Peer-Aggression Victimization and Perpetration in Middle School Youth: Estimating Prevalence and Frequency, Joint Trajectory Patterns, and Predictive Utility

The purpose of this study was to examine early adolescent trajectories of bullying/peer-aggression in terms of their prevalence, composition, and ability to correlate with concurrent delinquency. Three hypotheses were tested in a group of 1145 middle school students (49.6% male; mean age = 11.23 years) using longitudinal data spread out over three waves. The first hypothesis predicted that bullying/peer-aggression victimization would be significantly more prevalent and frequent than bullying/peer-aggression perpetration. The second hypothesis held that a semiparametric sequential process growth mixture modeling (GMM) analysis for two latent variables would identify pure victim and mixed victim–perpetrator trajectories but no pure perpetrator trajectories. The third hypothesis asserted that the trajectory models identified in the GMM analysis would differentially correlate with a change in delinquency, such that accelerating trajectories would be associated with a rise in delinquency and decelerating trajectories with a drop in delinquency. Analyses provided support for all three hypotheses: victimization was significantly more prevalent and frequent than perpetration; there were no pure perpetration trajectories, even after increasing the number of trajectories from 6 to 9; and accelerating trajectories were associated with a significant rise in delinquency from Wave 1 to Wave 3 and decelerating trajectories with a marginally significant decrease in delinquency from Wave 1 to Wave 3. These results highlight the value of studying change in the perpetration and victimization of peer-aggression as a way of understanding how bullying/peer aggression in early adolescence develops and contributes to the formation of other problems, such as delinquency.