A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning

Mei Ling Ong; Laura Lu; Sunbok Lee; Allan Cohen

doi:10.1007/978-3-319-07503-7_22

A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning

Mei Ling Ong, Laura Lu, Sunbok Lee, Allan Cohen

Department of Education

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

This chapter describes the property of estimates of points in a multidimensional space that is labeled by some as paradoxical, shows when this property of the estimates is present, and also shows that the paradoxical result is not flaw in estimation because estimates improve with additional information even when the paradox occurs. The paradox is that when a correct response to a test item is added to the string of responses for an examinee to previous items, at least one of the coordinates of the new estimated ?-point decreases compared to the estimate based on the initial string of responses. The information presented in the chapter shows that this can occur whenever the likelihood function for the estimates has a particular form. This form is present in many cases when the item responses for a test can not be described by simple structure. Results are presented to show that the additional response improves the estimate of the-point even though the paradoxical result occurs.

Original language	English
Title of host publication	Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society
Editors	L. Andries van der Ark, Daniel M. Bolt, Wen-Chung Wang, Roger E. Millsap
Publisher	Springer New York LLC
Pages	343-357
Number of pages	15
ISBN (Electronic)	9783319075020
DOIs	https://doi.org/10.1007/978-3-319-07503-7_22
State	Published - 2015
Event	78th Annual Meeting of the Psychometric Society, 2013 - Arnhem, Netherlands Duration: 22 Jul 2013 → 26 Jul 2013

Publication series

Name	Springer Proceedings in Mathematics and Statistics
Volume	89
ISSN (Print)	2194-1009
ISSN (Electronic)	2194-1017

Conference

Conference	78th Annual Meeting of the Psychometric Society, 2013
Country/Territory	Netherlands
City	Arnhem
Period	22/07/13 → 26/07/13

Bibliographical note

Publisher Copyright:
© Springer International Publishing Switzerland 2015.

Access to Document

10.1007/978-3-319-07503-7_22

Cite this

Ong, M. L., Lu, L., Lee, S., & Cohen, A. (2015). A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning. In L. A. van der Ark, D. M. Bolt, W.-C. Wang, & R. E. Millsap (Eds.), Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society (pp. 343-357). (Springer Proceedings in Mathematics and Statistics; Vol. 89). Springer New York LLC. https://doi.org/10.1007/978-3-319-07503-7_22

Ong, Mei Ling ; Lu, Laura ; Lee, Sunbok et al. / A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning. Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society. editor / L. Andries van der Ark ; Daniel M. Bolt ; Wen-Chung Wang ; Roger E. Millsap. Springer New York LLC, 2015. pp. 343-357 (Springer Proceedings in Mathematics and Statistics).

@inproceedings{7b9afc57258343b0acf5add22ade3cad,

title = "A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning",

abstract = "This chapter describes the property of estimates of points in a multidimensional space that is labeled by some as paradoxical, shows when this property of the estimates is present, and also shows that the paradoxical result is not flaw in estimation because estimates improve with additional information even when the paradox occurs. The paradox is that when a correct response to a test item is added to the string of responses for an examinee to previous items, at least one of the coordinates of the new estimated ?-point decreases compared to the estimate based on the initial string of responses. The information presented in the chapter shows that this can occur whenever the likelihood function for the estimates has a particular form. This form is present in many cases when the item responses for a test can not be described by simple structure. Results are presented to show that the additional response improves the estimate of the-point even though the paradoxical result occurs.",

author = "Ong, {Mei Ling} and Laura Lu and Sunbok Lee and Allan Cohen",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 78th Annual Meeting of the Psychometric Society, 2013 ; Conference date: 22-07-2013 Through 26-07-2013",

year = "2015",

doi = "10.1007/978-3-319-07503-7_22",

language = "English",

series = "Springer Proceedings in Mathematics and Statistics",

publisher = "Springer New York LLC",

pages = "343--357",

editor = "{van der Ark}, {L. Andries} and Bolt, {Daniel M.} and Wen-Chung Wang and Millsap, {Roger E.}",

booktitle = "Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society",

}

Ong, ML, Lu, L, Lee, S & Cohen, A 2015, A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning. in LA van der Ark, DM Bolt, W-C Wang & RE Millsap (eds), Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society. Springer Proceedings in Mathematics and Statistics, vol. 89, Springer New York LLC, pp. 343-357, 78th Annual Meeting of the Psychometric Society, 2013, Arnhem, Netherlands, 22/07/13. https://doi.org/10.1007/978-3-319-07503-7_22

A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning. / Ong, Mei Ling; Lu, Laura; Lee, Sunbok et al.
Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society. ed. / L. Andries van der Ark; Daniel M. Bolt; Wen-Chung Wang; Roger E. Millsap. Springer New York LLC, 2015. p. 343-357 (Springer Proceedings in Mathematics and Statistics; Vol. 89).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning

AU - Ong, Mei Ling

AU - Lu, Laura

AU - Lee, Sunbok

AU - Cohen, Allan

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - This chapter describes the property of estimates of points in a multidimensional space that is labeled by some as paradoxical, shows when this property of the estimates is present, and also shows that the paradoxical result is not flaw in estimation because estimates improve with additional information even when the paradox occurs. The paradox is that when a correct response to a test item is added to the string of responses for an examinee to previous items, at least one of the coordinates of the new estimated ?-point decreases compared to the estimate based on the initial string of responses. The information presented in the chapter shows that this can occur whenever the likelihood function for the estimates has a particular form. This form is present in many cases when the item responses for a test can not be described by simple structure. Results are presented to show that the additional response improves the estimate of the-point even though the paradoxical result occurs.

AB - This chapter describes the property of estimates of points in a multidimensional space that is labeled by some as paradoxical, shows when this property of the estimates is present, and also shows that the paradoxical result is not flaw in estimation because estimates improve with additional information even when the paradox occurs. The paradox is that when a correct response to a test item is added to the string of responses for an examinee to previous items, at least one of the coordinates of the new estimated ?-point decreases compared to the estimate based on the initial string of responses. The information presented in the chapter shows that this can occur whenever the likelihood function for the estimates has a particular form. This form is present in many cases when the item responses for a test can not be described by simple structure. Results are presented to show that the additional response improves the estimate of the-point even though the paradoxical result occurs.

UR - http://www.scopus.com/inward/record.url?scp=84915746718&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-07503-7_22

DO - 10.1007/978-3-319-07503-7_22

M3 - Conference contribution

AN - SCOPUS:84915746718

T3 - Springer Proceedings in Mathematics and Statistics

SP - 343

EP - 357

BT - Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society

A2 - van der Ark, L. Andries

A2 - Bolt, Daniel M.

A2 - Wang, Wen-Chung

A2 - Millsap, Roger E.

PB - Springer New York LLC

T2 - 78th Annual Meeting of the Psychometric Society, 2013

Y2 - 22 July 2013 through 26 July 2013

ER -

Ong ML, Lu L, Lee S, Cohen A. A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning. In van der Ark LA, Bolt DM, Wang WC, Millsap RE, editors, Quantitative Psychology Research - The 78th Annual Meeting of the Psychometric Society. Springer New York LLC. 2015. p. 343-357. (Springer Proceedings in Mathematics and Statistics). doi: 10.1007/978-3-319-07503-7_22

A comparison of the hierarchical generalized linear model, multiple-indicators multiple-causes, and the item response theory-likelihood ratio test for detecting differential item functioning

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this