THE METHOD OF ANALYSIS USED DETERMINES THE DEGREE OF ASSOCIATION BETWEEN A CLIENT'S CHARACTERISTICS AT ADMISSION AND AT DISCHARGE ON SLECTED OUTCOME CRITERIA--PRODUCTIVITY AT DISCHARGE (IN SCHOOL, IN A TRAINING PROGRAM, OR EMPLOYED), CRIMINAL JUSTICE INVOLVEMENT, AND SUCCESSFUL COMPLETION OF THE TREATMENT PROGRAM. THE PENNSYLVANIA DATA COLLECTION SYSTEM, WHICH EVALUATES ALL TREATMENT FACILITIES WITHIN THE STATE, PROVIDED THE DATA USED IN THE ANALYSIS. OF THE 4,738 DRUG TREATMENT CLIENTS INCLUDED IN THE STUDY, 62 PERCENT WERE JUVENILES UNDER 18 YEARS OLD AND 38 PERCENT WERE YOUNG ADULTS 18 OR 19 YEARS OF AGE. THESE TWO AGE GROUPS RECEIVED TREATMENT IN TWO DIFFERENT TREATMENT ENVIRONMENTS--OUTPATIENT DRUG-FREE PROGRAMS AND THERAPEUTIC COMMUNITIES. THE VARIABLES USED TO PREDICT TREATMENT SUCCESS (INDEPENDENT PREDICTOR VARIABLES) WERE DEMOGRAPHIC VARIABLES (RACE AND SEX), DRUG-USE VARIABLES, AND THE CLIENT STATUS AT ADMISSION REGARDING EDUCATION, EMPLOYMENT, CRIMINAL JUSTICE INVOLVEMENT, AND PREVIOUS TREATMENT HISTORY. THE PREDICTION OF TREATMENT SUCCESS WAS STUDIED FOR FOUR GROUPS: JUVENILES IN OUTPATIENT COUNSELING/PSYCHOTHERAPY FACILITIES, JUVENILES IN RESIDENTIAL THERAPEUTIC COMMUNITIES, YOUNG ADULTS IN OUTPATIENT UNITS, AND YOUNG ADULTS IN THERAPEUTIC COMMUNITIES. A MULTIVARIATE ANALYSIS WAS CONDUCTED, WITH THE SPECIFIC TECHNIQUE BEING THE STEPWISE REGRESSION PROCEDURE. TWO FEATURES WERE APPARENT IN THE DATA. THOSE RECEIVING TREATMENT IN OUTPATIENT COUNSELING AND/OR PSYCHOTHERAPY SEEMED TO HAVE DIFFERENT SETS OF SIGNIFICANT PREDICTOR VARIABLES THAN DID THE CLIENTS IN THERAPEUTIC COMMUNITIES, AND THE DIFFERENT SET OF PREDICTOR VARIABLES WAS RELATED TO THE TYPE OF CLIENTS TREATED AND THE MANNER IN WICH THEY WERE TREATED. SUPPORTING TABULAR DATA AND REFERENCES ARE PROVIDED. (RCB)