Strategic Sophistication in Strategy-Proof Matching Environments: Evidence from Large-Scale Surveys of Medical StudentsAdd to Calendar
12:00 pm – 1:00 pm
The University of Pennsylvania
A large literature in economics has developed and studied strategy-proof matching mechanisms. In these settings, reported preferences are used as an input to the matching algorithm, and participants are incentivized to report their preferences truthfully. However, recent work (e.g., Rees-Jones 2016; Hassidim, Romm, and Shorrer 2016; Shorrer and Sovagoy 2017) suggests that at least some students try to “game the system” through preference misrepresentation. This project aims to better understand the causes and correlates of this strategically unsophisticated behavior. We administer an online experiment to over 1700 medical students immediately after their participation in the 2017 medical residency match. This experiment elicits an incentivized measure of subjects’ beliefs about optimal behavior in the mechanism used for residency matching, and allows us to document widespread misunderstanding of incentives. We explore the role of cognitive ability, strategic positioning, overconfidence, ego utility, and the presence of trusted advice as explanations for this phenomenon.