An approach for clustering consumers by their top-box and top-choice responses

Cluster analysis is often used to group consumers based on their hedonic responses to products. We give a motivating example in which conventional cluster analyses converge on a solution where consumers do not agree on which products they like. We show why this occurs. We state a goal: to group together consumers who have a shared opinion of which products are delightful and which products are not delightful, apart from consumers who have a different opinion. To meet this goal, we code consumers' hedonic responses in ways inspired by top-k box analysis, then cluster consumers using b-cluster analysis. For comparison, we cluster consumers using two conventional methods. We interpret each cluster by focusing on which product(s) the cluster accepts and whether a large proportion of cluster members are aligned in accepting these products. Solutions from b-cluster analysis based on top-k box-inspired codings met our goal better than conventional approaches, indicating that these methods deserve further study.

Practical Applications

Cluster analysis outcomes are profoundly shaped by a researcher's decisions related to response coding and clustering algorithm. This paper highlights the importance of determining the goal of the cluster analysis first, then selecting a response coding and clustering algorithm to best meet this goal. Our stated goal is one that is frequently of interest in sensory evaluation but is not well met by conventional clustering approaches. The novel approaches that we give in this paper meet the goal and are available using software that is freely available in the public domain.