Peter Schmidt is Professor of Methodology for Social Research, University of Giessen. His research interests are the foundations and applications of structural equation models, analysis of panel data, and empirically testing rational choice theory. Applications include studies of national identity, inter-ethnic relations, and environmental behavior. He has published several books and papers on these topics.
Eldad Davidov is an assistant and a post-doctorate fellow in sociology in the University of Basel. His research interests are applications of structural equation models, analysis of panel data, and testing rational choice theory. Applications include studies of prejudice and discrimination in Europe, cross country comparisons, values and environmental behavior. He is publishing several papers on these topics.
The course extends the basic SEM course, and shows how to apply the structural equation modeling approach to longitudinal data using the AMOS computer program. In the first part we deal with autoregressive models. Different model specifications, including models with cross-lagged effects, are applied and tested with data from a longitudinal study on authoritarianism and anomia in Germany. In the second part, we introduce latent growth models (LGM) and hybrid/ALT models, applying the same data set. Topics in both parts include parameterization of autocorrelation, Socratic effect, multiple-group comparisons, latent means, MIMIC models and treatment of missing values.
To learn how to analyze panel data using autoregressive, growth and hybrid models by applying the software AMOS; and to understand the technical problems of these applications. The course is application-oriented rather than statistics-oriented.
SEM I (the Summer School course offered in the past or similar). We expect participants to have a very good knowledge of regression analysis and structural equation modeling, preferably AMOS.