Workshop on Structural Equation Modelling (SEM)

Click here to dowload the description of the Workshop

15/10/2016 - 09:30-13:00

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, social values, personality factors, and behavioural intentions. SEM combines factor analysis and path analysis by simultaneously estimating causal relations between latent constructs ('structural model') and relations between latent constructs and their corresponding manifest indicators ('measurement models'). Additionally, SEM allows the estimation and control of random and systematic measurement errors. Thus, SEM methodology allows an adequate modelling and empirical testing of measurement models and complex theoretical assumptions.

The course introduces theory and practice of SEM with different software packages (e.g. Mplus, AMOS, Lavaan, EQS) and discusses typical problems researchers have to deal with when adopting their own research models.

The course covers topics like:

• Specification problems (e.g. formative vs. reflective indicators; higher order factors, modelling strategies)
• Model evaluation (e.g. fit indices, modification indices, correction of attenuation, preconditions)
• Systematic measurement errors
• measurement equivalence in longitudinal and multiple group modelling
• mean structure modelling
• causality (e.g. autoregressive cross-lagged models, mediator models, non-recursive models, equivalent models)
• latent variable interaction effects
• bootstrapping and missing value treatment
• estimation problems, different estimators and categorical structural equation models
• comparison of different SEM software packages

There is a registration fee of €10. Interested parties can register through www.easyconferences.org.