Standard

Multiple imputation of missing data in multilevel models with the R package mdmb : A flexible sequential modeling approach. / Grund, Simon; Lüdtke, Oliver; Robitzsch, Alexander.

in: Behavior Research Methods, Jahrgang 53, Nr. 6, 12.2021, S. 2631-2649.

Publikationen: Beitrag in FachzeitschriftArtikel in FachzeitschriftForschungBegutachtung

Harvard

APA

Vancouver

Author

Grund, Simon ; Lüdtke, Oliver ; Robitzsch, Alexander. / Multiple imputation of missing data in multilevel models with the R package mdmb : A flexible sequential modeling approach. in: Behavior Research Methods. 2021 ; Jahrgang 53, Nr. 6. S. 2631-2649.

BibTeX

@article{4115bc4adbb148a7986d005aa89397f2,
title = "Multiple imputation of missing data in multilevel models with the R package mdmb: A flexible sequential modeling approach",
abstract = "Multilevel models often include nonlinear effects, such as random slopes or interaction effects. The estimation of these models can be difficult when the underlying variables contain missing data. Although several methods for handling missing data such as multiple imputation (MI) can be used with multilevel data, conventional methods for multilevel MI often do not properly take the nonlinear associations between the variables into account. In the present paper, we propose a sequential modeling approach based on Bayesian estimation techniques that can be used to handle missing data in a variety of multilevel models that involve nonlinear effects. The main idea of this approach is to decompose the joint distribution of the data into several parts that correspond to the outcome and explanatory variables in the intended analysis, thus generating imputations for a manner that is compatible with the substantive analysis model. In three simulation studies, we evaluate the sequential modeling approach and compare it with conventional as well as other substantive-model-compatible approaches to multilevel MI. We implemented the sequential modeling approach in the R package mdmb and provide a worked example to illustrate ist application.",
keywords = "Interaction effects, Missing data, Multilevel analysis, Multiple imputation",
author = "Simon Grund and Oliver L{\"u}dtke and Alexander Robitzsch",
note = "Springer",
year = "2021",
month = dec,
doi = "10.3758/s13428-020-01530-0",
language = "English",
volume = "53",
pages = "2631--2649",
journal = "Behavior Research Methods",
issn = "1554-351X",
publisher = "Springer",
number = "6",

}

RIS

TY - JOUR

T1 - Multiple imputation of missing data in multilevel models with the R package mdmb

T2 - A flexible sequential modeling approach

AU - Grund, Simon

AU - Lüdtke, Oliver

AU - Robitzsch, Alexander

N1 - Springer

PY - 2021/12

Y1 - 2021/12

N2 - Multilevel models often include nonlinear effects, such as random slopes or interaction effects. The estimation of these models can be difficult when the underlying variables contain missing data. Although several methods for handling missing data such as multiple imputation (MI) can be used with multilevel data, conventional methods for multilevel MI often do not properly take the nonlinear associations between the variables into account. In the present paper, we propose a sequential modeling approach based on Bayesian estimation techniques that can be used to handle missing data in a variety of multilevel models that involve nonlinear effects. The main idea of this approach is to decompose the joint distribution of the data into several parts that correspond to the outcome and explanatory variables in the intended analysis, thus generating imputations for a manner that is compatible with the substantive analysis model. In three simulation studies, we evaluate the sequential modeling approach and compare it with conventional as well as other substantive-model-compatible approaches to multilevel MI. We implemented the sequential modeling approach in the R package mdmb and provide a worked example to illustrate ist application.

AB - Multilevel models often include nonlinear effects, such as random slopes or interaction effects. The estimation of these models can be difficult when the underlying variables contain missing data. Although several methods for handling missing data such as multiple imputation (MI) can be used with multilevel data, conventional methods for multilevel MI often do not properly take the nonlinear associations between the variables into account. In the present paper, we propose a sequential modeling approach based on Bayesian estimation techniques that can be used to handle missing data in a variety of multilevel models that involve nonlinear effects. The main idea of this approach is to decompose the joint distribution of the data into several parts that correspond to the outcome and explanatory variables in the intended analysis, thus generating imputations for a manner that is compatible with the substantive analysis model. In three simulation studies, we evaluate the sequential modeling approach and compare it with conventional as well as other substantive-model-compatible approaches to multilevel MI. We implemented the sequential modeling approach in the R package mdmb and provide a worked example to illustrate ist application.

KW - Interaction effects

KW - Missing data

KW - Multilevel analysis

KW - Multiple imputation

U2 - 10.3758/s13428-020-01530-0

DO - 10.3758/s13428-020-01530-0

M3 - Journal article

VL - 53

SP - 2631

EP - 2649

JO - Behavior Research Methods

JF - Behavior Research Methods

SN - 1554-351X

IS - 6

ER -

ID: 1517803