Rationality

Evaluation, in the heady days when the United States was at war with poverty, when government was thought to be capable of solving social problems, was the hope for making decisions rational. The expectation for rationality ran like this: Specify a problem that required government action, develop a list of policies that might remedy the problem, evaluate the effectiveness and cost of each of the policy alternatives, compare the results, and choose the most effective, lowest cost policy. Effective policy solutions were assumed to be abundant. Evaluations producing comparable information on the relevant policies were expected to be easy to conduct. Decisions could wait to follow the delivery of findings. Political interests would yield to the correctness of empirically determined solutions. Evaluation findings would supplant democratic debate and deliberation with instrumentally determined decisions. The petty politics of personality, vested economic interests, backroom deals, arbitrariness, and arm twisting for votes would all become things of the past. Evaluation would light the way of rationality.

Of course, few evaluators or proponents of evaluation completely bought this idealized view of the impact of evaluation, but the idea that evaluation could make decisions more rational was widely held. Passing years have circumscribed expectations for evaluation to produce greater rationality. Evaluations are much less definitive than the idealized model would require, and politics is more vital for social decision making than might have been believed. Rarely do we have evaluations that provide head-to-head comparisons of alternative policies. Many policies have been shown to produce different effects among different groups of program participants, and effects often differ in size and significance from one place to another. Experiences of everyday evaluators, as well as systematic research on evaluation use, began to show that evaluation findings were less influential in the decision-making process even when they were available and clear. Evaluation, according to the [Page 359]analysis of a student of politics, Majone (1989), does not so much determine the best decision as add fuel to the arguments for and against certain policies.

Not only has experience diminished the promise of evaluation leading to rational decisions, the desirability of such an outcome has been reduced, if not entirely reversed. “Epistemocracy,” or the rule of knowledge that results from the model of rational decision making, has been severely challenged. Knowledge and expertise have been decried by some as simply guises for privilege. In democratic societies, issues about values and preferences must find avenues for debate and open, legitimate means for resolving differences that are too complex for technical solutions.

Melvin Mark predicts that the next great divide in the field of evaluation will separate evaluators who consider themselves neutral brokers of information from evaluators who act as advocates for a group or a set of policy preferences. Although the idealized model of rational decision making has succumbed to politics, the idea that empirical findings generated by systematic evaluations can and should influence public decisions made in modern democracies is being vigorously applied and vigorously contested.