General Linear Mixed Models of Longitudinal Studies: Small Samples, Varied Distributional Conditions, and Missing Data

General linear mixed models are commonly used in the analysis of unbalanced repeated measures designs (Verbeke & Molenberghs, 2000). Though the relative performance of linear mixed models in unbalanced designs has been widely contrasted with general linear modeling of data using varied approaches toward the imputation of missing data values, these studies have focussed largely on modeling under large sample sizes and standard distributional conditions. In the current paper, the use of general linear mixed modeling under unbalanced repeated measures conditions under small sample size, diverse distributional conditions, and a variety of missing data patterns is contrasted with modeling using missing value imputation methods, with the focus of documenting the characteristics of the model parameter estimates, their standard errors, and the rejection probabilities of the corresponding test statistic.