Examining the Effect of Weapons-Based Victimization on Depression: A Fixed Effects Approach to Account for the Effect of Offending and Low Self-Control

Abstract Prior research has identified that being the victim of a violent crime is associated with depression and several other mental disorders. However, these studies have relied on cross-sectional research designs that can overestimate effect sizes due to confounding or selection bias. Thus, it is unclear whether victimization itself causes a person to experience depression or whether it is caused by a third variable that causes victimization and depression, such as low self-control or offending. The purpose of this study is to address this theoretical and methodological issue by assessing whether victimization experiences involving a weapon have a causal impact on experiencing depression in adolescents using an innovative causal inference research design. The data for this study come from three waves of the restricted version of the National Study of Adolescent to Adult Health ( N = 46,249). This study uses a linear fixed effects regression model with individual slopes to determine whether different types of weapon-based victimization have a causal effect on experiencing depression. In addition, this study uses two novel sensitivity analyses to determine the robustness of the estimates to confounding and selection bias. The results found that individuals who experienced multiple types of weapon-based victimization had increased symptoms of depression even after accounting for the role of low self-control and offending. However, additional sensitivity analyses revealed that this relationship is not robust to confounding and selection bias from omitted variables. Future research should use other causal inference research designs to determine whether victimization has a causal impact on depression and other mental disorders.