Incorporating context effects in the multidimensional scaling of 'pick any/N' choice data

This paper presents a multidimensional scaling model that is estimated on p ick any/N choice data, and accommodates a broad range of context effects. T he methodology estimates a set of parameters capturing the direction and ma gnitude of the context effects, as well as the locations of brands and cons umers' ideal points in a joint multidimensional space. Numerical simulation suggests that our approach can recover the true configuration of brands an d ideal points in the presence of context effects with greater accuracy tha n a 'baseline' model that does not incorporate any context effects. To illu strate the methodology and its implications, we provide an empirical applic ation based on choice of supermarkets, from a consumer study conducted in t he Netherlands. (C) 1999 Elsevier Science B.V. All rights reserved.