D RA FT : D o no t c ite . A Case for the Use of Perception Measures

This paper utilizes corruption data from a laboratory experiment in order to examine the effectiveness of perceptions as valid empirical measures. Somewhat ironically, we find that perception measures are most useful in exactly those situations where they disagree with their objective counterparts. Our results indicate that perception-based measures may actually be preferable to the use of fact-based measures even when facts are readily available and can be measured perfectly. Our results also indicate that self-reported data deviates away from actual facts in systematic ways similar to those observed for the case of perceptions. ∗We would like to thank the Sutton Fund and the University of Arkansas for financial support. We also thank Kristine Brown, Cary Deck, Ori Heffetz, Youenn Loheac, and seminar participants at the University of Arkansas for valuable comments. Xirong Chen provided excellent research assistance. All errors are ours. Contact Info: Department of Economics, Sam M. Walton College of Business, Business Building 402, Fayetteville, AR 72701-1201, USA. Email: [email protected], [email protected].