Random effects choice models: seeking latent predisposition segments in the context of retail store format selection

The aim of this paper is to gain insight into the potential use of random effects choice models as analytical instruments. These instruments then may be used to identify the latent predisposition and benefit segments of a market in the context of store selection, specifically, in the context of the selection of store format. First, the theoretical bases, formulations, and practical aspects that make up the modelling technique in question are reviewed and revised. Second, an adaptation to detect latent segments with predispositions to shop at a store format, isolated from spatial coverage factors, is proposed. Third, the proposal is empirically tested in the case of the hypermarket. The results confirm the explanatory possibilities of this modelling approach and show that various segments with different levels of response to the hypermarket exist. They can be characterised ad hoc by means of demographic and sociological variables to facilitate their identification, access, and evaluation, with a view to possible marketing strategies.