Agrawal et al. 4 use a twin model approach to examine a specific environmental influence, religious services attendance, and its association with alcohol involvement in African American and White female younger adults. Religious attendance is conceptualized as a prosocial protective activity, as well as a family practice often initiated during childhood and one indicator of personal religious devotion in adulthood. As expected, the associations between religious attendance and alcohol involvement measures (ever drink, age of first drink, and DSM-5 AUD) were different depending both on the timing of religious attendance (i.e., childhood and/or adulthood) and on race/ethnicity. While African Americans reported higher religious attendance and lower drinking levels, the protective relationships between childhood religious attendance and alcohol involvement were observed in Whites only. Adult religious attendance was associated with lower odds of DSM-5 AUD in both African Americans and Whites, but the protective relationship in Whites extended to ‘ever drinking’ and ‘first alcohol use prior to age 18’. The modeling of the relationships between monozygotic and dizygotic twins allows for the variance in and covariance between alcohol involvement and religious