EVALUATING PROGRAMS FOR THE DISADVANTAGED IS DIFFICULT BECAUSE THERE IS NO CLEAR THEORETICAL OR OPERATIONAL DEFINITION OF THE TERM "DISADVANTAGED." THE PROLIFERATION OF SCHOOL PROGRAMS ALSO CREATES EVALUATION PROBLEMS, FOR OFTEN EVALUATION TAKES PLACE BEFORE THE PROGRAM IS EITHER FULLY IMPLEMENTED OR HAS FUNCTIONED FOR A SUFFICIENT PERIOD OF TIME. A DISTINCTION SHOULD BE MADE BETWEEN CRITICAL EVALUATION, WHICH WOULD DETERMINE THE CONTINUATION OF A PROGRAM, AND ONGOING EVALUATION, WHICH WOULD IDENTIFY ITS STRENGTHS AND WEAKNESSES. RESEARCHERS MUST RECOGNIZE THAT THOSE WITH VESTED INTERESTS IN THE SCHOOL POWER STRUCTURE WILL USE EVALUATIONS TO SUPPORT THEIR OWN VIEWPOINTS. EVALUATIONS SHOULD ESTABLISH SENSIBLE, RELEVANT, MEANINGFUL CRITERIA OF SUCCESS OR FAILURE AND RECOGNIZE THE TOTALITY OF GAINS AND LOSSES OF A GIVEN ENDEAVOR. THERE IS ALSO THE QUESTION OF WHAT IS TO BE CONSIDERED PROGRESS FOR CHILDREN WITH DEFICITS--WHETHER IT SHOULD BE DEFINED AS RESTORATION TO THEIR NORMAL BEGINNING LEVEL, OR AS ARRESTING ANY FURTHER DECLINE, OR AS RAISING THE LEVEL TO THAT OF NONDEFICIENT CHILDREN. ALSO, RESEARCHERS CAN NO LONGER MAINTAIN AN ISOLATION FROM POTENTIAL IMPLEMENTATION OF THEIR FINDINGS BUT MUST BECOME INVOLVED IN THE USES TO WHICH THEIR DATA ARE PUT. THIS ARTICLE WAS PUBLISHED IN "THE URBAN REVIEW," VOLUME 2, NUMBER 3, DECEMBER 1967. (NH)