Evaluating mitigation and calibration techniques for hypothetical bias in choice experiments

Due to the persistence of hypothetical bias in contingent valuation, some researchers have turned to choice experiments to value the multi-dimensions of non-marketed goods that may not be as prone to the bias. The recent empirical studies of choice experiments have found it may also be prone to hypothetical bias. This paper investigates what role mitigation and calibration techniques such as cheap-talk scripts and follow-up certainty questions can play in mitigating or calibrating for hypothetical bias in choice experiments. While these techniques are not as straightforward in choice experiments, due to their multiple-choice nature, the findings indicate that hypothetical bias may not be present when using a local quasi-public good in the valuation exercise. In addition, cheap-talk and follow-up certainty are found to reduce marginal willingness to pay estimates to be less than actual willingness to pay estimates.