In our Spring 1998 newsletter, we discussed a method for the analysis of liking data. The goal of this method was to discover attributes that drive liking. The method, called probabilistic unfolding, displays products and ideals as distributions. Preference data can also be unfolded to provide insights into ideal product characteristics, and can be used to improve products. In fact, preference data are more valuable than liking data to determine the basis for consumer hedonics using unfolding models. Unfortunately, preference experiments are often more expensive to conduct than liking experiments. In this report, an overview of the benefits of multivariate preference unfolding is given using new techniques that specify products and ideals as distributions rather than discrete points. The value of this approach in modeling preference data will be illustrated.
This technical report appears as:
Ennis, D. M. (1999). Multivariate Preference Mapping. IFPress, 2(2) 2-3.
Download the entire technical report here:
Not a Colleague? Click here to join for free!
This technical report also appears in our book, Tools and Applications of Sensory and Consumer Science.