@article {94, title = {Item exposure constraints for testlets in the verbal reasoning section of the MCAT}, journal = {Applied Psychological Measurement}, volume = {27}, number = {5}, year = {2003}, pages = {335-356}, abstract = {The current study examined item exposure control procedures for testlet scored reading passages in the Verbal Reasoning section of the Medical College Admission Test with four computerized adaptive testing (CAT) systems using the partial credit model. The first system used a traditional CAT using maximum information item selection. The second used random item selection to provide a baseline for optimal exposure rates. The third used a variation of Lunz and Stahl{\textquoteright}s randomization procedure. The fourth used Luecht and Nungester{\textquoteright}s computerized adaptive sequential testing (CAST) system. A series of simulated fixed-length CATs was run to determine the optimal item length selection procedure. Results indicated that both the randomization procedure and CAST performed well in terms of exposure control and measurement precision, with the CAST system providing the best overall solution when all variables were taken into consideration. (PsycINFO Database Record (c) 2005 APA ) (journal abstract)}, keywords = {Adaptive Testing, Computer Assisted Testing, Entrance Examinations, Item Response Theory, Random Sampling, Reasoning, Verbal Ability computerized adaptive testing}, author = {Davis, L. L. and Dodd, B. G.} } @article {308, title = {A comparison of item selection techniques and exposure control mechanisms in CATs using the generalized partial credit model}, journal = {Applied Psychological Measurement}, volume = {26}, number = {2}, year = {2002}, pages = {147-163}, abstract = {The use of more performance items in large-scale testing has led to an increase in the research investigating the use of polytomously scored items in computer adaptive testing (CAT). Because this research has to be complemented with information pertaining to exposure control, the present research investigated the impact of using five different exposure control algorithms in two sized item pools calibrated using the generalized partial credit model. The results of the simulation study indicated that the a-stratified design, in comparison to a no-exposure control condition, could be used to reduce item exposure and overlap, increase pool utilization, and only minorly degrade measurement precision. Use of the more restrictive exposure control algorithms, such as the Sympson-Hetter and conditional Sympson-Hetter, controlled exposure to a greater extent but at the cost of measurement precision. Because convergence of the exposure control parameters was problematic for some of the more restrictive exposure control algorithms, use of the more simplistic exposure control mechanisms, particularly when the test length to item pool size ratio is large, is recommended. (PsycINFO Database Record (c) 2005 APA ) (journal abstract)}, keywords = {(Statistical), Adaptive Testing, Algorithms computerized adaptive testing, Computer Assisted Testing, Item Analysis, Item Response Theory, Mathematical Modeling}, author = {Pastor, D. A. and Dodd, B. G. and Chang, Hua-Hua} } @article {67, title = {The effect of population distribution and method of theta estimation on computerized adaptive testing (CAT) using the rating scale model}, journal = {Educational \& Psychological Measurement}, volume = {57}, number = {3}, year = {1997}, note = {Sage Publications, US}, pages = {422-439}, abstract = {Investigated the effect of population distribution on maximum likelihood estimation (MLE) and expected a posteriori estimation (EAP) in a simulation study of computerized adaptive testing (CAT) based on D. Andrich{\textquoteright}s (1978) rating scale model. Comparisons were made among MLE and EAP with a normal prior distribution and EAP with a uniform prior distribution within 2 data sets: one generated using a normal trait distribution and the other using a negatively skewed trait distribution. Descriptive statistics, correlations, scattergrams, and accuracy indices were used to compare the different methods of trait estimation. The EAP estimation with a normal prior or uniform prior yielded results similar to those obtained with MLE, even though the prior did not match the underlying trait distribution. An additional simulation study based on real data suggested that more work is needed to determine the optimal number of quadrature points for EAP in CAT based on the rating scale model. The choice between MLE and EAP for particular measurement situations is discussed. (PsycINFO Database Record (c) 2003 APA, all rights reserved).}, keywords = {computerized adaptive testing}, author = {Chen, S-K. and Hou, L. Y. and Fitzpatrick, S. J. and Dodd, B. G.} } @article {98, title = {A simulation and comparison of flexilevel and Bayesian computerized adaptive testing}, journal = {Journal of Educational Measurement}, volume = {27}, number = {3}, year = {1990}, pages = {227-239}, abstract = {Computerized adaptive testing (CAT) is a testing procedure that adapts an examination to an examinee{\textquoteright}s ability by administering only items of appropriate difficulty for the examinee. In this study, the authors compared Lord{\textquoteright}s flexilevel testing procedure (flexilevel CAT) with an item response theory-based CAT using Bayesian estimation of ability (Bayesian CAT). Three flexilevel CATs, which differed in test length (36, 18, and 11 items), and three Bayesian CATs were simulated; the Bayesian CATs differed from one another in the standard error of estimate (SEE) used for terminating the test (0.25, 0.10, and 0.05). Results showed that the flexilevel 36- and 18-item CATs produced ability estimates that may be considered as accurate as those of the Bayesian CAT with SEE = 0.10 and comparable to the Bayesian CAT with SEE = 0.05. The authors discuss the implications for classroom testing and for item response theory-based CAT.}, keywords = {computerized adaptive testing}, author = {De Ayala, R. J., and Dodd, B. G. and Koch, W. R.} }