Justice and Law Enforcement:
Evaluation and the Research Process
Oct 15, 1979
The research process should be evaluated. The nature of our political system and institutions makes it inevitable that decisions affecting scientific research will be made by professional managers and elected officials who are not scientists. Scientists must seek an effective way to participate. The most effective and enduring participation may well take the form of assisting the professional analysts as they try to strengthen their methods of analysis and make them more relevant to decisions affecting research. The Comptroller General was instructed by legislation to develop and recommend to Congress methods for review and evaluation of government programs and activities carried out under existing law. Research programs have been evaluated, all too often, by people who know little about the research process, without benefit of a clear analytical framework, and using analytical tools which are inadequate to the task. An analytical framework is needed which will provide a more balanced context in which managers and elected officials can make decisions about research and which will also satisfy the public's legitimate demand for accountability. GAO is making a substantial investment in trying to advance the state of the art of evaluation in general and of science policy formulation. The best framework to build is the long-term investment model of high- risk, high payoff, and long-term investment. The research community would be well advised to make a modest investment of its own to advance the state of the art in this form of evaluation.