<p dir="ltr">Substance use behaviors vary substantially by gender identity, where transgender individuals use substances at higher rates than cisgender individuals, on average, and cisgender men additionally report high rates of use. Although research has explored these average differences, few studies have examined the mechanisms which underlie and shape these differences. Guided by theories on gender, social identity, social control and support, and stigma, the central aims of this dissertation are to investigate: (1) the potential for measurement variation in a commonly used scale that screens for alcohol use disorders (2) whether the relationship between lowered substance use with higher religious attendance holds for transgender individuals (3) the impact of gender conformity on alcohol use behaviors for transgender individuals. To answer these questions, I utilize data from the U.S. Transgender Population Health Survey paired with structural equation modeling (SEM), confirmatory factor analysis (CFA), and mediation and moderation techniques. Consistent with expectation, I found that the measurement model for the Alcohol Use Disorder Identification Test-Compact (AUDIT-C) does have measurement variation by gender identity. I also found no significant relationship between religious attendance and substance use behaviors among transgender individuals, while I did find the well-established negative relationship between religious attendance and substance use behaviors among cisgender individuals. Finally, I found that higher rates of gender conformity, specific to gender expression and presentation is related to higher rates of abstinence from alcohol use among transgender men. These results illuminate the importance of considering gender identity when studying substance use behaviors and indicate the need for future research to further investigate the underlying mechanisms by which higher rates of substance use persist in this community.</p>