ABSTRACT
A standard determines whether a given score or performance is good enough for a particular purpose (Norcini & Guille, 2002). The term “standard setting” refers to a process used to create boundaries between categories such as pass | fail, or honors | proficient | needs remediation. Standard setting is “central to the task of giving meaning to test results and thus lies at the heart of validity argument” (Dylan, 1996). Establishing credible, defensible, and acceptable passing or cut-off scores for examinations in health professions education can be challenging (Friedman, 2000; Norcini & Shea, 1997; Norcini & Guille, 2002). There is a large literature of standard setting, much of which is devoted to empirical passing score studies and comparisons of various standard-setting methods which are appropriate for selected-response tests or performance tests used in K-12 educational settings (Cizek, 2001, 2006; Cizek, Bunch, & Koons, 2004; Livingston, 1982; Norcini, 2003). This chapter will discuss key issues and decisions regarding standard setting, identify ways to assess the quality and consequences of resulting standards, and address special situations such as combining standards across subtests, setting standards for performance tests, and multiple-category cut scores. At the end of the chapter we provide detailed instructions for conducting six standard setting methods commonly used in health professions settings: Angoff, Ebel, Hofstee, Borderline Group, Contrasting Groups, and Body of Work.
