Abstract
The author proposes an alternative approach for measuring consumer preferences by using limited information. The approach differs from the current popular approaches in that once the nonpreferred choice alternatives are identified, no further information on them is obtained. Behavioral rationale as well as theoretical support underlying the proposed approach is provided. The author reports an empirical investigation and two Monté Carlo studies comparing the performance of the proposed approach with that of metric conjoint analysis, which is a corresponding approach based on “full” information, and an OLS approach that uses information on the preferred alternatives only. Finally, the implications, advantages, and limitations of the proposed approach are discussed.
Get full access to this article
View all access options for this article.
