Testing for Consistency in Contingent Ranking Experiments

This paper develops a set of simple tests for the logical consistency of responses to contingent ranking surveys based on the concepts of dominance, rank consistency, and transitivity of rank order. Nearly half of respondents occasionally fail to provide fully coherent responses, although only a small minority are systematically inconsistent. The presence of inconsistent rankings has a statistically significant effect on coefficient estimates, but not so on estimates of willingness to pay. Notwithstanding, the exclusion of inconsistent results does not do away with the problem of unstable coefficients across ranks which has been noted in the literature