Defense Date


Document Type


Degree Name

Master of Science



First Advisor

Dr. Terri N. Sullivan


Aggression has been demonstrated to pose a serious threat to the adaptive development of youth, with decades of research demonstrating the negative associations between aggression and other problem behaviors, both concurrently and predictively. However, despite this body of research, the current psychological literature continues to suffer from a lack of an overarching organizational framework from which to structure the construct(s) of aggression. Furthermore, existing discrepancies across the literature, particularly in the definitions of and outcomes associated with non-physical forms of aggression (e.g., relational aggression, social aggression), exacerbate the complexities facing prevention and intervention specialists. Insofar as research can isolate the unique subtypes of aggressive behaviors that best predict maladjustment outcomes, researchers can focus resources and efforts on those subtypes of aggression identified as being particularly relevant for prevention efforts. To this end, the purpose of the current study was to develop a measure that encompassed the structure of physical and relational aggression and to test competing structures of aggression based on the hypothesized relevant dimensions of mechanism of action (i.e., confrontational action vs. nonconfrontational action) and vehicle of harm (i.e., physical harm vs. relational/social harm) using confirmatory factor analyses. Additionally, this study examined relations between aggression subtypes and hypothesized correlates, including peer deviancy, delinquency, drug use, and social intelligence. Further, this study assessed both the factor structures and unique relations among aggression and its correlates separately for boys and girls, and identified unique structure and relations by gender. Participants included an urban, predominantly African American sample of 280 youth ages eleven through seventeen, who were sampled from an ongoing longitudinal study of violence, substance use, stress, and coping. As hypothesized, the mechanism of action and vehicle of harm dimensions did represent relevant conceptual distinctions in the structure of aggression. Although models did not reach objective standards for goodness of fit criteria, comparatively, the mechanism of action model best represented the structure of aggression for boys, whereas the vehicle of harm model best represented the structure of aggression for girls. Both boys and girls had significant positive correlations among their respective subtypes of aggression and other indicators of maladjustment, including peer deviancy, delinquency, and drug use. Overall, these findings confirm that structures of aggression tested were problematic for urban African-American youth, and suggest that further attention should be paid to disentangling those aspects of aggression that might be most relevant for addressing prevention and intervention efforts.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

June 2008

Included in

Psychology Commons