Protecting Mental Health Through Spiritual Awareness and Prayer: Insights From Robustness Testing

ABSTRACT Prior research finds that different kinds of religious engagement, including beliefs, practices, and affinities, are associated with lower rates of depressive symptoms. As rates of depression in the United States rise, and more Americans disaffiliate from formal religious organizations, it is worth revisiting whether these associations hold. Using a large sample of recurring cell phone surveys, we consider two common measures of religious “traits” from respondents’ intakes (religious service attendance and belief in God) and two measures of religious “states” collected in the moment (spiritual awareness of God or a higher power and daily prayer). Using specification curve analysis of multiple modeling strategies, we find that our state measures have the most robust relationship with lower depressive symptoms, and our trait measures are more sensitive to researcher choices in model specification. These results help us understand how personal religion and spirituality benefit adherents independently of engagement with formal religious institutions, and they advance the integration of theories of lived religion with work on mental health. They also demonstrate the use of a novel methodological tool for evaluating the extent to which key findings in the social scientific study of religion are robust to researchers’ design choices.