Standard

A Bayesian approach to more stable estimates of group-level effects in contextual studies. / Zitzmann, Steffen; Lüdtke, Oliver; Robitzsch, Alexander.

In: Multivariate Behavioral Research, Vol. 50, No. 6, 30.12.2015, p. 688-705.

Publication: Research - peer-reviewJournal articles

Harvard

Zitzmann, S, Lüdtke, O & Robitzsch, A 2015, 'A Bayesian approach to more stable estimates of group-level effects in contextual studies' Multivariate Behavioral Research, vol 50, no. 6, pp. 688-705. DOI: 10.1080/00273171.2015.1090899

APA

Zitzmann, S., Lüdtke, O., & Robitzsch, A. (2015). A Bayesian approach to more stable estimates of group-level effects in contextual studies. Multivariate Behavioral Research, 50(6), 688-705. DOI: 10.1080/00273171.2015.1090899

Vancouver

Zitzmann S, Lüdtke O, Robitzsch A. A Bayesian approach to more stable estimates of group-level effects in contextual studies. Multivariate Behavioral Research. 2015 Dec 30;50(6):688-705. Available from, DOI: 10.1080/00273171.2015.1090899

BibTeX

@article{d8570b1f886146d489d0f8e209e45ab7,
title = "A Bayesian approach to more stable estimates of group-level effects in contextual studies",
abstract = "Multilevel analyses are often used to estimate the effects of group-level constructs. However, when using aggregated individual data (e.g., student ratings) to assess a group-level construct (e.g., classroom climate), the observed group mean might not provide a reliable measure of the unobserved latent group mean. In the present article, we propose a Bayesian approach that can be used to estimate a multilevel latent covariate model, which corrects for the unreliable assessment of the latent group mean when estimating the group-level effect. A simulation study was conducted to evaluate the choice of different priors for the group-level variance of the predictor variable and to compare the Bayesian approach with the maximum likelihood approach implemented in the software Mplus. Results showed that, under problematic conditions (i.e., small number of groups, predictor variable with a small ICC), the Bayesian approach produced more accurate estimates of the group-level effect than the maximum likelihood approach did.",
keywords = "Aims and models, Bayesian estimation, multilevel modeling, structural equation modeling, latent covariate model, contextual analysis",
author = "Steffen Zitzmann and Oliver Lüdtke and Alexander Robitzsch",
year = "2015",
month = "12",
doi = "10.1080/00273171.2015.1090899",
volume = "50",
pages = "688--705",
journal = "Multivariate Behavioral Research",
number = "6",

}

RIS

TY - JOUR

T1 - A Bayesian approach to more stable estimates of group-level effects in contextual studies

AU - Zitzmann,Steffen

AU - Lüdtke,Oliver

AU - Robitzsch,Alexander

PY - 2015/12/30

Y1 - 2015/12/30

N2 - Multilevel analyses are often used to estimate the effects of group-level constructs. However, when using aggregated individual data (e.g., student ratings) to assess a group-level construct (e.g., classroom climate), the observed group mean might not provide a reliable measure of the unobserved latent group mean. In the present article, we propose a Bayesian approach that can be used to estimate a multilevel latent covariate model, which corrects for the unreliable assessment of the latent group mean when estimating the group-level effect. A simulation study was conducted to evaluate the choice of different priors for the group-level variance of the predictor variable and to compare the Bayesian approach with the maximum likelihood approach implemented in the software Mplus. Results showed that, under problematic conditions (i.e., small number of groups, predictor variable with a small ICC), the Bayesian approach produced more accurate estimates of the group-level effect than the maximum likelihood approach did.

AB - Multilevel analyses are often used to estimate the effects of group-level constructs. However, when using aggregated individual data (e.g., student ratings) to assess a group-level construct (e.g., classroom climate), the observed group mean might not provide a reliable measure of the unobserved latent group mean. In the present article, we propose a Bayesian approach that can be used to estimate a multilevel latent covariate model, which corrects for the unreliable assessment of the latent group mean when estimating the group-level effect. A simulation study was conducted to evaluate the choice of different priors for the group-level variance of the predictor variable and to compare the Bayesian approach with the maximum likelihood approach implemented in the software Mplus. Results showed that, under problematic conditions (i.e., small number of groups, predictor variable with a small ICC), the Bayesian approach produced more accurate estimates of the group-level effect than the maximum likelihood approach did.

KW - Aims and models

KW - Bayesian estimation

KW - multilevel modeling

KW - structural equation modeling

KW - latent covariate model

KW - contextual analysis

UR - http://www.tandfonline.com/doi/suppl/10.1080/00273171.2015.1090899/suppl_file/hmbr_a_1090899_sm8864.pdf

U2 - 10.1080/00273171.2015.1090899

DO - 10.1080/00273171.2015.1090899

M3 - Journal articles

VL - 50

SP - 688

EP - 705

JO - Multivariate Behavioral Research

T2 - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

IS - 6

ER -

ID: 571830