THE SUBJECTS OF THIS STUDY WERE 28 WOMEN WHO HAD LIVED AT THE GROUP HOME FOR ADOLESCENT GIRLS IN PHILADELPHIA (PA.), A SERVICE OF THE ASSOCIATION FOR JEWISH CHILDREN, BETWEEN 1959 AND 1969. MEAN AGE AT ADMISSION WAS 15.1 YEARS AND AVERAGE LENGTH OF STAY WAS 2.5 YEARS. THE GIRLS WERE OF AVERAGE INTELLIGENCE AND MOST WERE DIAGNOSED AS SUFFERING FROM A PERSONALITY DISORDER. THE OUTCOME STUDY INVOLVED EVALUATION ON PSYCHOLOGICAL AND AND SOCIAL CRITERIA AT ADMISSION AND FOLLOWUP. CASE HISTORIES, TRANSCRIPTS OF PERSONAL INTERVIEWS CONDUCTED IN THE HOME, QUESTIONNAIRES, AND PREDICTED OUTCOME RATINGS WERE USED FOR ASSESSMENT. 'BLIND' JUDGES ESTIMATED THE GROWTH POTENTIAL OF EACH GIRL IN EVERY CRITERIA FOR AN EQUAL PERIOD OF TIME, AND THIS SERVED TO ESTABLISH A BASELINE FOR GROWTH COMPARISON. EACH GIRL CONTRIBUTED 10 RECOLLECTIONS OF HER RESIDENCE EXPERIENCE AND THESE WERE RATED FOR SUBJECT, TONE, AND THEMATIC CONTENT. THE GIRLS ALSO EVALUATED THE IMPACT OF THE PROGRAM AND ITS COMPONENTS. THE RESULTS INDICATED THAT THE GIRLS MADE SIGNIFICANT GAINS IN OVERALL PSYCHOLOGICAL FUNCTIONING; HOWEVER, THE ACTUAL GAINS DID NOT EXCEED THOSE PREDICTED TO OCCUR WITHOUT TREATMENT. THIS SUGGESTED THAT FACTORS UNRELATED TO THE PROGRAM WERE RESPONSIBLE FOR GAINS. THE PREDICTED OUTCOME TECHNIQUE DEMONSTRATED THE ABILITY OF MENTAL HEALTH PROFESSIONALS TO ASSESS GROWTH POTENTIAL ON A RELATIVE BASIS. AS A GROUP, THE GIRLS WERE MOST ADEQUATE IN ACADEMIC AND OCCUPATIONAL PURSUITS, AS COMPARED TO HETEROSEXUAL RELATIONSHIPS, AND SELF-ATTITUDE WAS THE ONLY VARIABLE IN WHICH SIGNIFICANT GAINS FAILED TO OCCUR. VARIABLES PARTICULARLY ASSOCIATED WITH OUTCOME WERE AGE AT ADMISSION AND PSYCHOLOGICAL ADJUSTMENT. COMPARISON WITH THE RESULTS OF PREVIOUS INVESTIGATIONS WAS IMPEDED DUE TO NONSIMILARITY IN FUNCTION OF THE FORMER INVESTIGATIONS. A MAJORITY OF THE GIRLS HAD POSITIVE MEMORIES OF THE BUILDING, COMMUNITY, PEER GROUP, AND CASEWORK SERVICE; BUT THEY HAD NEGATIVE RECOLLECTIONS OF THE STAFF, PSYCHIATRIC SERVICE AND THE DIRECTOR. THE PROGRAM WAS PERCEIVED AS AT LEAST MODERATELY BENEFICIAL TO 59 PERCENT OF THE GIRLS, AND THE PEER GROUP WAS EVALUATED AS HAVING A POSITIVE IMPACT. RESULTS SUPPORT THE NEED FOR PROFESSIONALIZATION OF THE CHILD CARE ROLE, CONFIRM THAT INDIVIDUAL PSYCHOTHERAPY WAS NOT AGREEABLE TO THE CLIENTS, AND EMPHASIZE THE IMPORTANCE OF THE PEER GROUP AS TREATMENT RESOURCE. APPENDIXES INCLUDE A COPY OF THE QUESTIONNAIRE ADMINISTERED TO THE SUBJECTS, THE QUESTIONS ASKED OF THE WOMEN IN THE INTERVIEWS WITH GUIDELINES FOR THE JUDGES TO USE IN MEASURING THE ADJUSTMENT AND OTHER VARIABLES. REFERENCES ARE PROVIDED. (AUTHOR ABSTRACT MODIFIED)