The evaluation found that the juveniles who participated in the Halt program had the same reoffending pattern after 1 year as a control sample of juveniles who did not participate in the program. The seriousness and frequency of reoffending was similar for both groups. After 6 months, both groups showed fewer emotional, behavioral, relationship, and attention problems. The findings indicate, however, that juveniles with certain characteristics did have reduced reoffending related to program participation. These youth were first offenders who were aware of the consequences of their criminal behavior, had no or only mild emotional and behavioral problems, generally engaged in socially acceptable behavior, and were susceptible to group peer pressure. Positive effects from the program were greater for youth who completed a learning assignment or a punishment that involved a combination of community service and education. These findings suggest directions for modifying the Halt components and the selection for referral to the program of those juveniles most likely to benefit from it. The evaluation, which was conducted some 25 years after the first Halt Bureau was established, monitored almost 1,000 juveniles, half of whom were referred to Halt and half of whom received no intervention. The evaluation measured reoffending and attitudes before and at monitoring points. 55 references and appended questionnaire