An Investigation Into Religious Peer Effects

Abstract

The communal nature of religion suggests peer effects exist in religiosity, potentially through positive spillovers in religious participation and social formation of religious beliefs. Using seven measures of religiosity, we estimate positive and negative peer effects in each case using peer group composition. We use a simultaneous equation model to account for self-selection into religious peer groups, and while we find positive peer effects are insignificant, there are significant negative peer effects operating through non-religious friends. This suggests peers affect social formation of religious beliefs, rather than through positive spillovers in religious participation.