Religion and Risky Health Behaviors among U.S. Adolescents and Adults

Recent studies analyzing the effects of religion on various economic, social, health and political outcomes have been largely associational. Although some attempts have been made to establish causation using instrument variable (IV) or difference-in-difference (DID) methods, the instruments and the spatial and temporal variations used in these studies suffer from the usual issues that threaten the use of these identification techniques--validity of exclusion restrictions, quality of counterfactuals in the presence of spatial assortative sorting of people, and concern about omitted variable bias in the absence of information on family level unobservables and child-specific investment by families. During the adolescent years, religious participation might be a matter of limited choice for many individuals, as it is often heavily reliant on parents and family background more generally. Moreover, the focus of most of the studies has been on religious rites and rituals i.e., religious participation or on the intensity of participation. Using the National Longitudinal Study of Adolescent Health, this paper analyzes the effects of a broad set of measures of religiosity on substance use at different stages of the life course. In contrast to previous studies, we find positive effects of religion on reducing all addictive substance use during adolescence, but not in a consistent fashion during the later years for any other illicit drugs except for crystal meth and marijuana.