Decision-Making Formats: a Comparison on an Evaluative Task of Interacting Groups, Consensus Groups, the Nominal Group Technique, and the Delphi Technique.

Abstract

Using Steiner's (1972) model of group process and productivity, four decision-making formats: the interacting method, the consensus method, the nominal group technique, and the Delphi technique, were compared on two dimensions of an effective decision, quality and acceptance. A literature review indicated that a comparison of the effectiveness of the decision-making formats is difficult because of the prevalence of idiosyncratic modifications made to the formats. To preserve the integrity of the findings of the present study, no modifications were made in the intervention techniques. A total of 144 male undergraduate students, working in four-person, ad hoc groups solved the NASA Lost on the Moon evluative task. An unbiased quality measure was obtained by using the LSU scoring algorithm. Both self-report and behavioral measures of acceptance were determined. The results indicated that as predicted, groups using the Delphi method produced significantly better quality decisions than did the groups using the interacting method. There was a significant main effect for decision-making formats on the behavioral measure of acceptance. Post-ANOVA tests revealed that, as predicted, groups using the consensus method fostered more acceptance than did groups using the interacting method, and that groups using the nominal group technique fostered more acceptance than did groups using the Delphi technique. In addition, eight of the sixteen self-report measures of acceptance reached significance. Examination of these results indicated that the consensus method fostered the largest amount of self-reported acceptance, followed by the interacting method and the nominal group technique, while the Delphi technique fostered the least amount of acceptance. The findings of this study add support to the claims that the results of research using modified intervention techniques may be suspect. Implications for future research using quality and acceptance measures were discussed. It was concluded that the results of this study may be of assistance to the practitioner interested in selecting the appropriate decision-making format for evaluative problem-solving situations.