Showing 1 - 2 of 2 Research Library Publications

A Comparison of Strategies for Smoothing Parameter Selection for Mixed-Format Tests Under the Random Groups Design

Posted: | C. Liu, M. J. Kolen

Journal of Educational Measurement: Volume 55, Issue 4, Pages 564-581

 

Smoothing techniques are designed to improve the accuracy of equating functions. The main purpose of this study is to compare seven model selection strategies for choosing the smoothing parameter (C) for polynomial loglinear presmoothing and one procedure for model selection in cubic spline postsmoothing for mixed‐format pseudo tests under the random groups design.

Category:Assessment-Oriented Research, Reliability/Validity, Scoring

Enhancement of Assessment of Physician-Patient Communication Skills in the United States Medical Licensing Examination (USMLE)

Posted: | Ruth B. Hoppe, Ann M. King, Kathleen M. Mazor, Gail E. Furman, Penelope Wick-Garcia, Heather Corcoran–Ponisciak, Peter J. Katsufrakis

Academic Medicine: Volume 88 - Issue 11 - p 1670-1675

 

From 2007 through 2012, the NBME team reviewed literature in physician–patient communication, examined performance characteristics of the Step 2 CS exam, observed case development and quality assurance processes, interviewed SPs and their trainers, and reviewed video recordings of examinee–SP interactions.  The authors describe perspectives gained by their team from the review process and outline the resulting enhancements to the Step 2 CS exam, some of which were rolled out in June 2012.

Category:Product-Oriented Research, USMLE