Compositional perceptual mapping using chi‐squared trees analysis and correspondence analysis

Abstract

The construction of compositional perceptual maps is an important aspect of the management of brands. The use of correspondence analysis to construct such maps is not new, nor is the subsequent use of conventional hierarchical and non‐hierarchical cluster analysis to segment the brands. However, the application of chi‐squared trees analysis can be used to cluster both brands and attributes using the same principle of decomposing the χ2 statistic of the matrix of frequency of responses as in correspondence analysis. Used in conjunction, these two techniques allow the construction of an enriched compositional perceptual map.