THE SUBJECTS WERE 167 YOUTHS--41 FEMALE AND 126 MALE--HANDLED BY THE JUVENILE COURT PROBATION SERVICE OF FRESNO COUNTY, CALIF. OF THESE, 10 WERE ELIMINATED BECAUSE THEY HAD SERIOUS ASSAULTIVE BEHAVIOR OR WERE SUICIDAL, LEAVING 157 WHO WERE DIVIDED INTO FOUR GROUPS: (1) THOSE MATCHED WITH STUDENT INTERN PROBATION COUNSELORS USING THE CALIFORNIA YOUTH AUTHORITY'S COMMUNITY TREATMENT PROJECT (CTP) SCALE; (2) THOSE MATCHED AT THE MEDIAN USING THE FUNDAMENTAL INTERPERSONAL RELATIONS ORIENTATION (FIRO-B) SCALE; (3) THOSE MATCHED AT THE EXTREMES USING THE FIRO-B SCALE; AND (4) UNMATCHED CONTROLS. THE FACTORS MEASURED BY EACH OF THESE MATCHING SYSTEMS ARE DESCRIBED. BASICALLY THE FIRO-B PROVIDES A DIRECT OPERATIONAL MEASURE OF INTERPERSONAL COMPATIBILITY, WHILE THE CTP MODEL IS A COMPLEX SYSTEM OF CROSS-CLASSIFICATION TO DETERMINE EFFECTIVE PERSONALITY COMBINATIONS. THE STUDY LASTED 9 MONTHS; MOST OF THE SUBJECTS WERE IN COUNSELING FOR 7.5 TO 9 MONTHS, AND THE DEPENDENT VARIABLES WERE NUMBER OF DELINQUENT ACTS COMMITTED DURING THIS PERIOD. NO SIGNIFICANT DIFFERENCES WERE FOUND AMONG GROUPS. IT IS CONCLUDED THAT (1) MATCHING A FIRST OFFENDER SUBJECT WITH A STUDENT PROBATION OFFICER DID NOT INFLUENCE PROBATION OUTCOMES SIGNIFICANTLY, AND (2) PERSONALITY VARIABLES SHOWN BY DELINQUENTS ARE SO COMPLEX AND SO DIFFERENT THAT 'FIRST OFFENDERS' CANNOT BE CONSIDERED A HOMOGENEOUS GROUP AND STUDIES MUST BE DESIGNED TO STUDY SUBGROUPS. TABULAR DATA AND REFERENCES ARE PROVIDED. (GLR)