Estimating mean willingness to pay from dichotomous choice contingent valuation studies

Methods for estimating mean willingness to pay for some environmental goal are reviewed. Logistic regression analysis of data from dichotomous choice contingent valuation studies often models the willingness-to-pay curve poorly. We develop solutions to this problem. We also show how to model responses as a function of several covariates, and how to model the case in which a proportion of respondents is not willing to pay anything. Analytic and bootstrap methods for quantifying precision are developed. We illustrate the methods by using an example in which biodiversity losses due to acid rain deposition in Scotland are valued.