A SURVEY IS REPORTED IN WHICH A SAMPLE OF 882 MENTAL HEALTH FACILITIES IN NEW YORK WERE SENT QUESTIONNAIRES; 225 RESPONSES WERE RECEIVED. INFORMATION WAS OBTAINED ON THE LEVEL OF EVALUATION ACTIVITY, THE LEVEL OF INFORMATION CAPABILITY, AND THE ROLE OF THE EVALUATOR. DATA ON THE ROLE OF THE EVALUATOR WAS THE MOST DIFFUSE. ANOTHER SURVEY OF RESEARCH ACTIVITIES WAS CONDUCTED IN FEDERALLY FUNDED MENTAL HEALTH CENTERS. TABULATIONS FROM 189 OUT OF 325 CENTERS INDICATED THAT 60 PERCENT OF THE CENTERS EMPLOYED A PERSON WHOSE PRIMARY RESPONSIBILITY WAS RESEARCH AND EVALUATION. ALTHOUGH TRAINING WAS JUDGED IMPORTANT, FEW EVALUATORS WERE TRAINED IN THEIR FIELD. IN 1963, AN AMERICAN PUBLIC HEALTH ASSOCIATION SUBCOMMITTEE PROPOSED FOUR LEVELS OF STATISTICIANS THAT TAKE INTO ACCOUNT DIFFERING LEVELS OF RESPONSIBILITY AND PRIMARY CONCERNS: (1) ENTRY AND TRAINING LEVEL; (2) OPERATIONS METHODOLOGY LEVEL (DEVELOPMENT OF DATA COLLECTION AND HANDLING PROCEDURES AND DETERMINATION OF APPROPRIATE STATISTICS); (3) OPERATIONS METHODOLOGY LEVEL (SUPERVISING STATISTICAL STAFF, STATISTICAL THEORY, AND RESEARCH IN STATISTICS); AND (4) LEVEL INVOLVING THE FORMULATION AND DEFINITION OF RESEARCH QUESTIONS. IN 1968, A STUDY IDENTIFIED FIVE MAJOR GROUPS ENGAGED IN BOTH STATISTICS AND PUBLIC HEALTH. TRAINING NEEDS FOR MATHEMATICAL STATISTICIANS, EPIDEMIOLOGISTS, OTHER HEALTH PROFESSIONALS, PUBLIC HEALTH AND MENTAL HEALTH PROGRAM STATISTICIANS, AND HEALTH PROGRAM PLANNERS AND SYSTEM ANALYSTS WERE OUTLINED. NUMEROUS TRAINING PROGRAMS HAVE BEEN DEVELOPED, AND ONE CHARACTERISTIC OF THE PROGRAMS IS TO PROVIDE FIELD EXPERIENCES FOR UNIVERSITY STUDENTS. CONTINUING EDUCATION FOR PERSONS WHO HAVE ALREADY RECEIVED THEIR BASIC EDUCATION IS VIEWED AS ONE WAY FOR IMPROVING THE PROFICIENCY OF EVALUATORS. IT WILL BE NECESSARY TO TRAIN ADMINISTRATORS AND DIRECTORS OF SERVICE PROGRAMS, AS WELL AS METHODOLOGISTS OR OPERATIONS STATISTICIANS. NOTES ARE INCLUDED. (DEP)