Computerized adaptive testing with item clones

Cornelis A.W. Glas, Willem J. van der Linden

Research output: Book/Report › Report › Professional

18 Downloads (Pure)

Abstract

To reduce the cost of item writing and to enhance the flexibility of item presentation, items can be generated by item-cloning techniques. An important consequence of cloning is that it may cause variability on the item parameters. Therefore, a multilevel item response model is presented in which it is assumed that the item parameters of a three-parameter logistic model describing response behavior are sampled from a multivariate normal distribution associated with a parent item. In this approach to item calibration, only distributions of item parameters are estimated. Therefore, the savings in item calibration costs for the item cloning model are potentially enormous. A marginal maximum likelihood and a Bayesian item-calibration procedure are formulated. Further, a two-stage item selection procedure for computerized adaptive testing is, presented. First, a set of items cloned from the same parent item is selected to be optimal at the ability estimate. Second, a random item from this set is administered. Simulation studies illustrate the accuracy of the item pool calibration and ability estimation procedures. An appendix describes Bayes model estimates for the item cloning model.
Original languageEnglish
Place of PublicationEnschede
PublisherUniversity of Twente, Faculty Educational Science and Technology
Number of pages23
Publication statusPublished - 2001

Publication series

NameResearch Report TO/OMD
PublisherUniversity of Twente, Faculty of Educational Science and Technology
No.01-10

Fingerprint

Adaptive Testing
Clone
Calibration
Marginal Maximum Likelihood
Multivariate Normal Distribution
Logistic Model
Selection Procedures
Costs
Bayes
Model
Estimate
Flexibility
Simulation Study

Keywords

  • item clones
  • Computerized Adaptive Testing
  • IR-103568
  • marginal maximum likelihood
  • item shells
  • METIS-203898
  • Bayesian item selection
  • multilevel item response theory

Cite this

Glas, C. A. W., & van der Linden, W. J. (2001). Computerized adaptive testing with item clones. (Research Report TO/OMD; No. 01-10). Enschede: University of Twente, Faculty Educational Science and Technology.
Glas, Cornelis A.W. ; van der Linden, Willem J. / Computerized adaptive testing with item clones. Enschede : University of Twente, Faculty Educational Science and Technology, 2001. 23 p. (Research Report TO/OMD; 01-10).
@book{abac930116ea433591c2510245fbc96c,
title = "Computerized adaptive testing with item clones",
abstract = "To reduce the cost of item writing and to enhance the flexibility of item presentation, items can be generated by item-cloning techniques. An important consequence of cloning is that it may cause variability on the item parameters. Therefore, a multilevel item response model is presented in which it is assumed that the item parameters of a three-parameter logistic model describing response behavior are sampled from a multivariate normal distribution associated with a parent item. In this approach to item calibration, only distributions of item parameters are estimated. Therefore, the savings in item calibration costs for the item cloning model are potentially enormous. A marginal maximum likelihood and a Bayesian item-calibration procedure are formulated. Further, a two-stage item selection procedure for computerized adaptive testing is, presented. First, a set of items cloned from the same parent item is selected to be optimal at the ability estimate. Second, a random item from this set is administered. Simulation studies illustrate the accuracy of the item pool calibration and ability estimation procedures. An appendix describes Bayes model estimates for the item cloning model.",
keywords = "item clones, Computerized Adaptive Testing, IR-103568, marginal maximum likelihood, item shells, METIS-203898, Bayesian item selection, multilevel item response theory",
author = "Glas, {Cornelis A.W.} and {van der Linden}, {Willem J.}",
year = "2001",
language = "English",
series = "Research Report TO/OMD",
publisher = "University of Twente, Faculty Educational Science and Technology",
number = "01-10",

}

Glas, CAW & van der Linden, WJ 2001, Computerized adaptive testing with item clones. Research Report TO/OMD, no. 01-10, University of Twente, Faculty Educational Science and Technology, Enschede.

Computerized adaptive testing with item clones. / Glas, Cornelis A.W.; van der Linden, Willem J.

Enschede : University of Twente, Faculty Educational Science and Technology, 2001. 23 p. (Research Report TO/OMD; No. 01-10).

Research output: Book/Report › Report › Professional

TY - BOOK

T1 - Computerized adaptive testing with item clones

AU - Glas, Cornelis A.W.

AU - van der Linden, Willem J.

PY - 2001

Y1 - 2001

N2 - To reduce the cost of item writing and to enhance the flexibility of item presentation, items can be generated by item-cloning techniques. An important consequence of cloning is that it may cause variability on the item parameters. Therefore, a multilevel item response model is presented in which it is assumed that the item parameters of a three-parameter logistic model describing response behavior are sampled from a multivariate normal distribution associated with a parent item. In this approach to item calibration, only distributions of item parameters are estimated. Therefore, the savings in item calibration costs for the item cloning model are potentially enormous. A marginal maximum likelihood and a Bayesian item-calibration procedure are formulated. Further, a two-stage item selection procedure for computerized adaptive testing is, presented. First, a set of items cloned from the same parent item is selected to be optimal at the ability estimate. Second, a random item from this set is administered. Simulation studies illustrate the accuracy of the item pool calibration and ability estimation procedures. An appendix describes Bayes model estimates for the item cloning model.

AB - To reduce the cost of item writing and to enhance the flexibility of item presentation, items can be generated by item-cloning techniques. An important consequence of cloning is that it may cause variability on the item parameters. Therefore, a multilevel item response model is presented in which it is assumed that the item parameters of a three-parameter logistic model describing response behavior are sampled from a multivariate normal distribution associated with a parent item. In this approach to item calibration, only distributions of item parameters are estimated. Therefore, the savings in item calibration costs for the item cloning model are potentially enormous. A marginal maximum likelihood and a Bayesian item-calibration procedure are formulated. Further, a two-stage item selection procedure for computerized adaptive testing is, presented. First, a set of items cloned from the same parent item is selected to be optimal at the ability estimate. Second, a random item from this set is administered. Simulation studies illustrate the accuracy of the item pool calibration and ability estimation procedures. An appendix describes Bayes model estimates for the item cloning model.

KW - item clones

KW - Computerized Adaptive Testing

KW - IR-103568

KW - marginal maximum likelihood

KW - item shells

KW - METIS-203898

KW - Bayesian item selection

KW - multilevel item response theory

M3 - Report

T3 - Research Report TO/OMD

BT - Computerized adaptive testing with item clones

PB - University of Twente, Faculty Educational Science and Technology

CY - Enschede

ER -

Glas CAW, van der Linden WJ. Computerized adaptive testing with item clones. Enschede: University of Twente, Faculty Educational Science and Technology, 2001. 23 p. (Research Report TO/OMD; 01-10).

Access to Document