Home > Research > Publications & Outputs > Kernel Equating Using Propensity Scores for Non...

Electronic data

Links

Text available via DOI:

View graph of relations

Kernel Equating Using Propensity Scores for Nonequivalent Groups

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Kernel Equating Using Propensity Scores for Nonequivalent Groups. / Wallin, Gabriel; Wiberg, Marie.
In: Journal of Educational and Behavioral Statistics, Vol. 44, No. 4, 31.08.2019, p. 390-414.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Wallin, G & Wiberg, M 2019, 'Kernel Equating Using Propensity Scores for Nonequivalent Groups', Journal of Educational and Behavioral Statistics, vol. 44, no. 4, pp. 390-414. https://doi.org/10.3102/1076998619838226

APA

Wallin, G., & Wiberg, M. (2019). Kernel Equating Using Propensity Scores for Nonequivalent Groups. Journal of Educational and Behavioral Statistics, 44(4), 390-414. https://doi.org/10.3102/1076998619838226

Vancouver

Wallin G, Wiberg M. Kernel Equating Using Propensity Scores for Nonequivalent Groups. Journal of Educational and Behavioral Statistics. 2019 Aug 31;44(4):390-414. Epub 2019 Apr 2. doi: 10.3102/1076998619838226

Author

Wallin, Gabriel ; Wiberg, Marie. / Kernel Equating Using Propensity Scores for Nonequivalent Groups. In: Journal of Educational and Behavioral Statistics. 2019 ; Vol. 44, No. 4. pp. 390-414.

Bibtex

@article{3bcb8187ef714b7d8cb5d2044163d7f5,
title = "Kernel Equating Using Propensity Scores for Nonequivalent Groups",
abstract = "When equating two test forms, the equated scores will be biased if the test groups differ in ability. To adjust for the ability imbalance between nonequivalent groups, a set of common items is often used. When no common items are available, it has been suggested to use covariates correlated with the test scores instead. In this article, we reduce the covariates to a propensity score and equate the test forms with respect to this score. The propensity score is incorporated within the kernel equating framework using poststratification and chained equating. The methods are evaluated using real college admissions test data and through a simulation study. The results show that propensity scores give an increased equating precision in comparison with the equivalent groups design and a smaller mean squared error than by using the covariates directly. Practical implications are also discussed.",
author = "Gabriel Wallin and Marie Wiberg",
year = "2019",
month = aug,
day = "31",
doi = "10.3102/1076998619838226",
language = "English",
volume = "44",
pages = "390--414",
journal = "Journal of Educational and Behavioral Statistics",
issn = "1076-9986",
publisher = "SAGE Publications Inc.",
number = "4",

}

RIS

TY - JOUR

T1 - Kernel Equating Using Propensity Scores for Nonequivalent Groups

AU - Wallin, Gabriel

AU - Wiberg, Marie

PY - 2019/8/31

Y1 - 2019/8/31

N2 - When equating two test forms, the equated scores will be biased if the test groups differ in ability. To adjust for the ability imbalance between nonequivalent groups, a set of common items is often used. When no common items are available, it has been suggested to use covariates correlated with the test scores instead. In this article, we reduce the covariates to a propensity score and equate the test forms with respect to this score. The propensity score is incorporated within the kernel equating framework using poststratification and chained equating. The methods are evaluated using real college admissions test data and through a simulation study. The results show that propensity scores give an increased equating precision in comparison with the equivalent groups design and a smaller mean squared error than by using the covariates directly. Practical implications are also discussed.

AB - When equating two test forms, the equated scores will be biased if the test groups differ in ability. To adjust for the ability imbalance between nonequivalent groups, a set of common items is often used. When no common items are available, it has been suggested to use covariates correlated with the test scores instead. In this article, we reduce the covariates to a propensity score and equate the test forms with respect to this score. The propensity score is incorporated within the kernel equating framework using poststratification and chained equating. The methods are evaluated using real college admissions test data and through a simulation study. The results show that propensity scores give an increased equating precision in comparison with the equivalent groups design and a smaller mean squared error than by using the covariates directly. Practical implications are also discussed.

U2 - 10.3102/1076998619838226

DO - 10.3102/1076998619838226

M3 - Journal article

VL - 44

SP - 390

EP - 414

JO - Journal of Educational and Behavioral Statistics

JF - Journal of Educational and Behavioral Statistics

SN - 1076-9986

IS - 4

ER -