THE REPORT COMPARED 155 JUVENILES REFERRED TO 1 COUNTY'S JUVENILE DIVERSION UNIT IN 1978 WITH 312 SIMILAR JUVENILES FROM 1977 WHO COULD HAVE BEEN DIVERTED HAD THE STATE'S REVISED JUVENILE CODE BEEN IN EFFECT. THE TYPICAL OFFENDER WAS MALE, WHITE, AND BETWEEN 14.7 AND 15.1 YEARS OLD. OF THE DIVERSION GROUP, 20.3 PERCENT HAD SOME PRIOR CRIMINAL HISTORY, COMPARED WITH 36.2 PERCENT FOR THE CONTROL GROUP, AND 12.9 PERCENT WERE ADVISED AND RELEASED TO THEIR PARENTS. ALL BUT TWO DIVERTEES RECEIVED COUNSELING. OVER TWO-THIRDS (69 PERCENT) OF THE DIVERSION CASES WERE DISPOSED OF WITHIN 10 DAYS, COMPARED WITH ONLY 28.2 PERCENT OF THE COMPARISON GROUP'S. AVERAGE DIVERSION AGREEMENT LENGTH WAS 22.4 DAYS. RECIDIVISM WAS 10.3 PERCENT IN THE DIVERSION GROUP AND 11.9 PERCENT IN THE COMPARISON GROUP, AN INSIGNIFICANT DIFFERENCE. THE HIGHEST RECIDIVISM RATE (22.2 PERCENT) WAS ABOUT THOSE WHO HAD RECEIVED COMMUNITY SERVICE WORK ORDERS, SUGGESTING THAT MORE OBLIGATIONS PRODUCE A LOWER RECIDIVISM RATE. A MAILED QUESTIONNAIRE SURVEY OF THE DIVERSION GROUP RECEIVED A 38 PERCENT RESPONSE. MOST RESPONDERS UNDERSTOOD THEIR LEGAL RIGHTS AS WELL AS THE MEANING OF THE DIVERSION AGREEMENT, BUT HAD MIXED OPINIONS ABOUT THE COMMUNITY ACCOUNTABILITY TRAINING CLASS. THE PROGRAM EVALUATION RESULTS INDICATED BOTH THAT A NEED EXISTS FOR TRAINING CLASSES SPECIFICALLY RELATED TO THE OFFENSES COMMITTED AND THAT DIVERSION REDUCES COSTS AND TIME SPENT AND ACHIEVES THE SAME RESULTS AS EARLIER JUVENILE PROCESSING METHODS. TABLES OF RESULTS AND A FLOW CHART ON THE ACCOUNTABILITY TRAINING PROGRAM ARE INCLUDED. (CFW)