THE STUDY EXPLORES ASSOCIATIONS BETWEEN PAROLE SUCCESS OR FAILURE AND SUCH FACTORS AS PREVIOUS CRIMINAL RECORD, INTELLIGENCE, PSYCHIATRIC DIAGNOSIS, PAROLE PROGRAM, SOCIAL STATUS, PHYSICAL HEALTH, REPORT STATUS, RECREATIONAL ACTIVITIES, AND RELIGION. THE STUDY POPULATION INCLUDES 100 MEN WHO WERE RETURNED TO THE ELMIRA REFORMATORY FOR PAROLE VIOLATIONS DURING 1935, AND 100 MEN WHO SUCCESSFULLY COMPLETED PAROLE SUPERVISION AND WERE RELEASED UPON EXPIRATION OF THEIR SENTENCES DURING THE SAME YEAR. THE SUCCESS OR FAILURE OF THE 200 PAROLEES IS NOT DEPENDENT UPON ANY ONE PREDOMINATING FACTOR. PARTICULARLY SIGNIFICANT FACTORS (LISTED IN ONE ORDER OF IMPORTANCE) ARE RECREATIONAL ACTIVITIES DURING PAROLE, LENGTH OF SENTENCE, PHYSICAL HEALTH UPON RELEASE, NUMBER OF MONTHS EMPLOYED WHILE ON PAROLE, NUMBER OF JOBS, EMPLOYMENT STATUS, AGE AT TIME OF PAROLE, INTELLIGENCE QUOTIENT (IQ), DURATION OF EMPLOYMENT, TIME SERVED AT ELMIRA, RELIGIOUS OBSERVANCE, NUMBER OF PREVIOUS ARRESTS, OCCUPATIONAL LEVEL, REPORT STATUS, MOBILITY DURING PAROLE, AND PSYCHOLOGICAL CLASSIFICATION. EIGHT OF THE SIXTEEN FACTIONS CLOSELY RELATED TO PAROLE SUCCESS OR FAILURE ARE ASSOCIATED WITH THE ACTUAL PAROLE PERIOD. THE DATA SUGGEST THAT YOUNGER PAROLEES ARE MOST APT TO VIOLATE PAROLE THAN ARE OLDER PAROLEES, THE VIOLATIONS ARE FREQUENT AMONG MEN WITH MANY PREVIOUS ARRESTS FOR PETTY OFFENSES (VAGRANCY, USE OF DRUGS, INTOXICATION), AND THAT THE REFORMATORY'S CLASSIFICATION BOARD TENDS TO RELEASE POTENTIAL VIOLATORS EARLIER THAN NONVIOLATORS. NONVIOLATORS GENERALLY HAD HIGHER IQ'S THAN DID VIOLATORS. PERSONALITY ABERRATIONS WERE FOUND IN THE MAJORITY OF PAROLEES IN BOTH GROUPS. FAMILY TIES AND EMPLOYMENT ARE POSITIVELY RELATED TO PAROLE SUCCESS. THE TYPE OF NEIGHBORHOOD IN WHICH THE PAROLEE RESIDES AFTER RELEASE HAS LITTLE SIGNIFICANCE FOR PAROLE OUTCOME. IMPLICATIONS OF THESE AND OTHER FINDINGS FOR PAROLE SUPERVISION AND FOR FUTURE RESEARCH ARE DISCUSSED. SUPPORTING DATA, STUDY INSTRUMENTS, AND A BIBLIOGRAPHY ARE INCLUDED. (LKM)