Valuing Consumer Preferences with the CUB Model: A Case Study of Fair Trade Coffee

D'Elia and Piccolo (2005) have recently proposed a mixture distribution, named CUB, for ordinal data. The use ofsuch a mixture distribution for modelling ratings is justified by the following consideration: the judgment that asubject expresses is the result of two components, uncertainty and selectiveness. The possibility of relating theparameters of CUB models to covariates makes the formulation interesting for practical applicationsIn this case study, a sample of 224 fair‐trade coffee consumers were interviewed at stores. With this data‐set, CUBmodel split consumers, according to their preferences, in two different segments: one showing high price elasticity,and one with a low price elasticity. As regards the potential of the CUB model, it showed a considerable integrationcapacity with stochastic utility models, namely latent class models. Indeed, by using the segmentation factorsemerging from the CUB as covariates of segmentation in a latent class model and setting the number of classesequal to those emerging from the CUB, it was possible to estimate a model which not only validated the findings ofthe CUB but also allowed estimation of the WTP for the fair trade characteristic in the different groups.