S u m m a r y

 

Leskinen E (2005)

Analyzing longitudinal data using latent-variable models

Psykologia 40:476-488. Vammala ISSN 0355-1067.

 

Longitudinal studies enable a versatile examination of change and permanence in the phenomena studied. Analyses of longitudinal data demand advanced statistical methods and models. In this article we will focus on latent-variable models. These models can be used to examine the reliability of the measurements and the features of change and permanence in the latent variables in relation to time. Various comparisons according to group and level are also possible. The analyses are based on modeling the covariance structure and expected value structures of the measured variables (the sample covariance matrices and sample average vectors of the variables). The models chosen as examples are the confirmatory longitudinal factor models, simplex models and latent growth curve models, whose characteristics and applicability in longitudinal studies are demonstrated with examples of applications.

 

 

Esko Leskinen, Ph.D., Department of mathematics and statistics, P.O. Box 35, FIN-40014 University of Jyväskylä, Finland.