Colloquium

On Tuesday, November 15th at 3:30 pm, Dr. Jack McArdle of the University of Southern California will give a research talk titled: "Slowly Moving from Repeated Measures ANOVA to Dynamic BUT Structural Modeling" in A211 PDB.

ABSTRACT:

The predominance of Repeated Measures ANOVA (RANOVA) in longitudinal data analysis is considered. RANOVA is a readily available, and widely respected way to test mean changes over time, so it is a widely used technique in both observational and manipulation research. Controversies about the required covariance assumptions of the data (i.e., compound symmetry) have been largely settled by the use of an epsilon factor to correct the probability values. There is no doubt that RANOVA is a special and useful technique.

But the recent surge of activity in Longitudinal Structural Equation Models (LSEM) should not be ignored either. Although it is not often stated, the RANOVA can be thought of and fitted as a special case of the more general LSEM approach. That is, exactly the same parameter values and fit indices can be obtained from RANOVA or SEM programs. As soon as this basic RANOVA option is demonstrated in LSEM, other longitudinal modeling approaches become clear – including the recent surge of activity in latent growth curve modeling and latent change score analysis.

The need for these new approaches to dynamic analysis comes largely when we want to examine hypotheses about the individual differences in changes. This focus on the individual and their changes is not a formal property of RANOVA. The LSEM is not considered the final statement here, and Exact Differential Models or Chaos Models can be used instead. To clarify this first option, numerical examples are presented using standard SEM and SAS software. The key dynamic question arises – "What is your model for change?" The biggest surprise comes when many researchers have questions and ideas that are well beyond the RANOVA approaches they use, and the LSEM approaches would be much more suitable for evaluating their own ideas.