Longitudinal Study on Victimization of Relational Aggression and Peer Network Health: Moderating Effects of Gender and Antisocial Behavior

Introduction: Experiencing relational aggression (e.g., peer exclusion, untrue gossip) during adolescence can have negative consequences, including psychosocial issues and substance use. Adolescent girls may be particularly at risk, as they experience greater distress when having difficult interpersonal relationships than boys. In this study, we measured peer network health, asking adolescents about both protective and risky activities of three friends. This presentation examines how victimization of relational aggression (VRA) is associated with peer network health (PNH) across two years and the moderation of gender and antisocial behavior.

Methods: Participants were 244 adolescents, ages 13-14, and predominantly African-American from urban U.S. neighborhoods. Adolescents reported on their PNH, VRA, and antisocial behaviors every 6 months for 2 years. Latent growth models were used to assess change in PNH. VRA was added to the model as a predictor of intercept and slope of PNH. Two hypothesized moderators of the relationship of VRA to the intercept and slope of PNH, gender and baseline antisocial behavior (mean-centered) were tested, separately, by regressing the intercept and slope on the main effect of the moderator and the 2-way interaction with VRA. Next, the three-way interaction of VRA, gender, and antisocial behavior was tested.

Results: Significant linear and quadratic trends were found in PNH; scores decreased and the decreasing trend decelerated over time. VRA significantly predicted PNH intercept (estimate = -1.67, p-value = .002), yet was not associated with the linear or quadratic trend. Gender was a significant moderator (p = .002). For females there was a significant and negative association between baseline VRA and PNH (estimate = -2.81, p-value <.001), but not for males (estimate = 0.25, p = .734). Antisocial behavior also moderated the relationship (p = .040). A negative relationship between VRA and PNH was found for those who were at the mean or 1 SD below the mean on antisocial behavior (estimate = -1.28, p = .039 and estimate = -1.36, p = .028 respectively), yet for those 1 SD above the mean in antisocial behavior the association was not significant (estimate = -1.20, p = .055). The three-way interaction between VRA, gender, and antisocial behavior also was not significant (p = .085).

 

Conclusions: The association between VRA and PNH was likely already established prior to baseline assessment when adolescents were ages 13-14. Decreases in PNH were associated with other factors. For girls, efforts to reduce relational aggression and increase friendship support in prosocial activities during middle childhood may be useful for preventing social-emotional difficulties and substance use.