Random coefficient models for stated preference surveys

This paper combines the use of random coefficient discrete choice models th at allow for correlated errors in the form of unobserved preference heterog eneity with an investigation of the reliability and usefulness of rankings data. For the data we examine, we find significant unobserved preference he terogeneity in the rankings, which violates the underlying assumptions of t he rank-ordered legit model. Further, directly modeling unobserved preferen ce heterogeneity results in significant improvement in the precision of the parameter estimates and in estimated Willingness to Pay (WTP). The reliabi lity and usefulness of rankings data should he re-examined with econometric models not subject to the Independence of Irrelevant Alternatives (IIA). ( C) 2000 Academic Press.