Dichotomous Versus Trichotomous Choice - The Influence of the Elicitation Question Format on Contingent Valuation Results

This paper analyzes the possible advantages of the trichotomous choice elicitation question format in contingent valuation studies over the traditional dichotomous choice format. In the past, contingent valuation studies showed that mean stated willingness to pay (WTP) for a public project was systematically greater when using the dichotomous choice format instead of a payment card or the open-ended question format. Therefore, the dichotomous choice format is often alleged of overestimating the benefits from public projects. It might induce respondents to agree to a higher than their true WTP for a public project since it does not offer a suitable answering option in cases where the project is appreciated by respondents but the suggested payment is higher than their true WTP. As a solution, the use of the trichotomous choice format has been suggested, where respondents can accept the suggested project but reject the payment amount. We develop criteria to compare the performance of the two elicitation formats. We then test the performance of trichotomous versus dichotomous choice in two contingent valuation surveys in China and find that trichotomous choice, indeed, leads to lower mean WTP estimates than dichotomous choice and to regression models with a larger number of significant determinants. The findings of this study are particularly important in contexts where culture demands respondents to show support for public projects in personal interview settings