On heterogeneous latent class models with applications to the analysis of rating scores

Discovering the preferences and the behaviour of consumers is a key challenge in mar- keting. Information about such topics can be gathered through surveys in which the respondents must assign a score to a number of items. In this article we suggest a strat- egy to analyze such data and achieve this objective: it consists in identifying groups of consumers whose response patterns are similar and characterizing them in terms of pref- erences and covariates. We use latent class models allowing for heterogeneity of both latent class and within-class probabilities across individuals. We illustrate the proposed methodology using data about the preferences of Belgian households for supermarkets.