Around and beyond the cheap talk script in Choice Experiments

Using a sample of respondents interviewed face-to-face while accessing a natural park in Sardinia (Italy), we conduct a Discrete Choice Experiment to assess respondents’ willingness to pay for improved environmental quality of the site. We assess the impact of four different strategies to mitigate hypothetical bias (soft cheap talk, honesty priming, consequentiality scripts, and solemn oath) and two elicitation methods (direct and inferred evaluation methods). Results indicate that none of the strategies were significantly effective in reducing HB. Conversely, inferred valuation led to significantly lower WTP estimates. The effect was especially large for attributes of pure public nature, while attributes that include utility from indirect use are less affected by elicitation method. Overall, the study suggests that inferred valuation is more effective than other strategy in removing social desirability of respondents.