Mitigating hypothetical bias in willingness to pay studies: post-estimation uncertainty and anchoring on irrelevant information.

Abstract

One possible source of hypothetical bias in willingness to pay (WTP) estimates is response uncertainty, referring to subject's uncertainty about the value of the good under assessment. It has been argued that uncertainty can be measured using the post-valuation 'certainty question' that asks: 'How certain are you about your stated WTP?' and marks the degree of certainty on a quantitative or a qualitative scale. Research has shown that the self-reported certainty evaluations can help mitigate hypothetical bias and obtain increasingly accurate WTP estimates. These study reports present a simple test of reliability of post-valuation certainty assessment and then looks at the empirical evidence for clues regarding the general usefulness of certainty adjustment in mitigating hypothetical bias in WTP studies. We find that the post-estimation uncertainty scores are malleable, i.e., significantly correlated with entirely irrelevant information. We conclude that more robust evidence could justify the routine inclusion of certainty evaluation in WTP studies although in the meantime the interpretation of certainty-adjusted WTP values should be approached cautiously.