Modelling preference heterogeneity in stated choice data: an analysis for public goods generated by agriculture

Stated choice models based on the random utility framework are becoming increasingly popular inthe applied economics literature. The need to account for respondents’ preference heterogeneity insuch models has motivated researchers in agricultural, environmental, health and transporteconomics to apply random parameter logit and latent class models. In most of the publishedliterature these models incorporate heterogeneity in preferences through the systematic componentof utility. An alternative approach is to investigate heterogeneity through the random component ofutility, and covariance heterogeneity models are one means of doing this. In this paper we comparethese alternative ways of incorporating preference heterogeneity in stated choice models andevaluate how the selection of approach affects welfare estimates in a given empirical application.We find that a Latent Class approach fits our data best but all the models perform well in terms ofout-of-sample predictions. Finally, we discuss what criteria a researcher can use to decide whichapproach is most appropriate for a given data set.