LAUSR

ABSTRACT This research provides a test of general strain theory examining the relevance of strain recency, the role of victimization history for predicting anger development, and the relevance of anger development for mediating this relationship. Analyses utilized group-based trajectory modeling to elucidate patterns of anger development and also utilized multinomial logistic regression to examine the relevance of victimization history for predicting anger development. Analyses then utilized Poisson regression to examine the relevance of victimization for predicting violent offending at age 23 and to test for mediation. A three-group model best fit the data, with low, moderate, and high anger groups emerging. Victimization prior to age 16 significantly increased the risk of assignment to the moderate and high groups. More recent victimization demonstrated a significant effect on violent offending at the age of 23. Anger group assignment fully mediated the relationship between victimization history prior to ages 16 and 23 violent offending.