Longitudinal Predictions of Religious Change in Adolescence: Contributions from the Interaction of Attachment and Relationship Status

The purpose of the study was to test our (Granqvist, 2002a) emotional compensation and two-level correspondence hypotheses by examining longitudinal associations of attachment and change in romantic relationship status in relation to religious change. Questionnaires were filled out by 196 adolescents in Stockholm (mean age = 16 years); 72% completed the 15-month follow-up. Virtually no main effects on religious change as a function of either attachment or relationship status were obtained. However, in line with the interaction predictions, an insecure attachment history and insecure adult attachment in romantic relationships predicted increased religiousness for those who had experienced a separation and decreased religiousness for those who had formed a new romantic relationship. The pattern of relations for a secure attachment history and romantic attachment was in the opposite direction. The findings imply that attachment predicts religious change primarily in the context of changes in romantic relationships. These findings may explain why previous studies in the separate domains of attachment and relationship status in relation to religiousness have yielded weak or inconsistent results, and suggest that the two fields need to be integrated. Implications for romantic attachment theory and research in general are discussed.