The Impact of Misspecifying the Within-Subject Covariance Structure in Multiwave Longitudinal Multilevel Models: A Monte Carlo Study
Authors:
- 10 October 2007
- journal article
- research article
- Published by Taylor & Francis in Multivariate Behavioral Research
- Vol. 42 (3) , 557-592
Abstract
This Monte Carlo study examined the impact of misspecifying the 𝚺 matrix in longitudinal data analysis under both the multilevel model and mixed model frameworks. Under the multilevel model approach, under-specification and general-misspecification of the 𝚺 matrix usually resulted in overestimation of the variances of the random effects (e.g., τ00, ττ11 ) and standard errors of the corresponding growth parameter estimates (e.g., SEβ 0, SEβ 1). Overestimates of the standard errors led to lower statistical power in tests of the growth parameters. An unstructured 𝚺 matrix under the mixed model framework generally led to underestimates of standard errors of the growth parameter estimates. Underestimates of the standard errors led to inflation of the type I error rate in tests of the growth parameters. Implications of the compensatory relationship between the random effects of the growth parameters and the longitudinal error structure for model specification were discussed.Keywords
This publication has 38 references indexed in Scilit:
- Autoregressive Latent Trajectory (ALT) Models A Synthesis of Two TraditionsSociological Methods & Research, 2004
- Time Series Analysis2003
- Differences in early parent-child conversations about negative versus positive emotions: Implications for the development of psychological understanding.Developmental Psychology, 2002
- Children's competence and value beliefs from childhood through adolescence: Growth trajectories in two male-sex-typed domains.Developmental Psychology, 2002
- A comparison of two approaches for selecting covariance structures in the analysis of repeated measurementsCommunications in Statistics - Simulation and Computation, 1998
- General longitudinal modeling of individual differences in experimental designs: A latent variable framework for analysis and power estimation.Psychological Methods, 1997
- Detecting repeated measures effects with univariate and multivariate statistics.Psychological Methods, 1997
- Multilevel time series models with applications to repeated measures dataStatistics in Medicine, 1994
- Standard Errors and Sample Sizes for Two-Level ResearchJournal of Educational Statistics, 1993
- Models for Longitudinal Data with Random Effects and AR(1) ErrorsJournal of the American Statistical Association, 1989