The effect of elicitation methods on ambiguity aversion: an experimental investigation

In this paper we elicit preferences for the classical three-color Ellsberg Paradax employing three different methods, choices, minimal selling prices and maximal buying prices. The resulting data reveal a high frequency of preference reversals which have not been analyzed before in choice under uncertainty. Moreover, we analyze the effect of elicitation methods on the degree of ambiguity aversion. While there is no apparent difference in the attitude towards ambiguity between selling and buying prices we observe a rather distinct pattern of behavior for choices: Compared to choices, eliciting preferences by pricing tasks decreases the number of subjects being ambiguity averse in both tasks and increases the number of subjects being ambiguity neutral or prone. We argue that this difference between pricing and choice supports the hypothesis of comparative ignorance.