Article Text
PDF
Abstract
Background Research on the links between violence and social capital has produced mixed results and is mainly limited to Western countries.
Aims To assess the relationship of social capital to physical fights among unmarried boys aged 13–19 years living in three disadvantaged neighbourhoods in the outskirts of Beirut, controlling for variables associated with youth violence. Lebanon has a history of civil and cross-border war, which may influence the production of violence at the individual level.
Methods 674 boys were interviewed. The dependent variable was being involved in a physical fight in the last 3 months. The independent variable was social capital, classified into six categories: civic engagement and community development; locational capital; trust; reciprocity; hypothetical social support; and social network.
Results One fifth (20%) had been in a fight in the last 3 months. Youth who were members of a group, who had been victimised, who could ask for help from a family member/friend/neighbour when they had a problem, and who liked living in their neighbourhood were more likely to be involved in a fight.
Conclusion This research supports the literature that indicates that physical fights are positively related to both individual and social capital variables. Social capital had a detrimental effect on physical fights. This may be related to bonding in a gang-like way. While high levels of social capital are good for other health outcomes and thus should not be reduced, interventions that channel youth energy in positive ways are needed.
- Youth
- boys
- violence
- social capital
- Lebanon
- adolescent
- community
- developing nations
- public health
- violence
Statistics from Altmetric.com
Introduction
Adolescent violence is a major international public health concern with a significant health and social burden for all ages and socioeconomic groups.1 2 The WHO defines violence as ‘the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation’.2
Research suggests that there are multiple determinants of youth violence. These include personal (individual), social (family, peers, and school), and community factors. Individual factors include beliefs supportive of violence, social cognitive deficits, exposure to stressful life events, and religiosity. Family factors involve inappropriate parental behaviour, low emotional attachment to parents/caregivers, poor monitoring of children, exposure to violence, and poor family functioning. Peer/school factors are negative peer influences, low commitment to school, academic failure, and school environment. Finally, community factors comprise impoverished environments, high level of neighbourhood disruptions, and diminished economic opportunity.3–8 These factors have resulted in the acknowledgement that ‘violence is a joint product of the individual and situational context’ (Williams et al, p198).4 More recently, social capital has emerged as an important contributor to violence and safety.9 10
Physical fights are considered to be one of the most frequent demonstrations of interpersonal violence among youth.1 Literature indicates that 30–50% of youth have been involved in a physical fight in the past year, with differences by country.11–15 In all the above research, boys were more likely to have fought than girls. In addition, younger adolescents (measured by age or grade in school) are more likely to fight than older adolescents.16 17
Social capital is defined as the ‘stock of investments, resources and networks that produce social cohesion, trust and a willingness to engage in community activities’ (Lynch and Kapalan, p307).18 There are various studies showing strong links between social capital and health-related outcomes, including mortality,19 self-rated health,20 mental health,21 22 and risky behaviour such as drug abuse,23 sexually transmitted diseases,24 and smoking.25 With respect to violence, the research findings have been mixed. Some research has suggested that greater social capital is protective against violent crime.26 27 However, Wright and Fitzpatrick5 found social capital to be linked to higher frequency of fights and weapon use.
Research on violence—and its determinants among youth—is common in the USA and other Western countries; however, very little research on this topic has been conducted in the Arab region. The Arab region has been the site of many conflicts in the recent past. The potential for future violent conflict is also high given the stalemate on major strategic issues in the region, including the Arab–Israeli conflict, as well as the Western focus on control of ‘terrorism’. Structural violence of this kind condones and promotes individual violence. However, the strength of social networks in the Arab region and its collective culture where the extended family and community are involved in monitoring of behaviour, are often touted the protective factors for adolescents. Therefore, the assessment of the link between individual violence and social capital is particularly interesting in this context.
The present research aimed to assess the relationship of social capital to physical fights among unmarried boys aged 13–19 years living in three disadvantaged neighbourhoods in the outskirts of Beirut, Lebanon, controlling for other variables known to be associated with violence in youth. Lebanon has a history of both civil and cross-border war, which may also influence the production of violence at the individual level.
Methods
Community context
The data used for this analysis come from the Urban Health Study (UHS) conducted by the Center for Research on Population and Health at the Faculty of Health Sciences, American University of Beirut. The UHS examined the health of residents in three low-income suburbs of Beirut, with a focus on adolescents, women, and older persons, as well as the influences of the wider social and economic conditions on their lives. The main objective of this study was to provide policy-relevant analyses of the health consequences of population change in the context of relatively impoverished urban settings. The focus on the UHS was on Beirut as it is the largest urban area in Lebanon, and the capital. Three neighbourhoods, namely Nabaa, located in the eastern suburb of Beirut, and Hay El Sellom and Burj El Barajneh Camp, in the southern suburb, were chosen for the UHS. These neighbourhoods were selected within the urban poverty belt surrounding Beirut, after a thorough investigation of all the information available on Beirut's suburbs, several visits to the areas, and discussions with local non-governmental organisations and experts in the field. The main criteria for the choice of communities were their dense war displaced populations, prevailing poverty conditions, rural–urban mobility, and lack of basic infrastructure.
Each community has a unique history. Nabaa has grown as a result of rural–urban migration and internal displacement. More recently, foreign workers have come into Nabaa from surrounding Arab countries (Syria and Egypt), as well as from Asia and Africa. Today, and as a consequence of these population movements, this community is ethnically diverse and overcrowded. Nabaa has a reputation for being a ‘reservoir of labour’, a connotation for low social class.
Residents of the Burj El Barajneh Camp are refugees who originally came from Palestinian villages in waves since 1948 and settled in neighbourhoods, which currently carry their names. The poor conditions of the Camp may be attributed to the legal and political restrictions imposed on the Palestinian refugees by the State. The Lebanese State does not allow significant improvements in the Camp since it is seen as a temporary residence. Participants tell the story of how the housing conditions slowly changed over time from tents in 1948 to zinc roof rooms with limited numbers of primitive public toilet facilities. It was not until the late 1950s that they were allowed to build houses with cement rooftops.28 29 Despite these difficult conditions, the Camp has strong social ties among its residents.
Hay El Sellom was originally an olive grove. The area developed due to lack of work opportunities in Beqaa and the war front in Southern Lebanon, which resulted in several migrations of villagers from these two areas. People who came to live from the South perceived their stay as temporary. However, due to the prolongation of the Lebanese civil strife, the people stayed, seeking security and employment. Despite many years of settlement, the government still does not recognise the area in legal terms. The buildings are not registered and on official maps the area is labelled ‘olive groves’.30
Other socio-demographic characteristics of the three communities can be found in table 1.
Socio-demographic characteristics of the three communities
Methods
The data used in this analysis are from the UHS.31 This study was conducted by the Center for Research on Population and Health at the American University of Beirut in 2003. A stratified, two-stage probability sample of households was chosen from each of the three communities. The sample was chosen using a probability proportional to size sampling design and a detailed sampling frame constructed specifically for this study. In the first stage of selection, a sample of blocks which constituted the primary sampling units was randomly chosen from each community separately. A complete mapping and listing of households in these blocks (n=71) was then undertaken. In the second stage of selection, a sample of households was drawn systematically from each sampled block. In total, 2797 households were selected. All eligible adolescents found in each selected household were approached for interview.
In the first phase, interviewers recruited locally from the three communities and trained intensively by study investigators collected data pertaining to household and household members' socio-demographic characteristics during a face-to-face home interview with any adult respondent. During the second phase, detailed individual information was obtained by interviewing youth themselves.
Approximately 1300 (n=1294) adolescents were interviewed during the second phase of the survey. The intent was to interview youth separately, and this was requested on entry into the household. Generally, we were able to abide by this principle: 88.8% of the youth were interviewed with no one else present. Of the 11.2% who had someone present, 4.6% had someone present throughout the entire interview, and 6.6% throughout part of the interview. All the never married adolescents aged between 13 and 19 years that were identified in the first phase were approached for this survey. Although 13–19 years may seem like a wide range, the lower end is significant as it the age at which young people start to drop out of school (after 9th grade). In addition, inclusion criteria included only not married youth, ensuring that all our participants still lived at home. In order to further control for possible differences by age, we added age as a covariate in all analyses. Consent was obtained from both the parents/guardians and the adolescents themselves. Interviews were handled in the strictest confidence. Questions included general health, self-esteem, risky behaviours such as cigarette and narghile (water pipe) smoking, eating habits, fighting, education, employment, and social capital. The average response rate was 85%, with the highest in Burj El Barajneh (96.4%) and the lowest in Hay El Sellom (71.1%). The response rate in Nabaa was 85.6%. Selected variables from the phase 2 survey will be utilised for the analysis described herein.
Variables
The dependent variable was physical fighting: participants were asked whether or not they had been involved in a physical fight in the last 3 months.
The independent variable was social capital which was classified into six categories: civic engagement and community development; locational capital; trust; reciprocity; hypothetical social support; and social network.32 Each category was measured by a different set of questions. (1) Civic engagement and community development was measured by: youth membership in any group: social, recreational, sports, etc (yes, no); whether the respondent stated that he/she followed up on local issues (yes, no); and whether the respondent felt that he/she had control over things that took place in their neighbourhood (community control: yes, no). (2) Locational capital was measured by: the extent to which the respondent was happy living in the area (yes, somewhat, no); the extent to which youth felt safe walking alone at night (yes, no); and whether the respondent had been the victim of verbal harassment or criminal or non-criminal assault (neighbourhood victimisation: yes, no). (3) Trust was measured by the extent to which the respondent stated that he/she trusted people in the area (yes, no); and felt that one does not need to be vigilant and cautious when dealing with people (yes, no). (4) Reciprocity was measured by reciprocal exchange of non-material favours with neighbours (yes, no); reciprocal exchange of non-material with family (yes, no); reciprocal exchange of non-material favours with friend/neighbour/colleague (yes, no); and reciprocal exchange of material favours with friend/neighbour/colleague (yes, no). (5) Hypothetical social support was measured by the respondent's perception of whether he/she could ask for an urgent ride if needed from family/friend/neighbour (yes, no); ask to borrow LL15 000 (LL2045=€1) if needed from family/friend/neighbour (yes, no); and ask for help with a personal problem from family/friend/neighbour (yes, no). (6) Social network was measured by the presence of close relatives living nearby (yes, no); and the presence of close friends living nearby (yes, no) (table 2).
Background characteristics and associations between social capital variables and physical fights among boys in three impoverished communities in Beirut
A variety of socio-demographic variables were added as control variables. These included: community of residence, age (13–15 years, 16–19 years), school enrolment (yes, no), work (yes, no), living arrangements (with both parents, otherwise), family income (<3 600 000, 3 600 000–6 000 000, 6 000 001–10 200 000, >10 200 000LL), and literacy of head of household (parent literacy).
Other variables found to have been associated with physical fighting in the literature were included in the analysis. These included: school performance (very good, good, or average vs bad or very bad), religiosity measured through the combination of two questions: to what extent do you consider yourself a religious person, and to what extent do your religious beliefs affect the way you act (‘very much’ to both labelled very religious, ‘not at all’ to both labelled not religious, combinations of the above labelled in between), self-esteem measured through a set of eight items abstracted from Rosenberg's self-esteem scale,33 and stressful life events measured through a set of 12 items. Self-esteem items included: I have a number of good qualities; I am of equal worth with others; on the whole I am (not) a failure; I have much to be proud of; I'm satisfied with myself; I have self respect; I am beneficial to others; I am able to do things as well as other people. The self-esteem items were scored on a 1–3 scale (1=agree, 2=no specific opinion, 3=disagree). Stressful life events items included within the last year a new birth in the family; you were hospitalised; someone's death; parents' divorce or separation; mother or father lost a job; relative or friend moved into house; someone's travelled; hospitalisation of family member or relative; change in residence; parent taking a loan and having to pay it; family member marriage; other stressful event. The stressful life events items were scored on a 1–2 scale (1=yes, 2=no). The Cronbach α for the stressful life events scale was 0.471 (for boys). Responses to the eight self-esteem and 12 stress items were combined into a scale, and dichotomised at the mean (mean score was 21 for stressful life events scale and 11 for self-esteem).
All variables were elicited from the young men themselves, except for family income and literacy of head of household which were abstracted from the household questionnaire.
Analysis
Since boys were much more involved than girls in physical fights (20% compared to 4%), only boys (n=674 total; n=135 involved in a fight) were included in the analysis reported herein. All the analyses were weighted using a sampling weight constructed to adjust for both non-response as well as age and sex differences in the distribution of the adolescent sample (phase II) relative to the population (phase I).
Descriptive analysis was conducted for all the three communities to obtain general frequencies of the main variables. Bivariate analysis was conducted and subsequently only significant variables were included in the multiple logistic regression analysis. Since the outcome variable is dichotomous, a binary logistic regression model, adjusting for household clustering, was used in the analysis. The analysis was conducted using Stata V.11.
Results
The mean age of the sample of boys was 16 years; 65.4% were in school, and 38.5% were working. Forty-three per cent lived in households with a mean annual income less than LL6 000 000 (equivalent to €2934). One fifth (20.0%) had been involved in a physical fight in the last 3 months.
With respect to social capital, only 36% were members of any group. Although 63% claimed to know many or most people living in their area (data not shown), only 7% stated that they trust most or many persons living there. Only around 30% had a reciprocal exchange of service with a family member, a neighbour, or a friend. About 84% had relatives living near to them, and 90% had friends living nearby. Frequencies of all the individual and social capital variables are included in table 2.
Several of the individual and social variables were significantly related to the probability of engaging in a physical fight (table 2). Younger boys (26.4% vs 15.7%, p≤0.001), and those enrolled in school (25.5% vs 13.4%, p<0.000) were more likely to have been involved in a fight than their counterparts. Respondents having more stressful life events in the past year were also more likely to be involved in a fight than those having less (23.9% vs 13.0%, p<0.01). Most of the social capital variables were significantly and positively related to engaging in physical fights. Boys were more likely to have been involved in a fight if they were members of a group (28.9% vs 14.8%, p≤0.000); if they liked living in their neighbourhood (25.7% vs 11.7%, p≤0.000); if they had been victims of verbal harassment or criminal or non-criminal assault (49.2% vs 13.2%, p≤0.000); if they trusted most or many people in their neighbourhood (31.2% vs 19.2%, p<0.05); if they had reciprocated non-material services with neighbours, or friends and colleagues; and if they could ask for an urgent ride (29.3% vs 17.7%, p<0.01), or for help from a family member, friend, or neighbour (34.1% vs 16.9%, p≤0.000).
When all the above variables were considered together, most of the variables dropped out of the regression equation (table 3). Youth who had not been victimised were less likely to be involved in a fight (OR=0.190, 95% CI 0.140 to 0.346). Youth who were members of a group were twice as likely to have been involved in a fight than those who were not members of a group (OR=1.701, 95% CI 1.070 to 2.703). In addition, those who could ask for help from a family member, friend, or neighbour were also about twice as likely to have been involved in a fight than those who could not (OR=2.056, 95% CI 1.120 to 3.772). Finally, those that were somewhat happy living in their neighbourhood were less likely to have got into a fight than those who were very happy (OR=0.484, 95% CI 0.248 to 0.9427).
Influence on physical fights among boys in three communities of Beirut; results of the logistic regression
Discussion
One fifth of the boys in our sample had been involved in a physical fight in the last 3 months. It is unclear what the rate of fighting would be in this population in the past year, which is the most common measure in the literature. International literature suggests that between 44% and 55% of boys have been involved in a physical fight in the past 12 months.13 34
Results support the literature which indicates that physical fights are related to both individual and community variables and are more common among the young,16 17 and those exposed to stressful life events.35 The result on social capital and physical fights adds to the literature suggesting that social capital can be promotive of physical fights: membership in a group was positively correlated to fighting,5 suggesting the ‘downside’ of social capital—that is, gangs 36 37 However, most of the social capital indicators were not significantly related to physical fights, suggesting that some forms of social capital may not necessarily be associated with increased violence.
Our research found similar reasons for violence as did other studies38: when they hear someone say something bad about themselves or their families (especially mothers or sister); when they are angry and have no resources or skills to deal with the situation except to resort to violence; when the group perception is that walking away from violence is a sign of weakness and that violence is the solution to problems.
The research described herein has indicated that social capital is related to increased violence among young men in Beirut. The solution is not to decrease social capital, as its benefits have been documented.19–24 But, clearly, intervention is needed to channel youth energy and anger and disillusionment into positive conflict resolution or non-violent ways of expression. Several violence prevention programmes have proven to be effective.39 40 These have focused on building social skills such as conflict resolution.
Limitations of the research include the fact that it was cross-sectional, so causal directions (eg, between fighting and high levels of social capital) are not proven. In addition, face-to-face interviews and the use of female interviewers mean that a bias towards socially acceptable responses is possible.
This research has indicated the potential negative effect of social capital on youth violence, especially in the context of historical disadvantage and discrimination. More studies are needed to examine the increasingly common finding that high levels of social capital are associated with high levels of violence (the negative aspect of social capital). Is it merely the gang phenomenon that is at play here or are there additional psychological factors that stem from higher social interaction which can, in turn, promote violence—for example competitiveness, relative low self-esteem, and environment? Further qualitative research should engage youth in dialogue about the specific meaning and mechanism of their social capital that enhance the probability of violence.
What is already known on this subject
Adolescent violence is a major international public health concern.
There are multiple determinants of youth violence. These include personal, social, and community factors.
Social capital has emerged as an important contributor to violence, but results are mixed as to whether it is protective or promotive of violence.
What this study adds
Research on violence and its determinants among youth is common in the USA and other Western countries. This study extends such research to the Arab world, and contexts where structural violence is frequent.
Social capital was related to increased violence among young men in Beirut. This research has indicated the potential negative effect of social capital on youth violence, especially in the context of historical disadvantage and discrimination.
In impoverished environments, young people may form gang-like bonds in a gesture of solidarity to overcome the many obstacles in their lives.
Interventions need to channel youth energy and anger and disillusionment into positive conflict resolution or non-violent ways of expression.
Acknowledgments
The authors would like to thank the three communities of Burj El Barajneh, Nabaa, and Hay El Sellom for engaging in this collaborative research with us, and especially the youth, for giving us insight into their lives. The authors would like to thank reviewers for constructive comments that strengthened the manuscript.
References
Footnotes
Funding The research was supported by grants from the Wellcome Trust, the Mellon Foundation, and the Ford Foundation.
Competing interests None.
Ethics approval This study was conducted with the approval of the American University of Beirut Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.