Structural Equation Modelling (SEM) with R

Structural Equation Modelling (SEM) is a powerful tool to analyze latent variable models that are common in social sciences, e.g. the analysis of social and political attitudes or social values. SEM combines factor analysis and path analysis by simultaneously estimating relations between latent constructs and/or manifest variables, and also relations of latent constructs and its corresponding manifest indicators. Additionally, SEM allows the estimation and control for random and systematic measurement errors. Thus, SEM methodology allows an adequate modelling and empirical testing of measurement models and complex theoretical assumptions. Additionally, SEM is capable to compare these models between different groups, like countries or waves of a longitudinal study, or also social groups.

The course introduces to theory and practice of SEM on a general level. During the course, R software will be used.

Basic modelling techniques of SEM are explained and applied by exercises using free access social science data. Additionally, participants have got the possibility to use their own data for analyses. Daily assignments allow the application and transfer of SEM methodology to own research interests.

On the first day we go through the statistical and methodological basics of SEM, such as regression analysis and classical path model analysis. We show the advantages of SEM and present its fitting principles. Basic (visual and statistical) notations and statistical tests and indices will be mentioned. On the second day, we focus on the first step of building a SEM model, namely the usage of confirmatory factor analysis (CFA) for the creation of latent constructs. Comparison with other factor analysis methods will be presented. We go through different model-building (parametrisation) and model improving techniques as well. The practical implementation of this knowledge will be possible on the lab session. On the third day the latent variable model will be expanded with other, explanatory variables to get a more complex and interpretable model. These models are better for answering a scientific research question as they give more space for explanation. There will be possibility again for the application of these models, with emphasis on the interpretation and practical questions. On the fourth day, the class will be about multiple group comparison, which is – taking the different levels of testing into account – one of the most useful part of SEM. The class starts with the theoretical problems of multi-group comparison and connects these problems with SEM tools, that can help to decide the depth of the comparison. A step-by-step guide will be provided for the participant to be able to proceed such analysis. In the lab session, technical advises will be given about the measurement and realization of these models, especially about the interpretation of the different results. The last day is to summarize the knowledge studied during the week, with the help of a concrete example, where all the techniques will be applied. After going through a complex example, the consequences will be generalised, and other suggestions will be drawn. Finally, some practices will be presented for publication of papers, which include SEM methods.

Participants should understand basic principles of regression analysis and the meaning of regression results. A basic understanding of principal component analysis (explorative factor analysis) would be helpful. Participants should have some familiarity with software R to manage data.

By registering for this course, you confirm that you possess the knowledge required to follow it. The instructor will not teach these prerequisite items. If in doubt, contact the instructor before registering.