Measurements and characteristics of internalized homophobia among gay men

Abstract

This study investigated measurements and characteristics of internalized homophobia among gay men. The psychometric properties of a 20-item instrument, the Multi-Axial Gay Inventory - Men's Short Version (MAGI-MSV), were investigated. A convenience sample of 228 self-selected ethnically diverse HIV-negative or untested gay men seeking sexual risk-reduction counseling was utilized. Preceding factor analyses, the MAGI-MSV had a Cronbach alpha equal to .93. Following factor analyses, the instrument was reduced to 15 items and four distinct factors emerged. Three of the factors (homosexual self-assurance and worth, impacts of HIV/AIDS on homosexuality, and public appearance of homosexuality) were consistent with the proposed factors involved in the original construction of the MAGI-MSV. The fourth factor (extreme or maladaptive measures to eliminate homosexuality) was consistent with existing theory regarding internalized homophobia. Cronbach alpha equaled .91 for total scores in the new 15-item version of the MAGI-MSV, and three of the four observed distinct factors had Cronbach alphas above .80. The fourth factor (extreme or maladaptive measures to eliminate homosexuality) achieved a Cronbach alpha of .69. Logistic regressions were then used in an attempt to predict group membership among four dichotomous variables (depression status, relationship status, risky sexual behavior, and illicit substance use). Both total scores and weighted scores of the four observed MAGI-MSV factors were used in the prediction analyses. The set of all four MAGI-MSV factors considered together predicted depression status in both a factor total score design and a weighted factor score design. Depression status was best predicted by the factor "homosexual self-assurance and worth" in all models. The set of all four factors considered together did not predict relationship status, risky sexual behavior, or illicit substance use. These findings suggest that at least one factor of the 15-item MAGI-MSV may be a valid tool for predicting depression without the additional use of statistical transformations, weightings, or adjustments.