Political, Interpersonal, and Ethical Issues in Evaluation 69
needs, not more. As Carol Weiss has remarked, “Politics is the method we have as a nation to resolve differences and reach conclusions and decide policy issues. We don’t always like the way it turns out, but it’s an essential part of our system” (Weiss & Mark, 2006, p. 482).
Furthermore, evaluation serves a central purpose in our political system: accountability. Accountability means that government is accountable to the people, to the public that elects its leaders. Eleanor Chelimsky, the former director of the Program Evaluation and Methodology Division of the U.S. government’s General Accounting Organization, now the Government Accountability Office (GAO), which provides evaluative information to Congress, argues that “the ultimate client or user of our work is the public” (2008, p. 403). She views evaluation as central to democracy in making the work of government more transparent so that leaders may be held accountable. It is rare, she notes, for nondemocratic countries to have organizations within government that evaluate its work (Chelimsky, 2006).
Thus, evaluators do work in a political context, and evaluation exists, at least partly, to make government accountable. Of course, this system does not always work perfectly, but it is important for evaluators to recognize the potential for their role within this system. That potential is to provide information to the public and to stakeholder groups, be they policymakers, program managers, or groups lobbying for or against a particular cause. However, to do so well requires the eval- uator to have some understanding of that system and the complexities involved in evaluators’ interaction with it.
One reason why evaluators are sometimes reluctant to become involved in politics is their concern that the strength of evaluative findings lies in those findings being seen as independent or objective. In other words, policy actors and the public value evaluations because they think evaluators and their evaluations are not political, but instead are neutral and, as such, are providing information that is “untainted” by political views and beliefs. (Most evaluators recognize that, in fact, data and evalua- tions are inevitably influenced by values and that it is impossible to totally remove bias from data collection. We will discuss that issue further in later chapters on data collection. Suffice it to say here that evaluation is often valued by stakeholders because they perceive it to be objective.) Therefore, evaluators can be legitimately concerned with how their work within this political context may affect the perceived objectivity of their work. How can evaluators interact with those in the political environment to make sure their study addresses important issues and that the results get to the right people or groups without harming the perceived independence or objectivity of their work? There is no easy answer to this question, but we will describe several potential ways in which evaluators can work within the political system.