Unexpected yes- and no-answering behaviour in the discrete choice approach to elicit willingness to pay: a methodological comparison with payment cards.

When measuring willingness to pay (WTP) by contingent valuation surveys, several elicitation methods can be applied. The most common methods are the discrete choice (DC) approach and payment cards. The purpose of this study was to analyse the convergent validity of both approaches in order to investigate different kinds of answering behaviour in these approaches. Unexpected deviations of the answers in the DC approach from the answers provided on the payment cards were analysed, i.e. unexpected yes- (no-)answering was given when respondents stated to the DC question that they are (not) willing to pay a monetary amount while they stated the opposite on the payment cards. Furthermore, we analysed the feasibility of these two elicitation methods. Each of a group of 92 diabetic patients was asked to state their WTP for reductions of the risk of several diabetic complications by both elicitation methods in two surveys. Both elicitation methods were feasible. Compared with the WTP stated on the payment cards, we found unexpected yes- as well as no-answering behaviour in the DC approach which partly balanced each other. At low bids, there was a tendency that unexpected no-outweighed unexpected yes-answering behaviour. At high bids, unexpected yes- outweighed unexpected no-answering behaviour. Overall, unexpected yes-answering behaviour was predominating. Several explanations for these phenomena remain to be investigated.