BECAUSE THE UNIVERSITY OF WASHINGTON'S STUDY RELEASE PROGRAM WAS NOT DESIGNED FOR EVALUATION, THIS PRELIMINARY ANALYSIS WAS NECESSARY TO IDENTIFY PROGRAM BENEFITS AND FACTORS ASSOCIATED WITH INMATE SUCCESS. FOLLOWING A DESCRIPTION OF THE PROGRAM'S DEVELOPMENT, THE PROBLEMS ENCOUNTERED IN COLLECTING DATA ON THE 203 PERSONS WHO WERE ENROLLED IN 1974 ARE REVIEWED. ANALYSIS IS THEN MADE OF RECIDIVISM RATES, CRIMINAL HISTORIES OF THE INMATE-STUDENTS, ACADEMIC HISTORIES, AND BACKGROUND CHARACTERISTICS. IT WAS FOUND THAT 47 OF THE 203, OR 23 PERCENT, WERE RETURNED TO PRISON FROM THE PROJECT. ABOUT HALF WERE RETURNED FOR DRUG OR ALCOHOL VIOLATIONS, A QUARTER FOR A NEW OFFENSE, AND A QUARTER FOR VIOLATION OF PROJECT RULES, INCLUDING ACADEMIC FAILURE. OF THE 158 WHO ACHIEVED PAROLE STATUS, 20 RECIDIVATED. IT WAS FOUND THAT SOME WHO RECIDIVATED DURING THE PROJECT PERFORMED WELL ON PAROLE. PERSONS CONVICTED OF DRUG OFFENSES, AS OPPOSED TO A HISTORY OF DRUG USAGE, WERE LEAST LIKELY TO RECIDIVATE. A CROSS TABULATION FOUND THAT THOSE BACKGROUND CHARACTERISTICS MOST CORRELATED WITHIN PROJECT RECIDIVISM ARE ALL COMPONENTS OF A FAILURE TO DEVELOP POSITIVE SOCIAL BONDS. MOST HIGHLY CORRELATED WAS 'POOR RELATIONSHIP WITH FAMILY.' LESS CORRELATED WERE NO MILITARY SERVICE, NO DEPENDENTS, BEING ASIAN AMERICAN, CHICANO, OR AMERICAN INDIAN, AND DRUG-USE HISTORY. TO ADAPT TO THE PROJECT ONE MUST ADAPT TO THE AMERICAN MAINSTREAM. THIS IS NOT TRUE OF PAROLE. APPENDIXES PRESENT STUDY STATISTICS, ANALYSES, AND REFERENCES. (GLR)