Evaluated the efficacy of a self-selection recruitment process designed to attract fourth-, fifth-, and sixth-grade children into a school-based prevention program for children of alcoholics. Participants were 296 families comprising one child and either one or both parents. Family members' self- and collateral reports were used to assess parental problem drinking, family characteristics, and individual pathology. Analyses revealed that the recruitment process was not effective in recruiting children of alcohol-abusing parents. Furthermore, families in which children received parental consent to participate in the prevention program were indistinguishable from families whose children either showed interest without obtaining consent or showed no interest at all. Implications for recruitment strategies for future prevention programs for children of alcoholics are discussed.