Joining the conversation: Predictors of success on the United States Medical Licensing Examinations (USMLE)

Document Type


Publication Date



The purpose of this study was to evaluate whether models based on pre-­admission testing, including performance on the Medical College Admission Test (MCAT), performance on required courses in the medical school curriculum, or a combination of both could accurately predict performance of medical students on the United States Medical Licensing Examination (USMLE) Steps 1 and 2. Models were produced using stepwise linear regression and feed forward neural networks. Notable accuracy in predicting Step 1 and Step 2 scores were achieved from models integrating pre-admission variables with medical school coursework grades. Of interest, the coursework grades contributed far greater to these models than the pre-admission variables except the MCAT.


The copy of record is available from the publisher at http://www.nclca.org/tlar_back_issues/spring11vol16num1.pdf#page=13. Copyright © 2011 Learning Assistance Review. Reprinted with permission. All rights reserved.