A new technique for dealing with complex stimuli in Conjoint Analysis

This paper deals with the problem of a large number of multi-attributes stimuli in Conjoint Analysis. The aim of this paper is to critically discuss some specific aspects of the bridging technique originally proposed by Bretton and Clark and to propose an innovative approach based on the same philosophy but on a different estimation method. The new technique is based on several estimation steps. It is able to make the most of the orthogonality properties related to the experimental designs. Furthermore, a validation procedure for the bridging results has been proposed. This procedure allows answering to the general question on the reliability of performing a bridging technique.