Comparing risk elicitation in lotteries with visual or contextual framing aids

Eliciting risk preferences usually involves tasks that subjects may find complex, such as calculations of expected values and assessment of probabilities in multiple price lists (MPL). There is a serious concern that the decisions of the subjects may be driven by miscalculations or miscalibration of probabilities, rather than by their risk preferences. In this paper, we test whether introducing aids to the usual lottery choiceswould help to reduce the error rate and possibly change risk aversion elicitation. The experiment was run with subjectsfrom a rural area in Honduras. We compare the risk elicitation results of a multiple price list and two different treatments, one with visual aids (graphical representation of probabilities) and the other with contextual framing aids (bills to represent rewards and a distribution of ten beans between the two rewards to represent a lottery). Our results indicate that risk attitudes elicitation was affected with contextual framing aids, reducing risk aversion. For the treatment with visual aids we observe no effect.