DATA ON THE LASTING EFFECTIVENESS OF COMPENSATORY EDUCATION PROGRAMS ARE AMBIGUOUS. EVALUATIONS HAVE BEEN HAMPERED BY A LACK OF LONGITUDINAL STUDIES AND CONTROLLED EXPERIMENTS AND BY THE IMPRECISION OF THE STANDARD MEASURES. EFFECTIVE PROGRAMS, THOSE WHICH PRODUCE INCREASES IN LEARNING, SHOULD FEATURE (1) CURRICULUM ADAPTATION TO THE INDIVIDUAL NEEDS AND ENVIRONMENTAL REALITIES OF POOR AND/OR NEGRO CHILDREN, (2) INSERVICE TRAINING IN ATTITUDES AND CURRICULUM FOR TEACHERS WHO HAVE NOT RAISED STUDENTS' ACADEMIC PERFORMANCE TO ADEQUATE LEVELS, (3) CONCERN FOR HEALTH, WELFARE, AND FOOD NEEDS, AND (4) PARENT INVOLVEMENT. ALL THESE FEATURES ARE EXPENSIVE BUT HIGH EXPENDITURES ALONE CAN NOT GUARANTEE PROGRAM EFFECTIVENESS. THE MORE EFFECTIVE SCHOOLS PROGRAM IN NEW YORK CITY HAS BEEN UNABLE TO STOP THE PROGRESSIVE RETARDATION OF DISADVANTAGED YOUTH BECAUSE IT HAS LACKED QUALIFIED TEACHERS AND INDIVIDUALIZED INNOVATIVE INSTRUCTION. REPORTS FROM VARIOUS STUDIES STRESS TEACHER ATTITUDES AND EXPECTANCY OF SUCCESS OR FAILURE AS IMPORTANT VARIABLES IN A DISADVANTAGED CHILD'S ACHIEVEMENT. OTHER STUDIES HAVE FOUND THAT EFFECTIVE INDIVIDUALIZED INSTRUCTION--INTENSIVE READING TEAMS, TUTORS, AND HOMEWORK HELPERS--PRODUCES SIGNIFICANT ACADEMIC GAINS. ONE STUDY HAS SHOWN SIGNIFICANT I.Q. GAINS WHEN INFANT TUTORING STARTED AT 14 MONTHS. THE VASTNESS OF TITLE I PROGRAMS OFFERS THE CHANCE TO EVALUATE SYSTEMATICALLY WHAT TYPES OF COMPENSATORY PROGRAMS ARE MOST EFFECTIVE IN IMPROVING THE ACHIEVEMENT OF THE DISADVANTAGED STUDENT. THIS PAPER WAS PREPARED FOR THE NATIONAL CONFERENCE ON EQUAL EDUCATIONAL OPPORTUNITY IN AMERICA'S CITIES, SPONSORED BY THE U.S. COMMISSION ON CIVIL RIGHTS, WASHINGTON, D.C., NOVEMBER 16-18, 1967. (NH)