Reviewed methods of estimating theta suitable for computerized adaptive testing (CAT) and discussed the differences between Fisher and Kullback-Leibler information criteria for selecting items. Examined the accuracy of different CAT algorithms using samples from the National Assessment of Educational Progress. Results show when correcting for…

Item scores that do not fit an assumed item response theory model may cause the latent trait value to be estimated inaccurately. Several person-fit statistics for detecting nonfitting score patterns for paper-and-pencil tests have been proposed. In the context of computerized adaptive tests (CAT), the use of person-fit analysis has hardly been…

Item pools supporting computer-based tests are not always completely calibrated. Occasionally, only a small subset of the items in the pool may have actual calibrations, while the remainder of the items may only have classical item statistics, (e.g., "p"-values, point-biserial correlation coefficients, or biserial correlation…

Two methods for estimating the reliability of a computerized adaptive test (CAT) without using item response theory are presented. The data consist of CAT and paper-and-pencil scores from identical or equivalent samples, and scores for all examinees on one or more covariates, using the Armed Services Vocational Aptitude Battery. (TJH)

Performed a simulation study to demonstrate the dramatic impact of capitalization on estimation errors on ability estimation in adaptive testing. Discusses four different strategies to minimize the likelihood of capitalization in computerized adaptive testing. (SLD)

Compared five item selection rules with respect to the efficiency and precision of trait (theta) estimation at the early stages of computerized adaptive testing (CAT). The Fisher interval information, Fisher information with a posterior distribution, Kullback-Leibler information, and Kullback-Leibler information with a posterior distribution…

Parameter recovery in the graded-response model was investigated using the MULTILOG computer program under default conditions. Results from 36 simulated data sets suggest that at least 500 examinees are needed to achieve adequate calibration under the graded model. Sample size had little influence on the true ability parameter's recovery. (SLD)

This paper suggests several item selection criteria for adaptive testing that are all based on the use of the true posterior. Some of the ability estimators produced by these criteria are discussed and empirically criticized. (SLD)

A simulation study was conducted to investigate the application of expected a posteriori (EAP) trait estimation in computerized adaptive tests (CAT) based on the partial credit model and compare it with maximum likelihood estimation (MLE). Results show the conditions under which EAP and MLE provide relatively accurate estimation in CAT. (SLD)

The purpose of this paper is to describe the techniques used in establishing the concordance tables between the Test of English as a Foreign Language (TOEFL), paper and pencil (P&P), and computer-based testing (CBT) sections and total reported score scales. Listening, reading, and composite structure and essay scores plus a total score are…

Used simulations of computerized adaptive tests to evaluate results yielded by four commonly used ability estimation methods: maximum likelihood estimation (MLE) and three Bayesian approaches. Results show clear distinctions between MLE and Bayesian methods. (SLD)

In a simulation study of ability and estimation of differential item functioning (DIF) in computerized adaptive tests, Rasch-based DIF statistics were highly correlated with generating DIF, but DIF statistics tended to be slightly smaller than in the three-parameter logistic model analyses. (SLD)

The average Kullback-Keibler (K-L) information index (H. Chang and Z. Ying, in press) is a newly proposed statistic in Computerized Adaptive Testing (CAT) item selection based on the global information function. The objectives of this study were to improve understanding of the K-L index with various parameters and to compare the performance of the…

A new computerized mastery test is described that builds on the Lewis and Sheehan procedure (sequential testlets) (1990), but uses fuzzy set decision theory to determine stopping rules and the Rasch model to calibrate items and estimate abilities. Differences between fuzzy set and Bayesian methods are illustrated through an example. (SLD)

Derives three reliability estimates for the Bayes modal estimate (BME) and the maximum-likelihood estimate (MLE) of theta in computerized adaptive tests (CATs). Computes the three reliability estimates and the true reliabilities of both BME and MLE for seven simulated CATs. Results show the true reliabilities for BME and MLE to be nearly identical…