Evaluating Medical Students' Trajectory on Progress Tests Using Markov Chain Model

• In Event: Educational Measurement Issues in the Health Care Professions

Abstract

Though progress tests have been used for several decades in various medical education settings, most programs suffer from methodological limitations to assess progress tests results. In this study, we propose to use first-order Markov chain model to find the students’ transition probabilities from inferior state to a superior state. Also, Markov chain model is used to predict long-run students’ performance via the estimation of stationary distributions. We employed Markov chain to analyze 169 medical school students with 10 iterations of progress tests and United States Medical Licensing Examination (USMLE) Step 1 results. It is found that the estimated students’ steady-state based progress tests transition probabilities can significantly predict Step 1 test results.

Authors

• Ling Wang, Michigan State University

• Heather Laird-Fick, Michigan State University

• Chi Chang, Michigan State University

• Carol Parker, Michigan State University

• Robert Paul Malinowski, Michigan State University

• David J. Solomon, Michigan State University