TY - JOUR T1 - Computer adaptive testing for small scale programs and instructional systems JF - Journal of Applied Testing Technology Y1 - 2011 A1 - Rudner, L. M. A1 - Guo, F. AB -

This study investigates measurement decision theory (MDT) as an underlying model for computer adaptive testing when the goal is to classify examinees into one of a finite number of groups. The first analysis compares MDT with a popular item response theory model and finds little difference in terms of the percentage of correct classifications. The second analysis examines the number of examinees needed to calibrate MDT item parameters and finds accurate classifications even with calibration sample sizes as small as 100 examinees.

VL - 12 IS - 1 ER - TY - CHAP T1 - Implementing the Graduate Management Admission Test Computerized Adaptive Test T2 - Elements of Adaptive Testing Y1 - 2010 A1 - Rudner, L. M. JF - Elements of Adaptive Testing ER - TY - CHAP T1 - An examination of decision-theory adaptive testing procedures Y1 - 2009 A1 - Rudner, L. M. AB - This research examined three ways to adaptively select items using decision theory: a traditional decision theory sequential testing approach (expected minimum cost), information gain (modeled after Kullback-Leibler), and a maximum discrimination approach, and then compared them all against an approach using maximum IRT Fisher information. It also examined the use of Wald’s (1947) wellknown sequential probability ratio test, SPRT, as a test termination rule in this context. The minimum cost approach was notably better than the best-case possibility for IRT. Information gain, which is based on entropy and comes from information theory, was almost identical to minimum cost. The simple approach using the item that best discriminates between the two most likely classifications also fared better than IRT, but not as well as information gain or minimum cost. Through Wald’s SPRT, large percentages of examinees can be accurately classified with very few items. With only 25 sequentially selected items, for example, approximately 90% of the simulated NAEP examinees were classified with 86% accuracy. The advantages of the decision theory model are many—the model yields accurate mastery state classifications, can use a small item pool, is simple to implement, requires little pretesting, is applicable to criterion-referenced tests, can be used in diagnostic testing, can be adapted to yield classifications on multiple skills, and should be easy to explain to non-statisticians. CY - D. J. Weiss (Ed.), Proceedings of the 2009 GMAC Conference on Computerized Adaptive Testing. N1 - {PDF file, 203 KB} ER - TY - CHAP T1 - Implementing the Graduate Management Admission Test® computerized adaptive test Y1 - 2007 A1 - Rudner, L. M. CY - D. J. Weiss (Ed.), Proceedings of the 2007 GMAC Conference on Computerized Adaptive Testing. N1 - {PDF file, 135 KB} ER - TY - CONF T1 - An examination of decision-theory adaptive testing procedures T2 - Paper presented at the annual meeting of the American Educational Research Association Y1 - 2002 A1 - Rudner, L. M. JF - Paper presented at the annual meeting of the American Educational Research Association CY - New Orleans, LA N1 - {PDF file, 46 KB} ER - TY - ABST T1 - Item banking Y1 - 1998 A1 - Rudner, L. M. AB - Discusses the advantages and disadvantages of using item banks while providing useful information to those who are considering implementing an item banking project in their school district. The primary advantage of item banking is in test development. Also describes start-up activities in implementing item banking. (SLD) JF - Practical Assessment, Research and Evaluation VL - 6 N1 - Using Smart Source Parsing ER - TY - JOUR T1 - Computer testing: Pragmatic issues and research needs JF - Educational Measurement: Issues and Practice Y1 - 1990 A1 - Rudner, L. M. VL - 9 (2) N1 - Sum 1990. ER - TY - ABST T1 - Computerized adaptive tests Y1 - 1989 A1 - Grist, S. A1 - Rudner, L. M. A1 - Wise CY - ERIC Clearinghouse on Tests, Measurement, and Evaluation, no. 107 ER -