Date of Award

Summer 8-15-2013

Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Dominic J. Parrott

Second Advisor

Christopher C. Henrich

Third Advisor

John L. Peterson

Fourth Advisor

Monica H. Swahn


The study aimed to fill critical gaps in the literature on aggression toward sexual minorities (1) by examining a specifically prejudice-based model, intergroup threat theory (ITT), to explain aggression toward sexual minorities, and (2) by determining the extent to which alcohol facilitates prejudice-based mechanisms of aggression toward sexual minorities. Participants comprised a final sample of 161 heterosexual, undergraduate men of ages ranging from 18 to 30. Participants completed measures of perceived threat based on the ITT model (i.e., realistic threat, symbolic threat), antecedents to these threats (i.e., AIDS-related stigma, religiosity, adherence to traditional male gender role norm of antifemininity), and frequency of both alcohol-related and non-alcohol-related aggression toward sexual minorities. Structural equation modeling was used to identify indirect effects, to statistically test contrasts between these indirect effects, and to employ multigroup models in moderated-mediation analyses. Perceived threats of ITT were expected to mediate positive associations between (1) antecedents of perceived threat and (2) frequency of alcohol-related and non-alcohol-related aggression toward sexual minorities. Although the final sample was relatively small, this study found that, overall, indirect effects were mediated by symbolic threat more so than realistic threat. Additionally, indirect effects via symbolic threat on alcohol-related aggression were significant, but indirect effects on non-alcohol-related aggression were generally not significant. These results supported the view that perceived threats, particularly symbolic threat, of ITT, combined with the risk-enhancing role of alcohol, facilitates prejudice-based aggression in response to sexual minorities. Additional analyses showed that group differences based on (1) racial membership and (2) whether participants endorsed being religious were potential moderators. Moderated-mediation analyses showed that racial membership did not moderate indirect effects. However, among participants who reported being religious, realistic threat mediated associations between antecedents to threat rather than symbolic threat. Conversely, realistic threat was not a significant mediator among reportedly non-religious participants; however, symbolic threat was. Taken together, results showed that the type of perceived threat in response to a marginalized group, alcohol consumption, and social-group membership of perpetrators are critical when designing interventions to reduce prejudice-based aggression. Strengths, limitations, and other implications of the study were discussed.