Replication Data for: “Loss Aversion in Social Image Concerns” by Petrishcheva, Riener, Schildberg-Hörisch, Experimental Economics

This archive contains the complete replication files for the paper "Loss Aversion in Social Image Concerns" by Vasilisa Petrishcheva, Gerhard Riener and Hannah Schildberg-Hörisch. Our replication package is organized as follows. The "zTree" folder contains the experimental code along with the images of the Raven's Progressive Matrices we used in our experiment. They are organized into Easy and Hard parts. We used zTree 4.0.8 to conduct our experiment. You can find the original instructions of the experiment (in German) and their translated English versions in Appendix E of the paper. The "data-analysis" folder contains the do-file named "Analysis.do" that replicates the results we discuss in the order of appearance in the paper. In this do-file, we comment all steps extensively. "ExperimentMain.dta" is a clean dataset. All variables are labeled. The folder "build-dataset" contains the raw data as well as two do-files necessary to generate "ExperimentMain.dta". We used Stata 15 to analyze our data. Abstract Does loss aversion apply to social image concerns? In a laboratory experiment, we first induce social image in a relevant domain, intelligence, through public ranking. In a second stage, subjects experience a change in rank and are offered scope for lying to improve their final, also publicly reported rank. Subjects who care about social image and experience a decline in rank lie more than those experiencing gains. Moreover, we document a discontinuity in lying behavior when moving from rank losses to gains. Our results are in line with loss aversion in social image concerns.