Title

Social Cognitive Mediators and Moderators of the Relation Between Experiences of Community Violence and Adolescent Outcomes

Date of Award

2008

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.)

Department

Psychology/Clinical

First Advisor

Eric Dubow, PhD

Second Advisor

Dara Musher-Eizenman, PhD (Committee Member)

Third Advisor

Randall Leite, PhD (Committee Member)

Fourth Advisor

Jean Gerard, PhD (Committee Member)

Abstract

This study examined moderation and mediation effects of three types of social cognitions (perceived social support, future expectations, and normative beliefs about aggression) on the relation between environmental experiences of aggression/violence and adolescent outcomes (depression and aggression). Additionally, this study sought to determine if moderation and mediation effects were varied based on the type of environmental experience of aggression/violence (victimization or witnessing) and/or on the level of severity of these experiences (low-severity or high-severity). Using a short-term longitudinal design, 248 high school students in 9th through 12th grade were surveyed twice, three months apart, regarding victimization experiences, witnessing experiences, perceived social support, future expectations, normative beliefs, aggressive behavior, and depressive symptoms. Teachers completed surveys about students’ aggressive behavior and depressive symptoms.

A confirmatory factor analysis indicated that experiences of aggression/violence fit a four-factor model (low-severity victimization, high-severity victimization, low-severity witnessing, and high-severity witnessing) better than two or one factor models. Contrary to previous research, perceived social support did not moderate the relation between victimization and depression, but had a protective-reactive effect on the relations between victimization and aggression and between witnessing and aggression. Future expectations moderated experiences of aggression/violence (both victimization and witnessing) and adolescent outcomes (depression or aggression), but tended to have a protective-reactive effect on boys and younger adolescents and a protective-stabilizing effect on girls and older adolescents. Normative beliefs about aggression mediated the relation between victimization (both low and high severity) and aggression such that higher levels of victimization predicted higher levels of normative beliefs, which predicted higher levels of aggression. Normative beliefs did not mediate the relation between witnessing and aggression. Future expectations mediated the relation between high-severity witnessing and aggression such that increases in high-severity witnessing predicted lower future expectations that in turn predicted higher levels of aggression. Gender analyses indicated that the link between future expectations and aggression was only significant for girls. Future expectations did not mediate the relations between low-severity witnessing and aggression or victimization and aggression and was not a strong mediator between experiences of aggression/violence and depression. Implications for future research and clinical application are discussed.