Standard

On the treatment of missing data in background questionnaires in educational large-scale assessments : An evaluation of different procedures. / Grund, Simon; Lüdtke, Oliver; Robitzsch, Alexander.

in: Journal of Educational and Behavioral Statistics, Jahrgang 46, Nr. 4, 08.2021, S. 430-465.

Publikationen: Beitrag in FachzeitschriftArtikel in FachzeitschriftForschungBegutachtung

Harvard

Grund, S, Lüdtke, O & Robitzsch, A 2021, 'On the treatment of missing data in background questionnaires in educational large-scale assessments: An evaluation of different procedures', Journal of Educational and Behavioral Statistics, Jg. 46, Nr. 4, S. 430-465. https://doi.org/10.3102/1076998620959058

APA

Vancouver

Author

Grund, Simon ; Lüdtke, Oliver ; Robitzsch, Alexander. / On the treatment of missing data in background questionnaires in educational large-scale assessments : An evaluation of different procedures. in: Journal of Educational and Behavioral Statistics. 2021 ; Jahrgang 46, Nr. 4. S. 430-465.

BibTeX

@article{3d57875f52f5473fb059ff5d85d7783f,
title = "On the treatment of missing data in background questionnaires in educational large-scale assessments: An evaluation of different procedures",
abstract = "Large-scale assessments (LSAs) use Mislevy{\textquoteright}s “plausible value” (PV) approach to relate student proficiency to noncognitive variables administered in a background questionnaire. This method requires background variables to be completely observed, a requirement that is seldom fulfilled. In this article, we evaluate and compare the properties of methods used in current practice for dealing with missing data in background variables in educational LSAs, which rely on the missing indicator method (MIM), with other methods based on multiple imputation. In this context, we present a fully conditional specification (FCS) approach that allows for a joint treatment of PVs and missing data. Using theoretical arguments and two simulation studies, we illustrate under what conditions the MIM provides biased or unbiased estimates of population parameters and provide evidence that methods such as FCS can provide an effective alternative to the MIM. We discuss the strengths and weaknesses of the approaches and outline potential consequences for operational practice in educational LSAs. An illustration is provided using data from the PISA 2015 study.",
keywords = "Methodological research and development, missing data, multiple imputation, plausible values, measurement error, large-scale assessment",
author = "Simon Grund and Oliver L{\"u}dtke and Alexander Robitzsch",
year = "2021",
month = aug,
doi = "10.3102/1076998620959058",
language = "English",
volume = "46",
pages = "430--465",
journal = "Journal of Educational and Behavioral Statistics",
issn = "1076-9986",
publisher = "Sage",
number = "4",

}

RIS

TY - JOUR

T1 - On the treatment of missing data in background questionnaires in educational large-scale assessments

T2 - An evaluation of different procedures

AU - Grund, Simon

AU - Lüdtke, Oliver

AU - Robitzsch, Alexander

PY - 2021/8

Y1 - 2021/8

N2 - Large-scale assessments (LSAs) use Mislevy’s “plausible value” (PV) approach to relate student proficiency to noncognitive variables administered in a background questionnaire. This method requires background variables to be completely observed, a requirement that is seldom fulfilled. In this article, we evaluate and compare the properties of methods used in current practice for dealing with missing data in background variables in educational LSAs, which rely on the missing indicator method (MIM), with other methods based on multiple imputation. In this context, we present a fully conditional specification (FCS) approach that allows for a joint treatment of PVs and missing data. Using theoretical arguments and two simulation studies, we illustrate under what conditions the MIM provides biased or unbiased estimates of population parameters and provide evidence that methods such as FCS can provide an effective alternative to the MIM. We discuss the strengths and weaknesses of the approaches and outline potential consequences for operational practice in educational LSAs. An illustration is provided using data from the PISA 2015 study.

AB - Large-scale assessments (LSAs) use Mislevy’s “plausible value” (PV) approach to relate student proficiency to noncognitive variables administered in a background questionnaire. This method requires background variables to be completely observed, a requirement that is seldom fulfilled. In this article, we evaluate and compare the properties of methods used in current practice for dealing with missing data in background variables in educational LSAs, which rely on the missing indicator method (MIM), with other methods based on multiple imputation. In this context, we present a fully conditional specification (FCS) approach that allows for a joint treatment of PVs and missing data. Using theoretical arguments and two simulation studies, we illustrate under what conditions the MIM provides biased or unbiased estimates of population parameters and provide evidence that methods such as FCS can provide an effective alternative to the MIM. We discuss the strengths and weaknesses of the approaches and outline potential consequences for operational practice in educational LSAs. An illustration is provided using data from the PISA 2015 study.

KW - Methodological research and development

KW - missing data

KW - multiple imputation

KW - plausible values

KW - measurement error

KW - large-scale assessment

U2 - 10.3102/1076998620959058

DO - 10.3102/1076998620959058

M3 - Journal article

VL - 46

SP - 430

EP - 465

JO - Journal of Educational and Behavioral Statistics

JF - Journal of Educational and Behavioral Statistics

SN - 1076-9986

IS - 4

ER -

ID: 1419482