Jump to ContentJump to Main Navigation
Show Summary Details
In This Section

The International Journal of Biostatistics

Ed. by Chambaz, Antoine / Hubbard, Alan E. / van der Laan, Mark J.

2 Issues per year


IMPACT FACTOR 2016: 0.500
5-year IMPACT FACTOR: 0.862

CiteScore 2016: 0.42

SCImago Journal Rank (SJR) 2015: 0.495
Source Normalized Impact per Paper (SNIP) 2015: 0.180

Mathematical Citation Quotient (MCQ) 2015: 0.04

Online
ISSN
1557-4679
See all formats and pricing
In This Section

Estimating Multilevel Logistic Regression Models When the Number of Clusters is Low: A Comparison of Different Statistical Software Procedures

Peter C Austin
  • Institute for Clinical Evaluative Sciences
Published Online: 2010-04-22 | DOI: https://doi.org/10.2202/1557-4679.1195

Multilevel logistic regression models are increasingly being used to analyze clustered data in medical, public health, epidemiological, and educational research. Procedures for estimating the parameters of such models are available in many statistical software packages. There is currently little evidence on the minimum number of clusters necessary to reliably fit multilevel regression models. We conducted a Monte Carlo study to compare the performance of different statistical software procedures for estimating multilevel logistic regression models when the number of clusters was low. We examined procedures available in BUGS, HLM, R, SAS, and Stata. We found that there were qualitative differences in the performance of different software procedures for estimating multilevel logistic models when the number of clusters was low. Among the likelihood-based procedures, estimation methods based on adaptive Gauss-Hermite approximations to the likelihood (glmer in R and xtlogit in Stata) or adaptive Gaussian quadrature (Proc NLMIXED in SAS) tended to have superior performance for estimating variance components when the number of clusters was small, compared to software procedures based on penalized quasi-likelihood. However, only Bayesian estimation with BUGS allowed for accurate estimation of variance components when there were fewer than 10 clusters. For all statistical software procedures, estimation of variance components tended to be poor when there were only five subjects per cluster, regardless of the number of clusters.

Keywords: statistical software; multilevel models; hierarchical models; random effects model; mixed effects model; generalized linear mixed models; Monte Carlo simulations; Bayesian analysis; R; SAS; Stata; BUGS

About the article

Published Online: 2010-04-22



Citation Information: The International Journal of Biostatistics, ISSN (Online) 1557-4679, DOI: https://doi.org/10.2202/1557-4679.1195. Export Citation

Citing Articles

Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

[1]
Pau Baizan, Bruno Arpino, and Carlos Eric Delclòs
European Journal of Population, 2016
[2]
Rosa Duarte, José-Julián Escario, and José-Alberto Molina
Journal of Substance Use, 2015, Page 1
[3]
Ignacio Ferreira-González, Xavier Carrillo, Victoria Martín, José M. de la Torre Hernández, José Antonio Baz, Josep Navarro Manchón, Mónica Masotti, Ángel Cequier, Mérida Cárdenas, and Fernando Alfonso Manterola
Revista Española de Cardiología, 2015
[4]
Ignacio Ferreira-González, Xavier Carrillo, Victoria Martín, José M. de la Torre Hernández, José Antonio Baz, Josep Navarro Manchón, Mónica Masotti, Ángel Cequier, Mérida Cárdenas, and Fernando Alfonso Manterola
Revista Española de Cardiología (English Edition), 2015
[5]
Giorgio Di Gessa, Karen Glaser, Debora Price, Eloi Ribe, and Anthea Tinker
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 2015, Page gbv007
[6]
Nazim Habibov
Social Policy & Administration, 2013, Volume 47, Number 3, Page 262
[7]
Shahab Jolani, Thomas P. A. Debray, Hendrik Koffijberg, Stef van Buuren, and Karel G. M. Moons
Statistics in Medicine, 2015, Volume 34, Number 11, Page 1841
[8]
Daniel M. McNeish and Laura M. Stapleton
Educational Psychology Review, 2014
[9]
Luca Börger and Thomas D Nudds
Ecological Applications, 2014, Volume 24, Number 1, Page 121

Comments (0)

Please log in or register to comment.
Log in