Model Choice Can Obscure Results in Longitudinal Studies
Open Access
- 1 February 2009
- journal article
- research article
- Published by Oxford University Press (OUP) in The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences
- Vol. 64 (2), 215-222
- https://doi.org/10.1093/gerona/gln024
Abstract
This article examines how different parameterizations of age and time in modeling observational longitudinal data can affect results. When individuals of different ages at study entry are considered, it becomes necessary to distinguish between longitudinal and cross-sectional differences to overcome possible selection biases. Various models were fitted using data from longitudinal studies with participants with different ages and different follow-up lengths. Decomposing age into two components—age at entry into the study (first age) and the longitudinal follow-up (time) compared with considering age alone—leads to different conclusions. In general, models using both first age and time terms performed better, and these terms are usually necessary to correctly analyze longitudinal data.Keywords
This publication has 31 references indexed in Scilit:
- Is the Metabolic Syndrome, With or Without Diabetes, Associated With Progressive Disability in Older Mexican Americans?The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 2007
- Selecting the Best Linear Mixed Model Under REMLThe American Statistician, 2006
- A method for fitting regression splines with varying polynomial order in the linear mixed modelStatistics in Medicine, 2005
- Applying Linear Mixed-Effects Models to the Problem of Measurement Error in Epidemiologic StudiesCommunications in Statistics - Simulation and Computation, 2003
- Describing the Natural Heterogeneity of Aging Using Multilevel Regression ModelsInternational Journal of Sports Medicine, 1997
- Mixed‐effects regression models for studying the natural history of prostate diseaseStatistics in Medicine, 1994
- Modelling hearing thresholds in the elderlyStatistics in Medicine, 1991
- Application of hierarchical linear models to assessing change.Psychological Bulletin, 1987
- Multilevel mixed linear model analysis using iterative generalized least squaresBiometrika, 1986
- A general model for the study of developmental problems.Psychological Bulletin, 1965