Testlet-Based Adaptive Mastery Testing

Abstract

In mastery testing, the problem is to decide whether a test taker must be classified as a master or a nonmaster. The decision is based on the test taker’s observed test score. Well-known examples of mastery testing include testing for pass-fail decisions, licensure, and certification. A mastery test can have both fixed-length and variable-length forms. In a fixed-length mastery test, the performance on a fixed number of items is used for deciding on mastery or nonmastery. Over the last few decades, the fixed-length mastery problem has been studied extensively by many researchers (e.g., De Gruijter & Hambleton, 1984; van der Linden, 1990). Most of these authors derived, analytically or numerically, optimal rules by applying (empirical) Bayesian decision theory (e.g., DeGroot, 1970; Lehmann, 1986) to this problem. In the variable-length form, in addition to the action of declaring mastery or nonmastery, the action of continuing to administer items is available also (e.g., Kingsbury and Weiss, 1983; Lewis & Sheehan, 1990; Sheehan and Lewis, 1992; Spray & Reckase, 1996).