This article discusses the similarities and dissimilarities between research and evaluation, which are two clearly differentiated disciplines despite their similarity in concepts, tools, and methods. The purpose of research is to enlarge the body of scientific knowledge; the purpose of evaluation is to provide useful feedback to program managers and entrepreneurs. In this article I examine the central characteristics of research and evaluation (validity, generalization, theory and hypotheses, relevance, and causality) and the different roles those characteristics play in each. I discuss the different functions of evaluation and research, and propose some criteria for fulfilling the different demands of evaluation and research. And I argue that the constant pressure to examine evaluations by the criteria of research prevents evaluation from becoming an independent discipline and delays the development of standards and criteria that are useful to evaluators.