The Information Product Methods: A Unified Approach to Dual-Purpose Computerized Adaptive Testing

Abstract

This article gives a brief summary of major approaches in dual-purpose computerized adaptive testing (CAT) in which the test is tailored interactively to both an examinee’s overall ability level, $\mathrm{\theta }$ , and attribute mastery level, $\mathbf{\alpha }$ . It also proposes an information product approach whose connections to the current methods are revealed. An updated comprehensive empirical study demonstrated that the information product approach not only can offer a unified framework to connect all other approaches but also can mitigate the weighting issue in the dual-information approach.