Sexual Violence and Homophobic Bullying: Victimization Trajectories Predict Perpetration

Aggression among adolescents is prevalent and associated with a wide array of deleterious consequences. There are many longitudinal studies illustrating how peer victimization, covariates, and consequences evolve over time during adolescence. Some findings have demonstrated the interconnectedness of victimization and perpetration trajectories using a social learning theory framework. However there is a clear dearth of developmental research examining pathways to bias-based aggression, specifically aggression rooted in sexism and homophobia. Given the prevalence of sexual violence (SV) and homophobic bullying (HB) in schools, understanding the epidemiology is critical to protecting students and creating a safe and positive learning climate.

The present 3-wave longitudinal study aimed to fill this gap by examining the temporal relation between SV victimization and perpetration as well as the relation between HB victimization and perpetration. The sample consisted of 3,064 high schoolers from the U.S. Midwest. Multi-group latent growth curve modeling was utilized to analyze the data. Students were grouped by perpetration status (perpetrator or non-perpetrator) at Waves 2 and 3. For SV victimization, students who identified as SV perpetrators had a significant intercept (η₁ = 1.79, SE = .119, p < .001) and a significant positive slope (η₂ = .158, SE = .073, p < .05). Alternatively, those who did not perpetrate SV had a lower intercept (η₁ = 1.31, SE = .010, p < .001) and a significant negative slope (η₂ = -.02 SE = .060, p < .05). These results suggest that SV perpetrators had experienced more SV victimization at Wave 1 and that they followed a different trajectory. For HB victimization, students who identified as HB perpetrators had a significant intercept (η₁ = 1.77, SE = .074, p < .001) and a non-significant slope (η₂ = .043, SE = .053, p < .41). Alternatively, those who did not perpetrate had a lower intercept (η₁ = 1.34, SE = .012, p < .001) and a significant negative slope (η₂ = -.06, SE = .006, p < .001). The results suggest that those who engaged in HB perpetration had experienced more HB victimization at Wave 1 and that they had a flat, albeit different trajectory than their peers who had not been victimized. Wald tests of parameter constraints suggested all of the reported differences were significant.

These longitudinal data demonstrate that the connection between victimization and perpetration extends to SV and HB. It is reasonable to speculate that adolescents who are victimized learn to engage in aggression in the future, including when the aggressive behavior is based in gender and sexual-orientation biases. These findings have several important implications for prevention science, as victimization likely has cascading consequences over time, which are harmful to both the individual and others. First, it should be recognized that students who have been victimized are at higher risk of perpetrating violence. Additionally, the unique nature of these acts as bias-based behaviors calls for prevention efforts that address sexism and homophobia in schools. Preventing the initial SV and HB victimization could diminish future aggression and should therefore be the focus of school-based policies, programs, and interventions.