Measuring consequentiality: A “Knife-Edge” versus Continuous Approach

A survey is consequential to a respondent if they believe their answer could influence the policy being addressed in the survey and if they will have to pay for the policy if implemented. We show that separating out respondents who find the survey inconsequential, even by very simple metrics such as a single question, goes a long way if the goal is to improve willingness to pay estimates. Using various follow up questions, we develop multiple thresholds to classify respondents into groups based on whether or not their responses satisfy the consequentiality criteria. Independent of the threshold, we find that respondents in the inconsequential group have a willingness to pay that is insignificantly different from zero. For those in the consequential group, marginal willingness to pay does not significantly depend on the threshold. These results lend additional support to the ‘knife-edge’ hypothesis. To provide additional insights we explore consequentiality using a hybrid choice model and find that the likelihood of payment consequentiality increases with income while respondents who find the survey policy consequential are more likely to be in favor of the policy. Key Words: consequentiality, stormwater management, stated preferences, hybrid choice models, generalized structural equation method