Associations of Supernatural Beliefs and Depressive Symptoms in Three Nationally Representative Samples

People who engage in religious practices are generally healthier than those who do not. Despite the growing research on these associations, few studies have examined how afterlife and other “supernatural” beliefs are associated with mental health. One study found that beliefs in the afterlife were associated with fewer psychological symptoms (Flannelly, Koenig, Ellison, Galek, & Krause, 2006). Follow up studies have suggested that this association may depend on the positive or negative nature of such afterlife beliefs (e.g., Flannelly, Ellison, Galek, & Koenig, 2008). Similarly, images of God as angry have been associated with worse mental health (e.g., Silton, Flannelly, Galek, & Ellison, 2014), further suggesting that negative supernatural beliefs would be associated with more psychological symptoms. However, these studies have not examined supernatural beliefs such as the belief in angels or demons. The present study analyzes the link between mental health, specifically depressive symptoms, and a variety of religious afterlife and supernatural beliefs. For the present study, we identified three nationally representative datasets from the Association of Religion Data Archives with items which assessed depressive symptoms and afterlife and supernatural religious beliefs: the Portraits of American Life Study (PALS), the Baylor Religion Survey, and the National Study of Youth and Religion (NSYR). PALS is a panel study of over 2,500 adults. The Baylor Religion Survey is a multi-year study of over 1,500 adults. The NSYR is a multi-year telephone survey of over 3,000 teenagers. We are analyzing associations of supernatural and afterlife beliefs with depressive symptoms in each dataset using basic correlational analyses and regression. Initial results suggest an association of supernatural beliefs with more depressive symptoms. Complete descriptions of beliefs and depressive symptoms in each sample will be presented along with predictive findings. The implications of our findings for the field and for health will be discussed