Josephson, Gordon J2013-11-082013-11-0820042004Source: Dissertation Abstracts International, Volume: 66-05, Section: B, page: 2825.http://hdl.handle.net/10393/29119http://dx.doi.org/10.20381/ruor-19603Existing research suggests that gay men may be more vulnerable to depression than are heterosexual men, and that gender-related personality traits may be important in understanding this vulnerability. In the present study the association between two gender-related personality traits (agency and unmitigated communion) and depressive symptoms was examined in an Internet sample of 510 gay identified men, aged 18 years or older. Potential mediators of this association were also examined including: recalled peer harassment, interpersonal behavior, and self-discrepancies in the gender-related personality trait of agency. Participants completed the Extended Personal Attributes Questionnaire (EPAQ), the Unmitigated Communion Scale, the victim subscale from the Olweus Bully/Victim Questionnaire Revised (worded retrospectively), the Checklist of Interpersonal Transactions Revised, and the Centre for Epidemiological Studies Depression Scale. Self-discrepancy was calculated by repeating the agency subscale of the EPAQ from the perspective of ideal self and subtracting the actual score from the ideal score. Zero-order correlations indicated that among gay men who were less agentic than their ideal, the gender-related personality traits of agency and unmitigated communion were associated with depressive symptoms, as were recalled peer harassment, interpersonal behavior characterized as unassured and submissive, and self-discrepancies in agency. Two models were tested with half the sample and a respecified model was validated with the second half. The final model indicated that when all variables were considered jointly, agency and unmitigated communion were not directly related to depressive symptoms but rather, the effects of agency and unmitigated communion were mediated by unassured and submissive interpersonal behavior, and the effect of agency was also mediated by self-discrepancies in agency (p < .05). Thus, how the participants interacted with close others, and the extent that they were not as agentic as they wished, were more important in predicting their depressive symptoms than were the gender-related personality traits of agency and unmitigated communion. The final model also indicated that recalled peer harassment has both a direct impact on depressive symptoms and an indirect impact on depressive symptoms through unassured and submissive interpersonal behavior. Implications for future research, anti-bullying initiatives, and interventions with depressed gay men are discussed.124 p.enPsychology, Clinical.Thesis