Abstract
This study tests interparental boundary problems (IBPs), parent hostility with adolescents, and adolescent hostility with parents within a reciprocal influence model and tests each as risk factors for adolescent aggression problems. Prospective, longitudinal analyses were conducted with multi-informant data from 768 adolescents and their families, from 6(th) to 9(th) grade. Guided by spillover and social learning perspectives, our findings suggest that IBPs have a robust, negative influence on both parent and adolescent hostility. In turn, adolescent hostility was the best predictor of global adolescent aggression problems. Two indirect effects were found that link IBPs and adolescent aggression problems; however, findings indicate that adolescent hostile behavior in the family is the key risk indicator for adolescents' later aggression problems. Model invariance tests revealed that this model was not different for boys and girls, or for adolescents in families with two biological parents and youth in families with two caregivers (e.g. stepparent families).