Top of the Batch Interviews and the Match

Document pages: 24 pages

Abstract: Most doctors in the NRMP are matched to one of their most-preferredinternship programs. Since various surveys indicate similarities acrossdoctors preferences, this suggests a puzzle. How can nearly everyone get aposition in a highly-desirable program when positions in each program arescarce? We provide one possible explanation for this puzzle. We show that thepatterns observed in the NRMP data may be an artifact of the interview processthat precedes the match. Our analysis highlights the importance of interactionsoccurring outside of a matching clearinghouse for resulting outcomes, and castsdoubts on analysis of clearinghouses that take reported preferences at facevalue.