Religious salience and adolescent substance use revisited: the role of beliefs in moral relativism and moral contextualism

ABSTRACT

Much prior research has established an inverse relationship between religiosity and delinquency among adolescents, with particularly noteworthy effects observed for illicit substance use. These patterns correspond with the expectations of social bond theory and situational action theory, which posit that commitment to conventional religious norms can reduce youths’ exposure to, and inform their experiences with, delinquent opportunities. However, little empirical attention has been given to youths’ beliefs surrounding the nature of morality itself, which are theoretically expected to moderate the protective effects of religious salience. Specifically, while religious salience might have a general controlling influence on adolescents’ use of cigarettes, alcohol, and marijuana, holding beliefs in moral relativism (i.e., the absence of definitive rights and wrongs) or moral contextualism (i.e., that rights and wrongs are fluid and dynamic) might weaken these inhibiting effects. Using data on a nationally representative sample of adolescents from the National Study of Youth and Religion (N = 3,170), this study’s analyses reveal that the association between religious salience and substance use is conditioned by moral contextualism but not moral relativism, though this interactive relationship is found only for alcohol use and drunkenness.

KEYWORDS:

• Religiosity
• substance use
• moral beliefs

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Data from the NSYR are publicly available via the Association of Religion Data Archives (ARDA): https://doi.org/10.17605/osf.io/wnqt9.

2. U.S. households without a home phone number were rare in 2002–2003 when these data were collected, thus mitigating the extent to which this represents a limitation of the sampling procedure (NSYR 2008). However, because this sample is drawn using household phone numbers, institutionalized youth are excluded by design. Though this is a limitation of many surveys of adolescents administered to population-based samples, the findings are not generalizable to high-risk youth housed in confinement facilities.

3. Parent interviews were conducted primarily with either the mother or father, though interviewers asked about the availability of the mother first. Stepparents, resident grandparents, resident partners of parents, and other resident parent-like figures were also eligible to complete this portion of the survey. Since the teenager portion of the survey was administered immediately following the parent portion, adolescents were asked to confirm that they were in a place in the home that prevented their parents from overhearing their responses.

4. The overall cooperation rate for eligible households was 81%, and the Wave 1 response rate generated via the American Association of Public Opinion Research (AAPOR) RR4 calculator was 57%. As noted in the NSYR (2008, 4) documentation, ‘the NSYR’s final response rate represents among the highest possible given the methodology employed and prevailing social conditions and available technologies.’ Comparative analyses of key variables from the weighted NSYR data to U.S. Census data as well as to several contemporary nationally representative datasets revealed no evidence of sampling or nonresponse bias (Smith and Denton 2005).

5. While the removal of youth who had dropped out of school allows us to account for school-related risk factors for substance use, this decision arguably renders the analytic sample somewhat biased. This issue should be considered when interpreting the findings and generalizing them to the entire population of non-institutionalized youth in the U.S.

6. It should be noted that the use of sampling weights, while beneficial in this research context, precludes the calculation of certain statistics. These include conventional t- and z-tests for the equality of means and proportions, some diagnostic tests for linear regression, and model fit statistics based on maximum likelihood assumptions in categorical data analysis (e.g., Model chi² statistics, BIC and AIC, pseudo R² values, etc.).

7. Approximately 45.1% of respondents who are coded as 1 on the moral relativism item are also coded 1 on the moral contextualism measure. Similarly, nearly 45.4% of participants coded as 1 on the moral contextualism item also reported agreement with the question on moral relativism. In the sample overall, 21.5% of participants are coded as 1 on both measures.

8. An examination of the variance inflation factor (VIF) values revealed that multicollinearity is not a concern in these analyses (mean VIF = 1.49; highest VIF = 4.02). The only variables with VIF values higher than 1.76 are the three mutually exclusive categories of religious affiliation.

Additional information

Notes on contributors

Peter S. Lehmann

Peter S. Lehmann is an Assistant Professor in the Department of Criminal Justice and Criminology at Sam Houston State University. His research interests include juvenile justice and delinquency, racial and ethnic disparities in punishment, school discipline and safety, and public opinion on crime and criminal justice policy.

Ghady Hbeilini

Ghady Hbeilini is a doctoral student at the Department of Criminal Justice and Criminology at Sam Houston State University. He is interested in conducting comparative research, research on illicit organizations, terrorism, and Queer criminology and the treatment of LGBTQIA+ individuals.