The Health, Education and Welfare (HEW) Office of Youth Development's National Strategy for Youth Development model was promoted as a community-based planning and procedural tool for enhancing positive youth development and reducing delinquency. To test the applicability of the model as a function of delinquency level, the program's Impact Scales were administered to 1,561 males and females, aged 10-19, who comprised a roughly 6 percent systematic school system sample. The sample was divided into three groups representing low, medium, and high self-reported delinquency (SRD). Stepwise multiple regression analyses were run separately for the full sample and within each subsample. Results showed prediction of the SRD variation was poorest within the high SRD subsample. Normative Peer Pressure appeared as a major predictor in the full sample and the mid and high SRD samples. Youth Perceived Negative Labeling by Teachers was a major predictor in all data sets. Normlessness appeared as a major predictor for the full sample, and the low and mid SRD subsamples. Model confirmation was generally best when considering the general youth sample as was intended in the model's development. Other variables need to be investigated for more refined delinquency prediction within samples of highly delinquent youth. (JAC)