Date Available


Year of Publication


Document Type



Arts and Sciences



First Advisor

Tamara Lynette Brown


Alcohol use, because of its many negative consequences, is the number one health problem facing college students. Because of this, researchers have looked for factors associated with reduced drinking. Religiousness is one such factor. Religiousness is a complex, multidimensional construct, and while it tends to be negatively associated with alcohol use, research progress has been slow due to the tendency of researchers to poorly operationalize this construct and to design studies that fail to go beyond the bivariate relationship of religiousness and alcohol use. In order to address these shortcomings, this study will assess two dimensions of religiousness, religious commitment/motivation and religious consequences, and will test a model, presented by Koenig et al., (2001), that postulates religiousness works through mental health in order to reduce alcohol use. More specifically, this study will test optimism as a possible mediator and moderator of the relationship between religiousness and alcohol use. This study used archival data from 260 (202 female and 58 male) Caucasian, Christian, undergraduate college students who completed a battery of surveys that included measures of religiousness, optimism, and alcohol use. A factor analysis was conducted on one measure of religiousness, the short form of the Faith Maturity Scale. Also, optimism was tested as both a mediator and a moderator for both dimensions of religiousness in predicting alcohol use. Findings indicated optimism is not a significant mediator of the religiousness-alcohol use relationship because optimism did not meet the preconditions for a mediator as it was not associated with alcohol use in this sample. Also, optimism was not a significant moderator of religious commitment/motivation, but it did moderate the relationship of religious consequences and alcohol use. Finally, the two dimensions of religiousness interacted in predicting alcohol use. While both dimensions of religiousness were negatively associated with alcohol use throughout the findings, gender was a significant moderator in all significant interactions. Several implications follow from this study. First, greater specificity is needed regarding Koenig et al.s (2001) model specifically in regards to which third variables are associated with which health outcomes and to whom the model applies. Second, this study highlights the importance of a multidimensional assessment of religiousness. Finally, this study indicates specificity is needed regarding what religious interventions will be helpful for which genders.