THE MAIN OBJECT OF THE FIRST STAGE OF THE PROJECT (1961-1969) WAS TO DETERMINE IF CERTAIN TYPES OF JUVENILE OFFENDERS COULD REMAIN IN THEIR HOME COMMUNITIES WITHOUT PRIOR COMMITMENT TO AN INSTITUTION IF THEY WERE GIVEN INTENSIVE SUPERVISION AND TREATMENT WITHIN A SMALLSIZED PAROLE CASELOAD. ALL ELIGIBLE YOUTHS WERE RANDOMLY ASSIGNED TO EITHER THIS EXPERIMENTAL GROUP OR TO A CONTROL GROUP IN A TRADITIONAL PROGRAM. BY 1967-1968 IT HAD BEEN FOUND THAT AT LEAST ONE THIRD OF THE EXPERIMENTAL GROUP HAD BEEN INVOLVED IN DELINQUENT ACTS, AND THAT SIMILAR RESULTS HAD OCCURRED IN THE TRADITIONAL PROGRAM. IN THE SECOND PHASE, YOUTHS WERE ASSIGNED TO A RESIDENTIAL SETTING, NOT DIRECTLY TO THE COMMUNITY ITSELF. THESE YOUTHS WERE FURTHER CLASSIFIED ACCORDING TO THE APPROPRIATENESS OR THE INAPPROPRIATENESS OF THAT TREATMENT. FINDINGS FROM THE 1969-1974 EXPERIMENT SUGGEST THAT CAREFUL DIAGNOSIS AND APPROPRIATE PLACEMENT OF INDIVIDUALS MAY LEAD TO A REDUCTION OF DELINQUENT BEHAVIOR FOR RESIDENTIAL AND COMMUNITY-BASED PROGRAMS ALIKE. (AUTHOR ABSTRACT MODIFIED)