Surprising robustness of the self-explicated approach to customer preference structure measurement

The authors introduce customized conjoint analysis, which combines self-explicated preference structure measurement with full-profile conjoint analysis. The more important attributes for each respondent are identified first using the self-explicated approach. Full-profile conjoint analysis customized to-the respondent's most important attributes then is administered. The conjoint utility function on the limited set of attributes then is combined with the self-explicated utility function on the full set of attributes. Surprisingly, the authors find that the self-explicated approach by itself yields a slightly (but not statistically significantly) higher predictive validity than does the combined approach.