A TOTAL OF 45 SUBJECTS WERE SELECTED AT RANDOM FROM THE INMATE POPULATION OF A MEDIUM-SECURITY CORRECTIONAL FACILITY FOR THE REHABILITATION OF OFFENDERS RANGING IN AGE FROM 18 TO 26 YEARS. ALL INMATES WHO HAD PARTICIPATED IN GROUP COUNSELING FOR A 6-MONTH PERIOD WERE DERIVED FROM ENTRIES IN SUBJECTS' PROGRAM RECORDS AS WELL AS FROM EVALUATIVE INFORMATION PROVIDED BY THE GROUP LEADERS. RESULTS INDICATED THAT GROUPS LED BY PSYCHOLOGISTS HAD BETTER ADJUSTMENT IN THEIR WORK SETTINGS THAN THOSE GROUPS LED BY OTHER LEADERS. SIMILARLY, MOTIVATION WAS HIGHER IN THE PSYCHOLOGIST-LED GROUPS. THE SIGNIFICANT DIFFERENCES AMONG THE DEGREE OF EFFECTIVENESS OF GROUP LEADERS WAS ATTRIBUTED TO THE DIFFERENTIAL ROLE THE GROUP LEADER HAD IN THE INMATE'S FINAL RECOMMENDATION FOR PAROLE. THUS, THE INMATE'S PERCEPTION OF THE GROUP LEADER AS A THREAT TO HIS FREEDOM NEGATED THE EFFECTIVENESS OF THE GROUP COUNSELING PROCESS. THE WIDE USE OF GROUP COUNSELING WITH ITS CONCOMITANT EXPENDITURES NECESSITATES THE IMPROVEMENT OF THE PROGRAMS, AND THIS IMPROVEMENT SHOULD START WITH THE TRAINING OF GROUP LEADERS. IT IS ALSO SUGGESTED THAT MAJOR CONSIDERATION BE GIVEN TO THE JOB FUNCTIONS OF STAFF PERSONNEL WHO LEAD GROUPS. TRAINING SHOULD BE UNDERTAKEN WITH PERSONNEL WHOSE PARTICIPATION WILL NOT CONFLICT WITH THEIR REGULAR JOB FUNCTIONS, AND WHOSE PRESENCE IN THE GROUP WILL NOT IMPEDE ACTIVE, HONEST INMATE PARTICIPATION. REFERENCES ARE GIVEN. (MJW)