SETTING STANDARDS FOR MULTIPLE-CHOICE ITEMS IN CLINICAL REASONING

Standard setting is a critical component in licensing decisions. In this article, it is argued that Gross's modification of the Nedelsky procedure is a valid approach to setting standards when clinical reasoning is measured using multiple-choice questions. By defining minimum competence in terms of how candidates think about problems, the Nedelsky approach more closely reflects the construct than do competing procedures. An example application taken from the Medical Council of Canada's Qualifying Examination is used to show that the procedure leads to credible standards, that the assumptions on which the process is based are reasonable, and that the standards are consistent over time.