Should reference alternatives in pivot design SC surveys be treated differently?

Abstract

Analysts are increasingly making use of pivot style Stated Choice (SC) data in the estimation of choice models. These datasets often contain a reference alternative whose attributes remain invariant across replications for the same respondent. This paper presents some evidence to suggest that the standard specification used for such data may not be appropriate. As such, our analysis shows differences not only in the specification of the observed part of utility between the reference alternative and hypothetical SC alternatives, but also suggests differences in the error terms.