Readspeaker menu

Workshop 5: Prof. Dr. Yves Rosseel - Structural Equation Modeling with lavaan

The aim of this workshop is to provide a concise overview of the theory of structural equation modeling (SEM), with a special focus on topics that may be of interest to researchers in the field of educational measurement. In particular, there will be a strong emphasis on confirmatory factor analysis (CFA), both with continuous and binary/ordinal data. All theory shall be accompanied by practical hands-on sessions.

Content and method of instruction

The workshop will be based on lecture-style presentations interchanged with practical sessions. All analyses will be done with the R package 'lavaan'. After a brief general introduction to SEM, several specific topics will be discussed: meanstructures and multiple groups, measurement invariance, dealing with missing data and non-normal data, CFA with binary and ordinal data, and the connection with item response theory (IRT).

Intended audience

Participants should have a solid understanding of regression analysis and factor analysis. A basic knowledge of SEM is recommended, but not required. Basic knowledge of working with the R language and environment are recommended.


Participants need to bring their own laptop. They must have installed the latest version of R ( and the R package `lavaan' (see for instructions). Participants are encouraged to bring their own dataset(s).

Background reading for this workshop:
Rosseel, Y. (2012). lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48(2), 1-36.


Yves Rosseel

Yves Rosseel obtained his PhD from Ghent University, Belgium. He is now an associate professor at the Department of Data Analysis, Faculty of Psychology and Educational Sciences, Ghent University.  His research interests include neuroimaging data analysis, connectivity and causality, and structural equation modeling. He is the author of lavaan, an R package for structural equation modeling.

For further information please click here.