What Are the Risk Factors for Youth Deconversion from Religion?

• In Event: Poster Session 11
  In Poster Session: PS 11 Topic - Moral Development

Abstract

Introduction. Recent research has found that religiousness seems to decline through the adolescent years (Hardy & Longo, 2020). Nevertheless, there appears to be a relationship between supportive religious communities and teen religiousness such that more supportive religious communities appear to have more religious teens. Yet, few studies have sought to longitudinally examine the relationship between religious community support and adolescent deconversion. Most studies qualitatively explore retrospective accounts of reasons for leaving religion, many of which pertain to felt lack of religious community support. The current study is the first that we know of to use measures of religious community support to prospectively predict later deconversion.
Method. Data came from a large, representative sample of U.S. teens (N = 3370; National Survey of Youth and Religion; NSYR; Smith & Denton, 2005). In total, there were four waves of data. In Wave 1, participant age ranged from 13-17, Wave 2 age ranged from 16-20, Wave 3 age ranged from 18-23, and Wave 4 age ranged from 23-28. Religious community support was measured via four variables at Wave 1: the general support adolescents felt from their church, adolescents’ church engagement, adolescents’ views of their church’s ability to communicate, and adolescents’ views of hypocrisy within their church community. Deconversion was operationalized as anyone who was not religious for at least one wave after having been religious for at least two consecutive waves.
Results. We sought to address these gaps in the literature, via a multivariate logistic regression analysis, to see if church engagement, communication, hypocrisy, and support variables prospectively predicted later deconversion. To begin, we ran correlations between all independent and dependent variables (see Table 1). We then ran our logistic regression equation (see Table 2), and found that church engagement (OR = .70, p < .001) and church communication (OR = .71, p < .001) negatively predicted later deconversion. In other words, per unit increase in church engagement and church communication, there were 30% and 29% reductions respectively in the likelihood of later deconversion. Interestingly, there was no evidence that church support (OR = 1.04, p = .564), nor church hypocrisy (OR = 1.02, p = .69) prospectively predicted deconversion.
Discussion. These findings corroborate earlier findings that supportive religious communities may stave off later deconversion (Hardy & Longo, 2020). Indeed, both a church’s capacity to be engaging to the teen and the church’s ability to effectively communicate with the teen prospectively predict later deconversion i.e., these variables appear to reduce the risk for later deconversion.

Authors

• Dorian Hatch, Brigham Young University
  Presenting Author

• Justin Hendricks, Brigham Young University
  Non-Presenting Author

• Sam A Hardy, Brigham Young University
  Non-Presenting Author