In this brief discussion of the evaluation of compensatory education, the author analyzes why it has been difficult to discover whether compensatory education has succeeded; in addition, why the causes of success or failure are difficult to assess. In particular, the author points out that the political and economic circumstances of programs make evaluation difficult, and that problems of evaluative research in general, research design, and the static approach to assessment have posed formidable obstacles. Despite the many problems in the design, implementation, and evaluation of compensatory education programs and the equivocal status of much of the evaluation effort, educators are nonetheless constantly called upon to make judgements and policy decisions based upon the experiences so far. The author presents some insights drawn from these research evaluation experiences. (Author/JW)