Managing Delphi Surveys Using Nonparametric Statistical Techniques*


  • Roy C. Schmidt

    1. Department of Information and Systems Management, School of Business and Management, The Hong Kong University of Science and Technology, Clearwater Bay, Sai Kung, Hong Kong, email:
    Search for more papers by this author
    • 1

      Roy Schmidt An assistant professor of information systems in the School of Business and Management, The University of Science and Technology, Hong Kong. He has a B.A. and B.S. from the University of Maryland, an M.B.A. from St. Mary's University, San Antonio, and a Ph.D. in MIS from Indiana University. Before entering academia, Dr. Schmidt served in the U.S. Air Force for over 22 years. His research interests include support of the strategic decision process, executive information use, information systems project management, and information systems project risk management.

  • *

    The author would like to thank Dan Couger for his kind cooperation in using the data as presented in his previous paper (Couger, 1988b) to illustrate the techniques advocated. Thanks also to a blind reviewer and the associate editor for insightful comments that strengthened and streamlined the paper.


Information systems researchers have often turned to a variant of the Delphi survey technique to support their research of key issues in their field. Two particular weaknesses of past studies using this approach have been a lack of a definitive method for conducting the research and a lack of statistical support for the conclusions drawn by the researchers. In this paper, the author presents a method, based on nonparametric statistical techniques, to conduct ranking-type Delphi surveys, perform analysis, and report results. The author takes this one step further by illustrating an actual analysis of a Delphi survey. The analysis is compared to results that were presented without the benefit of the author's approach. This paper shows that use of the advocated approach can streamline and strengthen studies, improve the validity of results, and thus better serve the consumers of the research findings. Since the ranking-type Delphi is so popular among information systems researchers, a consistent method is needed to apply to their data collection, analysis, and reporting of results. This paper provides such a method in concise form and illustrates the use of the method in a manner affording comparison between it and previous practice.