The predictive validity of the Child and Adolescent Functional Assessment Scale (CAFAS) is investigated using the data set generated by the national evaluation of the demonstration service grants funded by the Center for Mental Health Services. Logistic regressions were performed separately for contact with the law and poor school attendance, which were both assessed at 6 months postintake. Other variables included in the model besides the CAFAS total score at intake were age, gender, and family risk factors. The results show that the CAFAS total score at intake was a positive predictor of the likelihood of contact with the law and poor school attendance, even after controlling for age, gender, and risk factors. Furthermore, the CAFAS total score was predictive even after excluding scores on CAFAS subscales, which may have been influenced by absenteeism or delinquency. These findings are consistent with recent research indicating that the CAFAS predicts recidivism in juvenile delinquents.