Objective. This article examines the association between religious attendance and disaster recovery in Mississippi and Alabama. Methods. We use ordinary least squares regression to determine the effect of sociodemographic variables, social network size, and religious attendance on one's selfdescribed level of disaster recovery. Results. We find a robust association between frequent religious attendance and a greater level of recovery. Somewhat surprisingly, we also find a strong relationship between religious nonattendance and a greater level of recovery. However, these results differ by race. For whites (but not for blacks), nonattendance is associated with a greater degree of recovery, while for blacks (but not for whites), frequent attendance is correlated with a greater degree of recovery. For both whites and blacks, the size of one's social network does not affect disaster recovery. Conclusion. While according to previous research, religious attendance is associated with benefits based upon social networks and community engagement, we find that those who are strongly connected to their religious organizations recover more, but it is not directly connected to one's social network size.People who regularly attend assemblies of religious organizations have greater community engagement and are less likely to be socially isolated (Hovey et al., 2004). These benefits, suggests the social science research evidence thus far, are not related to religious belief, but instead are associated with regular attendance at religious organizations (Shor and Roelfs, 2013). 1 Furthermore, community engagement often confers other benefits on individuals, including psychological benefits (Cavendish, 2000), health benefits (Moren-Cross and Lin, 2006), and more resilience to trauma and disaster (Nakagawa and Shaw, 2004). Specifically, studies have found a connection between the level of community engagement (including the size of one's social network) and the ability to recover from natural and man-made disasters and that this effect differs between whites and African Americans. The current article examines this relationship to see whether social networks, religious attendance, and