years later, all originally targeted adolescent, now young adult, women were again contacted. For those completing wave 1, the participation rate was 84% at wave 2, and some earlier nonparticipants were added. Median age at baseline and follow-up was 15.8 and 21.8 years, respectively. Comprehensive data profiles on alcohol use, alcoholism risk factors, and religious variables were obtained. When compiled as a cross-sectional data set, N = 4,002 female offspring cases were available; when both waves of data were available (true for 84% of cases), the later data were used. Previous attrition analyses of this sample were based on geosocial information derived from census block data. Results indicated some minor demographic differences between participants and nonparticipants in the initial data collection, whereas those lost to follow-up were found to have lower income, non-White race, and more paternal history of alcoholism. Attrition bias on outcomes, however, was found to be not significant.
