Article Text

Download PDFPDF

92 Social networks and violence among male youth in marginalized communities
  1. Gabrielle Corona1,
  2. Nicolás Kass2,3,
  3. Elizabeth Miller2,3,
  4. Kathleen Carley Alison Culyba2
  1. 1University of Pittsburgh, Graduate School of Public Health
  2. 2Children’s Hospital of Pittsburgh
  3. 3University of Pittsburgh School of Medicine
  4. 4Institute for Software Research, Carnegie Mellon University, Pittsburgh, PA


Statement of Purpose Male youth residing in low-resource neighborhoods experience high rates of violence. Prior research in marginalized communities suggests that a singular adult presence may not universally confer protection amidst high levels of community violence. This study aimed to describe adolescent-adult support network structure and quality and analyze associations between network properties and violence exposure among male youth in low-resource neighborhoods.

Methods/Approach 45 male youths, ages 13–21, completed social network surveys as part of a community-based violence prevention study in Pittsburgh, PA to define family-, school-, and community-based adult supports and past 30-day violence exposure (operationalized as any/none). Logistic regression examined associations between 1) perceived strength of overall emotional and instrumental support, 2) strength of social support type, 3) network size, and violence exposure. Wilcoxon rank-sum tests examined associations between network density and violence exposure.

Results Mean participant age was 17.6 years (SD=1.5); 77.8% were African American. In the past 30 days, 84% reported witnessing violence, 60% perpetrating violence, and 69% being victimized. Increased immediate and extended family support were associated with significantly higher odds of witnessing violence (OR: 4.91, 95%CI: 1.25–19.63; OR: 3.99, 95%CI: 1.23–13.01, respectively). Mean number of adult supports was 4.8 (range:1–14). Increased number of adult supports was associated with higher odds of witnessing violence (OR: 2.17, 95%CI: 1.01–4.66). Network density (Median=0.50) was significantly inversely associated with witnessing violence (z=2.32, p=0.02).

Conclusion Findings highlight complex associations between adolescent-adult support network structure, relationship quality, and violence. Associations between family support and witnessing violence may reflect families providing increased support to help youth navigate community violence. Inverse associations between network density and violence exposure suggest that network connectivity may act as a protective factor against witnessing violence.

Significance/Contribution to Injury and Violence Prevention Science Results can guide interventions to leverage adult support networks and protect youth from violence.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.