are considered; 1951-64, 1965-80, 1981-92 and 1992-2002. The findings show that the extent of seasonality differs between periods as well as between the five population subgroups though the phenomenon becomes less prominent over time in all cases. There is a tendency of an increased number of conceptions among mountainous populations during summer, irrespective of religion or socio-economic status, possibly partly due to environmental factors (i.e. seasonal workload, domestic organisation of extended families, etc). Nevertheless, the mountainous populations differ regarding the intensity and duration of this phenomenon. By contrast, in Dion, a lowland Christian Orthodox population, conceptions increase after Easter and remain elevated until June.

as the repercussions of immigration on the characters' lives and societal dynamics. Moreover, by integrating theories by Pierre Bourdieu and pertinent literature on social class, cultural capital, and habitus, the study delves into the film's portrayal of social class and inequality, illuminating the intricate interplay between cultural assets, social stratification, and the quest for distinction. Through a rigorous examination of the film, the study underscores the ways in which Rawuh (2019) intersects with and contributes to ongoing discourses on immigration, labor, and global economic forces, accentuating the film's Unravelling the Cinematic Tapestry of Immigration: A Socio-Cultural Disccection of Indonesian Migrant Labor in Rawuh (2019) Introduction Immigration has become a prominent issue in Indonesia, with a significant number of people leaving the country to work as migrant laborers in other parts of the world. The issue of immigration has been explored in Indonesian cinema, with a number of films focusing on the experiences of migrant workers and their families. This article aims to examine the portrayal of immigrants in Indonesian cinema, with a specific focus on the local film entitled Rawuh (2019). The local Indonesian film Rawuh (2019), directed by Agung Kurwiawan, presents a compelling narrative centered around Jumarno, a materialistic cattle broker, and the anticipated return of his eldest son, Zainu, from working as a migrant worker in Malaysia for over two years. The film delves into the complex dynamics of social class, materialism, and the impact of migrant labor on family and community relationships. Rawuh (2019) has made remarkable strides in film festivals both locally and internationally. The film's realistic portrayal of social class, materialism, migrant labor, and gender dynamics in a rural community setting has resonated with audiences and critics alike, earning it numerous accolades and recognition at various film festivals. Some notable achievements of Rawuh (2019) Through an analysis of Rawuh (2019), this article contributes to the broader discourse on immigration in Indonesia, shedding light on the complex and often ambiguous representations of migrant laborers in Indonesian cinema, especially representing Jember, Indonesia. By engaging with existing literature and using the framework of discourse analysis, this article offers a nuanced and insightful analysis of the portrayal of immigrants in Rawuh (2019), and the larger socio-political and cultural forces that shape this representation. The achievements of Rawuh (2019) warrant a thorough analysis using both Bourdieuian lense, as this theoretical framework provide valuable insights into the socio-cultural dynamics present in the film. Pierre Bourdieu's theories on social class, habitus, and capital can offer a comprehensive understanding of the film's exploration of materialism and social stratification (Bourdieu, 1984;Bourdieu & Passeron, 1977). Specifically, the film highlights the tension between different forms of capital, as Jumarno's pursuit of material wealth stands in stark contrast to the unexpected outcome of Zainu's migrant labor experience. Rawuh 2019 invites viewers to reflect upon the role of habitus in shaping the characters' attitudes and behaviors, as well as how these dispositions influence their perception of success and social standing (Bourdieu, 1977). By analyzing Rawuh (2019) through the lens of Bourdieu's theoretical framework, scholars can gain a deeper understanding of the film's thematic elements and its socio-cultural implications. This approach allows for a more comprehensive examination of the characters' motivations, behaviors, and the impact of their actions on their community, while also shedding light on the broader societal context within which the narrative unfolds. --- Research Method This study employs a methodical, multi-faceted approach to analyze the representation of immigrants in the Indonesian film Rawuh (2019) and its socio-cultural implications. The research methodology, drawing on Creswell's (2013) qualitative research principles, is organized ISSN 2963-0835 Ni Luh Ayu Sukmawati -OJS -ROLLING into distinct stages to ensure a comprehensive and nuanced understanding of the subject matter. Initially, an extensive literature review will be undertaken to survey existing research on Indonesian cinema, immigration, and migrant labor. This literature review aims to provide the necessary contextual foundation for the analysis of Rawuh (2019) by examining academic studies on the depiction of immigrants in Indonesian cinema and scholarly work on the wider socio-political and cultural forces that influence these representations (Creswell, 2013). Subsequently, a detailed film analysis of Rawuh (2019) will be conducted, concentrating on the portrayal of immigrants and the themes of materialism, social class, and migrant labor. The film's narrative structure, character development, visual elements, and cinematography will be meticulously analyzed to identify the salient messages and themes conveyed. This analysis will serve as the foundation for the subsequent application of Bourdieuian. In the third stage, Pierre Bourdieu's theoretical framework on social class, habitus, and capital (Bourdieu, 1984(Bourdieu,, 1986) ) will be applied to scrutinize the representation of materialism and social stratification in Rawuh (2019). The study will examine the interplay between various forms of capital and the role of habitus in shaping the characters' attitudes, behaviors, and perceptions of success and social standing. The framework of discourse analysis will be utilized to examine how the film's portrayal of migrant workers intersects with broader discourses on immigration, labor, and global economic forces. The contribution or challenge of Rawuh (2019) to existing narratives on these topics will be assessed. In the penultimate stage, semi-structured interviews with filmmakers, critics, scholars, and experts in the fields of Indonesian cinema and immigration studies will be carried out to gather their insights and opinions on Rawuh (2019) and its portrayal of immigrants, in line with Creswell's (2013) recommendations for qualitative research. Finally, the key findings and insights garnered throughout the ISSN 2963-0835 Ni Luh Ayu Sukmawati -OJS -ROLLING research process will be encapsulated and discussed, focusing on the implications of the study for the wider discourse on immigration in Indonesia and the representation of migrant laborers in Indonesian cinema. Additionally, areas for future research and potential limitations of the study will be delineated. --- Finding and Discussion --- Portrayals Of Immigration and Migrant Labor An in-depth analysis of Rawuh (2019) focusing on portrayals of immigration and migrant labor can be conducted through a close examination of the film's mise-en-scène, which encompasses various aspects of the visual presentation, such as setting, lighting, costume, and props. Drawing on the works of Bordwell ( 2008) and Galt and Schoonover ( 2010), this analysis explores how these elements contribute to the overall narrative and thematic development of the film, providing insight into the experiences of Indonesian migrant workers and the impact of immigration on their families and communities. --- Setting The rural setting of Rawuh ( 2019) is integral to the portrayal of immigration and migrant labor. The agricultural community, with its limited economic opportunities and the omnipresent desire for upward social mobility, serves as a backdrop for the characters' motivations to seek work overseas (Galt & Schoonover, 2010). The contrast between the rural environment and the imagined prosperity of Malaysia, where Jumarno's son Zainu works as a migrant laborer, highlights the driving forces behind the decision to immigrate in search of better opportunities. Jumarno's fixation on the return of his eldest son, Zainu, further emphasizes his unyielding pursuit of material wealth. Throughout the film, Jumarno eagerly awaits Zainu's return, as he believes that the money Zainu has earned as a migrant laborer will significantly contribute to the family's economic capital. Jumarno's anticipation is fueled by the potential increase in social standing that this additional wealth may bring. Ni Luh Ayu Sukmawati -OJS -ROLLING He envisions a better life for his family, where they can enjoy the fruits of their labor and gain recognition within the community. Moreover, Jumarno's pursuit of material wealth manifests in his interactions with other community members. He often boasts about his business successes and the accomplishments of his son as a migrant laborer, seeking validation and acknowledgment from his peers. This behavior highlights Jumarno's desire to use his material wealth as a means of asserting his social position and garnering respect from others. In addition, Jumarno's pursuit of material wealth is further complicated by the film's portrayal of the socio-cultural dynamics within the community. The film illustrates how the community's values and norms prioritize material success and social standing, further reinforcing Jumarno's ambitions. This context reveals the extent to which Jumarno's pursuit of wealth is shaped by the societal expectations and pressures that surround him. Jumarno's pursuit of material wealth in Rawuh ( 2019) is a multifaceted aspect of his character, reflecting his ambitions to accumulate economic capital and improve his social standing. His relentless drive for wealth is evidenced through his profession as a cattle broker, his fixation on Zainu's return, and his interactions with community members. By exploring these dimensions, the film offers a compelling portrayal of the complex interplay between individual aspirations and the socio-cultural forces that shape them. Tabel 1.1 The Complex Interplay Among Social Class, Habitus, and Capital in Jumarno Habitus Attitudes, Behaviours, and Disposition --- Jumarno's Pursuit of Materialism --- Economic Capital Wealth and Income --- Cultural Capital --- Recognition and Prestige --- Social Class Position in Society ISSN 2963-0835 Ni Luh Ayu Sukmawati -OJS -ROLLING Jumarno's aspirations to improve his family's social standing are evident in his constant efforts to accumulate wealth through cattle trading. This demonstrates his habitus, shaped by the rural environment and its limited economic opportunities. Similarly, Zainu's decision to work as a migrant laborer in Malaysia reflects the shared habitus among the community members, who view migration as a path to financial success and social mobility. Bourdieu's concept of habitus refers to the socialized norms and dispositions that individuals develop as a result of their upbringing and environment (Bourdieu, 1990). The rural setting and the limited economic opportunities available to the characters contribute to the formation of their habitus, which is characterized by a strong desire for upward social mobility. This desire leads the characters to seek work overseas, as they perceive it as a means to accumulate economic and social capital. The concept of cultural capital, as introduced by Pierre Bourdieu (1986), plays a significant role in understanding the social dynamics and aspirations of the characters in Rawuh (2019). Cultural capital can be defined as the set of cultural assets and knowledge that individuals can use to improve their social standing. Through various examples in the film, we can observe how the characters seek to accumulate cultural capital as a means to elevate their social status and distinguish themselves from others within their community. In one notable scene from the film as provided in figure 9, Jumarno and his wife decide to organize an exclusive circumcision ceremony for their second son, diverging from the more common practice of mass circumcision events in their community. This decision to opt for a private and exclusive ceremony, performed by a doctor using an advanced and painless laser technique, can be interpreted as an attempt to accumulate cultural capital (Bourdieu, 1986). The choice of a more expensive and modern method of circumcision reflects the family's desire to adopt urban values and practices that they believe will set them apart from their neighbors and elevate their social status (Bourdieu, 1984). Ni Luh Ayu Sukmawati -OJS -ROLLING Jumarno's wife's, Sulastri excessive preparations for the ceremony, such as purchasing an extravagant amount of groceries for catering purposes, further highlights their pursuit of distinction (Bourdieu, 1984). This display of wealth showcases the family's improved economic situation, primarily due to Zainu's earnings from his work in Malaysia, and their intention to signal their elevated social status to the community. Moreover, the presence of a renowned local religious leader (Kyai) at their son's circumcision ceremony serves as another example of the family's efforts to accumulate cultural capital (Bourdieu, 1986). Inviting the respected Kyai not only signifies the family's religious devotion but also acts as a marker of their social standing within the community. The film critically examines social class and inequality by portraying the challenges faced by rural communities, such as Jumarno's family, who grapple with scarce resources and opportunities (Suryahadi et al., 2009). Jumarno's attempts to accumulate wealth often encounter setbacks, like losing cattle to illness, which underscores the persistent social and economic inequalities within Indonesia (Yeoh & Ramdas, 2014). Moreover, the film highlights the contrasting lifestyles of those who have successfully migrated, such as Zainu's friend, who returns with motorcycle and fashionable clothes, representing the potential rewards of migration and the existing disparities within the community (Silvey, 2004). However, there are cultural differences that emerge, such as Zainu returning to Jember with a wife and two children, with the latter being born out of wedlock, which goes against traditional customs that require marriage to be officiated with witnesses and parental presence (Smith-Hefner, 2005). This cultural discrepancy reveals that not all immigrants can bring pride in the context of cultural capital (Bourdieu, 1986). In fact, there are realities that need to be faced when returning to one's hometown. Some scenes also show that Zainu struggles to make a living in Malaysia (Hugo, 2007). This nuanced portrayal of the challenges faced by migrants highlights that while migration can lead to economic gains and social mobility, it can also result in cultural clashes and unanticipated difficulties (Carling, 2002). In this way, the film provides a comprehensive and critical exploration of social class and inequality, illustrating the complex and multifaceted experiences of individuals impacted by migration. --- ISSN 2963-0835 Ni Luh Ayu Sukmawati -OJS -ROLLING Challenging Assumptions: A Critical Analysis of Migrant Worker The portrayal of migrant workers in Rawuh (2019) intersects with and contributes to existing discourses on immigration, labor, and global economic forces in several ways. By delving into the experiences of Indonesian migrant workers and their families, the film provides a nuanced and realistic representation of the complex dynamics that underlie international migration. Firstly, Rawuh (2019) contributes to the discourse on immigration by showcasing the motivations behind individuals' decisions to migrate. As discussed earlier, the film highlights the limited economic opportunities and the desire for upward social mobility that drive character like Zainu to seek work overseas (Suryahadi et al., 2009). This portrayal aligns with existing research on the "push" and "pull" factors that influence migration decisions, such as poverty, unemployment, and the prospect of better opportunities elsewhere (Carling, 2002). Secondly, the film's depiction of the challenges faced by migrant workers, including cultural clashes and struggles to earn a living, adds to the discourse on labor and migration. The experiences of Zainu and his friend demonstrate that the realities of working abroad can be far from the idealized expectations that many migrants hold (Hugo, 2007). This portrayal is consistent with research on the often difficult and exploitative conditions that many migrant workers face, particularly in low-skilled jobs (Silvey, 2004). Furthermore, Rawuh (2019) highlights the impact of global economic forces on migration and the perpetuation of social and economic inequalities within Indonesia (Yeoh & Ramdas, 2014). The film shows that rural communities, like Jumarno's, are disproportionately affected by these forces, with limited resources and opportunities contributing to the cycle of poverty and inequality. The experiences of Zainu and his friend, as portrayed in Rawuh (2019), align with the findings of Lindquist ( 2010), who notes that Indonesian migrant workers, particularly from rural areas like Jember, often face harsh realities and exploitative conditions in their destination countries, such as Malaysia. This situation is particularly prevalent among low-skilled workers who are vulnerable to deception and abuse by unscrupulous recruitment agencies. This perspective supports existing literature on the role of global economic forces in shaping migration patterns and exacerbating social and economic disparities both within and between countries (Carling, 2002). Rawuh --- ISSN 2963-0835 Ni Luh Ayu Sukmawati -OJS -ROLLING

This article addresses the representation of Indonesian migrant labor experiences, particularly from Jember, in local cinema by presenting a multifaceted analysis of the film Rawuh (2019). The study emphasizes the film's depiction of socio-cultural dynamics and its contribution to existing discourses on immigration and migrant labor. Utilizing close textual analysis and scrutinizing mise-en-scène elements such as setting, lighting, costume, and props, the paper probes the film's representation of the rural community's aspirations for upward social mobility, as well

INTRODUCTION When compared to renting, home-ownership is associated with positive outcomes at the individual level. There is quantitative evidence that home-owners have higher life satisfaction (e.g. Zumbro, 2014), mental health (Manturuk, 2012), and ontological security (e.g. Saunders, 1990); and that children of home-owning parents do better at school (e.g. Haurin et al, 2002;Green and White (1997). 1 The literature typically attributes this tenure gap to three absolute benefits of home-ownership (Zumbro, 2014). First, people have a natural possessive instinct and a desire to mark out their own territory which home-ownership fulfils (e.g. Saunders, 1990;Lindblad and Quercia, 2015). Second, home-ownership improves living conditions because home-owners have a greater financial stake in their home than renters (Galster, 1983). Third, homeownership can offer greater security, as home-owners cannot be involuntarily moved from their home by a landlord. Some sociologists (e.g. Saunders, 1990) contend that this preserves the 'ontological security' of home-owners. These absolute benefits have arguably influenced governments to make expanding rates of homeownership a key policy goal. For example, a press release from the U.S. Department of Housing and Urban Development (2000) cited the tenure gap in educational outcomes as justification for policies supporting home-ownership (Harkness and Newman, 2003). Similarly, in the UK, where this paper is based, policy documents have invoked the absolute benefits in justifying policies to expand homeownership rates (see Gurney, 1999) which currently sit at 65 percent (English Housing Survey, 2015) In this paper, we contend that the tenure gap may actually be attributable, at least in part, to the relative benefits of home-ownership. The study of the relative benefits of consumption has emerged as a key area of research across the social sciences. 2 The idea that relative income and relative consumption matters is not new. For example, Adam Smith (1776) recognised that there were certain 'necessities' without which an individual would feel shame. Given that housing is the largest form of consumption, it is striking that so little attention has been devoted to examining the importance of relative housing conditions. To facilitate a thorough theoretical and empirical study, we limit our scope to the relative benefits of housing tenure in terms of subjective well-being. We test for two types of relative benefits. First, we follow previous authors in conceptualising homeownership as a social norm: home-owners benefit from being considered 'normal' by society, versus renters who are considered 'abnormal' 3. We hypothesise that if home-ownership is a social norm, then the magnitude of the relative benefits of home-ownership should be positively related to the homeownership values of relevant others: the stronger the social norm of home-ownership among one's friends and family, the higher subjective well-being home-owners will have, and the lower subjective well-being renters will have. To our knowledge, we are the first to additionally conceptualise home-ownership as a positional good. Home-owners not only benefit from being 'normal', but also through being considered to have higher relative wealth, and in turn, higher status than renters. If home-ownership is a positional good, then we hypothesise that the subjective well-being of owners (and renters) should be negatively related to the home-ownership rates of relevant others. When home-ownership is expanded, this reduces the wealth that home-ownership (and renting) signals, thereby decreasing the status and subjective well-being of the original home-owner (or renter). The crucial point about relative benefits-as opposed to absolute benefits-is that they necessarily come attached with relative costs. By conforming to the social norm, home-buyers benefit from being 'normal', but it comes at the expense of renters who are increasingly 'abnormal'. By signalling their wealth, home-buyers benefit from increased status, but because status is an inherently relative concept, this comes at the expense of others whose status necessarily decreases. Therefore, if the tenure gap in outcomes is due to the relative benefits of home-ownership then expanding rates of home-ownership will have a much weaker overall effect on societal well-being, than if they were due to the absolute benefits. Understanding whether the tenure gap in outcomes is primarily attributable to the absolute or the relative benefits is vital because the policy implications of the two logics are very different. The paper is structured as follows. First, we conceptualise home-ownership as a social norm, and develop our first hypothesis. Next, we additionally conceptualise home-ownership as a positional http://mc.manuscriptcentral.com/cus [email protected] Urban Studies good, and form hypothesis two. We then outline the data and methodology before presenting and discussing our results. We conclude with policy implications and an agenda for future research into the relative benefits of home-ownership and housing more generally. --- III. LITERATURE REVIEW AND THEORETICAL FRAMEWORK --- Home-Ownership as a Social Norm Of the three papers that have formally conceptualised and tested the relative benefits of homeownership, all have adopted a social norm framework (Gurney, 1999;Cohen et al, 2009;Knight, 2002). There is no standard definition of a'social norm' but definitions tend to imply three distinct elements (Hechter and Opp, 2001): they should be behavioural regularities; there should be a sense of 'oughtness' with them; and there should be sanctions associated with conforming or deviating. Sanctions can be rewards or punishments. Significantly, according to the economic model of Akerlof (1980), the strength of the sanctions associated with conforming/not conforming will be in proportion to the strength of 'oughtness' among one's'relevant others'. In this paper, we define'relevant others' as those persons an individual interacts with, seeks the respect of, or compares themselves to. Generally, these will encompass an individual's friends, family, work colleagues and neighbours. Homans (1974, p.150) recognises the influential role this group plays in enforcing social norms: "If a group can offer much in the way of friendships to its members, it can exert much control over them, since it can deprive them of much if they don't conform". So is home-ownership a social norm in the UK? It is certainly a behavioural regularity. Despite recent declines, in 2013-14, 65 percent of households owned their own homes (English Housing Survey, 2015). Furthermore, there is a sense of oughtness to do with home-ownership. Becoming a homeowner has been variously described as a 'rite of passage' (Dupuis and Thorns, 1998) and a 'badge of citizenship' (Murie 1998). Gurney (1999) examined the language used in landmark housing policy documents, and among homeowners in Bristol, to identify three discrete discourses which normalise home-ownership. First, the dwellings of home-owners are imbued with the warmth and security of <unk>home' whilst renters' dwellings are described in more spartan terms. Second, home ownership is associated with a set of values that constitute a 'good citizen'. Third, homeownership is viewed as meeting a deep and natural desire for independent control of one's living environment. This third discourse is likely to be age dependent (Knight, 2002). For students and unmarried young adults, renting is likely to be considered 'normal' but past a certain point-Knight suggests the age of 30renting privately becomes a deviation from schedule and 'abnormal'. All three discourses work together to cast home-ownership as 'normal' and renting as 'abnormal'. The norm against social rental is likely to be stronger still. Not only are social renters on the wrong side of Gurney's three discourses, but they are also perceived as being dependent on the state (e.g. Robinson, 2013). The strength of the sanctions associated with (not) conforming will depend on the strength or 'oughtness' of the social norm among relevant others. To illustrate, take 1950's Britain. There were social norms against both homosexuality and holding your knife and fork in the wrong hand but the strength of the social norm-and the sanctions involved -were much greater in the case of the former than the latter. Similarly, the norm in favour of home-ownership -and the sanctions associated with renting or owning-will vary in strength between different groups of relevant others. Consistent with this logic, Cohen et. al (2009) conducted a longitudinal study of 919 low-moderate income renters in the USA, and found the home-ownership values of 'important others' to be associated with stronger homeownership intentions. Respondents were more likely to intend to buy a home if they thought 'important others' were of the opinion that the respondent should buy a home. Furthermore, stronger home-ownership intentions were associated with a higher likelihood of becoming a home-owner in the future. http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies Sanctions can be rewards or punishments, and can be imposed directly or indirectly. If a renter's relevant others regularly discuss the merits of home-ownership or the disadvantages of renting then the renter may internalise this discourse, and feel a sense of shame for being 'abnormal'. Inversely, an owner may feel a sense of pride and confidence (see Gurney, 1999). These are referred to as internal sanctions. Relevant others could also affect an individual's subjective well-being directly through granting admiration or deference to owners, or through withholding respect from renters. These are defined as external sanctions. Sanctions could also be more material. A person of high rank can expect to be treated favourably by other individuals with whom he might engage in social and economic interactions (Hirsh, 2005;Frank, 2007). For all the sanctions above to exist, an individual's housing tenure must be known by their relevant others. This may seem unlikely given housing tenure is intangible. However, the status that parents' derive from their children's level of education (Solnick and Hemenway, 1998) indicates that something need not be tangible to carry relative benefits, it just needs to be known by others. Our contention is that if someone undergoes a tenure transition, then considering the social significance of the life event, this information is likely to be dispersed among their relevant others. Because of the accompanying sanctions, 'being normal', and social status more generally, is important for a number of outcomes including adult health (see Wilkinson and Pickett, 2006 for a review), educational outcomes (e.g. Hoff and Pandey, 2004) and, most relevantly, subjective wellbeing (e.g. Clark, 2003). The social norm literature therefore predicts that an individual's housing tenure indicates, in part, whether they are 'normal' or 'abnormal'. The extent of the relative benefits (or costs) associated with being a home-owner (or renter) depend on the strength of the homeownership values of relevant others. Thus, we have our first hypothesis: if home-ownership is a social norm, the home-ownership values of relevant others will be positively related to the subjective wellbeing of owners, but negatively related to the subjective well-being of renters. But what about the home-ownership rates of relevant others? If more of one's relevant others are homeowners, does this impact the status of owners and renters? --- Home-Ownership as a Positional Good Veblen's (1899) theory of conspicuous consumption, Duesenberry's (1949) relative income theory and Hirsh's (1976) theory of positional goods all make the point that utility depends on relative, as well as absolute, consumption. Relative consumption matters because it signals the consumer's relative wealth. Relative wealth, in turn, matters because it indicates one's power over others (Csikszentmihalyi and Rochberg-Halton, 1981) and, in many cases, one's natural ability (Frank, 2007). It is therefore a key determinant of social status. Being able to purchase one's own home requires a greater level of wealth than renting in the private sector, which in turn requires a greater level of wealth than renting social housing. Thus becoming a home-owner signals an increase in relative wealth, The average home owner is higher status, better paid, better educated, richer and more middle class... consequently the change from tenant to home owner increases the likelihood that the individual will be taken to be well paid, well educated and middle class. (Marcuse, 1975 p. 195) However, as the proportion of the population who can access home-ownership increases, the relative wealth that home-ownership signals will decrease (Forrest and Murie, 1983). By the same logic, expansion of home-ownership will also decrease the relative wealth with which renting is associated. This process has been most noticeable in the social housing sector. Between 1981 and 2011, the size of the social rental sector decreased from 31% to 18% as skilled workers moved to home-ownership (Jones and Murie, 2006). Whereas council tenants in England in the 1950s and into the 1960s were considered by society to be relatively affluent, now the perception is of an economically inactive underclass (Watt, 2008). The same logic also applies to private renting. Those who move from private renting to homeownership need capital. This means that while relative to other owners, they are likely to be poor; compared to other renters, they are likely to be wealthy. By leaving the private rental sector, they will lower the relative status of those left behind. In the Spanish context, Vera-Toscana and Ateca-Amestoy ( 2008) found that renting in a predominantly home-owners' neighbourhood had a significant negative effect on housing satisfaction "It does not seem to be the fact of being an owner vs. nonowner what causes satisfaction or dissatisfaction" they noted "but the fact of being the renter surrounded by home-owners". We therefore propose that home-ownership is a positional good, which Frank (1985) defines as a good "whose value depends relatively strongly on how they compare with things owned by others." Becoming a home-owner will increase an individual's status as it signals an increase in their relative wealth, but the price will be paid by everyone else whose status necessarily declines. We thus have our second hypothesis: if home-ownership is a positional good, then home-ownership rates among relevant others will be negatively related to the subjective well-being of both home-owners and renters. --- IV. DATA AND METHODOLOGY In this paper, we test both hypotheses using a quantitative methodology. We recognise this approach has limitations. Notably, we cannot explore what form sanctions take, nor who imposes them. However, we can more confidently test the causal relationships hypothesised above, and quantify their strength. Data is drawn from the British Household Panel Study (BHPS). Between 1991 and 2008, the BHPS conducted annual interviews on a nationally representative sample of the same 5500 private households, expanding to 9000 households by 2008. The BHPS includes a wide range of demographic variables. Our analysis is limited to the three waves which included information on the importance attached to home-ownership; 1998, 2003 and 2008. Our econometric analysis is based on the following specification: <unk> <unk> = <unk>(<unk> <unk> + <unk> <unk> + <unk> <unk> + <unk> ) (1) According to expression (1), for each individual i, subjective well-being at time t is labelled by <unk> <unk>, and is considered to be a function of relevant others' home-ownership values, V; relevant others' home-ownership rates or consumption, C; and a range of socio-demographic control variables, X. We adopt a fixed-effects regression analysis, which means that for each individual i, we estimate the effect of changes in the independent variables, from one wave to the next, on changes in the dependent variable. In doing so, a fixed-effects regression controls for all time invariant and individual-specific unobservables <unk>i, for example, baseline wealth, or how one understands the life satisfaction scale. In the past decade, subjective well-being indicators have gained legitimacy among researchers and policymakers. Subjective well-being has both an evaluative (cognitive/judgemental) component, and an experiential (emotional/affective) component, which when assessed together, are at least moderately correlated (Pavot and Diener, 1993). To proxy for the evaluative component of subjective well-being, we use responses to the question "How dissatisfied or satisfied are you with your life overall?" on a scale of 1-7. There is a large literature pointing to the validity of life satisfaction measures. See Diener et al (2013) for a review. The experiential component of subjective well-being is proxied for using the General Health Questionnaire (GHQ). The GHQ-12 reflects overall mental well-being. It is constructed from the responses to twelve questions (administered via a selfhttp://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies completion questionnaire) covering feelings of strain, depression, happiness, inability to cope, anxiety-based insomnia, and lack of confidence, amongst others. Responses are made on a four-point scale of frequency of a feeling in relation to a person's usual state: "Not at all", "No more than usual", "Rather more than usual", and "Much more than usual". The GHQ is widely used in medical, psychological and sociological research, and is considered to be a robust indicator of the individual's psychological state. This paper uses the Caseness GHQ score, which counts the number of questions for which the response is in one of the two "low well-being" categories. This count is then reversed so that higher scores indicate higher levels of subjective well-being, running from 0 (all twelve responses are "low subjective well-being") to 12 (no responses are "low subjective well-being"), after which the majority of respondents record a score of 12. In the absence of any literature to suggest otherwise, we assume both the evaluative and experiential components to move together in our analysis. Note that both our hypotheses are concerned with the effect of relevant others' home-ownership values/rates on those people who have stayed in the same tenure over time (as opposed to those people who have changed tenure). To this end, all regressions are run separately for the two groups; i) renters and ii) owners. We then split the renters sample into private and social renters in order to empirically distinguish between the sanctions associated with the two tenures. Throughout the analysis, we define those adults (>20 years old) who live in the same household as home-owning parents as renters, contrary to the BHPS which defines them as owners. Observations which relate to individuals aged under 21 are excluded from our analysis. One key step in our methodology is to define relevant others. Relevant others, or'reference groups' have been variously defined as people in the same country (e.g. Easterlin, 1974); region (e.g. Clark, 2003); and of a similar age (McBride, 2001). Others have used multiple criteria, such as Van de Stadt et al (1985) who define the reference group as people of a similar education level, age, and employment status. We follow Ferrer-i-Carbonell ( 2005) in identifying this group as those people with similar education (higher education, medium education or low education), inside the same age bracket (21-30; 31-40; 41-50; 51-65; 65+), and living in the same region (North UK 4 or South UK). We use these criteria for several reasons 5. We use the same BHPS data to quantify the attributes of relevant others. When we split the full sample of individuals according to the above criteria, we obtain thirty groups of relevant others, varying in size from 64 to 1012. Thus in each of the three waves, every individual in the sample is both an individual in their own right (i.e. we are interested in what affects their subjective well-being) and part of one group of'relevant others' (i.e. according to our hypotheses, they affect other people's subjective well-being). By growing older, becoming educated, or relocating, an individual can change their relevant others from one wave to the next. In each one of the three waves, individuals were asked to rate how important it was to own their own home on a scale of 1-10. This question is reproduced below, as it appeared to the interviewer (figure 1). INSERT FIGURE 1 ). Similarly, we use the home-ownership rates among relevant others as an indicator of the consumption of relevant others, C, with scores ranging from 18 percent to 93 percent. Note that the reference group is assumed to be exogenous, which is standard in empirical work (Ferrer-i-Carbonell, 2005). We control for changes in age, marital status, employment status, individual household income, housing costs (mortgage/rent), housing problems which have been found to influence subjective wellbeing (damp and noise), year, region, year*region interaction terms (which control for changes in house prices/rents at the regional level) along with other variables listed in appendix 2. Importantly, we also control for the mean household income of relevant others, in order to ensure that the effect of changes in relevant others' home-ownership values and consumption are independent of changes in relevant others' income6. The data used in this paper were extracted using the Add-On package PanelWhiz v4.0 (Oct 2012) for Stata. PanelWhiz was written by Dr. John P. Haisken-DeNew ([email protected]). The PanelWhiz generated DO file to retrieve the BHPS data used here and any Panelwhiz Plugins are available upon request. Any data or computational errors in this paper are our own. Haisken-DeNew and Hahn (2010) describe PanelWhiz in detail. --- V. RESULTS --- Relevant others' home-ownership rates (or consumption) We first describe the effect of relevant others' home-ownership rates on the subjective well-being of owners and renters. The coefficients are reported in the second row of table 1. Consistent with the positional good hypothesis (hypothesis one), we find that for home-owners, changes in homeownership rates among relevant others are negatively related to changes in life satisfaction (p<unk>0.05). The magnitude of this effect is substantial: an increase in relevant others' home-ownership rates from the 25 th percentile (56 percent) to the 75 th percentile (85 percent) of the sample, would lead, on average, to a decrease in owners' life satisfaction of -.0.11. To put this effect into context, Zumbro (2014) found becoming a home-owner with a low financial burden (in Germany) led to an increase in life satisfaction of only 0.09 (and this was on a 10 point scale!). As another benchmark, Fujiwara (2013) found that when respondents in BHPS began reporting damp and neighbour noise, each was associated with a -0.05 decrease in life satisfaction, and these were the two most damaging housing problems. The coefficient on relevant others' consumption is also negative when GHQ caseness is used as a dependent variable, but the relationship is not statistically significant. For renters, we find no negative impact of relevant others' home-ownership rates on individual subjective well-being7. This holds even when we split renters into social renters and private renters -see appendix 4. Together, these findings imply that home-ownership is a positional good, but only for owners. --- INSERT TABLE 1 HERE --- Relevant others' home-ownership values http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies Our results also imply that home-ownership is a social norm. The coefficients on the third row of table 1 show that, for owners, changes in the home-ownership values of relevant others are positively related to changes in subjective well-being, both in terms of life satisfaction (p<unk>0.01) and GHQ caseness (p<unk>0.05). The magnitude of these effects are also meaningful. If the importance of homeownership to a home-owner's relevant others increased from the 25 th percentile (7.68) to the 75 th percentile (8.08), this would lead to an increase in the home-owner's life satisfaction of 0.06 and an increase in GHQ caseness of 0.15. For renters, the negative effects of relevant others' home-ownership values on subjective well-being are also statistically significant and substantial. If the importance of home-ownership to a renter's relevant others increased from the 25 th percentile to the 75 th percentile, this would lead to a decrease in the renter's life satisfaction of -0.12 (p<unk>0.05) and a decrease in GHQ caseness of -0.18 (p<unk>0.1)8 Together, these results imply that the relative benefits associated with home-ownership as a social norm are distributed on a zero sum basis: a strengthening of home-ownership values increases the subjective well-being of 64 percent of respondents (owners) and decreases the subjective well-being of 36 percent of respondents (renters), but the magnitude of the decrease is approximately double the magnitude of the increase.9 When we split the renter sample into social and private renters (see appendix 4), the coefficients on relevant others' values remain negative for both private and social renters, but are only statistically significant for social renters: an increase in the home-ownership values of relevant others has a negative effect on the mental health and life satisfaction of social renters. If the importance of homeownership to a social renter's relevant others increased from the 25 th percentile to the 75 th percentile, this would lead to a decrease in their life satisfaction of -.12 (p<unk>0.1), and a decrease in their GHQ Caseness of -0.4 (p<unk>0.05). To put this latter effect into context, the effect on becoming unemployed on male GHQ caseness is -0.85 (Clark and Georgellis, 2012). --- VI. ROBUSTNESS CHECKS We also conducted two additional regressions to test further whether home-ownership is a social norm. First we examined whether the effect of becoming a home-owner was moderated by changes in the home-ownership values of relevant others. If, in the period that I move from renting to owning, home-ownership becomes more important to my relevant others, then according to the social norm hypothesis, this should increase the magnitude of any positive home-ownership effect on my subjective well-being. To test this hypothesis, we used the same specification (equation 1) but included two additional variables; i) a dummy variable indicating whether the individual was owner (=1) or renter (=0) and, ii) this dummy variable interacted with relevant others' home-ownership values (<unk> <unk> ). We included both owners and renters in the sample, to identify those individuals who moved in between tenures. This model is very similar to that adopted Clark (2003) in testing for unemployment as a social norm. As can be seen in the figures below (full regression results are shown in appendix 5) the effect of becoming a home-owner on subjective well-being increases as the homeownership values of one's cohort strengthen. For instance, if everybody's cohort attached a high importance to home-ownership (i.e. 75 th percentile of 8.08), then becoming a home-owner would lead, on average, to a small increase in subjective well-being (+0.05 life satisfaction; +0.03 GHQ Caseness). But if everyone's cohort attached low importance to home-ownership (i,e 25 th percentile of 7.68), then becoming a home-owner would, on average, lead to a small decrease in subjective wellbeing (-0.07 life satisfaction;-0.3 GHQ Caseness). It should be noted though, that by looking at people who have changed tenure, we heighten the risk of confounding variables. For instance, among groups http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies of relevant others with stronger home-ownership values, the gap in neighbourhood/housing quality between the home-ownership sector and the private rental sector may be larger, and this may explain why the home-ownership effect on subjective well-being is larger among these groups. --- INSERT FIGURE 2 If home-ownership is a social norm, then we would also expect increases in the importance that relevant others attach to home-ownership to be associated with increases in the importance that the individual (renter or owner) attaches to home-ownership. We conducted an additional regression using the same specification (1) but replacing subjective well-being W with the individual's homeownership values, on the same scale of 1-10. Using the full sample (owners and renters together), we find changes in the home-ownership values of relevant others to be positively related (p<unk>0.05) to changes in the home-ownership values of the individual (see table 2, below) which again implies that home-ownership is a social norm. --- INSERT TABLE 2 HERE One potential concern is that the home-ownership social norm is correlated with other social norms which may have opposite effects on the subjective well-being of owners and renters 10. It could be, for instance, that home-ownership values are correlated with levels of racism. If there were more ethnic minorities renting than owning, then this may explain why the subjective well-being of home-owners (renters) are positively (negatively) related to the home-ownership values of relevant others. This would not, however, explain the findings presented in figure 2 (or appendix 5), which show that the increase in subjective well-being associated with becoming a home-owner depends on the homeownership values of relevant others, keeping income, education, household structure and all timeinvariant variables (e.g. race/ethnicity/family background) constant. As noted, this regression is not perfectly robust. Therefore, there is still a possibility that these other norms are confounding variables. Moving house is also a potential source of omitted variable bias. For instance, a strengthening of relevant others' values may be associated with home-owners moving to a better neighbourhood, thus improving their subjective well-being. To test for this, we excluded those respondents who moved house between waves. That is, we only looked at the effect of changes in relevant others' homeownership values and consumption on changes in the subjective well-being of those respondents who stayed in the same house from one wave to the next. For owners, the results fit even more closely with our hypotheses (see appendix 6); all four coefficients of interest are statistically significant with the expected signs. On the other hand, for renters, none of the four coefficients of interest are statistically significant, although the coefficients on relevant others' values remain negative. One potential concern raised by Ferrer-i-Carbonell ( 2005) is multi-collinearity. The three variables used to construct the relevant others (age, education, and region) are also included in regressions as variables in their own right. This could lead to multi-collinearity. To test for this, we conducted the regressions in table 1, but this time without age, education, region and region*year variables. This leads to similar conclusions as the ones presented in Table 1 (see appendix 7) and indicates that multicollinearity is not a problem. Some may object to our treatment of life satisfaction and GHQ caseness as cardinal. With regard to life satisfaction, this concern has been addressed by Ferrer-i-Carbonell and Frijters (2004), who have shown that ordinal and cardinal approaches usually lead to qualitatively very similar results. With regard to GHQ Caseness, the concern is more pertinent. Therefore, we ran regressions without assuming cardinality in GHQ Caseness. More specifically, we collapsed GHQ Caseness into two values; 0 (for GHQ Caseness values 1-11) and 1 (for GHQ Caseness value 12), and ran several nonlinear regressions, consisting of; i) two pooled logit models for renters and owners, and ii) two fixed 10 Thank you to an anonymous referee for raising this point. http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies effect logit models for owners and renters. All logit regressions used a similar specification to table 1 11. With regard to the pooled logit model (appendix 8, columns 1 and 2), relevant others' homeownership values are positively related to the (binary) GHQ Caseness of home-owners (p<unk>0.05). None of the other relevant coefficients are statistically significant. For a respondent to be included in the fixed effect logit regressions, their GHQ Caseness score must have changed over time from 1-11 to 12, or the reverse. Setting this criterion reduces the sample size of owners by approximately 75 percent, and of renters by almost 90 percent. It is perhaps unsurprising therefore, that none of the relevant coefficients are statistically significant (see appendix 8, columns 3 and 4), although the effect of relevant others' home-ownership values on the (binary) GHQ Caseness of owners is nearly statistically significant (p=.13). Finally, some maybe concerned that those absolute benefits of home-ownership which are not controlled for in <unk> <unk> (security and autonomy, in particular), maybe driving the effects on subjective well-being that we attribute to the relative benefits. We cannot discount this possibility. However, for this to happen, these absolute benefits would need to meet two criteria. First, they would need to vary over time within an individual who stays in the same tenure. Second, and most notably, this variance over time would need to be correlated with the variance in relevant others' home-ownership rates/values. If, for example, changes in autonomy or security were driving the'social norm effect' that we observe, then increases in relevant others' home-ownership values would need to be negatively related to renters' autonomy/security, but positively related to owners' autonomy/security (and both associations would need to be independent of changes in stigma, shame, pride, or any other'relative benefit'). It seems unlikely that the absolute benefits of home-ownership would meet this second criterion. Or to put it another way, it seems likely that the relative benefits of homeownership explain subjective well-being over and above the absolute benefits of homeownership. Nevertheless, it is undoubtedly a limitation of this paper that we do not directly control for all of the absolute benefits of home-ownership, and one which future research should address, if possible. --- VII. DISCUSSION AND CONCLUSION In this paper, we conceptualised home-ownership as both a positional good and a social norm. Based on this original conceptual framework, we proposed that the status of owners and renters should depend on both the home-ownership rates and values of relevant others. Hypothesis one drew from social norm theory and predicted that if sanctions are dealt out on a binary basis of whether the individual is 'normal' (owner) or

Much attention has been devoted to examining the absolute benefits of home-ownership (e.g. security and autonomy). This paper by contrast is concerned with conceptualising and testing the relative benefits of homeownership; those benefits that depend on an individual's status in society. Home-ownership has previously been analysed as a social norm, implying that the relative benefits (costs) associated with being an owner (renter) are positively related to relevant others' home-ownership values. The theoretical contribution of this paper is to additionally conceptualise home-ownership as a positional good, implying that the status of both home-owners and renters is negatively related to relevant others' home-ownership consumption. The empirical contribution of this paper is to quantitatively test for these relative benefits in terms of subjective well-being. We run fixed effects regressions on three waves of the British Household Panel Study. We find that i) a strengthening of relevant others' home-ownership values is associated with increases (decreases) in the subjective well-being of home-owners (renters), and ii) an increase in relevant others' home-ownership consumption decreases the life satisfaction of owners but has no effect for renters. Overall our findings suggest that i) the relative benefit of home-ownership are both statistically significant and of a meaningful magnitude, and ii) homeownership is likely to be both a social norm and a positional good. Without explicitly recognising these relative benefits, policymakers risk overestimating the contribution of home-ownership to societal well-being.

to be negatively related to renters' autonomy/security, but positively related to owners' autonomy/security (and both associations would need to be independent of changes in stigma, shame, pride, or any other'relative benefit'). It seems unlikely that the absolute benefits of home-ownership would meet this second criterion. Or to put it another way, it seems likely that the relative benefits of homeownership explain subjective well-being over and above the absolute benefits of homeownership. Nevertheless, it is undoubtedly a limitation of this paper that we do not directly control for all of the absolute benefits of home-ownership, and one which future research should address, if possible. --- VII. DISCUSSION AND CONCLUSION In this paper, we conceptualised home-ownership as both a positional good and a social norm. Based on this original conceptual framework, we proposed that the status of owners and renters should depend on both the home-ownership rates and values of relevant others. Hypothesis one drew from social norm theory and predicted that if sanctions are dealt out on a binary basis of whether the individual is 'normal' (owner) or 'abnormal' (renter), then the strength of these sanctions should depend on the strength of the norm among relevant others. Consistent with this hypothesis, we found that if relevant others attach greater importance to home-ownership, this increases the subjective well-being of home-owners and decreases the subjective well-being of renters. Conceptualising home-ownership as a social norm, however, does not fully capture the nature of the relative benefits of home-ownership. Hypothesis two drew on the positional goods theory to contend that sanctions are also dealt out by relevant others according to status. One determinant of status is relative wealth. By becoming a home-owner, one can signal that they are worth a certain amount. When home-ownership is expanded, this reduces the wealth that home-ownership signals, thereby decreasing the status and subjective well-being of the original home-owner. Consistent with this logic, we found that as home-ownership rates among relevant others increased, the life satisfaction of homeowners decreased. http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies Hypothesis two also predicted that as more of a renter's relevant others become home-owners, remaining a renter would signal a greater level of relative deprivation. We found no evidence to support this logic. Our results therefore imply that home-ownership is a social norm for both owners and renters, but is only a positional good for owners. We considered two possible explanations for this discrepancy. First, owners and their relevant others may be more materialistic than renters and their relevant others, meaning relative wealth matters more to the status of owners than renters. We compared the importance that owners and renters attach to 'having a lot of money' (measured using responses to question 'b' in figure 1) but renters actually rated this domain as more important than owners (results not shown). Another possible explanation, and one deserving of further examination, is that housing tenure matters more as an indicator of relative wealth for owners than renters, who may signal their wealth through alternative forms of conspicuous consumption (e.g. cars, clothes). Even with this caveat, our findings suggest that in the UK, the relative benefits of home-ownership are statistically significant and of a meaningful magnitude. These relative benefits could (at least) partially explain why, according to some studies, home-owners have higher levels of subjective wellbeing and ontological security than renters. They may also explain why demand for home-ownership is higher in some countries than others. Furthermore, Rowlands and Gurney (2000) found homeownership to be a social norm among children. Seeing as status is a determinant of educational outcomes (e.g. Hoff and Pandey, 2004), these relative benefits of home-ownership may even explain part of any tenure gap in child educational outcomes (Haurin et al, 2002;Green and White, 1997). If researchers do not control or test for these relative benefits when examining the tenure gaps in outcomes, they risk attributing their effect to the absolute benefits of home-ownership. We therefore echo the call of other researchers (e.g. Manturuk, 2012;Lindblad & Quercia, 2015) to not only examine whether home-ownership affects subjective well-being, but also test the pathways through which any effect occurs. These relative benefits provide less justification for the promotion of home-ownership than the oftcited absolute benefits, as for every home-owner that benefits from being 'normal', other renters pay the price for being increasingly 'abnormal'. Similarly, for every home-buyer that benefits from an increase in status, other owners pay the price in the form of lower status. That is not to say that the relative benefits of home-ownership are necessarily distributed on a zero sum basis12. The subjective well-being that home-owners derive from conforming to the social norm may outweigh the psychological costs that renters incur from deviating (although our findings do suggest they offset each other). Similarly, the psychological benefits that a home-owner derives from signalling an increase in their relative wealth may not be entirely offset by the psychological costs that others incur from a decrease in social status. Even if these relative benefits are distributed on a zero sum basis, it does not mean that the UK, or any other country that promotes home-ownership, should become tenure neutral. There are many other arguments both for and against home-ownership13. Nevertheless, by providing evidence to support the existence of these relative benefits, we cast doubt on the significance those absolute benefits of home-ownership, which are often cited to justify state support of home-ownership Our findings also imply that in the UK at least, the welfare gap between owners and renters is wider than previously thought. Not only do home-owners benefit financially, through state led tax breaks, at the expense of renters; they also benefit from having higher status than renters, and from 'being normal'. The negative sanctions associated with 'being abnormal' are particularly acute for social renters. Future research should examine the extent and form of these sanctions, and the role that media coverage and government rhetoric play in augmenting them. This is an important area of research as http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies these sanctions may undermine regeneration initiatives, and may also play a role in explaining why social housing tenants have higher than predicted levels of unemployment (Hills, 2007). Finally, there are two broader research questions that arise out of our findings; 1. Who are the'relevant others' and what sanctions do they mobilise to enforce homeownership as a social norm? A notable limitation of our study is that does not shed light on who an individual's relevant others are. We defined relevant others as people of a similar age, education and region but this does not fully capture the complexity of an individual's relevant others. Future (qualitative) research should explore whose opinion counts when individuals are deciding which tenure to choose; is it family, friends, colleagues, neighbours, or media or some other group? Future (qualitative) research could also explore how relevant others' values and consumption influence individual subjective well-being (or housing tenure choices). Are renters openly mocked/ostracised by their home-owning relevant others? Or is it subtler, in that past a particular age, renters feel that they are deviating from the social norm of ownership? Knight (2002) was the last paper to address these questions in the UK context, but since then home-ownership rates have declined, implying that the social norm of home-ownership may have weakened, thus more research is needed. Among relevant others, home-owners have a particular interest in maintaining and enforcing the social norm of home-ownership, as they are beneficiaries. If UK home-ownership rates continue to decline, it will be in the interests of fewer people to sustain the normalising discourse of homeownership. Renters -who cannot access home-ownership-may increasingly challenge the idea that home-ownership is 'normal'. Home-ownership may therefore become less of a social norm 14, and the relative benefits of home-ownership may diminish, further undermining the appeal of homeownership. There could, in short, be a tipping point causing the decline in home-ownership rates to accelerate. That said, any challenge to the idea that home-ownership is 'normal' is likely to face considerable resistance from other beneficiaries. Developers, the financial services industry, the real estate industry, planners, road builders all benefit from high rates of home-ownership (Buchholz, 2002). It is therefore in the interest of these parties to portray home-ownership as 'normal' through marketing, speeches, policy and lobbying or other means. The UK government has also played a significant role in normalising home-ownership. Future research should build on the discourse analysis of Gurney (1999) to examine the role of these different actors, including home-owners themselves, in portraying home-ownership as 'normal'. --- What are the relative benefits of housing more generally? In this paper we have focussed on home-ownership but other housing characteristics are also likely to determine an individual's status. In the UK, the ongoing 'housing crisis' is usually defined in terms of a lack of supply. However, what if house size is a positional good, as proposed by Frank (2007)? What if people not only want enough space to facilitate those activities that enhance their subjective well-being, but also want more space than their relevant others? Then we are stuck in what Frank (2007) defines as a 'positional arms race' 15. In the short term and at the individual level, it is rational to purchase a larger house for increased status. But at the collective level it reduces the status of everyone else, who feels pressure to buy a larger house too, thereby eliminating the status advantage of the original status seeker. Research on the relative vs absolute benefits of income could provide a template for future research in this area (e.g. Ferrer-i-Carbonell, 2005). 14 Logic implies that home-ownership will still remain a positional good in this case. 15 This is essentially the prisoner's dilemma applied to status seeking. http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies One's neighbourhood may also be a positional good. Much of the 'neighbourhood effects' debate is centred on whether there are material benefits (e.g. educational opportunities) associated with livingand growing up-in a wealthier neighbourhood. Much less research has examined the relative benefits of living in a neighbourhood with high socio-economic status. If, by living in a relatively wealthy neighbourhood, one signals they are relatively wealthy, this signal will increase their status. Differences in neighbourhood status may partly explain spatial disparities in outcomes such as education, health and subjective well-being. Furthermore, as Phe and Wakeley (2000) posited in their theoretical model, it may also explain new trends in residential location such as gentrification and abandonment. Summarising, this paper has demonstrated that i) the subjective well-being of owners and renters depends on the home-ownership values of relevant others, thus implying that in the UK, homeownership is a social norm, and ii) the subjective well-being of owners is also negatively related to the home-ownership rates of relevant others, implying that for owners, home-ownership is also a positional good. Together, these findings suggest that home-owners enjoy relative benefits at the expense of renters: first, through being considered 'normal' by society versus renters who are considered 'abnormal', and second through being considered wealthier than renters. --- http://mc.manuscriptcentral.com/cus [email protected] --- Urban Studies Figure 1 Appendix 2: Variable description

Much attention has been devoted to examining the absolute benefits of home-ownership (e.g. security and autonomy). This paper by contrast is concerned with conceptualising and testing the relative benefits of homeownership; those benefits that depend on an individual's status in society. Home-ownership has previously been analysed as a social norm, implying that the relative benefits (costs) associated with being an owner (renter) are positively related to relevant others' home-ownership values. The theoretical contribution of this paper is to additionally conceptualise home-ownership as a positional good, implying that the status of both home-owners and renters is negatively related to relevant others' home-ownership consumption. The empirical contribution of this paper is to quantitatively test for these relative benefits in terms of subjective well-being. We run fixed effects regressions on three waves of the British Household Panel Study. We find that i) a strengthening of relevant others' home-ownership values is associated with increases (decreases) in the subjective well-being of home-owners (renters), and ii) an increase in relevant others' home-ownership consumption decreases the life satisfaction of owners but has no effect for renters. Overall our findings suggest that i) the relative benefit of home-ownership are both statistically significant and of a meaningful magnitude, and ii) homeownership is likely to be both a social norm and a positional good. Without explicitly recognising these relative benefits, policymakers risk overestimating the contribution of home-ownership to societal well-being.

Introduction Pervious boundaries have witnessed narcotics like Kush and Cocaine locating their path illicitly into the nation in immense quantities. To additionally synthesize aforementioned, domestic formation is directly also rife with Kush and fructose talc, which is essentially illicitly processed and dispersed all over Sierra Leone (2023). In the previous, narcotic utilization was a familiar trait among derelict juveniles 'Raray bois'(Criminal boys) who affianced in this behavior for multiple grounds, amidst them lightening exhausting circumstances, hoisting bravery for involvement in crook engagements and as stupefying narcotics to assist them slumber. Nonetheless, extremely lofty jobless levels in Sierra Leone, accompanied with destitution and hopelessness, have generated in augmenting figures of juvenile nationals also transforming to utilizing narcotics as a direction of amusement (2022). This has been magnified by the prepared handiness and reasonability of these narcotics to adolescents and juveniles as they relate within their social bands. The prevailing posture of enormous rise in costs of eatable stuffs might further have compounded the circumstance as juvenile, --- 76 involving those of formal education bracket; spot themselves stuck in this occurrence (2023), with constrained entertaining hustles, incredibly in prominent density cultural communities. The lofty occurrence of narcotic utilization amidst Sierra Leonean Juveniles and its fatal upshots oblige extra Sierra Leonean sociological research into its causation suitable to accommodate further appropriate estimates to curb this social malaise. Albeit no extensive communal sociological inquiry of narcotic utilization exists in Sierra Leone, this social malaise has been recounted broadly [3]. The upshots of narcotic utilization amidst Sierra Leonean juveniles involve delinquent intimacy attitude with its transpiring peril of contracting social disease [4], the mitigated opportunities of acquiring job consequent of academy maverick and augmented brutality and destitution. Albeit a literature review of narcotic utilization by Sierra Leonean juveniles offers multiple justifications for this phenomenon, they are extremely usually analyzed in a fragmental style and not perceived in their social bond inside a broader sociological construction. The grounds for narcotic utilization amidst juveniles furnished in this sociological inquiry, expressly the susceptibility of juveniles, the downward fulfillment of speculative conducts, destitution and joblessness and extensive communal drives, are previously not appropriately located inside an extensive conceptual structure that could deliver as a cornerstone for instructions to curb this trouble. The literature nonetheless depicts an increasing interest for the demand to perceive the causation of narcotic utilization amidst juveniles generally. This sociological inquiry makes a contribution by depicting that the inquiry of the grounds for the predicament inside a wider social ambience which permits for comprehending them in their social bond provides regarding its solving. But also would it develop in enhanced approaches to curb narcotic utilization among Sierra Leonean juveniles, it would also curb pertained predicaments like brutality and destitution and would consequently impact favorably on Sierra Leonean Juvenile's objective and cognitive health [5]. --- Literature Review Narcotic and alcohol abuse in Sierra Leone are alarming and a root source or patron to multiple economic, health and social predicaments distressing the populace. Narcotic reliance statistics depict that narcotic ingestion (Kush and marijuana) in Sierra Leone doubles the Mano River Basin norm and second to none in the Basin. The norm bracket of narcotic reliance in Sierra Leone is twelve and twenty four years and growing [6]. Sierra Leone is among the top three narcotics and alcohol abusers in the Mano River Basin. For every hundred folks, fifteen have a narcotic predicament and for every five Leones in circulation, five Leones are associated to the narcotic abuse predicament [7]. The principal narcotics being abused are 'Kush', marijuana and cocaine. Collectively, the triad narcotics recorded for over eighty six percent of all cases handled for narcotic abuse in 2023. Between individuals treated for jones, thirty eight percent were handled for 'Kush' reliance, preceded by marijuana at twenty three percent and cocaine at nineteen percent. The prevailing information at Mano River Basin threshold proposes that illegal narcotic utilization among extremely juvenile folks is restricted to a diminutive splinter group who explore with narcotics at an extremely initial social bracket. Frequent utilization among the extremely juvenile is strange quite as is partially contemplated in the Mano River Basin data on folks observing narcotic prescription. Among those who commence utilizing narcotics at an extremely juvenile social bracket, social manifest proposes that prematurely trail with hallucinatory narcotics, involving shag and booze, is linked with an augmented peril of breeding narcotic troubles subsequently in life. Constant narcotic utilization among the under fifteens is terribly often stumbled among a greatly dubious cohort group of the social population, in whom narcotic utilization is merged with other synchronic or antecedent cognitive and social upsets and might holistically be a tag of social predicaments or of a rudimentary acoustic social route that requires community intervention [8]. Narcotic social abuse is a universal challenge with injurious impacts on health, wealth and safety of nations. In Sierra Leone narcotic abuse has been linked with felony, interpersonal brutality, perilous carnal conduct (with nursed augmented peril of AIDS obtainment and STI prevalences), adverse health druggies and psychological effect to their households. Nicotine and booze are holistically the most usually utilized narcotics amongst Sierra Leonean youth. These two narcotics are also the most tried with amongst youth. Because they are both legal, vast consider them tolerable and 'lenient'. This notwithstanding substantial health and social effect linked with them. Despite harmful narcotic abuse is usual in Sierra Leone, 'Kush' is the most usually utilized illegal narcotic amongst juvenile. An inquiry by [1] concluded that of Sierra Leonean Portloko youths had ever utilized at least one illegal narcotic such as marijuana and tobacco. This digit is the highest in the district. Granting the medical and social trauma generated by these narcotics, it is significant to comprehend the stretch of their utilization amongst sub social classes and prospect the efficient paths to curb them. The Micro Network of Social Peer Pressure Both qualitative and quantitative sociological research depicts the relevance of social peer pressure as a pivotal element for narcotic utilization amongst Sierra Leonean juveniles. On this detail, it was discovered that juveniles social approval by their peers are immensely treasured [9]. Surrendering to cohort social pressure to utilize narcotics was also mentioned by [10] who discovered that social peers had immense regard respect for the opinions of their social associates of their peer cohorts. Interviewees aforesaid that they would rather talk over their social troubles with someone of their age bracket or cohort than with tutors, guardians or parents or any household associate. Sociological research conducted by [11] disclosed that peers were regarded the primary suppliers of assistance and that they comprised the most dominant elements generating narcotic utilization amongst juveniles. Social peer pressure as element for narcotic utilization is also inferred by an inquiry embarked by [12] who discovered, with an eighty eight point nine percent accuracy, that juveniles who disclosed deep-rooted dipsomania were essentially probably to have comrades who utilize booze or liquor. Based on [13], adolescents were three to six times further potentially to have utilized liquor and to have been involved in satiate boozing in the following month if they had comrades who utilized liquor. Their social inquiry depicted that adolescents were gifted their foremost booze by comrades rather than other folks. --- Inadequate Functional Modeling The significance of the impression of parents who utilized narcotics on their adolescent children was also demonstrated by [14]. They discovered that parents utilization of narcotics in terms of smoking nicotine and cannabis and consuming liquor, and emphasize parents utilization of liquor and cannabis bore an important interaction to the consistency of their adolescent children's utilization of illicit narcotics. Relating the impression of parental conduct on their children, [15] disclosed that parental conduct of discipline, lodged in parent's social education of their children's social engagements and restriction is essentially predictive of lesser thresholds of liquor and nicotine utilization. They also mentioned that overt but not covert marital violence was linked with the utilization of the revealed narcotics. --- Ecosystem of Cultural Community Tolerance Cultural community antisocial conduct, in this context, refers to adults' utilization or peddling of illicit narcotics and their having been arrested by the police in terms of illicit narcotics, whilst the theory'subjective adult norms' refers to adults encouraging the utilization of cannabis, and accepting to their adolescent children's consumption of liquor and smoking of cigarettes. [16] discovered that cultural community antisocial conduct, run down communities and community affirmation could be significantly linked to secondary school students utilization of domestic hard liquor in the provinces or western urban Freetown. Nonetheless, assertion conveyed as acknowledgement and admission of duty perfect, reported an important interaction to adolescents declining from boozing domestic liquor. An important interaction between narcotic utilization and cultural community liberality was depicted by [17] who mentioned associations between adolescents who had been stupefied and cultural communities where liquor could be accessed comfortably and where police stop to react to juvenile's utilization of liquor. --- Destitution and Joblessness Based on [18], destitution is an indicator of narcotic utilization hence the latter might present as a mechanism of subsisting with the commonplace coercion of destitution. The functional function that joblessness plays in narcotic utilization should, based on [19], be perceived in terms of its interaction to morale failure and social retrogression generating from the destitution that joblessness ignites. The existing lofty joblessness threshold of 3.60 % among 18 and above, as revealed at the end of 2023 [20] presents an interpretation for the broadly observed destitution in Sierra Leone. Narcotic utilization by adults, geared at escaping frustration, could impress adolescents who chase adulthood to put up similar social conduct which, in turn, could negatively impress their peers, generating an augmented mitigation in opportunities of acquiring job. Basically, it is destitution and joblessness and the absence of any recreational centers that leave young folks with an emotion of despondency and mediocrity. If they perceive no anticipation of ever acquiring a noble employment and sufficient earning to reside a traditional existence, pursuing oblivion via narcotics or liquor becomes a tempting escape route [21]. --- Social Brutality Literature discloses reciprocity between social brutality and narcotic utilization by Juveniles. An expansive social research project headed by [22] at Secondary Schools in Freetown, Western Urban Sierra Leone disclosed that respondents from this metropolis considered narcotic utilization to be an essential element for brutality, which depicted, previously, in physical brutality geared towards cultural communities. The connection between the utilization of narcotics and brutality would interpret that the roots of social brutality disclosed in the mentioned inquiry, involving challenged cultural communities, tribal conflicts, social cliques and tribal tensions, would also connect to narcotic utilization since each of them has the ability to augment the depression observed by the juvenile at divergent cultural domains, thereby amplifying juveniles susceptibility to narcotics. [23] speculate that lofty thresholds of communal stress would relate to augmented narcotic utilization by adolescents. In this respect they reveal contemporary 78 stressors which happen in cultural communities, expressly existing brutality and felony, lofty levels of joblessness, automation transformations and the HIV contagion. --- Macro-Structure Lofty thresholds of cultural communal anxiety in the social space would induce narcotic utilization by juveniles [24], either rightly or discursively. Whether nuisances integrate right mainsprings for narcotic utilization amongst juveniles or bring about the rights mainsprings in the microstructure to come into social role and sustain on each other, they should not be perceived as being dissimilar to vistas in the macrostructure. Sierra Leone's contemporary technological social indulgence introduced a current and fair-minded standard network which has been given opportunities for previously deprived Sierra Leonean juveniles. Ultimately, this social digital freedom, because of its immensity, could induce vulnerable adolescents to demand adulthood further passionately and to precede improper "adult" conduct such as boozing and narcotizing. Holistic standards which are recently stressed, expressly those of being in dominion of employment skills, being motivated and stipulating on one's right could also impact narcotic utilization amongst Sierra Leonean juveniles. In themselves these social standards are impressive, but if juveniles become impulsive in their chase of them, they could ignite augmented depression which would establish them further susceptible to narcotic utilization [25]. Further critical pertaining relating to narcotic utilization by juveniles relates the difficulty observed by the government to establish adequate employments to uplift financially disadvantaged cultural communities in the social space of existence. Adequate employment chances would impact impressively on the eradication of destitution and mitigate inappropriate role epitome by parents and friends who utilize narcotics to evade from their hopelessness. It could be revealed that destitution would rightly or discursively have a gloomy impact on the mental health of juveniles discovering themselves at the epic of cultural community, thereby establishing them further susceptible to pursue social solace in the utilization of narcotics. --- Problem Statement Comprehending elements for the proliferation of narcotic utilization amongst young folks in Freetown, Sierra Leone. --- Social Goals This sociological inquiry intends to explore the rudimentary mainsprings of the juvenile narcotic abusers and their connection with their households and to recognize troubles and demands that young narcotic abusers are encountering. The central emphasis is to recognize the crux of narcotic abuse and to empower the household functional function for prevention of narcotic abuse in the cultural community. Sierra Leone has a grave narcotic usage enigma, as disclosed in the literature. Narcotic abuse imposes economic, health and social costs on persons, households, cultural community and economy in entirety. At the individual threshold, narcotic abuse has been associated to depression, brutal conduct and divergent shapes of felony, involving numerous accidental and premeditated injuries. Cultural community loses the productivity and energies of folks impacted by narcotic abuse. At the macro threshold, prevention and treatment costs linked with narcotic abuse are phenomenal. In Sierra Leone, evidence on the stretch, socio-cultural impact of narcotic abuse as well as its prevention is fragmented and further often not lodged within a comprehensive conceptual framework that could make it simpler to establish approaches and initiatives for curbing the narcotic abuse challenge. Despite much sociological research has been done on the matter, diminutive attempt has been formulated to place all this social fact in a consistent account that will locate in advance the stretch, and impact of the trouble and notate future arbitrations and the developing of initiatives. Ordinarily, the social goal of this sociological inquiry is to identify the mainspring of narcotic abuse, lodging attention into Kush as the principal narcotic that is being abused in the Freetown municipality. Kush (also known as K2) is a domestic narcotic that has allegedly come into widespread utilization in Sierra Leone since 2022, predominantly in the slum communities, although it is perceived to be emerging in other locations in Sierra Leone as well. The narcotic is mostly portrayed as a mixture of numerous components, and these may differ, nevertheless the main engaging component of the narcotic is fentanyl, tramadol and formalin, mostly whipped with 'Jamba' (local Krio word for Marijuana). --- Methodology The sociological inquiry methodology encompassed a review of books, direct social dialogue, focus cohort dialogues and websites. A random selection was done to choose fourteen 'Kush' addicts who dwell in the streets of Freetown for dialoguing. The volunteers were between the ages of fifteen to twenty two subsequently, amongst the 'Kush' addicts were eight males and six males, and among the respondents were a twenty year old pregnant lady. The principal question directed to them was the factor why they left their homes to dwell in the streets and the reason why they take drugs. The response was similar; ten of them revealed household troubles, involving unequal treatment with family members whom they inhabit with after their parents are deceased or when parents get separated or divorced. Nonetheless, five disclosed 79 their desire to do a trial with narcotics like liquor and tobacco then eventually commence experimenting robust narcotics like 'Kush'. These five 'Kush' addicts came from impressive households with both parents to look after them but the interest to utilize narcotics propelled them to dwell in the streets. Dialoguing if they can return home to their relatives and be embraced by their relatives, all the five 'Kush' addicts reacted impressively since they departed willingly without being pursued out. However, they even disclose that their relatives mostly come to cajole them to return home and engage a progressive social existence. Dialoging why they rejected to return home and dwell with family they all disclosed that at home their family will not provide them cash to purchase narcotics that is why street social existence is better as they can enlist with gangs and commence looting folks to get cash and sustain their narcotic appetite. Notwithstanding, the nine respondents also disclose that their family did not heed about their welfare, even purchasing secondary school uniforms or assist them to go to secondary school, after observing that sort of bad treatment at home, they turned to street social existence. --- Result and Discussion Majority of the male 'Kush' addicts disclosed that they are not timid to loot or to murder if the prey opposes with his or her properties. Majority of the respondents disclosed that they become very brave when they are under the 'Kush' inducement rather than when they are reasonable. They routinely commit felony when they are under the inducement of 'Kush' they become gallant and further stubborn. The female respondents revealed that they are not able to loot folks; consequently, they lean on men to loot and purchase the 'Kush' in exchange for sexual intercourse. Anybody who brings the 'Kush' will be rewarded with intercourse on that day, they also revealed that they do not have a constant street lover; a street lover is casual for period when he delivers eatables or 'Kush'. The social connection is rigorously sexual intercourse and no love among them. Dialoguing if they utilize safety measures, vast of the respondents disclosed they do not utilize condoms and does not even comprehend where to purchase them. Nonetheless, the whole respondents was conscious that the vast generate of sexually transmitted diseases is unsafe sexual intercourse. They reveal they are not cowed of sexually transmitted diseases because everybody is ultimately going to pass on but individuals pass on in divergent directions. Dialoguing if they understand their HIV status, twelve of the 'Kush' addicts said it is not imperative to comprehend and they will no way go for testing, the prevailing two involving the pregnant lady disclosed testing HIV assured when they got tested after crumbling unwell. The two mentioned they are not taking any medicine after they concocted the tablets with 'Kush' and gas them. Dialoguing if they will go and receive the tablets afresh to drink, the 'Kush' addicts disclosed they will go and receive the tablets but gas them subsequently. Nonetheless, eight of the respondents indicate eatable as comfort, they sometimes go for pair or further days lacking eatable, and despite they afford to purchase eatable, ultimately the principal interest is 'Kush' narcotic than eatable. They also disclose that when they are under the inducement the narcotic will make them extremely hungry, so they prefer shoplifting to get something to eat rather than utilizing their paper money on eatable. It was also disclosed that all the cash which they receive is only utilized to purchase 'Kush' narcotic; they propone to wear shabby patches than purchase something to wear. The 'Kush' addicts also disclose they are not apprehensive of being apprehended as they always got captured and acquitted, they also revealed that Pademba Road Correctional Center is also another domicile as they have colleagues there and it is better to be there. Nine 'Kush' addicts regret utilizing 'Kush' and detail street social existence as appalling and bleak. Nonetheless, the five respondents disclosed that street social existence is better than home, they are excited in the streets as they can make cash thus simply by looting folks their laptops, MiPad, mobile phones and cash. Dialoging them where they usually loot folks, they revealed the beach areas, markets, narrow streets and lorry parks They usually loot folks at knife point, their aim is to collect cash and other valuable but not to harm or murder, they can only jab if the sufferer aspires to withstand or coerce them. Dialoging how they take the terrible drug, all the participants revealed that they smoke the 'Kush'. Dialoging them if there are health professionals who drop by for health education, totally the respondents were steadfast that they never come across anybody from the health agency endeavoring or willing to converse with them about health connected affairs. Dialoging how the endure the frosty weather in streets, three of the fourteen respondents disclosed that when they are under the inducement, they do not feel any coldness as they slumber until the next day without waking up at fixed moments in the course of the night. The rest mentioned they do have coverings which they conceal in fixed locations and only utilize them at night, they disclosed that 'Kush' addicts slumber in divergent domains, ultimately folks have a propensity of stripping others coverings and go to their location were an individual will not discover them, even if an individual discover the location cannot retrieve it because they will clique assault the individual. Completely the four females disclosed that they do not have commands over coverings as the male will retrieve them utilizing compel or muscle. Nonetheless, the only path to survive the frosty is to share a covering with a man in exchange for sexual intercourse. The male respondents disclosed they occasionally clash for females as they are fewer than them, occasionally the clash will end via a social contract that addicts give each other rounds that night to have sexual intercourse with a similar woman. Four males can have unsafe sexual intercourse with one female per night without any clash as long they consent on giving each other rounds. Another respondent disclosed that the 'Kush' will make him to evolve a savage sexual appetite and usually clash during the day as women will also be inspecting for cash in the streets. He disclosed that the only path to satiate his appetite is via masturbation. The respondent also divulged he occasionally force himself on a woman (Rape) as occasionally the women will request for cash or 'Kush' foremost prior permitting him to engage in any action. The other male respondent also consented, revealing they always rape especially at night, especially when a woman complains of weariness or request for assistance cash. Another male respondent also divulged that these women are their duties, 'Kush' addicts always protect them in all their demands, even if they rape them, they do not take the matter to the police. Concept reveals that narcotic abuse has numerous upshots. Despite in Sierra Leone individuals comprehend much about the adverse impacts of liquor and tobacco on person; cultural community and society at large, there is still cocaine, nicotine, heroin, and numerous other narcotics. At a conceptual threshold, narcotics impact the person, cultural community, and municipality as a whole. All its adverse effects cripple all agencies of the economy, involving the health agency. This agency reviews literature on the upshots of narcotic abuse to the person, family or cultural community and municipality at large. On the onslaught it must be considered that this dichotomy is only made to structure the researcher's sociological discourse, the queues separating person, cultural community aftermaths of narcotic abuse are completely blurred. At the emergence, it is crucial to pinpoint that literature is obvious that the upshots of narcotic abuse vary between women and men, who insinuates any treatment or intervention program must zoom in gender parities. For instance, in a presentation (that is based on a survey conducted between April 2022 and June 2023) made to the Sierra Leone Government by the National Drug Law Enforcement Agency the following upshots of narcotic utilization and abuse on women and girls were noted: (a) Seclusion and repudiation were pivotal mainsprings of women narcotic abuse. (b) Women were further probably to abuse over the counter pills. (c) Women narcotic consumers were forty six percent further prone to be sufferers of physical abuse involving rape and incest. (d) Dependence among women happened further swiftly than men. Social conduct and engaged in sexual practices further persistently than those who had not utilized this narcotic. In this sociological inquiry, the relationship between narcotic utilization and sexual engagement, brutality for both males and females in metropolitan domains common among 'Kush' addicts. As narcotics tend to motivate instinctive social conduct and damage one' decision, some inquiries have illustrated that those on narcotics are further prone to engage in perilous sexual conduct such as having sexual intercourse without condoms or having numerous partners leading them to contract HIV and Sexually Transmitted Diseases. The psychopharmacological impacts of some narcotics raised the sexual engagement of narcotic consumers, subsequently grooming them to perilous sexual conduct. For instance, the association between narcotic utilization and augmented sexual appetite and recklessness, granting increase to perilous sexual conduct and augmented vulnerability to contracting HIV and other sexually transmitted infections. Nonetheless, 'Kush' addicts divulged having narcotics prior engaging in sexual intercourse. Multiple ladies have become pregnant as the psychopharmacological impacts of narcotics compromise their social existence. Further social impacts of narcotic abuse to a juvenile involve poor educational accomplishments, welfare reliance joblessness, felony, reliance, destitution, marginalization, social exclusion and brutal conduct. Additionally these backing all and sundry, they successively preserve narcotic ('Kush') abuse. Upshots There is agreement among all inquiries appraised that within a cultural household or family milieu, narcotic abuse has grave defeatists impacts, with brutality being the most important enigma. Narcotic utilization is broadly enmeshed in interactive brutality (involving gender-lodged brutality and sexual spank), school brutality, and mostly direct to an unfavorable teaching or knowledge community. At home brutality is dispensed predominantly against women associates, siblings or parents, wives, while at school peers and tutors are the overriding victims. Narcotic utilization substantially marshals to household dysfunctional bonds and social disintegration, economic deprivations and torment, augmented drifts attached with medical and other treatment services for narcotic consumers not able to assist them. Narcotic abuse is also linked with indigent academic output and ambitions and 81 protracted stay at school. Notwithstanding, narcotic utilization is linked with academic challenges, truancy and quitting out of school. For instance, Marijuana utilization, which is a narcotic of option among Sierra Leonean juveniles has been depicted that it holistically encroaches with education, squat stretch recollection and cognitive dexterity. Furthermore, 'Kush' had unpleasant impacts on physical, memory, and economic well-being, and restricted f ree privileges via school maverick and apprehension of narcotic consumers in the Western Urban Freetown. Similarly, 'Kush' utilization prompt household conflict, with gloomy upshots on children, involving abandonment and deficient birth aftermaths. At a cultural community threshold, respondents associated 'Kush' utilization to augmented levels of felony, brutality and graft, which sap cultural community solidarity. Upshots to cultural community narcotic abuse definitely signify the sparks, ingenuity and aptitudes of the juveniles are not exploited. Narcotic abuse is associated to joblessness, felony, stagnation and even untimely demises. To the cultural community this implies prior productivity and economic growth. In essence, narcotic abuse has a mediated or unmediated social posture on the economy. Other vast costs such as narcotic connected brutality, slashes, murders, ailment, law application and adrift productivity prevail immensely unspecified. With augmented levels of felony, brutality and graft, this undermined cultural community solidarity. Despite these social statistics do not straightly enunciate to the juvenile; they direct to a grave national or cultural community weight. --- Conclusion Freetown cultural community is on the verge of social crumble due to the menace of narcotics (Kush as principal narcotic) that has obstructed its accomplishment and growth. The domestic precinct is losing millions via vandalization and robbery, predominantly power lines and structures made of aluminum, steel, iron, zinc and copper. Multiple folks are now worried to do their business in Freetown due to the lofty level of felony, the police failed to control the felony level despite they endeavor to bring all the offenders to book but sufficient must be concluded to get rid of felony absolutely. Recommendations There is necessity for social effort which conveys the significance of household or family duty to the nurturing of children; the household or family is the most significant structure to train a child about the harms and perils of narcotics. The mosque and church also has a vital role to play by training the children good morals from 'Elaykayhu' (community Islamic center) or Sunday school right via the juvenile phase. The juvenile phase is the most crucial as young adults endeavor to test further at this phase. The household or family must observe all their children activities and routine, uplifting them to uphold the five daily Salah or prayers and go to church on Sundays. Most social researchers and scholars miss out the significance of prayer and Allah when they are committing to paper pedagogical details; they inclined focus further on cognitive rationality as the conclusive social solution for troubles. Allah is the ultimate solution for everything, these children must be motivated to cherish Allah God and comprehend His intent for mankind so that they will also acquire better social existence. The household or family social function is vital for a child conduct; numerous narcotic addicts come from fragmented households or families, even if the household family is wealthy but if harmony and affection does not prevail then the child is probably to go off-track. Sociological researchers usually undercut the effect of spiritual intercession as a lasting therapy to the scourge of narcotic abuse. There is grave demand for cultural communities to construct further mosques or Islamic colleges or churches and bible colleges were young juveniles are inspire to attend free of cost and get further religious information about good conduct and devotional education rather than speculative education. Speculative education is al so significant but they can acquire that in schools and colleges. This approach will have an impressive impact in mitigating the menace of narcotic abuse in our cultural communities. The household or family must play the most significant social function then proceeded by the mosque or church, the mosque or church and the household or family must work collectively in exchanging notions on how to construct the social existence of their children.

Sierra Leone's illegal narcotic utilization enigma has augmented considerably amidst the ultimate dual decades [1]. Of specific concern has been the rising drift in narcotic or drug utilization among Sierra Leonean adolescents and juveniles [2]. Frequently utilized narcotics in Sierra Leone include Kush or K2, Marijuana, Tramadol, and to familiar stretch Cocaine. In this sociological inquiry, the researcher discusses worsen narcotic utilization amidst the ultimate dual decades in Sierra Leone, and how it is a grave imminent community health apocalypse. This sociological inquiry intents to scan the rudimentary incentives of the juveniles narcotic abusers and their rapport with their families and to pinpoint dilemmas and necessitates that juvenile narcotic abusers are encountering. The essential insistence is to diagnose the crux mainsprings of narcotic abuse and to entrust the family functional function for neutralization of narcotic abuse in the cultural community. The objective of this sociological inquiry is to diagnose the central root of narcotic abuse, situating attention into 'Kush' as the principal narcotic that is being abused amongst adolescents and juveniles addicts. The inquiry methodology entailed a review of books, direct social dialogues, focus cohort social dialogues and internet sites. A random selection was done to choose fourteen addicts who live in the streets of Freetown for questioning. A critical review of narcotic utilization mitigation programs right through the nation is required to secure that they are customized to adolescents and juveniles, a springing critical population for whom these benevolences are indispensable. Invigorating social safety groundwork, narcotic regulative schemes and statute implementation is desperately enforced to resist this imminent communal health apocalypse. The inquiry summarized that to accomplish this, a multi-stakeholder crucial path for cultural community consciousness, jurisdictive transformations to secure punitive measures for suppliers and increased juvenile offenders social contingencies are imperative.

Introduction At the end of 2018 there were approximately 27,000 people living with HIV (PLHIV) in Australia with the number of new diagnoses remaining stable over the past five years at under 1000 per year [1]. Men who have sex with men (MSM) continue to be the main HIV risk exposure group in Australia, accounting for over 60% of new cases in 2017 [2]. The national, self-reported HIV prevalence rate among MSM in Australia was 8.2% in 2019 [3]. Most HIV-related epidemiological research does not differentiate between gay men and bisexual men. In Australia, new HIV cases are recorded in the national surveillance database according to mode of HIV transmission. The transmission category "male to male sex" includes gay and bisexual men if the known route of transmission was sex with another man. However, sexual identity information is not collected which means there is no means to disaggregate data according to sexuality. Even the category "male to male sex" does not adequately capture sexual identity as gay and bisexual men may be listed within the other transmission categories, such as 'injecting drug use' [1]. Social research related to HIV prevention will sometimes differentiate between gay and bisexual men, particularly with respect to whether or not bisexual men engage with safe-sex messaging or whether there are differences between gay and bisexual men in safe-sex behaviours [4]. However, it is more common for gay and bisexual men to be treated as a singular group in behavioural surveillance surveys related to HIV prevention and in social research related to people living with HIV (PLHIV) [5,6]. There are a range of terminologies used to capture this group: "men who have sex with men", "gay and bisexual men", "gay men, bisexual men, and other men who have sex with men" [7,8]. Even where bisexual men are acknowledged and named as part of the group, the small number of bisexual-identified men relative to the number of gay men means that the experiences of bisexual men are often marginalised or hidden in research or not readily visible in current data collection. While this may not be problematic in population-level studies of HIV transmission, it can be problematic in smaller scale studies which seek to understand human behaviour, safe sex decision making, sexual cultures or the experiences of PLHIV. There are many reasons why the experiences of bisexual men living with HIV may be different to that of gay men. Historically, bisexual men have held a complex position in relation to HIV. When the epidemic first emerged in the 1980s, bisexual men were often demonised in media and public discourse as they were considered to be the source of HIV transmission to women (who at the time were often positioned as "innocent victims" of HIV) and the heterosexual community in general [7][8][9]. Alongside this, epidemiological research concerned with the risk of HIV transmission to heterosexual communities [10][11][12] often implicitly reinforced the notion that bisexual men were problematic due to being a transmission 'link' between gay men and heterosexual women. This complex history means bisexual men may experience a different burden of stigma associated with HIV than gay-identified men. In addition, there is often limited space created for bisexual people within lesbian, gay, bisexual, transgender and queer (LGBTQ) communities [13], which in Australia have been at the forefront of HIV activism and community support [14]. Bisexual men living with HIV are often more isolated from services for PLHIV or less connected to others living with HIV as a result [15,16]. Social support acts as a buffer against the negative impact of HIV-related related stigma for PLHIV [17]. Peer support programs, and connection with other PLHIV, may also help to reduce the burden of stigma through creating social interactions where HIV is acknowledged and normalised [18,19] which in turn may reduce negative self-perception associated with HIV [20]. This paper explores the experiences of bisexual men in a sample of Australian PLHIV and compares gay and bisexual men on a range of measures including perceived stigma, social connectedness and mental health. The aim is to better understand the unique needs of the groups most impacted by HIV in Australia in order to target programs to improve health and wellbeing. --- Methods These data are drawn from the 2015-2016 round of a repeated, cross-sectional survey of people living with HIV (PLHIV) in Australia, the HIV Futures 8 survey, which is conducted every 2.5 years. Full details of the study protocol and method have been published elsewhere [21]. Ethics approval for the study was granted through the La Trobe University College of Science, Health and Engineering Human Ethics Committee (S15-100). PLHIV, aged 18 years or older and residing in Australia were invited to self-complete a survey in either online or hardcopy form. The survey was advertised widely on social media (namely Facebook), dating apps targeting gay and bisexual men, through HIV organisations, gay community media, HIV-related media, HIV and sexual health clinics, and relevant community events. --- Measures The survey instrument comprised approximately 250 items containing both closed and open questions. Standard demographic items were used to collect information about age, gender, sexual identity, area of residence, education, employment status and income. HIV-related items included year of diagnosis, current use of antiretroviral treatment, and result of most recent viral load test. Physical and mental health-related items included: diagnoses of mental health conditions over lifetime, use of medication for a mental health condition in the last six months, general health, emotional wellbeing, resilience and social support. General and mental health were measured using subscales from the RAND 36-Item Short Form Survey (SF-36) [22]. Subscale scores range from 0 to 100 and higher scores indicate better health. Resilience was measured with the 10-item Connor-Davidson Resilience Scale [23], with scores ranging from 0 to 40 and higher scores indicating greater resilience. Participants were also asked to describe their overall wellbeing using a single-item categorical measure (excellent, good, fair and poor). Perceived social support was measured using the 10-item Index of Social Support, in which participants were asked to indicate the extent to which they agreed with statements relating to friendship and support [24,25]. Summary scores ranged from -30 to 30 and higher scores indicated greater social support. Connectedness to the LGBTQ community was measured using a 6-item adapted version of Frost and Meyer's Connectedness to the LGBT Community Scale [26]. Scores ranged from 6 to 24 and higher scores indicated a stronger sense of community connection. Participants were also asked how much time they spent with other HIV-positive people using a four-point scale (none, a little, some, a lot). Data on health behaviours relating to smoking, alcohol consumption and non-medical drug use were collected. Alcohol consumption was measured with the Alcohol Use Disorders Identification Test-Consumption (AUDIT-C); scores ranged from 0 to 12 and higher scores indicated a 1 3 higher likelihood of risky alcohol use [27]. Data on nonmedical drug use was collected by asking participants which drugs they had used at least once in the past 12 months. Data on smoking were collected by asking about frequency of tobacco consumption. Perceived HIV-related stigma was measured using the 40-item Berger HIV Stigma Scale [28]. Scores were calculated for the overall scale (score range 40-160) as well as four established subscales: personalized stigma (relating to the way other have reacted to their HIV status; score range 18-72), concerns about HIV disclosure (score range 10-40), negative self-image (score range 13-52), and concern with public attitudes about people living with HIV (score range 20-80). Higher score indicated greater perceived stigma. This paper also reports qualitative findings from an opentext question that aimed to elicit people's experiences of HIV-related stigma using specific terms (guilt, shame, fear) that have been associated with stigma in the literature on stigmatised health conditions [29,30]: What, if at all, has been your experience of feeling guilty, shameful or fearful about HIV (including fears about how others may react to you being HIV-positive)? --- Data Analysis Quantitative analyses were conducted using Stata/SE Version 16.1, and p <unk> 0.05 was used to determine statistical significance. Chi square tests and t tests were used to compare bisexual men with gay men on sociodemographic characteristics, health and wellbeing, social support and community connection, and perceived HIV-related stigma. Univariable and multivariable linear regression analyses were conducted to determine which variables were associated with perceived HIV stigma (Berger HIV Stigma Scale total score). Variables that were statistically significant in binary comparisons were block entered into the multivariable model. Out of the 48 bisexual men who participated in the survey 44 (92%) provided a response to the open text question. Open-text data were analysed using inductive thematic analysis to identify themes relating to the impact of HIV-related stigma on the lives of bisexual men [31]. Data were closely read, initially coded and then grouped into higher level themes for reporting [32]. We only examined responses from bisexual men rather than comparing gay and bisexual men because short text responses provide just a small amount of information and insight into the issues of concern, hence capacity for a nuanced qualitative comparison between two groups is limited. Our analysis of these data is intended to complement the quantitative findings by providing descriptions of how HIV-related stigma is experienced by this sample of bisexual men. Quotations are presented without demographic characteristics in order to protect anonymity. --- Results --- Sample HIV Futures 8 was completed by 872 participants, which represented approximately 3.5% of the total number of people living with HIV in Australia in 2015 when survey data were collected [33]. This paper focuses on a subsection of participants, men who identified as bisexual (n = 48, 6.1% of male participants) or gay (n = 681, 87.0% of male participants). The sample includes cisgender men, transgender men and gender diverse participants who identified as gay. There were eight people who identified as gay and transgender or gender diverse (grouped with gay men for the purposes of this analysis given the focus is on sexual identity), while no bisexual men identified as transgender or gender diverse. Table 1 shows few differences between the samples of gay men and bisexual men in demographic characteristics. There were no statistically significant differences between gay men and bisexual men in: country of birth (<unk> 2 = 0.12, p = 0.73), Aboriginal and Torres Strait Islander status (<unk> 2 = 1.69, p = 0.19), place of living (city/rural) (<unk> 2 = 0.01, p = 0.90), highest level of education (<unk> 2 = 0.36, p = 0.84), employment status (<unk> 2 = 7.44, p = 0.19), reported financial stress (<unk> 2 = 3.46, p = 0.06) or relationship status (<unk> 2 = 1.20, p = 0.75). Compared to gay men, bisexual men had a higher mean age (55.2 years compared to 49.9, t = 2.81, p = 0.005) and were more likely to report their annual income to be AUD$30,000 or less (<unk> 2 = 14.16, p = 0.007), which at the time of the survey was approximately half the annual average household income in Australia [34]. Of those in a regular relationship, bisexual men were more likely than gay men to report that their primary partner was female (although due to small sample sizes in partner gender categories we cannot report on a statistical comparison) (Table 1). --- Quantitative Findings --- Health Status There were no differences between gay men and bisexual men with respect to: years living with diagnosed HIV (t = 0.74, p = 0.46), use of ART (<unk> 2 = 1.15, p = 0.28), HIV viral load detectability (<unk> 2 = 0.86, p = 0.36) or lifetime diagnosis of a mental health condition (<unk> 2 = 2.58, p = 0.11). Compared to gay men, bisexual men reported poorer general health (t = 2.85, p = 0.004), poorer emotional wellbeing (t = 2.22, p = 0.03) and a higher likelihood of using medication for a mental health condition in the previous 6 months (<unk> 2 = 7.00, p = 0.008) (Table 2). While there were no statistically significant differences between gay and bisexual men in alcohol consumption (t = 0.08, p = 0.94) or daily tobacco use (<unk> 2 = 0.24, p = 0.89), bisexual men were less likely than gay men to report non-medical drug use (<unk> 2 = 5.64 p = 0.02) (Table 2). --- Social Support and Stigma There were no statistically significant differences between gay men and bisexual men in reported levels of resilience (t = 1.45, p = 0.15). However, compared to gay men, bisexual men reported lower social support (t = 4.17, p <unk> 0.001), less connection to LGBTQ communities (t = 3.91, p <unk> 0.001) and were less likely to spend time with other PLHIV (<unk> 2 = 10.4, p = 0.02). Bisexual men also reported higher levels of perceived HIV-related stigma across all four stigma subscales-personalised stigma (t = 3.33, p <unk> 0.001), disclosure (t = 4.33, p <unk> 0.001), negative selfimage (t = 4.31, p <unk> 0.001) and concern about public attitudes (t = 3.48, p <unk> 0.001) as well as the total stigma score (t = 3.86, p <unk> 0.001) (Table 2). In the multivariable model, higher total stigma scores were associated with identifying as bisexual (B = 9.25 [1.45,17.05], p = 0.02), poorer emotional wellbeing (B = -0.17 [-0.28, -0.06], p = 0.003), lower social support (B = -0.64 [-0.83, -0.45], p <unk> 0.001), lower LGBTQ community connection (B = -0.81 [-1.27, -0.35], p <unk> 0.001), and less time spent with PLHIV (B = -9.26 [-15.66, -2.86], p = 0.005) (Table 3). --- Qualitative Findings --- Stigma as Social Isolation Among Bisexual Men Bisexual men's responses to the open-text question revealed a profound sense of isolation that was linked to HIV stigma. Participants did not mention bisexuality specifically in relation to their social relationships, rather themes were related to loss of connection with friends and partners as well as withdrawal from the social world. Several men spoke about retreating from their social lives after their HIV diagnosis due to fear of rejection, as one participant stated, "Since diagnosis, I have very rarely socialised. I very rarely leave my home". Another stated that since their HIV diagnosis and a period of ill-health which required leaving work they had never regained their social networks. They wrote, "I still feel apart from mainstream humanity". The theme of social isolation extended to romantic and sexual relationships. Many responses related to people feeling that no-one would accept them as a partner and fearing the consequences of disclosing their HIV status, for example the following participant wrote, "I feel that no-one worthy will want me as a partner therefore I will be alone for the rest of my life. I am tired of being asked am I 'clean' by people who want casual sex, this straight away makes me feel like 'damaged goods'. I don't tell people that care for me or love me [about] my status because I feel it will disappoint them and they will look at me differently." Other participants described a fear of rejection or questioning their own self-worth as a major reason why they avoided seeking sexual or romantic relationships, as one participant said, "I yearn for love but who wants me???" Another participant explained that they similarly have a fear of rejection that they linked to their sense of shame about their HIV diagnosis and fear of unwanted disclosure. They wrote, "I can't start a relationship as the fear of telling someone I love that I have HIV that I would lose them (sic) and that they would tell their family and friends. I feel shameful about having HIV and fearful of people finding out. Don't see much of a future for me-especially no chance of a relationship". Some participants reported breakdowns in relationships with their spouse or partner after disclosing their HIV status. In some cases, this affected relationships with children or grandchildren. These experiences caused significant distress, and for some remained a source of anguish many years later as one participant wrote, "My wife left me... She told the two children and their partners without my knowledge and without my being there when they were told... One partner of a son is still having trouble coping with [my HIV status] and this is starting to interfere with [me] seeing my grandson". --- Disclosure and Vigilance Disclosure, and fear of unwanted disclosure, of HIV status was commonly described by participants as feeling like they held a perpetual sense of vigilance, or social and emotional guardedness, related to fear of rejection. One participant described this, stating: "Everyone talks about their issues, but I feel worried to expose my feelings like normal people. Constant need to be aware of what I'm talking about." Another stated that they "cannot confide in any particular friend as I believe the risk of my disclosure would result in an 'open secret" Participants explained that they often managed this fear by trying to minimise disclosure, although this required actively avoiding disclosure, as the following participant stated, "I tell no-one. That way they do not have the pressure on them. Once told you cannot un-tell". Some participants located this fear of disclosure in particular settings, most commonly workplaces: "I have felt fearful at work of having my HIV status disclosed without my consent. I have had this happen to me and then my role at work changed after being 'outed' with weak excuses as to why". --- Discussion The findings from this study suggest that bisexual men living with HIV in Australia experience a greater sense of social isolation than gay men living with HIV and that this is associated with HIV-related stigma. Bisexual men also felt less connected to LGBTQ communities than gay men and were less likely to have contact with other PLHIV. Given community connectedness and peer support is known to be a buffer against the negative impact of stigma on mental health [35,36], these findings suggest that HIVpositive bisexual men may be more vulnerable than gay men to poorer mental health and emotional wellbeing as a result of HIV-related stigma. These findings are significant as they point to a need for social and support programs that are uniquely targeted to bisexual men. Bisexual men may not experience the same benefits from social support programs that broadly target gay and bisexual men as a group. The finding that bisexual men report poorer emotional wellbeing and greater social isolation than gay men is consistent with previous research on bisexuality in Australia and internationally [37][38][39][40]. The reasons for this are not well understood. It is possible that poorer mental health among bisexual people is related to a sense of disconnection from both heterosexual and LGBTQ communities or a response to biphobia and commonly-held negative stereotypes about bisexuality [41,42]. For bisexual men who are living with HIV, the impact of HIV-related stigma may further exacerbate social disconnection or isolation which in turn may have a negative impact on mental health. A sense of social isolation and lack of support was a core theme in responses to the open text (qualitative) survey questions in this study. These participants linked isolation from friends, families or communities to their HIV status, describing fractures in relationships that occurred as a result of their HIV diagnosis. Experiences of isolation or rejection were especially profound for participants where diagnosis with HIV led to relationship breakdown with partners or children. While it is likely that many gay men may have also experienced a sense of disconnection or rejection from people in their lives as a result of their HIV diagnosis [43], the findings of this study suggest that gay men may be more likely than bisexual men to have access to networks of support and community among other PLHIV or LGBTQ communities. Bisexual men may have less support than gay men to manage the negative impact of HIV-related stigma, which can result in social isolation and loneliness [44]. Vigilance around disclosure of HIV status was a common experience for bisexual men in this study as revealed in open text responses. It is likely that this sense of vigilance and fear about disclosure has a wearing effect on mental health for many PLHIV [45]. Again, without a network of peers, many men in this survey may have limited experience of feeling supported as a person with HIV and limited opportunity to let down their guard in a social setting. Peer support and connection plays a role in reducing the negative impact of HIV-related stigma as it offers PLHIV a sense of solidarity and connection. While peer support interventions may not address stigma at a structural level, they are likely to improve quality of life for PLHIV [20]. --- Conclusion and Implications for Practice Understanding how bisexual men's experience of living with HIV differs from gay men may help provide more tailored support for bisexual PLHIV. It is notable that bisexual men in this survey reported poorer emotional wellbeing, less social connection and higher perceived levels of HIV-related stigma than gay men. This suggests that bisexual men living with HIV may have unique support needs that are not adequately addressed through existing programs [46], and that social and community connection remains a crucial and unmet need for bisexual men living with HIV. Mechanisms to increase social connection should be explored. This most likely includes dedicated initiatives run by and for bisexual PLHIV given their unique experiences with respect to connecting with both the LGBTQ and HIV communities. More research is needed to understand the drivers of poor mental health outcomes for bisexual people in Australia, and the mechanisms to challenge and address these. Together with existing research into HIV-related stigma, this would allow a more nuanced understanding of the experiences of bisexual men in relation to HIV. Related to this, tackling biphobia and challenging negative stereotypes about bisexuality will reduce the corrosive impact of biphobia on mental health and wellbeing, something that adds an extra burden to the experience of HIV-related stigma and discrimination for bisexual men who are living with HIV. This should be a consideration in policy related to LGBTQ equality and social justice as well as community-level and individuallevel support programs. --- Data Availability Data for the HIV Futures study is available on request. Requests can be made to the corresponding author. Appropriate ethics approval will be sought to share data on request. --- Author Contributions JP led the development of the survey instrument and data collection, conceived of the analysis and co-wrote the paper with SA; SA undertook analysis of the open-text data and co-wrote the paper with JP; TL undertook quantitative data analysis and wrote the findings section of the paper; GB, JR, AB, MC and AL were actively involved in development of the survey instrument and data collection; all authors reviewed the paper and provided critical commentary on the analysis and text. --- Declarations Conflict of interest The authors declare no conflict of interest. --- Ethical Approval Ethics approval for the study was granted through the [removed for blind review]. --- Informed to Consent All participants completed a survey utilising an online or hardcopy form, completion indicated implied consent as per ethics approval for this study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Social research with people living with HIV (PLHIV) rarely distinguishes between gay men and bisexual men. However, bisexual men may have unique experiences of HIV-related stigma and distinct support needs. In this paper, findings are presented from a cross-sectional survey of Australian PLHIV, which included the Berger (HIV) stigma scale. A total of 872 PLHIV completed the survey, of which 48 (6.0%) were bisexual men. Bisexual men reported higher levels of internalised HIV-related stigma, greater negative self-image and poorer emotional wellbeing than gay men. Bisexual men also reported less social support, less connection with lesbian, gay, bisexual, transgender and queer (LGBTQ) communities, and less connection with other PLHIV. Analysis of data from an open-text question revealed feelings of social isolation and fear of rejection was associated with participant's HIV diagnosis. Study findings suggest that existing social supports for PLHIV may not adequately address the unique support needs of bisexual men.

Background Breast cancer is by far the most common cancer among women both in developed and developing regions, with an estimated 1.38 million new cancer cases diagnosed worldwide in 2008 (23% of all cancers) [1]. In recent years, both incidence of and mortality from breast cancer have declined in the United States; between 1999 and 2006, incidence rates decreased by 2.0% per year, and mortality decreased by 1.9% annually between 1998 and 2006 [2]. However, the incidence of breast cancer is steadily rising in developing countries [3][4][5][6]. In China, the incidence of breast cancer rose from 126,227 cases in 2002 [7] to over 169,000 in 2008 [1]. Preventive screening for breast cancer can increase the number of individuals diagnosed at an early stages and reduce mortality. In China, the main methods of breast cancer screening are clinical breast examination (CBE), mammography, high-frequency ultrasound, and magnetic resonance imaging (MRI). However, due to the lack of large population-based screening evaluations in China, there is no consistent evidence indicating which screening method or combination of methods is best for the country. To date, nationwide breast cancer screening has never been implemented routinely in China, and in places where routine breast cancer screening exists, its usage varies with area-based socioeconomic status (SES) [8,9]. Individuals living in poorer areas are less likely to seek cancer screening compared with individuals living in wealthier areas [10]. Studies over the past several decades have indicated that individuals living in less developed areas often had worse general health compared with individuals living in relatively developed areas [11][12][13][14][15][16][17]. The majority of studies investigating associations between SES and breast cancer stage at diagnosis have also documented socioeconomic and geographic disparities, with higher incidence of late stage breast cancer in lower income areas [18][19][20][21][22][23][24]. Regional disparities in SES are significant in China, where economic development in eastern cities generally started earlier and has been faster than that in the country's interior and in rural areas [25]. Breast cancer survival rates decline with delayed diagnosis. The Surveillance Epidemiology and End Results (SEER) showed that 98.3% of women treated for earlystage breast disease survive five or more years, while only 83.5% of women diagnosed with regional breast cancer and 23.3% of those with distant breast cancer survive beyond five years of the initial diagnosis [2]. Thus, as improvement in breast cancer prognosis would have a significant impact on countries' health budgets, and SES disparities result in different breast cancer outcomes, strategies are urgently needed that target higher-risk areas and resonate with higher-risk population subgroups. Breast cancer prevention and control has been prioritized by China's Ministry of Health, but so far, there has been no systemic study of the relationship between SES and breast cancer stage in the country. In this study, based on the Nationwide Multicenter 10-year (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008) Retrospective Clinical Epidemiological Study of Breast Cancer in China, directed by the Cancer Hospital/Institute, Chinese Academy of Medical Sciences (CICAMS), we explored the effects of both area level factors, namely SES, and individual demographic characteristics on breast cancer stage at diagnosis. --- Methods --- Hospital selection, case sampling, and data collection Hospital selection, case sampling, and data collection methods have been previously described in detail [26]. Briefly, all of China was stratified into seven geographic areas (north, northeast, central, south, east, northwest, and southwest). Then, purposive sampling was used to choose one tertiary public cancer hospital in each region with the following characteristics. Firstly, participant hospitals were the leading public cancer hospitals and regional referral centers providing pathology diagnosis, surgery, radiotherapy, medical oncology, and routine follow-up care for patients with breast cancer. Secondly, their patient source had to include the entire study region. Finally, since patients tend to seek better medical service in big cities, the hospitals were each located in a major city. The selected hospitals provided us with the medical records of breast cancer patients diagnosed between 1999 and 2008. For each year in each hospital, one month was randomly selected, and all inpatient cases for that month were reviewed (January and February were excluded from the random selection to eliminate any confounding effects of China's largest annual holiday). The hospital records were reviewed in each local hospital by local clerks who had been trained systematically in Beijing. Then, standard case report forms (CRF) designed by CICAMS [26] was used to extract the primary medical reports of every qualified patient, including general information, risk factors, diagnostic imaging tests, therapy models, and pathologic characteristics. The reliability and validity of CRF had been assessed by proceeding a pilot study and CRF was regarded as reliable and valid. After collection, the raw variables were coded and sorted into different categories for analysis. Cancer stage at diagnosis was categorized into six groups by archiaters referring to pathologic reports and using the American Joint Committee on Cancer (AJCC) TNM System (0, I, II, III, IV, and unknown stage or not applicable). The cases admitted between 1999 and 2002 were staged using the fifth version (revised in 1997) [27], and the cases from 2003-2008 were staged with the sixth version (revised in 2002) [28]. If individuals were diagnosed with more than one primary breast tumor at the same time, we included only the record of the tumor at the most advanced stage. We excluded tumors with an unknown or not applicable stage, and grouped the breast cancer stages of the other cases into two categories: "early" (stages 0 & I) or "non-early" (stages II & III & IV). The study protocol was approved by the Cancer Foundation of China Institutional Review Board and data were stripped of all personal identifiers, per the board's approved procedures. --- Area-based socioeconomic indicators Area-based SES measures derived from census data have been shown to be related to various health outcomes independent of individual SES [11,12,22,29,30]. Because there is a lack of accepted SES measures in China, we referred to the socioeconomic measures of "percent of the population living below poverty line" and "percent of the population <unk> 25 years of age without a high school diploma" in the United States [31] to devise our own socioeconomic indicators (SEIs) reflecting the economic and education status of the seven areas. These are "GDP per capita," "percentage of health-service expenditure in the regional/provincial public affairs general budget (HSE percentage)," "ratio of urban to rural population (PU/PR ratio)," and "percentage of illiteracy in females aged 15 and over (FPI percentage)." The SEIs of each district or province with a selected hospital were obtained from annual reports issued by the National Bureau of Statistics of the People's Republic of China from 1999 to 2008, except that data on the PU/PR ratio was available only from 2005 to 2008. These SEIs were used to represent the SES of each area. --- Data Analysis Descriptive statistics were used to summarize the demographic and pathologic characteristics of the study population. The mean and standard deviation of quantitative variables were calculated, and the differences among the seven areas were analyzed with one-way ANOVA. Differences in distribution of variables were examined using Mantel-Haenszel chi-square tests or Fisher's exact tests. The single SEI values among the seven areas were compared using one-way ANOVA followed by the Student-Newman-Keuls (SNK) test, a post hoc tests that measures significance among multiple groups. Those areas between which there were no statistically significant differences were combined, forming ordinal categorical levels of SEI status. Trend chi-square tests were used to examine associations between stage at diagnosis and the SEIs status. Additionally, area-based SES was calculated by cluster analysis, an exploratory data analysis tool which sorts objects into groups such that the degree of association between two objects is maximal if they belong to the same group and minimal otherwise. During cluster analysis, the k-means clustering was applied with k = 3, so that the seven regions were classified into three levels based on four SEIs. Binary logistic regression was performed stratified by area-based SES to further explore the association between individual demographic characteristics and breast cancer stage at diagnosis (non-early stage vs. early stage) in SES subgroups. SPSS statistical software version 17.0 was used to analyze the data. Statistical significance was assessed by two-tailed tests with an <unk> level of 0.05. --- Results --- SEIs, individual demographic and pathologic characteristics by area A total of 4,211 eligible cases were included in our study. In general, the differences in SEIs among the seven areas were significant (P <unk> 0.001) (Table 1). During the period studied, the mean of GDP per capita across the seven areas was 13,048 <unk> 5,567 yuan per person, ranging from 8,050 <unk> 3,599 in the southwest to 36,146 <unk> 17,217 in northern China. More than 70% of breast cancer patients were 40-69 years old when diagnosed. Cases in the north (42.1%), northeast (32.8%), and south (29.8%) were more likely to be overweight at diagnosis (BMI <unk> 25.00 kg/m 2 ) than were those in other areas (3.9%-18.8%). Women in eastern, northwestern, and southwestern China were mainly occupied in manual work (80.4%, 54.0%, and 62.7%, respectively), while more women were homemakers in central China (13.2%) than in other areas (average across all areas = 4.0%). Information on patients' education levels was generally unavailable in north, northeast, and southwest China. Among the other four areas, we found that women in central China had higher education levels, with 27.5% receiving higher education (university and above). The vast majority of cases in all seven areas were married. All the above demographic characteristics of breast cancer cases were significantly different among the seven areas (Table 2, P <unk> 0.001). Invasive ductal carcinoma was the dominant pathologic subtype, ranging from 70.1% of cases in the northeast to 92.4% in central China. Among 3555 cases that were tested for estrogen receptor (ER) and progesterone receptor (PR), 3534 had test results included in their records. About half of these (47.6%) were ER + and PR+, and 32.0% were ER-and PR-. In northwest China, only 282 (58.4%) cases were tested for ER/PR and 38.7% of them were ER + and PR+; both of these rates were far lower than in the other areas. Human epidermal growth factor receptor 2 (Her-2) information was available for 2849 patients, the majority of whom (2113, 65.0%) were Her-2 negative. The north had a high testing rate (609/641, 95.0%) and the highest positive rate (286/609, 47.0%) for Her-2. All of the pathologic characteristics of breast cancer cases were significantly different across the seven areas (Table 3, P <unk> 0.001). --- SEIs, area-based SES, and breast cancer stage at diagnosis For the analysis of factors affecting breast cancer stage at diagnosis, we excluded 717 cases with an unknown or not applicable stage, so that the final number of cases in the following analyses was 3494. We found an inverse relationship between breast cancer stage and both GDP per capita and ratio of urban to rural population, while percentage of illiteracy in females aged 15 and over was positively related to breast cancer stage (P for trend <unk> 0.001) (Table 4). However, there was no significant trend correlation between percentage of health-service expenditure in the regional/provincial public affairs general budget and breast cancer stage (c 2 for trend = 3.307, P = 0.081). The area-based SES based on the four SEIs was negatively correlated with breast cancer stage: 14.8% (95/641) of cases in the highest SES areas were diagnosed at stages III or IV, while 20.4% (417/2024) of cases in high SES areas and 25.5% (389/ 1528) in low SES areas were diagnosed at these late stages (c 2 for trend = 80.79, P <unk> 0.001). Patients in Beijing and Guangzhou, where routine breast cancer screening has been implemented for several years [32,33], were more likely to be diagnosed at stages 0 or I (27.9% and 17.5%) than were patients in other areas (Table 3). --- Association between individual demographic characteristics and breast cancer stage at diagnosis stratified by area-based SES With the exception of percentage of health-service expenditure in the regional/provincial public affairs general budget, each of the SEIs was associated with breast cancer stage (Table 4). To further explore the association between individual demographic characteristics and breast cancer stage at diagnosis, we used binary logistic regression and implemented stratified analysis by area-based SES (Table 5). In north China, the only area categorized as highest SES, education information was unavailable for 50.3% of the cases, so we were unable to analyze the effect of individual demographic characteristics in this subgroup. In high SES areas, education appeared to influence the cancer stage at diagnosis. Compared with the patients who had attended university, those only attended middle and/or high school had about a three-fold higher risk of being diagnosed at a late stage. In low SES areas, working women (manual, white-collar, or other) were less likely to be diagnosed at late breast cancer stages than were homemakers (OR: 0.18-0.66). Other demographic characteristics, including age at diagnosis and marital status, were not associated with breast cancer stage at diagnosis. --- Discussion Increasing social inequalities in health in China, coupled with growing inequalities in income and wealth, have focused attention on social class as a key determinant of population health. Our study focused on breast cancer stage at diagnosis as an indicator of health care access and quality, and we found a significant relationship between cancer stage and area-based SES. To our The single SEI values among the seven areas were compared using one-way ANOVA followed by the Student-Newman-Keuls (SNK) test, a post hoc test for significant differences among multiple groups. The areas between which there were no statistically significant differences were combined to form subgroups. a, b, c, and d represent different subgroups for each variable; if two or more regions are followed by the same letter, there is no significant difference between them. (1) : yuan per person per year (mean <unk> SD, 1999-2008) (2) : percentage of health-services expenditure in the regional/provincial public affairs general budget (%) (mean <unk> SD, 1999-2008) (3) : ratio of urban to rural population (mean <unk> SD, 2005-2008) (data is available in the National Bureau of Statistics of the People's Republic of China only for 2005 to 2008) (4) : percentage of illiteracy among women aged 15 and over (%) (mean <unk> SD, 1999-2008) knowledge, ours is the first study to tie SES to health status within a developing country. In this nationwide multicenter study of breast cancer in China, area-based SES was categorized into three levels: highest, high, and low. The Beijing district in north China was the only highest SES area, and also had the highest GDP per capita, percentage of health-service expenditure in the regional/provincial public affairs general budget, and ratio of urban to rural population, and the lowest percentage of illiteracy in females aged 15 and over. Northeast, south, and east China were high SES areas, while central, northwest, and southwest China were low SES areas. This classification accords with the pattern of economic growth and regional inequality in China during the reform era [25]. We found that cases living in low SES areas were more likely to be diagnosed at a later breast cancer stage than were those in high SES areas. Exactly how area-based socioeconomic conditions influence the stage at which an individual is diagnosed with breast cancer is complex. One explanatory factor is that individuals residing in areas with high GDP per capita are likely to have more personal income compared with those in low GDP per capita areas. In the United States, lack of ability to pay for screening services is often implicated as the reason why individuals ( 1) : carcinoma in situ: ductal carcinoma in situ, lobular carcinoma in situ, microinvasive ductal carcinoma, and Paget's disease invasive ductal carcinoma: pure invasive ductal carcinoma and mixed invasive ductal carcinoma other invasive carcinomas: invasive labular carcinoma, tubular carcinoma, medullary carcinoma, mucin carcinoma, and other invasive carcinomas (2) : according to the AJCC/UICC TNM system (3) : ER/PR status was based on all breast cancer cases tested for ER/PR (4) : Her-2 status was based on all breast cancer cases tested for Her-2 with low incomes, or those living in the poorest areas, have lower screening rates and are more likely to be diagnosed with cancer at a late stage [34]. In the present study, we observed that the patients from hospitals located in Beijing and Guangdong were diagnosed at an earlier stage than others. This may partly be explained by the routine breast cancer screening programs that have been implemented in these two cities for several years [32,33]. However, the screening programs in a city may not reflect the screening situation of the whole region. To evaluate the effect of routine screening on early diagnosis, a comprehensive prospective study is needed to compare districts with and without screening programs. Low 0.62 <unk> 0.08 Hunan, Shannxi, Sichuan : GDP per capita: yuan per person per year (mean <unk> SD, 1999-2008) (2) : percentage of health-services expenditure in the regional/province public affairs general budget (%) (mean <unk> SD, 1999-2008) (3) : ratio of urban to rural population (mean <unk> SD, 2005-2008) (data is available in the National Bureau of Statistics of the People's Republic of China only for 2005 to 2008) (4) : percentage of illiteracy among women aged 15 and over (%) (mean <unk> SD, 1999-2008) (5) : Area-based SES is a composite measure synthesized by cluster analysis of the four SEIs Local governments with a high percentage of healthservice expenditure in their regional/provincial public affairs general budgets (HSE percentage) may provide better medical services, giving residents more access to health care such as breast cancer screening, diagnosis, and treatment, which in turn results in earlier diagnosis. Barrya and Breen found that residence in economically and socially distressed or medically underserved neighborhoods tends to increase the likelihood of late-stage cancer diagnoses [19]. However, we observed no clear trend association between HSE percentage and breast cancer stage at diagnosis. HSE percentage may not be a direct indicator of the level of medical service of an area, since HSE percentage may be inconsistent with an area's economic development. For instance, while northeast China had high GDP per capita, it was the lowest HSE percentage of the areas studied. In contrast, we found an association between earlier breast cancer stage at diagnosis and higher ratio of urban to rural population (PU/PR ratio). In China, there is wide SES gap between cities and rural areas, and so urban/rural status is highly correlated with SES. A higher PU/PR ratio hints that more individuals in the area live in cities, and thus may have more access to medical services and ability to afford medical insurance than do residents of rural areas. This would in turn make them more likely to seek prompt medical care for any health problems. These differences may contribute to disparities in breast cancer stage at diagnosis. Several studies have shown that patients with health insurance are less likely to be diagnosed at a late stage of breast cancer [22,35], an effect similar to that found for other cancers, such as colorectal cancer [34,36]. In this study, area-based education was measured as percentage of illiteracy in females aged 15 and over (FPI percentage), emphasizing education of females. We found a trend association between FPI percentage and stage at diagnosis: the higher the FPI percentage in an area, the later the stage of breast cancer at which women in the area were diagnosed. After stratified Women with lower education level may also have higher risk of alcohol-related death and diseases [37]. Low education itself may also create barriers to receiving recommended screening, since low health literacy, low general literacy, and language barriers impact an individual's ability to navigate the medical service system, understand screening options and recommendations, and communicate with healthcare professionals [38]. Working women in low SES areas were more likely to be diagnosed at early stages, which may result from their high breast cancer screening attendance relative to that of homemakers. The results from our previous study [39] and a study by Damiani et al. [40] show that education and occupation were positively associated with breast cancer screening attendance, which in turn may influence breast cancer stage at diagnosis. However, after adjusting for area-based SES, we found no significant associations between other individual demographic characteristics and breast cancer stage at diagnosis. Thus, area-based SES appeared to be the main, underlying factor influencing breast cancer stage at diagnosis. Baquet et al. similarly found that SES predicted the likelihood of a group's access to education, certain occupations, and health insurance, as well as income level and living conditions, all of which are associated with a person's chances of being diagnosed with late stage disease and of surviving cancer [41]. Moreover, Singh et al. [42] and Launay et al. [43] found that at every stage of diagnosis, breast cancer patients from lower-income areas had lower 5-year relative survival rates did than did those from higher-income areas. The presence of additional illnesses and treatment disparities may contribute to these differences [44,45]. Compared with less developed areas, more developed areas possess better detection techniques, and their cases have the ability to pay more for more accurate tests and more effective therapy. For instance, we found that in Beijing, the highest SES area, more than 95% of cases were tested for ER/PR/Her-2. In most of the low SES areas, however, only about half of cases were tested for Her-2. Even so, this pattern was in agreement with the guidelines of the Breast Health Global Initiative (BHGI), which contend that Her-2 measurement is problematic in limited-resource settings due to the high cost of immunohistochemical analysis, fluorescence in situ hybridization, and trastuzumab therapy, and recommend introducing this test only at the maximal-resource level [46]. The present study is the first geographically epidemiologic study of breast cancer in China. Although we only included women with breast cancer attended to these 7 regional referral hospitals, we expect the effect of SES and later stage of diagnosis will be much stronger than we found in this study (if we included those women attended at local hospitals, such county hospitals). Our findings thus provide a baseline for understanding the country's patient and tumor characteristics. Our evaluation of associations between breast cancer stage and area-based SES, as well as individual demographic characteristics stratified by areabased SES, may help to identify high risk areas and the most important and controllable regional and individual factors influencing breast cancer stage at diagnosis. However, this was an ecological study, so we must recognize the possibility of ecological fallacy. In addition to the individual-level variables examined in this study, it would be useful to examine the effects of other variables, such as access to breast cancer screening services, alcohol use, family history of breast cancer, and lack of exercise, as they may affect breast cancer stage at diagnosis by influencing women's attitudes and behaviour to breast cancer prevention and screening. However, their effects could not be analyzed in this study due to the high proportion of missing data; we plan to test the effects of these variables on breast cancer stage in a future study. Another limitation is that we lacked data on individual SES, which was likely strongly correlated with both education level and area-based SES. Although our method of calculating area-based SES may not have assessed actual socioeconomic conditions with complete accuracy, our results for area-based SES accord with the pattern of economic growth and regional inequality in China during the reform era [25]. Thus, we think this indicator has some validity. However, a more comprehensive, validated area-based SES measurement is needed that takes into consideration factors such as concentration of poverty, health insurance coverage, proportion of the population with blue collar jobs, unemployment rate, median household income, and median value of owner-occupied houses [47]. Finally, area-based SES was measured at a large geographic scale in our study; it would be instructive to study area-based SES differences among smaller geographic areas in a future study. --- Conclusion In summary, the effects of area-based SES are complex and multidimensional. Women in lower SES areas may be more likely to ignore symptoms for a variety of economic, social, and cultural reasons. Additionally, women in lower SES areas tend to be less educated and may not understand disease progression, and so may not seek treatment until breast cancer has developed into a late stage. WHO's national cancer control programmes suggest that research to determine the most cost-effective cancer control strategies is especially relevant in developing countries, perhaps even more so than in industrialized countries. However, the available research data for effective and efficient cancer control decision-making are extremely limited in these countries. Our results suggest that in China and other developing countries with similar socioeconomic conditions, breast cancer control programs should focus on ensuring adequate access to screening and improving breast cancer awareness and education. This would facilitate diagnosis at an early stage, particularly for populations living in socioeconomically disadvantaged areas, and would in turn decrease breast cancer mortality and improve surviving patients' quality of life. --- Abbreviations SES: Area-based socioeconomic status; SEIs: Socioeconomic indicators; SEER: Surveillance Epidemiology and End Results; HSE: percentage Percentage of health-service expenditure in the regional/provincial public affairs general budget; PU/PR: ratio Ratio of urban to rural population; FPI: percentage Percentage of illiteracy in females aged 15 and over; CICAMS: Chinese Academy of Medical Sciences; AJCC: American Joint Committee on Cancer; ER: Estrogen receptor; PR: Progestin receptor; Her-2: Human epidermal growth factor receptor 2; CRF: Case report form; CBE: Clinical breast examination; MRI: Magnetic resonance imaging; BHGI: Breast Health Global Initiative. Authors' contributions QW helped to analyze, and interpret the data. She also drafted the initial manuscript. JL helped to design the study and interpret the data. SZ helped design the CRF to collect the data. JYL was the local PI who helped with data collection and performed critical revisions of the manuscript. JHF, YP, and RH helped with the data management and analysis. BNZ helped to design the study and is the clinical PI of the study. BZ, HJY, XMX, ZHT, HL, and JJH were the local PIs who helped with data collection. YLQ was the PI of this study and assisted in designing the study and performed critical revisions of the manuscript. All authors read and approved the final manuscript. --- Competing interests The authors declare that they have no competing interests.

Background: Although socioeconomic status (SES) has been focused on as a key determinant of cancer stage at diagosis in western countries, there has been no systemic study on the relationship of SES and breast cancer stage at diagnosis in China. Methods: The medical charts of 4,211 eligible breast cancer patients from 7 areas across China who were diagnosed between 1999 and 2008 were reviewed. Four area-based socioeconomic indicators were used to calculate area-based SES by cluster analysis. The associations between area-based SES and stage at diagnosis were analyzed by trend chi-square tests. Binary logistic regression was performed to estimate odds ratios for individual demographic characteristics' effects on cancer stages, stratified by area-based SES.The individual demographic and pathologic characteristics of breast cancer cases were significantly different among the seven areas studied. More breast cancer cases in low SES areas (25.5%) were diagnosed later (stages III & IV) than those in high (20.4%) or highest (14.8%) SES areas (c 2 for trend = 80.79, P < 0.001). When areabased SES is controlled for, in high SES areas, cases with less education were more likely to be diagnosed at later stages compared with more educated cases. In low SES areas, working women appeared to be diagnosed at earlier breast cancer stages than were homemakers (OR: 0.18-0.26).In China, women in low SES areas are more likely to be diagnosed at later breast cancer stages than those in high SES areas.

INTRODUCTION The transition from secondary to tertiary education is challenging for first-time entering students (FTENs) for a variety of reasons. According to Ajani and Gamede, these students experience "personal stress and emotional maladjustment in their new surroundings" 1 as a result of these reasons. Numerous academics have done transition studies from various angles; nevertheless, the author borrowed concepts from a few experts. Mgqwashu et al. explored how rural students handle the transition to university, as well as how earlier cultural and educational experiences influence their higher education trajectories. 2 They argued that university practices, to be successful in the arts should they choose to continue their education in the Further Education and Training (FET) phase of the Department of Basic Education. 11 The author is currently teaching first-year Arts and Culture students at the institution. The program focusing on Arts and Culture at this institution was relaunched two years ago. That is, in 2021, as a brand-new initiative. Most of the prospective students for admission into the program come from rural schools. 12 Sadly, most South African rural schools do not include subjects related to the arts in their FET programs; some of them that do include a few of these subjects are in the process of phasing them out of their curricula for a variety of reasons known to them. 13 As a result of the elimination of arts subjects from the FET Band, there is a knowledge gap between the content covered in the GET phase and the material covered in the FET Band itself. This puts the quality of the requirements for entering universities at risk. Because of this, most students coming from these rural schools are accepted into Arts and Culture programs based on auditions and interviews rather than on the content that they would have learned in the FET phase. Some of the students who had participated in one or more art subjects in school were not passionate about continuing with them but were primarily motivated by the need to earn good grades in grade twelve to be admitted to the university. As learners in school, they were almost entirely dependent on their teachers for the knowledge and direction that they required throughout their education. Following the receipt of their school endorsement, they were turned down in the programs they applied for to pursue their studies and opted for acceptance into the Arts and Culture program, which they last studied in the senior phase of the GET Band. As a result of the difficulties caused by the pandemic in 2020, in the following years, the newly admitted students went through a period of abrupt changes in the methods of teaching and learning. They were first exposed to a fully online mode of learning, which required them to make a significant shift in their learning methods. 14 Some of these students do not know much about technology, so any problems they have with it are hard for them to understand. 15 Because they are in the process of becoming educators (researchers and lesson developers) who will pass on knowledge and skills to empower their students, this sudden change caught them off guard. 16 They are moving away from the idea of learning by rote and recalling information to the idea of becoming information providers who must use their creative skills to share what they know. Because of these changes, the author was able to see that her first-year students in the Arts and Culture class had a hard time accepting new ways to learn with technology at the time. The study thus sought to answer the following question: What strategies should be used to support FTENs Arts and Culture students transitioning from rural contexts to HEIs? --- LITERATURE REVIEW Theoretical Background The paper is based on Bronfenbrenner's ecological systems theory, which was later changed to bioecology. 17 This theory states that individuals are still influenced by their environment while their personal characteristics become important, particularly through interactions between the individual and other influencers in each of the systems. Human development, according to his theory, is a lifelong process that reflects an individual's understanding of the environment and his or her relationship to it. According to Bronfenbrenner, to determine the developmental influences of the environment on the individual, one must closely observe the individual. 18 Furthermore, Bronfenbrenner states that 'there are environments in which individuals participate in activities, and each of them has a role in that environment.' 19 He defines roles as an individual's actions, the expected actions of society on the individual, and the relationship of other people's actions to the individual. In his assertions, he explained that individuals' roles may shift over time, resulting in role transitions and ecological shifts. 20 On the other hand, Maree, in his explanation of Erikson's psychosocial development theory, states that the importance of gaining insight into core aspects of people's identity development, particularly during their early years, promotes counselors' understanding of how clients experienced their early years and successfully or unsuccessfully navigated their transition to adulthood. 21 According to Maree, such understanding also enhances counselors' capacities to assist clients in effectively adapting to new circumstances and finally entering the world of work. 22 Because lecturing student teachers is a unique form of education, the above theories supplement the goal of this text. They help people develop not only academically but also intellectually, psychologically, emotionally, and so on. Teaching is a lifelong learning process that involves the continuous delivery of knowledge to future generations. It is critical in the sense that if the foundations are weak, the consequences will be disastrous for many people. As a lecturer, these close interactions with students provide a suitable platform for observing activities in the relationship between teacher, student, and subject for optimal development. With continuous engagement, development occurs gradually, resulting in transitory alterations. During this interaction, demonstrated conduct and attitudes have a significant part in attaining the goal of transition in the educational setting. --- Empirical Review Transitioning affects students differently depending on whether they live in rural or urban communities but this paper focuses on rural students. 23 Academics define rurality differently based on the geographical location of South African schools and the benefits they receive. Ganss investigated how students from rural communities define the transition to higher education by examining their common lived experiences in the first term of university and the difficulties they face upon entering university. 24 Timmis and Muhuro investigated students contextualized digital practices before, during, and after joining the university, based on Southern African Rurality in Higher Education, a longitudinal study of rural students' negotiating of transitions to and through HE in South Africa (SARiHE). 25 They contend that students from rural communities suffer significant obstacles when confronted with technical systems, procedures, and practices that they have little or no prior experience with because of 'ongoing coloniality.'26 This is a significant barrier because many students in rural areas do not have access to such facilities due to a lack of resources and connectivity challenges in their remote locations. Although some students had access to cell phones before entering university, the requirement to be technologically proficient presents significant difficulties in transitioning to university digital tools, making them feel marginalized. On the other hand, Timmis et al., focus on the students who have faced unique disadvantages because of colonialism and the ongoing effects of apartheid. 27 In their claims, they talk about problems that students face when applying to, getting into, and being a part of the university. For example, students from rural areas are not given enough credit for their ability to change higher education, universities are not ready for students from rural areas, and ideas like decoloniality and cognitive justice are not talked about enough. They also talk about how universities could meet the needs of rural students by recognizing their cultural practices and bringing them into conversations about the curriculum as key agents in university change. In a nutshell, effective student transitioning entails not only exposing students to newly introduced avenues but also identifying and incorporating their earlier experiences. If transition is not handled properly, it can have a variety of negative repercussions on a person's life, including psychological, emotional, and physical effects. Bowles et al., identify variables that should be examined as first-year students transition to the university as both intrinsic (what students bring with them) and extrinsic (what the university provides). 28 This demonstrates that, while students require assistance, both students and university agents have roles to play in ensuring an effective transition. Students must be willing to seek help and show dedication to their studies as well as the university culture. Furthermore, universities must implement techniques to help students acquire study skills, time management skills, research skills, and assignment writing skills in their first year by providing learning support services that are open to all students. Bowles et al., conclude their research by stating that universities should focus not only on encouraging key intrinsic measures, such as student effort towards their studies and transition by seeking the assistance of academic staff but also on the key extrinsic variable of "LEARNING@UNI's web-based resources, which appear to be a viable and useful tool provided for successful learning." 29 Geter investigated how social workers perceive the impact of social support strategies in the Midwestern United States by examining the complexities of shared decision making, reliance on self-determination, and common goal setting when implementing resources for youth transitioning from foster care to independent living. 30 He defines social support as both verbal and nonverbal communication. But, on the other hand, individual developmental training is suggested by Nolan et al. 31 This researcher agrees with these researchers that individual attention is essential, although group work is crucial in the teaching and learning setting since it enhances presentation skills and the effectiveness of imparting knowledge. Before demonstrating competency in subject matter mastery, students require counselling, physical aid, emotional and constructive criticism, as well as cognitive or behavioural feedback within a social group. According to Geter, subject matter mastery includes frequency of contact, access to resources and services, quality of support that develops resiliency, and fostering self-determination, decisionmaking, and goal achievement. 32 As a result, students require multiple but distinct forms of assistance to develop holistically and handle transitions maturely. Even though what the students used for the study transition focuses on teaching and learning, the support strategies they require are consistent with other scholars' research. This means that any type of transition requires some kind of assistance. Support from facilitators is essential for first-year university students to help them cope with unexpected anomalies that contradict their earlier experiences. Ajani and Gamede recommend that all stakeholders should provide adequate and necessary support services for the students to integrate them appropriately into the system and make a significant effort to adjust to their new environment. 33 The competence of lecturers stimulates students' academic advancement, so they must produce learning materials that allow them to display appropriate capabilities in their teaching and learning sessions and to "foster students' dynamic beliefs." 34 This effective, flexible learning encourages lecturers to create an innovative learning environment that meets the demands of the community. Lecturers should be able to adjust current information to "offer on-demand learning content that supports repetition and flexibility while providing for individual student preferences." 35 "In any language program, social interaction is a powerful tool to improve communicative competence." 36 Evans and Morrison, concur with these scholars, adding that "lecturers must design opportunities that provide collaborative support for students in the assessment writing process to fulfil their role in transition support," implying that language is an important element of any interaction. 37 In addition to the claims above, the role of lecturers in the well-being of undergraduates should encompass not just academic work, but also psychological aspects. 38 They define the role as lecturer support, benevolence, lecturer competency, lecturer availability, interaction, and the lecturer's attitude toward their work. As a result, the role of the academic as a subject matter expert (content facilitator, assessor, resource provider) has expanded to include the roles of designer, metacognition facilitator, process facilitator, advisor, and co-learner, Othman affirms. 39 --- RESEARCH METHODOLOGY Methodology is a research strategy and action plan (research design that shapes chosen research methods) that translates ontological and epistemological principles into guidelines that show how research is to be conducted as well as principles, procedures, and practices that govern research. 40 The interpretive paradigm and the qualitative approach go well together and were used in this research paper. The paradigm "includes interconnected components such as ontology, which seeks an answer to reality; epistemology, which is concerned with how researchers seek to discover this reality by observing the world; and methodology, which is a design process for conducting the study, identifying methods to be used that define means of data collection." 41 The strength of the interpretive paradigm stems from its naturalistic approach, which is based on natural modalities of human communication. So, the interpretivist paradigm and qualitative methods would let researchers dig deeper by looking into experiences and perceptions of a specific social situation, and interpretivists would use these experiences to build and interpret their understanding based on the data they collected. 42 This research was carried out through self-study and practice. Self-study is the investigation of one's own practice with the goal of improving it, articulating, and confirming one's professional expertise, and contributing to the body of knowledge of teacher education. 43 --- Population and Sampling The raw and original information was drawn from the researcher's 2022 first-year students' experiences, with whom she interacts daily. Moser and Korstjens define purposeful sampling as "the selection of participants based on the researchers' judgment of which possible participants will be most informative." 44 Each participant completed a consent form stating that they were participating voluntarily and that they had the right to withdraw at any time. Participants are from several ethnic groups, such as Xhosas, Zulus, Sothos, and others, but Xhosas and females predominated the lot. --- Research Instruments To collect data, open-ended qualitative questionnaires, document analysis and participant observation were used. "Participant observation is a method for collecting data through the involvement in and observation of a group or individuals over an extended period of time." 45 Adejimi, Oyediran, & Ogunsanmi state that "further qualitative data capture can be included in the questionnaire to give the research richer and more meaningful results, particularly when such research involves personal opinions." 46 Document analysis is a methodological process for assessing and evaluating printed and electronic documents (computer-based and Internettransmitted). 47 Data Collection Procedure A Google Forms link was used to disseminate an open-ended questionnaire to first-year Arts and Culture students from diverse linguistic backgrounds enrolled in 2022. The use of Google Forms to deliver electronic survey questions allows for data collection from any platform and access to respondent systems at any time and from any location. 48 The cultural diversity of the participants indicates that the difficulties of transitioning are not restricted to a specific cultural group but may have a negative impact on any individual. Consequently, the dominance of Xhosa participants in the group does not imply that Xhosa people are the most affected by transitioning; rather, the majority of Xhosa students are inhabitants of the affected community. Dake & Gyimah state that an open-ended questionnaire is administered to elicit students' genuine emotions, perspectives, and grievances because it permits students to provide valuable free-form responses with frequent and consistent identification of relevant keywords for simpler interpretation. 49 Furthermore, they assert that an open-ended questionnaire approach to qualitative feedback allows learners to provide insightful free-form responses without a predetermined list of responses. A desire to eliminate bias led to the use of a Google Forms link. Therefore, respondents were kept anonymous to encourage students' participation and reduce the possibility of bias since the author is also their lecturer. Despite the author's efforts to engage students, only 17 out of 100 Arts and Culture students who were given questionnaires responded. The respondents included five Zulus, two Sotho students, and ten Xhosas. Since most students are from rural areas, technological illiteracy may account for the lowest number of respondents. The alternative possibilities could be a delay in response due to network issues caused by electricity load shedding or the liberty to participate voluntarily. Data were also gathered from observation reports on interactions during teaching and learning activities and from the analysis of student scripts during the assessment period. --- Data Analysis In self-study, the data-generation and data-analysis processes are typically mutually reliant and might even occur concurrently. 50 Data was analyzed concurrently with data collection as the teacher education practice continued. In most qualitative techniques, coding (the process of data reduction) is an element of data organization. 51 Based on these arguments, thematic analysis was deemed the most appropriate for any study attempting to find meaning through interpretations. 52 The data was classified into themes and examined. --- FINDINGS AND DISCUSSIONS In the sections that follow, the findings are presented and the themes that came out of the data analysis are discussed. --- Insufficient Exposure to Advanced and Modern Cultural Activities Students in rural schools do not get enough experience with advanced and modern art activities, and as a result, they are insufficiently prepared for university education. Due to their diverse cultural backgrounds, the students' exposure differs. Some were not exposed at all, while others gained elementary knowledge from tour visits. Mak and Fancourt advocate that "arts and cultural participation include performing arts (dancing, singing), visual and literary arts (textiles, painting, writing stories), and culture (visiting museums, galleries, the theater, heritage sites)." 53 Music and dance, for example, are vital for both entertainment and religious purposes in family events or rituals. Rural school students may not be exposed to galleries, museums, or theater sites, or they may have barely visited them, but they participate in cultural activities in their communities. Numerous students are exposed to community cultural activities as participants and viewers. --- "I participate in reed dance with other kids in my community. (Participant 6) As participants, students volunteer for local community activities, but most students attend cultural family activities as well as adhere to religious values, beliefs, and norms. --- We keep that aid in developing a familiarity or understanding of the norms, values, and beliefs. (Participant 10) Therefore, students learn about cultural values through art activities such as dance, music, poetry, painting, traditional clothing, and so on. Community members like our culture when we are performing a ritual they will participate too. (Participant 17) Participants' responses demonstrate limited exposure to cultural events that emerge as entertainment and artifacts that are exchanged as presents among family members for customary purposes. --- Lack of Enthusiasm and Passion for Arts Subjects All first-year Arts and Culture students learned any two of these art forms in the General Education and Training (GET) band. They enjoyed arts topics as students during their school years since some teachers were enthusiastic about teaching them; however, some teachers lacked enthusiasm and passion. The learners were carrying out exercises on their own, without the assistance of the teacher. --- We were practicing by ourselves, we were in groups. (Participant 3) As a result, practical activities took precedence over theoretical ones. (Participant 1) I devote more time to practice. (Participant 17) On the positive side, teachers would allow students to bring their cultural knowledge and share it with other students through practical tasks.Some of these students felt motivated by their school experiences to pursue careers in the arts. Choral singing and other cultural events at school expose students to art knowledge. --- I have experience of singing and dancing. (Participant 12) The teacher or the choir conduct was always present in the practice. (Participant 4) In most FET schools these activities were the only sources of arts knowledge. Sadly, some students highlighted that they were unable to continue with art activities after passing grade nine because many rural schools do not offer arts subjects in the FET band. There are only a few students who indicated that they did one of the art courses in the FET band. Because of this several students did not choose arts and culture as their first choice for pursuing their studies but chose it because they were not admitted to their preferred courses. Some students applied for the degree out of passion, although they did not take arts subjects in the FET schools. --- It affected me because I didn't study music at my high school, and I don't know Music in Arts and Culture. (Participant 10) It was my second option; I was interviewed for auditions of music. (Participant 8) Students face difficulties while transitioning to university after this three-year break without exposure to these arts subjects. They struggle not only with emotional, psychological, and technical adjustment to the new context but also with content knowledge gaps of these subjects. --- I lost the background of arts and culture or the basics. (Participant 11) Students suggest that arts subjects should be compulsory in the FET schools to prepare students for their further education. --- Arts and culture should be taught in high school as a subject from grades 10 to 12, and it should be in all classes to instill learners' interests and abilities because some choose what they do not even have any ideas about and fail or drop out. (Participant 5) I would change arts and culture to be a major subject in every school because many people love and enjoy this subject except me. (Participant 7) Despite participants' concerns about the enthusiasm and passion of teachers who teach arts subjects in schools, they demonstrate a strong desire to learn arts subjects in an appropriate and effective manner. --- Impact of Transitioning on Students' Learning The transition from secondary to university negatively impacted students' learning on all levels-emotionally, psychologically, physically, and intellectually, as a result of virtual fatigue. The change in environment was not bad at all, but the methods of teaching and learning changed drastically. 54 --- Not bad but change of environment and new people in life. (Participant 16) According to Herkulaas and Oosthuizen, "Good content design is at the heart of student transition, but other aspects such as access to resources help with the transition into university." 55 Unfortunately, these students were the victims of the 2020 COVID pandemic, yet they transitioned to university amidst the challenges that came with the pandemic. The change in the mode of delivery negatively affected a vast number of students. The shift to online learning brings issues such as technological faults and incompetence in online services. 56 It was difficult to run a classroom successfully because you could not identify reality. Students would complain about "the inability to access or use online learning and teaching tools." 57 Several students would log in and go about their personal business instead of listening to the lesson presentation. Although you would want to interact with students personally as a lecturer, the enormous number of students connected at the same time makes it difficult to keep your students' attention. 58 Online learning was a challenge to me because I would miss some of the classes and sometimes the lesson would end without understanding. (Participant 2) The reliance on online teaching had severe implications due to the non-accessibility of proper monitoring devices, owing to ignorance, irresponsibility, and peer pressure, and many students ignored their studies and opted for plagiarism. As a result, students copy and paste information from websites and then share it as a group during online assessments. Similarly, Nwosu and Chukwuere assert that "attitudes of students towards plagiarism are influenced by factors such as trying to do quick work, influences from peers and family, availability of online information, a lack of understanding of plagiarism, poor learning, reading ability, and performance goals." 59 Such actions compromise the quality of content knowledge acquired by students, who subsequently lack creative thinking and academic writing skills. --- RECOMMENDATIONS The findings presented and discussed above clearly demonstrate that difficulties in transitioning to university exist because of unresolved recurring challenges in the preceding years. This study recommends that the Department of Education require all schools to offer any of the arts disciplines to minimize the three-year gap for students who do not study arts subjects in the FET. The support strategies do not suggest the necessity for intervention solely when students begin their studies at the university but also their readiness prior to university entrance. Proper selection of subjects for FET admissions will mitigate the number of students who choose Arts and Culture as their last alternative after being denied their intended courses. Such gaps demonstrate the critical necessity of assisting students in selecting subjects that will allow them to pursue their professional goals once they complete ninth grade. Students must be guided in terms of their previous grade-level ability in those specific areas. Parents must also monitor their children's abilities beginning in grade 10 so that they can be led to subjects in which they excel. Also, aptitude exams for grade ten admissions could establish the authenticity of grade nine results as well as the learners' competency in the courses they want to study further. The above-mentioned strategies may help determine the degree of competency for students advancing to grade twelve. For students entering university from various backgrounds with different abilities, the university Directorate of Learning and Teaching provides many support services that assist the teaching and learning processes. Lecturers should work effectively with the center to gain access to any information that would enhance learners' progress. There are tracking systems that give information about students' profiles obtained at the beginning of the year through services. Lecturers should acquire this information ahead of time to develop their learning materials and identify the types of students they are teaching. These lecturers should base their teaching on this understanding. Even their teaching approaches should be relevant and customized to the students they have. During the teaching and learning process, lecturers must use methods that consider the different ways that students learn. Students' support needs differ depending on their talents and skills. Lecturers should send students to writing centers to be trained in computer skills, reading, and writing abilities, presentation skills, time management skills, and so on. The lecturer should monitor the students' development on an ongoing basis. It is also vital to obtain feedback from the support team after each intervention and to report to the team on the progress noticed by the lecturer for further intervention, if necessary. Baseline assessments prior to lesson presentation help track development and decide on new teaching and learning strategies. Groupwork is widely recognized as the most effective way to help students improve their presentation skills. It allows students to learn from their peers at their own pace and level of comprehension in a conducive environment, but the lecturer is essential as a facilitator who advises and provides advice when necessary. These sessions also help lecturers identify students with insufficient skills so that they can seek intervention as soon as feasible, as well as support them individually. Lecturers, on the other hand, should encourage students to use library services. Lecturers should assign students projects that require them to read articles and draw their own conclusions. These assignments should be academically focused, with students conducting research, reading, and writing to publish articles when they finish their studies. Finally, communication between schools and universities for career counseling is required even before students apply for admissions to assist them in selecting appropriate courses for themselves. --- CONCLUSION This was a study of the researcher's practice that investigated the trajectories of Arts and Culture students transitioning from rural secondary contexts to higher education. The paper sought strategies that the researcher can employ to assist Arts and Culture first-year students to mitigate the negative impact on students transitioning to HEIs. To assist students in successfully transitioning from secondary to higher education, a collaboration between students, lecturers, and support team services at the university, with lecturers being mindful of the different types of students they have, can provide students with the necessary abilities to adapt to their new environment. In teacher education, group work is an effective method to impart teaching skills, but individualization allows the lecturer to assess students' strengths and limitations and support them based on their abilities and preferences. The question is: Can individualization only be effective in assisting students transition to the first-year level at the university? This motivated the researcher to undertake research on teaching methodologies that can be used to prepare students in arts subjects in the FET phase for the transition from secondary to higher education. --- BIBLIOGRAPHY --- ABOUT AUTHOR Nonceba Cynthia Mbeshu-Mhlauli is a researcher who aspires to indigenize the teaching and learning of arts subjects. Although the South African curriculum encourages the incorporation of indigenous knowledge into the arts curriculum, its implementation remains challenging due to a number of unknown and ongoing factors. Consequently, she seeks teaching strategies that can assist her in attaining her objective. Nonceba Cynthia Mbeshu-Mhlauli holds a Master of Music Education and teaches Music, Arts and Culture at Walter Sisulu University.

This is a self-study of practice investigating the trajectories of Arts and Culture students transitioning from rural secondary contexts to Higher Education Institutions (HEIs). The main objective of this paper was to propose support strategies for first-year university students majoring in Arts and Culture at a specific HEI. First-time entering students (FTENs) struggle to transition from secondary to postsecondary education for a variety of reasons. The paper was based on Bronfenbrenner's bioecological systems theory, which states that individuals are still influenced by their environment but become important through interactions with other influencers in each system. This research inquiry utilized an interpretive paradigm and a qualitative approach through a self-study of practice, and data was gathered using a qualitative questionnaire, observations, and document analysis. The data was thematically analyzed concurrently with data collection. Participants were selected from several ethnic groups, such as Xhosas, Zulus, Sothos, and others. They were selected purposefully, but Xhosas and females predominated. The findings revealed insufficient exposure to advanced and modern cultural activities; a lack of enthusiasm and passion for arts subjects; and the negative impact of transitioning on students' learning. These hiccups have a detrimental effect on students' transitions to HEIs. The paper recommended these support strategies: capacity building of students with proficient skills to transition effectively; use of teaching and learning strategies that cater for students' different learning styles; and consideration of students' backgrounds and indigenous knowledge. These variables may help first-year Arts and Culture students transition smoothly into higher education.