Journal article
Properties of Ability Estimation Methods in Computerized Adaptive Testing
Journal of educational measurement, Vol.35(2), pp.109-135
06/1998
DOI: 10.1111/j.1745-3984.1998.tb00530.x
Abstract
Simulations of computerized adaptive tests (CATs) were used to evaluate results yielded by four commonly used ability estimation methods: maximum likelihood estimation (MLE) and three Bayesian approaches-Owen's method, expected a posteriori (EAP), and maximum a posteriori. In line with the theoretical nature of the ability estimates and previous empirical research, the results showed clear distinctions between MLE and the Bayesian methods, with MLE yielding lower bias, higher standard errors, higher root mean square errors, lower fidelity, and lower administrative efficiency. Standard errors for MLE based on test information underestimated actual standard errors, whereas standard errors for the Bayesian methods based on posterior distribution standard deviations accurately estimated actual standard errors. Among the Bayesian methods, Owen's provided the worst overall results, and EAP provided the best. Using a variable starting rule in which examinees were initially classified into three broad ability groups greatly reduced the bias for the Bayesian methods, but had little effect on the results for MLE. On the basis of these results, guidelines are offered for selecting appropriate CAT ability estimation methods in different decision contexts. This article is based on a doctoral dissertation (Wang, 1995) that was directed by Walter P. Vispoel. We thank Roberto de la Torre for his assistance in developing software for the computer simulation analyses and Rebecca Zwick and the anonymous reviewers for their helpful comments on the original manuscripts. Correspondence concerning this article should be addressed to the first author.
Details
- Title: Subtitle
- Properties of Ability Estimation Methods in Computerized Adaptive Testing
- Creators
- Tianyou Wang - ACTWalter P. Vispoel - University of lowa
- Resource Type
- Journal article
- Publication Details
- Journal of educational measurement, Vol.35(2), pp.109-135
- Publisher
- Blackwell Publishing Ltd
- DOI
- 10.1111/j.1745-3984.1998.tb00530.x
- ISSN
- 0022-0655
- eISSN
- 1745-3984
- Number of pages
- 27
- Language
- English
- Date published
- 06/1998
- Academic Unit
- Center for Advanced Studies in Measurement and Assessment; Psychological and Quantitative Foundations
- Record Identifier
- 9984371283602771
Metrics
4 Record Views