Main Data
Author: Wim J. van der Linden, Cees A. W. Glas
Editor: Wim J. van der Linden, Cees A. W. Glas
Title: Elements of Adaptive Testing Theory and Practice
Publisher: Springer-Verlag
ISBN/ISSN: 9780387854618
Edition: 1
Price: CHF 110.90
Publication date: 01/01/2010
Category: Sozialwissenschaften
Language: English
Technical Data
Pages: 438
Kopierschutz: DRM
Geräte: PC/MAC/eReader/Tablet
Formate: PDF
Table of contents
The arrival of the computer in educational and psychological testing has led to the current popularity of adaptive testing---a testing format in which the computer uses statistical information about the test items to automatically adapt their selection to a real-time update of the test taker's ability estimate. This book covers such key features of adaptive testing as item selection and ability estimation, adaptive testing with multidimensional abilities, sequencing adaptive test batteries, multistage adaptive testing, item-pool design and maintenance, estimation of item and item-family parameters, item and person fit, as well as adaptive mastery and classification testing. It also shows how these features are used in the daily operations of several large-scale adaptive testing programs.
Table of contents
Part I Item Selection and Ability Estimation14
1 Item Selection and Ability Estimation in Adaptive Testing15
1.1 Introduction15
1.2 Classical Procedures17
1.2.1 Notation and Some Statistical Concepts17
1.2.2 Ability Estimators19
1.2.3 Choice of Estimator20
1.2.4 Classical Item-Selection Criteria23
1.3 Modern Procedures24
1.3.1 Maximum Global-Information Criterion25
1.3.2 Likelihood-Weighted Information Criterion27
1.3.3 Fully Bayesian Criteria28
1.3.4 Bayesian Criteria with Collateral Information30
1.3.5 Bayesian Criteria with Random Item Parameters33
1.3.6 Miscellaneous Criteria35
1.3.7 Evaluation of Item-Selection Criteria and Ability Estimators36
1.4 Concluding Remarks39
2 Constrained Adaptive Testing with Shadow Tests43
2.1 Introduction43
2.2 Review of Existing Methods for Constrained CAT45
2.2.1 Item-Pool Partitioning45
2.2.2 Weighted-Deviation Method45
2.2.3 Maximum Priority Index Method45
2.2.4 Testlet-Based Adaptive Testing46
2.2.5 Multistage Testing46
2.2.6 Evaluation of Existing Approaches47
2.3 Constrained CAT with Shadow Tests48
2.4 Technical Implementation49
2.4.1 Basic Notation and Definitions50
2.4.2 IP Model for Shadow Test51
2.4.3 Numerical Aspects53
2.5 Four Applications to Adaptive Testing Problems54
2.5.1 CAT with Large Numbers of Nonstatistical Constraints55
2.5.2 CAT with Response-Time Constraints55
2.5.3 CAT with Item-Exposure Control59
2.5.4 CAT with Equated Number-Correct Scores62
2.6 Concluding Remarks65
3 Principles of Multidimensional Adaptive Testing68
3.1 Introduction68
3.2 Literature Review69
3.3 Multidimensional Item Selection and Scoring70
3.3.1 Prior Density71
3.3.2 Likelihood Function72
3.3.3 Posterior Density74
3.3.4 Item Selection76
3.3.5 Posterior Inference79
3.4 Example80
3.4.1 Initialization80
3.4.2 Item Selection81
3.4.3 Provisional Ability Estimation82
3.4.4 Item Selection and Scoring Cycle82
3.5 Discussion84
3.6 Appendix: Computational Formulas84
4 Multidimensional Adaptive Testing with KullbackLeibler Information Item Selection87
4.1 Multidimensional IRT model88
4.2 Bayesian Estimation of bold0mu mumu *89
4.3 KullbackLeibler Information91
4.3.1 Mutual Information94
4.4 Item Selection Using KL Information94
4.4.1 Posterior Expected KullbackLeibler Information95
4.4.2 KL Distance between Subsequent Posteriors97
4.4.3 Mutual Information98
4.5 Relationship between Selection Criteria99