This article presents results from a psychometric evaluation of the Internalized Homonegativity Inventory (IHNI; Mayfield, 2001), a widely used scale for measuring internalized heterosexism in gay men. The IHNI was initially reported to consist of 3 correlated factors—Personal Homonegativity, Gay Affirmation, and Morality of Homosexuality—and our work represents the first attempt to test this assertion using latent variable modeling. Confirmatory factor analysis was conducted on data from 230 gay men recruited through Amazon Mechanical Turk. To improve model fit, and consistent with theoretical rationale, 2 gay affirmation items were rearranged to load onto different factors. Statistical evidence then indicated that the items adequately measured their respective factors. Bifactor modeling further suggested that although a single factor did not fully represent the construct, reliable variance on the IHNI largely originated from a unidimensional continuum on which individual differences in gay men’s internalized heterosexism are distributed. These results advance additional validity evidence related to the IHNI’s factor structure and offer some support for computing a mean total score and ipsatized subscale scores in applied research. Future validation studies should investigate response processes, consequences of test use in research and clinical settings, as well as relations to theoretically and clinically relevant criterion variables. Similarly, further work is needed to explore effective ways to use the IHNI in psychological assessment, psychotherapy, and other therapeutic capacities.