Background: Poor recruitment and high attrition may invalidate results of research studies. This paper describes successful recruitment and retention strategies in a school-based substance use prevention trial and explores factors associated with intervention attendance and retention. Methods: A total of 384 parent-child dyads from 15 schools in the New York Metropolitan area participated in a control trial, testing the efficacy of parent-training to prevent youth substance use. Assessments were completed immediately post-intervention and 6-, 12-, and 24-month post-intervention. Logistic regression analyses were used to determine which familial and study characteristics predicted attendance in the intervention and retention by parents and youth. Results: 84% of intervention parents attended 4 of the 5 workshops; 83% of control parents attended their single workshop. Intervention attendance was predicted by parent job status, but this was not significant after controlling for other family factors. Retention rates ranged from 87% to 91% over the 2 years. No family characteristics predicted retention, but time since baseline and attendance at treatment workshops and the control workshop did. For children, age at baseline and ethnicity predicted retention, but this did not remain significant in the adjusted model. Conclusion: Intervention attendance was high and retention rates far exceeded the minimum standard of 70% retention in behavioral studies. Recruitment and retention strategies were effective for different family constellations. Efforts to maximize participation in both treatment and control interventions are critical to retention in longitudinal trials. (Contains 4 tables.)