Recent years have seen the establishment of so-called discrete choice modeling methods. Included among these is the choice based conjoint. The major advantage of this method is that it presents a realistic decision-making situation. The respondent is requested to choose one product from a set of products. However, he also has the option of not choosing any of the products presented. In other words, he can explicitly reject them all. This results in a realistic approximation of the situation faced by the consumer at the point of sale.
One drawback of this procedure is the relatively low information density achieved for each individual participant. Individual utility values can only be determined at the end of the study based on all of the obtained survey data. Using the so-called Hierarchical Bayes Estimation Method, the missing information is supplied at the individual level based on the structures in the group as a whole. However, this requires that a certain minimum level of homogeneity be present within the group. If this minimum level is not present, subgroups exhibiting similar decision structures can be created in a parallel estimation and segmentation process (Latent Classes Method).
As with the adaptive conjoint, the choice based conjoint is also best carried out with the aid of a computer. This method lends itself especially well to the use of graphical elements. Additional advantages of this method include its ability to reveal the extent to which attributes have a mutual influence on one another (measurement of interactions). Undesired combinations of attribute levels can be excluded when compiling the stimulus examples.
