Introduction Millions of children witness violence and are victims of violence each year. Previous research suggests that this is a risk factor for perpetrating violence. There is a paucity of studies that examine factors that protect violence-exposed youth from perpetrating violence.

Methods This study used a panel design to measure the effects of exposure to violence on the perpetration of violence. It examined the protective effects of social support and school social capital on the risk of exposure to violence, using multivariate logistic regression modelling. The sample was weighted to reflect a national population.

Results The median age of the sample was 15. When considering risk factors only, those who ‘saw someone shoot or stab another person’ were at 4.77 times (95% CI 3.19 to 7.13) greater risk for perpetrating interpersonal violence. In the full model (risk and protective factors, (N=8375)), those with lower school social capital were at 2.43 (95% CI 1.15 to 5.15) to 2.91 (95% CI 1.02 to 8.29) times greater risk of perpetrating violence compared with those with the highest school social capital; adding the protective factors into the model reduced the odds of perpetrating violence from 4.77 times to 3.47 times (95% CI 1.97 to 6.11) (p<0.001).

Conclusion On a national level, the protective effects of school social capital could translate to a substantial reduction of violence. School-wide policies and programmes that reach all adolescents in a school and promote social capital should be pursued as a strategy to prevent the perpetration of interpersonal violence.