A GROUP OF 92 MALE JUVENILES WHO HAD BEEN CHARGED WITH FELONIES WAS REFERRED FROM THE BOYS' INTAKE DEPARTMENT OF THE WAYNE COUNTY JUVENILE COURT IN DETROIT, MICHIGAN. IT WAS DIVIDED INTO CONTROL AND TREATMENT GROUPS. THE MEAN AGE WAS 13-14 YEARS, THE MEAN GRADE WAS 8TH, AND THE MEAN READING LEVEL WAS 5TH GRADE. THE 45 COLLEGE STUDENTS WERE ALL TAKING AN ADVANCED PSYCHOLOGY COURSE IN FIELD WORK AT EITHER WAYNE STATE UNIVERSITY OR WAYNE COUNTY COMMUNITY COLLEGE. THEY WERE TRAINED BEFORE THE PROGRAM AND HAD ACCESS TO SKILLED HELP DURING THE PROGRAM. EACH WAS ASSIGNED ON A ONE-TO-ONE BASIS TO A BOY FROM THE TREATMENT GROUP. THE COMPANION COUNSELING DID NOT PREVENT FURTHER DELINQUENCY, OR IMPROVE READING ABILITY AND SCHOOL GRADES. THE COUNSELING DID, HOWEVER, IMPROVE PERFORMANCE ON THE JESNESS PERSONALITY SUBSCALES OF IMMATURITY AND SOCIAL ANXIETY. THOSE WHO HAD LONGER COUNSELING PERIODS (6 TO 8 MONTHS) ALSO IMPROVED ON THE JESNESS PERSONALITY INDEX OF WITHDRAWAL. THOSE COUNSELED FOR MORE THAN 3 MONTHS TENDED TO BECOME LESS DEPRESSED AND MORE SATISFIED WITH THEMSELVES. THIS WAS REFLECTED IN SCHOOL BEHAVIOR, AS RATED BY SCHOOL COUNSELORS, AND IN BEHAVIOR AT HOME, AS RATED BY PARENTS. REASONS WHY RECIDIVISM ALONE WAS NOT USED AS A MEASURE OF COUNSELING EFFECTIVENESS ARE DISCUSSED. THE SELECTION AND ADMINISTRATION OF THE PERSONALITY INVENTORIES IS DETAILED. THE SECOND SECTION DISCUSSES THE EVALUATION AND VALIDATION OF AN INSTRUMENT TO DISTINGUISH AND SELECT POTENTIALLY SUCCESSFUL VOLUNTEERS. THE INSTRUMENT WAS DEVELOPED BY THE OAKLAND COUNTY JUVENILE COURT VOLUNTEER CASE-AID RESEARCH STAFF IN PONTIAC, MICHIGAN. IT WAS HYPOTHESIZED THAT THE VOLUNTEER'S COMPETENCY COULD BE ASSAYED AND PREDICTED BY OBSERVING THE CONTINUOUS DECISIONMAKING PROCESS OCCURRING IN THE VOLUNTEER-YOUNGSTER RELATIONSHIP. TO REPLICATE THE PROCESS, THE CRITICAL INCIDENT TECHNIQUE WAS EMPLOYED. A CRITERION GROUP OF 28 VOLUNTEERS WITH AT LEAST 6 MONTHS OF EXPERIENCE (ON THE ASSUMPTION THAT THE MORE SUCCESSFUL VOLUNTEER IS MORE EXPERIENCED) PRETESTED AN INSTRUMENT, CONSISTING OF A SELECTION OF ALTERNATIVES TO EACH OF 51 ITEMS. THE RESULTING 32-ITEM INSTRUMENT WAS THEN ADMINISTERED TO THE 82 STUDENT VOLUNTEERS, DIVIDED INTO TWO GROUPS ACCORDING TO EXPERIENCE. THE EXPERIENCED GROUP VALIDATED 30 OF THE 32 ITEMS ON THE CRITICAL INCIDENT RESPONSE TEST. VARIABLES SUCH AS EDUCATION BUT NOT SEX OF THE VOLUNTEER AFFECTED SCORES ON THE TEST. WHITE VOLUNTEERS SCORED HIGHER, BUT THEIR HIGHER EDUCATION LEVEL MAY ACCOUNT FOR THIS. EXPERIENCED GROUP SCORES WERE SIGNIFICANTLY HIGHER ON ONLY 5 ITEMS. IT IS CONCLUDED THAT THE VALUE OF THE TEST AS AN EVALUATIVE MEASURE OF VOLUNTEER PERFORMANCE IS QUESTIONABLE; HOWEVER , IT WAS USEFUL AS A MEANS OF TRAINING VOLUNTEERS. FURTHER RESEARCH IS SUGGESTED. AN APPENDIX GIVES LETTERS TO PARENTS AND SCHOOL COUNSELORS ASKING COOPERATION PLUS TEST DATA. EXTENSIVE TABULAR DATA, DETAILS OF THE STATISTICAL ANALYSES, AND A BIBLIOGRAPHY ARE INCLUDED.