Having identified the "8 percent problem" group, the next phase of data exploration aimed at broader definition of the problem, using knowledge of juvenile offender risk prediction factors to guide the analysis. The starting point for the next set of analyses focused on differences between the three major study groups, based on age at final disposition of the initial referral. These groups were nonrecidivists (one referral only during the 36-month follow-up); low-rate recidivists (two or three referrals during the follow-up period); and chronic recidivists (four or more referrals during the follow-up period). The factors identified and measured were school behavior/performance, substance abuse, family problems, and delinquency patterns. The analysis clearly showed that chronic recidivists had a higher average number of composite problem factors than the other two groups. Substance abuse was the most important factor for chronic recidivists age 16 or older at their initial referral. The study concludes that it may be possible to reduce significantly the number and proportion of chronic juvenile recidivists in Orange County through a coordinated program of proactive and aggressive early intervention and treatment of young, high-risk juvenile offenders and their families. the goal of such efforts should be to break the escalating pattern of criminal and antisocial behavior that leads to chronic delinquency and adult criminal careers. Specific recommendations for a program strategy are offered. 7 tables and 24 references