A before-and-after experimental-control group design, with subjects matched, paired, and randomly assigned to experimental and control groups, was developed and applied to data on delinquents from intake to followup. The results of the comparison of experimental and control groups indicated that vocational rehabilitation did not enhance significantly delinquent boys' vocational development, self-esteem, and attitude towards authority. Vocational rehabilitation made no significant difference in reducing recidivism. Perhaps vocational rehabilitation itself is not an effective treatment for delinquent boys; yet its beneficial effects with a lower caseload than the one studied and an increased intensity of probation supervision have been demonstrated in other studies. The program studied may be at fault. There is a high turnover rate; perhaps the succession of new counselors caused the program to suffer. Many clients asserted at retesting that the program had helped them, though the success was not reflected in the objective data. The discrepancy may be a statistical one; it may be the result of other programs received by the control group. The study tells us, however, that certain types of treatment are effective only with certain types of delinquents. It is not yet known what types will be benefited from vocational rehabilitation. 4 tables and 36 references. (Author abstract modified)