Biased estimates in discrete choice models: the appropriate inclusion of psychometric data into the valuation of recycled wastewater

The introduction of measurement bias in parameter estimates into non-linear discrete choice models,as a result of using factor analysis, was identified by Train et al. (1987). They found that the inclusionof factor scores, used to represent relationships amongst like variables, into a subsequent discretechoice models introduced measurement bias as the measurement error associated with each factorscore is excluded. This is an issue for non-market valuation given the increase in popularity ofincluding psychometric data, such as primitive beliefs, attitudes and motivations, in willingness to payestimates. This study explores the relationship between willingness to pay and primitive beliefsthrough a case study eliciting Perth community values for drinking recycled wastewater. The standarddiscrete decision model, with sequential inclusion of factor scores, is compared to an equivalentdiscrete decision model, which corrects for the measurement bias by simultaneously estimating theunderlying latent variables using a measurement model. Previous research has focused on the issue ofbiased parameters. Here we also consider the implications for willingness to pay estimates.