TWO DIFFERENT EMPHASES OF PROGRAM EVALUATION ARE IDENTIFIED - THE 'MEASUREMENT COMPONENT' WHICH INVOLVES COLLECTING DATA ON AND MEASURING OBJECTIVE OUTCOME/IMPACT FACTORS SUCH AS REDUCTION IN RECIDIVISM RATES, AND THE 'JUSTICE COMPONENT' WHICH STRESSES SUBJECTIVE FACTORS SUCH AS THE QUALITY OF HUMAN RELATIONS. CERTAIN INHERENT CHARACTERISTICS OF VOLUNTEER PROGRAMS (CLIENT/VOLUNTEER MATCHING, ONE-TO-ONE ASSIGNMENT, VOLUNTEER INCULCATION OF LOW EXPECTATIONS FOR SUCCESS, ARE POINTED TO AS SUBJECTIVE FACTORS AT WORK IN THESE PROGRAMS. IT IS ARGUED THAT THE RESPONSIBLE FACTORS FOR A PROGRAM TEND TO PLACE MORE IMPORTANCE ON THE RELATIONSHIP BETWEEN THE VOLUNTEER AND THE CLIENT OVER THE OBJECTIVE OUTCOME OF THAT RELATIONSHIP. ALSO CITED ARE THE NEGATIVE ATTITUDES TOWARD PROGRAM EVALUATION CAUSED BY LOW LEVEL STRUGGLES FOR PROGRAM SELF-DETERMINATION AND SURVIVAL. THE AUTHOR THEN WEIGHS THE RELATIVE ADVANTAGES OF EIGHT EXPERIMENTAL DESIGNS FOR EVALUATING THESE PROGRAMS - ONE-SHOT CASE STUDY, ONE GROUP PRE-AND POST-TEST DESIGN, STATIC GROUP COMPARISON, UNDER-MATCHED (EX POST FACTO) DESIGN INTERRUPTED TIME SERIES DESIGN, MULTIPLE TIME SERIES DESIGN, POST-TEST ONLY DESIGN, AND CLASSICAL DESIGN. IT IS RECOMMENDED FUTURE PROSPECTIVE FUNDING GRANTEES, AS A PRIOR CONDITION OF FUNDING, BE REQUIRED TO PROVIDE A QUANTITATIVE DESCRIPTION OF THEIR ULTIMATE GOAL (WHICH SHOULD BE THE REDUCTION OF CRIME) AND THEIR MEANS OF ACHIEVING THE GOAL (THE PROGRAM). IT IS ALSO SUGGESTED THAT THE GRANTEE, OR AN INDEPENDENT EVALUATOR, CAN ALSO PRODUCE THE DESCRIPTION OF THE GOAL AS PART OF A FINAL REPORT ON THE PROGRAM. OBJECTIVE OUTCOME CRITERIA AND PROGRAM EVALUATION COULD THEN BE EXPECTED (AND WOULD BE) BASED ON THESE STATEMENTS OF GOALS AND MEANS. SAMPLE MODEL DATA COLLECTION FORMS FOR BASELINE (PROBLEM DESCRIPTION), OUTCOME (GOAL DESCRIPTION), AND VOLUNTEER/ CLIENT RELATIONSHIP DATA ARE APPENDED. A SHORT BIBLIOGRAPHY IS ALSO INCLUDED.