Relatore
Tenko Raykov - Michigan State University
Abstract
This talk is concerned with two helpful aids in the process of choosing between models in repeated measure investigations: (i) interval estimates of proportions explained variance in longitudinally followed variables; and (ii) individual case residuals associated with these variables. The discussed methods are developed within the frameworks of latent growth curve modeling (structural equation or latent variable modeling) and allow obtaining (a) confidence intervals for the R-squared indices of repeatedly administered measures, as well as (b) subject-specific discrepancies between model predictions and raw data on the observed variables. In addition to facilitating evaluation of local model fit, the outlined approach is useful for the purpose of differentiating between plausible models stipulating different patterns of change over time, in particular also in empirical situations characterized by (very) large samples and high statistical power that are becoming increasingly more frequent in complex sample design studies in the behavioral, health, and social sciences. The procedures are similarly applicable in cross-sectional investigations, as well as with general structural equation models (covariance structure models), and extend the set of means available for model evaluation by these ‘local’ goodness of fit indices for latent change/latent growth curve models. The methods are illustrated using data from a nationally representative study of older adults.
Organizzazione: Silvia Bianconcini e Silvia Cagnone.