Children of alcoholics have been identified as a group at risk for a variety of psychological and social problems, including difficulty in interpersonal relationships, low self-esteem, poor academic performance, anxiety, and eventual alcoholism. There is growing recognition of the need to offer preventive intervention in order to reduce or eliminate these tendencies. The proposed research would develop and study such an intervention. The intervention would consist of a composite of several features of existing programs, including education, social and emotional support, and social skill building in a small (6-8) group context. The unique and significant contributions of this research is that it would assess the effectiveness of this model using a controlled design and would also assess the differential impact of this intervention with two different student populations (urban, low-income, black and suburban, middle-income, white). Two schools from each of these two areas will participate. After students have volunteered to participate, assessment will be done, and one school from each area will be randomly assigned to receive a group program immediately. All students (control and experimental) will be assessed following the intervention and the two control hold-out schools will then receive the intervention and a third assessment. One year follow-up on all students will also be completed. The design will be repeated for a total of 5 groups from each of the two areas. Outcome measures include assessment of self-concept, anxiety, locus of control, perceived social support, school attendance, peer involvement, alcohol knowledge, and alcohol use. Process measures include frequency of group attendance and peer ratings of group involvement.