A JUVENILE WHO ADMITS GUILT FOR A CRIME AND VOLUNTARILY CHOOSES TO BE DIVERTED CAN BE REFERRED TO CAB FROM THE JUVENILE COURT. THE PROGRAM IS BASED ON THE ASSUMPTION THAT JUVENILES WOULD BE LESS LIKELY TO COMMIT FURTHER CRIMES IF THEY ARE HELD ACCOUNTABLE FOR THEIR BEHAVIOR ON A COMMUNITY LEVEL. DURING THE EVALUATION PERIOD FROM 1973 TO 1976, 711 YOUTHS WERE PROCESSED BY THE JUVENILE COURT. THE CAB WAS OPERATING FROM 1974 TO 1976 AND HEARD 356 CASES. THESE GROUPS OF CASES ARE USED FOR ANALYSIS. THE EVALUATION DESIGN IN THIS STUDY CONSISTS OF A SERIES OF NONEQUIVALENT CONTROL GROUP DESIGN COMPARISONS AND COMPARISONS OF JUVENILE OFFENDERS' RECIDIVISM AND RECIDIVISM PREDICTIONS. THREE MEASURES WERE CHOSEN: INDIVIDUAL YOUTHS' POLICE CONTACT HISTORIES, THE TOTAL NUMBER OF JUVENILE CONTACTS WITH THE CRIMINAL JUSTICE SYSTEM, AND THE REPORTED OCCURRENCE OF RESIDENTIAL BURGLARY, LARCENY, AND AUTO THEFT IN SEATTLE, WASH., BY CENSUS TRACT. WHEN THE THREE OFFENSES WERE COMBINED ACROSS CAB AREAS AND COMPARED WITH CITY TOTALS, THE OFFENSE RATES SHOWED AN INCREASE OF 7.2 PERCENT IN 1976 FOR CAB AREAS AND 13.4 PERCENT IN NONCAB AREAS. THE CAB TARGET AREAS ALSO SHOW DECREASES IN BOTH PART I UNIFORM CRIME REPORTS CRIMES AND TOTAL POLICE CONTACTS. POLICE CONTACTS INCREASED BY 12 PERCENT IN NONCAB AREAS. ONLY 17 OUT OF 150 CAB CLIENTS COMMITTED SUBSEQUENT OFFENSES WITHIN 6 MONTHS AFTER THEIR APPEARANCE BEFORE THE BOARD, COMPARED TO A SLIGHTLY HIGHER RATE OF RECIDIVISM FOR YOUTHS IN NONCAB AREAS. IT IS CONCLUDED THAT THE IMPOSITION OF AN APPROPRIATE SOCIAL SANCTION, EITHER FINANCIAL OR COMMUNITY SERVICE, CAN RESULT IN REDUCED RECIDIVISM RATES FOR JUVENILE OFFENDERS AND INDIRECTLY IN DECREASES IN REPORTED CRIME RATES FOR PRIMARILY JUVENILE CRIMES. REFERENCES ARE PROVIDED.