Adaptive Self-Explication of Multiattribute Preferences

Abstract

The authors propose a Web-based adaptive self-explicated approach for multiattribute preference measurement (conjoint analysis) with a large number (ten or more) of attributes. The proposed approach overcomes some of the limitations of previous self-explicated approaches. The authors develop a computer-based self-explicated approach that breaks down the attribute importance question into a ranking of attributes followed by a sequence of constant-sum paired comparison questions. In the proposed approach, the questions are chosen adaptively for each respondent to maximize the information elicited from each paired comparison question. Unlike the traditional self-explicated approach, the proposed approach provides standard errors for attribute importance. In two studies involving digital cameras and laptop computers described on 12 and 14 attributes, respectively, the authors find that the ability to correctly predict validation choices of the proposed adaptive approach is substantially and significantly greater than that of adaptive conjoint analysis, the fast polyhedral method, and the traditional self-explicated approach. In addition, the adaptive self-explicated approach yields a significantly higher predictive validity than a nonadaptive fractional factorial constant-sum paired comparison design.