Affiliation:
1. Human Resources Research Organization (HumRRO)
Abstract
AbstractComputer adaptive testing (CAT) is an increasingly common mode of test administration offering improved test security, better measurement precision, and the potential for shorter testing experiences. This article presents a new item selection algorithm based on a generalized objective function to support multiple types of testing conditions and principled assessment design. The generalized nature of the algorithm permits a wide array of test requirements allowing experts to define what to measure and how to measure it and the algorithm is simply a means to an end to support better construct representation. This work also emphasizes the computational algorithm and its ability to scale to support faster computing and better cost‐containment in real‐world applications than other CAT algorithms. We make a significant effort to consolidate all information needed to build and scale the algorithm so that expert psychometricians and software developers can use this document as a self‐contained resource and specification document to build and deploy an operational CAT platform.
Reference65 articles.
1. Babcock B. &Weiss D. J.(2009).Termination criteria in computerized adaptive tests: Variable‐length CATs are not biased. InD. J.Weiss(Ed.) Proceedings of the 2009 GMAC Conference on Computerized Adaptive Testing. Retrieved fromwww.psych.umn.edu/psylabs/CATCentral/.
2. Controlling the speededness of assembled test forms: A generalization to the three‐parameter lognormal response time model;Becker B.;Journal of Educational Measurement,2023
3. Distributions of the Kullback-Leibler divergence with applications
4. A Monte Carlo Approach for Adaptive Testing With Content Constraints
5. Adaptive EAP Estimation of Ability in a Microcomputer Environment