A small but growing literature focuses on the links between religion and family violence. Several recent studies report that regular religious attendance is inversely related to abuse among both men and women. After outlining a series of theoretical arguments regarding possible direct and indirect links between religious involvement and domestic violence, we analyze these relationships using data from Wave 1 of the National Survey of Families and Households (NSFH-1). Among the key findings: (1) regular religious attendance is inversely associated with the perpetration of domestic violence; (2) among men, this protective effect is evident only among weekly attenders, whereas among women, the protective effect also surfaces among monthly attenders; (3) although the estimated net effects of religious attendance are generally somewhat larger in models of self-reports of domestic violence, this link also remains strong and statistically significant in models of partner reports of violence; and (4) moreover, the inverse association between religious attendance and abuse persists even with statistical controls for measures of (a) social integration and social support, (b) alcohol and substance abuse, and (c) low self-esteem and depression. We conclude by discussing a number of implications of these findings, and by identifying several promising directions for future research.