DURING THE PERIOD AUGUST 1969 THROUGH MARCH 1971 ALL BOYS AGED 15 TO 17 RECEIVED BY THE YOUTH AUTHORITY WERE RANDOMLY ASSIGNED TO EITHER THE O. H. CLOSE SCHOOL, THE KARL HOLTON SCHOOL, OR ONE OF THE CONTROL CALIFORNIA YOUTH FACILITIES. THE STAFF AT CLOSE USED TRANSACTIONAL ANALYSIS (TA) IN ITS PROGRAM WHILE HOLTON SET UP A BEHAVIOR MODIFICATION (BMOD) PROGRAM BASED ON A TOKEN ECONOMY. BY THE END OF THE PERIOD, 541 BOYS HAD BEEN SENT TO HOLTON, 589 TO CLOSE; 988 WERE AVAILABLE FOR PAROLE FOLLOW-UP BUT COMPLETE TEST DATA WERE AVAILABLE FOR ONLY 904. THE BOYS, TYPICAL OF THOSE SEEN BY THE YOUTH AUTHORITY, HAD A MEDIAN AGE OF 16.6 YEARS. THEY HAD A HISTORY OF SCHOOL PROBLEMS, AN AVERAGE READING ABILITY AT THE EIGHTH GRADE LEVEL, AN AVERAGE MATH ABILITY AT THE SIXTH GRADE LEVEL, AND POLICE RECORDS SHOWING A VARIETY OF OFFENSES RANGING FROM MURDER TO INCORRIGIBILITY. BURGLARY AND AUTO THEFT WERE THE MOST COMMON OFFENSES. THE TA GROUP SHOWED THE GREATEST ATTITUDE CHANGE. THEY HAD POSITIVE ATTITUDES TOWARDS THEMSELVES, THE INSTITUTION, AND THEIR COUNSELORS. THE BMOD GROUP HAD THE GREATEST MEASURABLE IMPROVEMENT IN BEHAVIOR. BOTH GROUPS HAD SIGNIFICANTLY LOWER RECIDIVISM RATES THAN THOSE FOUND IN THESE PARTICULAR SCHOOLS BEFORE THE PROGRAM AND THOSE REPORTED IN THE CONTROL INSTITUTIONS. BEFORE THE TA EXPERIMENT, CLOSE'S RECIDIVISM RATES RANGED FROM 46 TO 49 PERCENT; DURING THE EXPERIMENT, RATES RANGED FROM 31 TO 34 PERCENT. THE FIGURES FOR HOLTON WERE 40 TO 42 PERCENT BEFORE, 31 TO 33 PERCENT AFTER. IT IS RECOMMENDED THAT THE BEST FEATURES OF THE TWO PROGRAMS BE COMBINED FOR STILL GREATER REHABILITATION EFFECTIVENESS. TABLES PRESENT ARE APPENDED. (GLR)