ABSTRACT
Examinations used to make pass/fail or other classification decisions have historically been referred to as mastery tests. Mastery tests have also been linked with the concept of criterion-referenced testing (e.g., Glaser, 1963; Hambleton, 1984; Hambleton & Novick, 1973; Popham, 1984) for two reasons. First, mastery tests are usually assumed to be content referenced—that is, the criterion for mastery is defined in terms of requisite levels of knowledge and skills measured by the test (Ebel, 1962; Popham, 1984). Second, a fixed point on the score scale of choice—the cut score—defines mastery. In principle, everyone in the examinee population could pass or fail the test, depending on their score relative to the cut score. 1
