Detecting Starting Point Bias in Dichotomous-Choice Contingent Valuation Surveys

We examine starting point bias in CV surveys with dichotomous choice payment questions and follow-ups, and double-bounded models of the WTP responses. We wish to investigate (1) the seriousness of the biases for the location and scale parameters of WTP in the presence of starting point bias; (2) whether or not these biases depend on the distribution of WTP and on the bids used; and (3) how well a commonly used diagnostic for starting point bias—a test of the null that bid set dummies entered in the right-hand side of the WTP model are jointly equal to zero—performs under various circumstances. Because starting point bias cannot be separately identified in any reliable manner from biases caused by model specification, we use simulation approaches to address this issue. Our Monte Carlo simulations suggest that the effect of ignoring starting point bias is complex and depends on the true distribution of WTP. Bid set dummies tend to soak up misspecifications in the distribution assumed by the researcher for the latent WTP, rather than capturing the presence of starting point bias. Their power in detecting starting point bias is low.