Selecting a linear mixed model for longitudinal data

Repeated measures analysis of variance, covariance pattern model, and growth curve approaches

Siwei Liu, Michael J. Rovine, Peter Molenaar

Research output: Contribution to journalArticle

72 Citations (Scopus)

Abstract

With increasing popularity, growth curve modeling is more and more often considered as the 1st choice for analyzing longitudinal data. Although the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this article we compare 3 approaches for analyzing longitudinal data: repeated measures analysis of variance, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this article, we first indicate the kinds of data that are appropriately modeled by each and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of Akaike information criterion and Bayesian information criterion in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.

Original languageEnglish (US)
Pages (from-to)15-30
Number of pages16
JournalPsychological Methods
Volume17
Issue number1
DOIs
StatePublished - Mar 2012

Fingerprint

Linear Models
Analysis of Variance
Growth
Patient Selection
Longitudinal Data
Growth Curve
Research Personnel
Modeling

All Science Journal Classification (ASJC) codes

  • Psychology (miscellaneous)
  • History and Philosophy of Science

Cite this

@article{c5883eb6f1644739b2e7a8bd8670b5d3,
title = "Selecting a linear mixed model for longitudinal data: Repeated measures analysis of variance, covariance pattern model, and growth curve approaches",
abstract = "With increasing popularity, growth curve modeling is more and more often considered as the 1st choice for analyzing longitudinal data. Although the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this article we compare 3 approaches for analyzing longitudinal data: repeated measures analysis of variance, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this article, we first indicate the kinds of data that are appropriately modeled by each and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of Akaike information criterion and Bayesian information criterion in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.",
author = "Siwei Liu and Rovine, {Michael J.} and Peter Molenaar",
year = "2012",
month = "3",
doi = "10.1037/a0026971",
language = "English (US)",
volume = "17",
pages = "15--30",
journal = "Psychological Methods",
issn = "1082-989X",
publisher = "American Psychological Association Inc.",
number = "1",

}

Selecting a linear mixed model for longitudinal data : Repeated measures analysis of variance, covariance pattern model, and growth curve approaches. / Liu, Siwei; Rovine, Michael J.; Molenaar, Peter.

In: Psychological Methods, Vol. 17, No. 1, 03.2012, p. 15-30.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Selecting a linear mixed model for longitudinal data

T2 - Repeated measures analysis of variance, covariance pattern model, and growth curve approaches

AU - Liu, Siwei

AU - Rovine, Michael J.

AU - Molenaar, Peter

PY - 2012/3

Y1 - 2012/3

N2 - With increasing popularity, growth curve modeling is more and more often considered as the 1st choice for analyzing longitudinal data. Although the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this article we compare 3 approaches for analyzing longitudinal data: repeated measures analysis of variance, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this article, we first indicate the kinds of data that are appropriately modeled by each and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of Akaike information criterion and Bayesian information criterion in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.

AB - With increasing popularity, growth curve modeling is more and more often considered as the 1st choice for analyzing longitudinal data. Although the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this article we compare 3 approaches for analyzing longitudinal data: repeated measures analysis of variance, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this article, we first indicate the kinds of data that are appropriately modeled by each and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of Akaike information criterion and Bayesian information criterion in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.

UR - http://www.scopus.com/inward/record.url?scp=84868285072&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84868285072&partnerID=8YFLogxK

U2 - 10.1037/a0026971

DO - 10.1037/a0026971

M3 - Article

VL - 17

SP - 15

EP - 30

JO - Psychological Methods

JF - Psychological Methods

SN - 1082-989X

IS - 1

ER -