A well-known evaluator once said, "Evaluation — more than any science — is what people say it is, and people currently are saying it is many different things" (1). Ask an economist what program evaluation is, and you will get a very different answer than if you asked a psychologist; and they would both differ from what an educator might say. Indeed, the field is so large and diverse, and the use of the term program evaluation is so ubiquitous, that it is often difficult to discern any common threads. Yet common threads do exist, and I would like to point to some of them in the articles of this special issue of Preventing Chronic Disease.
Five common concerns are woven throughout the literature on program evaluation (2). First is a concern with how to construct valid knowledge. This concern has both a philosophical component and a methodological component; the philosophical component concerns the kinds of things we can know about programs, and the methodological component concerns the designs, measures, and analyses that we use to create and organize data. Second is a concern with how we place value on evaluation results. One often hears it said that the data speak for themselves, but that is rarely the case. This concern articulates the many theoretical and practical tools we have available to help us in this valuation. Third is a concern with how programs change. After all, program evaluation is intended to be a very practical area of study, one that aims to make a real difference in people's lives. If we do not know the leverage points for program change, we cannot apply evaluation results to gain that leverage. Fourth is how to use evaluation results in the policy process. This concern is about how to get our results to the stakeholders who influence those leverage points in a way that helps stakeholders make use of the data. Fifth and finally is the paramount concern with how to organize evaluation practice, given the implications of all the preceding issues for what the evaluator actually does in any given evaluation. This fifth concern is always a matter of tradeoffs, for one can never do everything well in a single evaluation.