Estimating Linking Functions for Response Model Parameters

Michelle D. Barrett, Wim J. van der Linden

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Parameter linking in item response theory is generally necessary to adjust for differences between the true values for the same item and ability parameters due to the use of different identifiability restrictions in different calibrations. The research reported in this article explores a precision-weighted (PW) approach to the problem of estimating the linking functions for the common dichotomous logistic response models. Asymptotic standard errors (ASEs) of linking for the new approach are derived and compared to those of the mean/mean and mean/sigma linking methods to which it has a superficial similarity and to the Haebara and Stocking and Lord response function methods. Empirical examples from a few recent linking studies are presented. It is demonstrated that the new approach has smaller ASE than the mean/mean and mean/sigma methods and comparable ASE to the response function methods. However, when some of the item parameters have large estimation error relative to the others, all current methods appear to violate the rather obvious requirement of monotone decrease in ASE with the number of common items in the linking design while the ASE of the PW method demonstrates monotone decrease with the number of common items. The PW method also has the benefits of simple calculation and an ASE which is additive in the contribution of each item, useful for optimal linking design. We conclude that the proposed approach to estimating linking parameters holds promise and warrants further research.

Original languageEnglish
Pages (from-to)180-209
Number of pages30
JournalJournal of educational and behavioral statistics
Volume44
Issue number2
DOIs
Publication statusPublished - 1 Apr 2019
Externally publishedYes

Fingerprint

logistics
ability
Values

Keywords

  • dichotomous response models
  • estimating linking functions
  • identifiability restrictions
  • item calibration
  • linking design
  • parameter linking

Cite this

Barrett, Michelle D. ; van der Linden, Wim J. / Estimating Linking Functions for Response Model Parameters. In: Journal of educational and behavioral statistics. 2019 ; Vol. 44, No. 2. pp. 180-209.
@article{823bae35ec43468d8d90eabe4e51d82c,
title = "Estimating Linking Functions for Response Model Parameters",
abstract = "Parameter linking in item response theory is generally necessary to adjust for differences between the true values for the same item and ability parameters due to the use of different identifiability restrictions in different calibrations. The research reported in this article explores a precision-weighted (PW) approach to the problem of estimating the linking functions for the common dichotomous logistic response models. Asymptotic standard errors (ASEs) of linking for the new approach are derived and compared to those of the mean/mean and mean/sigma linking methods to which it has a superficial similarity and to the Haebara and Stocking and Lord response function methods. Empirical examples from a few recent linking studies are presented. It is demonstrated that the new approach has smaller ASE than the mean/mean and mean/sigma methods and comparable ASE to the response function methods. However, when some of the item parameters have large estimation error relative to the others, all current methods appear to violate the rather obvious requirement of monotone decrease in ASE with the number of common items in the linking design while the ASE of the PW method demonstrates monotone decrease with the number of common items. The PW method also has the benefits of simple calculation and an ASE which is additive in the contribution of each item, useful for optimal linking design. We conclude that the proposed approach to estimating linking parameters holds promise and warrants further research.",
keywords = "dichotomous response models, estimating linking functions, identifiability restrictions, item calibration, linking design, parameter linking",
author = "Barrett, {Michelle D.} and {van der Linden}, {Wim J.}",
year = "2019",
month = "4",
day = "1",
doi = "10.3102/1076998618808576",
language = "English",
volume = "44",
pages = "180--209",
journal = "Journal of educational and behavioral statistics",
issn = "1076-9986",
publisher = "SAGE Publications",
number = "2",

}

Estimating Linking Functions for Response Model Parameters. / Barrett, Michelle D.; van der Linden, Wim J.

In: Journal of educational and behavioral statistics, Vol. 44, No. 2, 01.04.2019, p. 180-209.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Estimating Linking Functions for Response Model Parameters

AU - Barrett, Michelle D.

AU - van der Linden, Wim J.

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Parameter linking in item response theory is generally necessary to adjust for differences between the true values for the same item and ability parameters due to the use of different identifiability restrictions in different calibrations. The research reported in this article explores a precision-weighted (PW) approach to the problem of estimating the linking functions for the common dichotomous logistic response models. Asymptotic standard errors (ASEs) of linking for the new approach are derived and compared to those of the mean/mean and mean/sigma linking methods to which it has a superficial similarity and to the Haebara and Stocking and Lord response function methods. Empirical examples from a few recent linking studies are presented. It is demonstrated that the new approach has smaller ASE than the mean/mean and mean/sigma methods and comparable ASE to the response function methods. However, when some of the item parameters have large estimation error relative to the others, all current methods appear to violate the rather obvious requirement of monotone decrease in ASE with the number of common items in the linking design while the ASE of the PW method demonstrates monotone decrease with the number of common items. The PW method also has the benefits of simple calculation and an ASE which is additive in the contribution of each item, useful for optimal linking design. We conclude that the proposed approach to estimating linking parameters holds promise and warrants further research.

AB - Parameter linking in item response theory is generally necessary to adjust for differences between the true values for the same item and ability parameters due to the use of different identifiability restrictions in different calibrations. The research reported in this article explores a precision-weighted (PW) approach to the problem of estimating the linking functions for the common dichotomous logistic response models. Asymptotic standard errors (ASEs) of linking for the new approach are derived and compared to those of the mean/mean and mean/sigma linking methods to which it has a superficial similarity and to the Haebara and Stocking and Lord response function methods. Empirical examples from a few recent linking studies are presented. It is demonstrated that the new approach has smaller ASE than the mean/mean and mean/sigma methods and comparable ASE to the response function methods. However, when some of the item parameters have large estimation error relative to the others, all current methods appear to violate the rather obvious requirement of monotone decrease in ASE with the number of common items in the linking design while the ASE of the PW method demonstrates monotone decrease with the number of common items. The PW method also has the benefits of simple calculation and an ASE which is additive in the contribution of each item, useful for optimal linking design. We conclude that the proposed approach to estimating linking parameters holds promise and warrants further research.

KW - dichotomous response models

KW - estimating linking functions

KW - identifiability restrictions

KW - item calibration

KW - linking design

KW - parameter linking

UR - http://www.scopus.com/inward/record.url?scp=85059330029&partnerID=8YFLogxK

U2 - 10.3102/1076998618808576

DO - 10.3102/1076998618808576

M3 - Article

VL - 44

SP - 180

EP - 209

JO - Journal of educational and behavioral statistics

JF - Journal of educational and behavioral statistics

SN - 1076-9986

IS - 2

ER -