TO BE RELEVANT, OUTCOME MEASUREMENT MUST ADDRESS THREE ASPECTS OF PROGRAM MANAGEMENT: CONTINUOUS MONITORING OF PROGRAM QUALITY; DEMONSTRATION OF PROGRAM EFFECTIVENESS; AND DECISIONS ABOUT PROGRAM MODIFICATIONS AIMED AT IMPROVING EFFECTIVENESS. MONITORING IS USED TO DETECT SPECIFIC STRENGTHS AND WEAKNESSES IN A PROGRAM. INEXPENSIVE OUTCOME INDICATORS, SUCH AS RATINGS OF CLIENT STATUS BEFORE AND AFTER SERVICE AND MEASURES OF CLIENT SATISFACTION, ARE PRACTICAL FOR THE QUANTITATIVE ASPECTS OF MONITORING AND MAY BE SUPPLEMENTED BY QUALITATIVE FOLLOWUP CONTACTS WITH SELECTED CLIENTS. DEMONSTRATING OVERALL PROGRAM EFFECTIVENESS IS MORE DIFFICULT. ESTIMATES OF EFFECTIVENESS REQUIRE NORMATIVE STUDIES AMONG GROUPS OF SIMILAR PROGRAMS. HOWEVER, THE DEVELOPMENT OF OUTCOME NORMS IS IN ITS INFANCY. OUTCOME EVALUATIONS CONCERNED WITH AIDING ADMINISTRATIVE DECISIONMAKING GENERALLY ARE TIME-LIMITED EFFORTS RATHER THAN ONGOING MONITORING SYSTEMS. COMPARATIVE OUTCOME STUDIES SUPPORT TWO TYPES OF PROGRAM DECISIONS: THE ADMINISTRATIVE POLICY DECISION TO OFFER A NEW OR SUBSTITUTE SERVICE AND THE CLINICAL DECISION PROCESS OF ASSIGNING CLIENTS TO SERVICES. TEN ISSUES MUST BE RESOLVED IN PLANNING AN OUTCOME STUDY: SELECTION OF SERVICES TO BE COMPARED; ASSIGNMENT OF SUBJECTS TO SERVICES; SELECTION OF SUBJECTS; OBTAINING INFORMED CONSENT FROM SUBJECTS; DESCRIPTION OF THE SAMPLE; DESCRIPTION OF SERVICES PROVIDED; SELECTION OF PREDICTOR VARIABLES; SELECTION OF OUTCOME MEASURES; ANALYSIS OF DATA; AND FEEDBACK OF RESULTS INTO THE DECISION PROCESS. GUIDELINES FOR RESOLVING EACH ISSUE ARE OFFERED (INCLUDING EXPLANATIONS OF FOUR TYPES OF SUBJECT ASSIGNMENT DESIGN). A SAMPLE GLOBAL ASSESSMENT SCALE (A METHOD FOR ASSESSING CLIENT STATUS) AND A SAMPLE CLIENT SATISFACTION QUESTIONNAIRE ARE PROVIDED. TOGETHER WITH A LIST OF REFERENCES. (AUTHOR ABSTRACT MODIFIED--LKM)