On the evaluation of procedural systems in laboratory experiments

Conclusion This article has criticized various aspects of the Thibaut and Walker research. Thibaut and Walker have greatly contributed to the study of procedural justice by conceptually isolating issues and suggesting ways they can be empirically studied. Our criticisms in Section I point out weaknesses in the design of particular experiments but do not question the appropriateness of laboratory research on the operating characteristics of procedural systems. In this realm, Thibaut and Walker's work will provide the starting point for future research.We do question the appropriateness of laboratory research for answering subjective questions, as outlined in Section III. While we agree that experimental research can indicate how people feel about a system, we do not think that objective measurement of the incidence of such subjective feelings tells us anything about the objective qualities of the systems considered. The assertion that one system is superior to another because some people feel it is so goes beyond the limitations of the experiments and the data. This type of overgeneralization is seriously misleading when it is used to support policy recommendations, particularly when other factors, such as the costs we discuss in Section II, are not considered.We wish to thank Dr. Jacqueline Macaulay, Professor Stewart Macaulay, and Dr. Neil Vidmar for their comments on an earlier draft of this article. Responsibility for its contents rests solely with us.