Chapter 9 312 Appendix A: Method for measuring loss aversion In the non-parametric method, estimating loss aversion requires a referencepoint (denoted ) which separates outcomes in the task into gains and losses. In this study the reference point was 0 euro, i.e., status quo, and outcomes were denoted as compared to this reference point (i.e., +€20 and €-20). Furthermore, the method requires specifying an amount The gauge outcome was set to €100. As can be seen in the demo task, the method was operationalised with 3 choice lists to elicit indifferences. Providing a full formal rationale for the method would be beyond the scope of this paper, but Table A1 gives an example. Table A1. Indifferences elicited in the non-parametric method, where denotes a gamble yielding with probability 0.5 and otherwise and the example indifferences yield a loss aversion coefficient of . General notation Goal Example Indifference 1: Mixed prospect Eliciting Indifference 2: Certainty equivalence – gains Eliciting Indifference 3: Certainty equivalence - losses Eliciting Köbberling and Wakker (2005) Loss aversion coefficient Appendix B: Additional results This Appendix contains a set of additional results, not presented in the main text, including: 1. Kaplan-Meier survival curves for the experiment 2. Mean sliders completed, stratified by conditions and incentive type 3. Regression analyses controlling for demographics 4. Regression analyses demonstrating the (lack of) effect of economic preferences on persistence and earning. B1. Survival-analysis As the figures presented closely represent survival analysis, additional KaplanMeier survival curves were estimated. A Mantel-Heanszel test indicated no significant difference between the two conditions (p=.20), which is also observable by the crossing lines in figure 6.
RkJQdWJsaXNoZXIy MTk4NDMw