Solutions to many market research, product development and quality assurance problems require the use of various types of product maps. Sometimes it is of interest to know whether products differ without regard to liking and preference. For instance, product maps are used to guide quality assurance, cost reduction formulations, and ‘me-too’ product development. It is natural to think of products as points and when two products are less similar than two other products that they are further apart in some product space. In this report we will show how this idea has significant limitations, and that a more meaningful interpretation of similarity data can be made when products are treated as distributions.
This technical report appears as:
Ennis, D. M. (2001). Probabilistic Multidimensional Scaling. IFPress, 4(3) 2-3.
Colleagues can download the entire technical report here:
Probabilistic Multidimensional Scaling
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.