Testing Willingness-to-Pay Models of Discrete Choice Contingent Valuation Survey Data

This paper considers models of willingness to pay (WTP) based on dichotomous choice contingent valuation survey data and examines the effect of the survey design on tests of goodness of fit. Simulations show that eliciting single binary responses result in tests that have low power in detecting departures from the maintained distribution of WTP, unless the sample size is truly large. The power of the tests improves if a follow-up question and a moderate number of sets of bid values are used, and if the bids are not placed on the extreme tails of the WTP distribution.