Identifying patterns of religiosity in adults from a large UK cohort using latent class analysis

Background: Religiosity is a complex, multifaceted construct, comprising a variety of beliefs and behaviours. Much of the previous research that seeks to quantify religiosity has made use of variable-centred approaches, which place individuals on a continuum of religiosity. However, alternative approaches provide a way to examine different types of religiosity, represented by heterogeneous classes of religious (non)belief. The aim of this study was to apply a person-centred approach to understanding religious beliefs.  Methods: : The present study used latent class analysis to identify classes of belief and non-belief in mothers (n=12429) and their partners (n=9953) from the Avon Longitudinal Study of Parents and Children, a large cohort study based in the UK. For this, we used a range of religious belief indicators. We also examined a number of socioeconomic factors, to identify differences between classes, using logistic regression.  Results: : We identified four different classes of religiosity which we named Highly Religious, Moderately Religious, Agnostic, and Atheist, with similar configurations in both men and women. We also found that class membership was associated with several socioeconomic factors.  Conclusions: : The findings provide an insight into different patterns of religiosity in adults in the UK. These classes could be used as exposures in future studies of religiosity and how it relates to a variety of outcomes in both mothers, partners, and their offspring.