Navigation

Skiplinks

Kontakt

Geschäftsführung und Koordination

Rea Kodalle

S 1-130

+49 (0)441 798-5481

Direktorin

Prof. Dr. Gisela Schulze

A01 1-132

+49 (0)441 798-2175

Koordination der Programme für Studierende/Evaluation

Robert Mitschke

Wissenschaftliche Hilfskräfte

Lukas Brüggen und Sandra Langhop

Details zum Termin:

Workshop: Introduction to Structural Equation Modeling with R (Dr. M. Murat Ardag)

3GO-Workshop 3GO + Graduiertenakademie 3GO

In words of Yves Rosseel, the developer of the R package lavaan (latent variable analysis), “In the social sciences, structural equation modeling (SEM) is often considered to be the mother of all statistical modeling. It includes univariate and multivariate regression models, generalized linear mixed models, factor analysis, path analysis, item response theory, latent class analysis, and much more. SEM can also handle missing data, non-normal data, categorical data, multilevel data, longitudinal data, (in)equality constraints, and on a good day, SEM makes you a fresh cup of tea.” Put simply: there is almost nothing you cannot do with SEM.

This workshop is designed to be a practitioner’s guide. After going over the fundamentals of the outlined topics below, we will get to focus on their application in the R environment. There is almost no math involved; however we will get to understand the conceptual reasoning over the math going on in the background.

Target Audience are the graduate students at the following departments: Social Sciences, Educational Sciences, Business Administration & Economics, Psychology, Biology.

Please note that a general familiarity with R is a prerequisite for attending the workshop. I can provide prep material for R before the workshop.

Day 1
Morning session(s)
Recap: Simple & multiple linear regression
Path analysis: going beyond the regression framework to test hypothesized mechanisms 
Recap: Principal component analysis vs. exploratory factor analysis
Applications in the R environment
Afternoon session(s)
Confirmatory factor analysis: are the operationalizations valid and reliable?
Basics of item response theory: eliminating the bad items in a measurement model and seeing the general psychometric qualities of the scale with graded response model
Structural equation modeling: finding the associations between the operationalized constructs
Applications in the R environment
Day 2
Morning session(s)
Multiple group models: exact vs. approximate measurement invariance across groups, SEM trees & forests
Non-recursive structural equation models for (pseudo-)causal relationships: testing causality with cross-sectional survey data
Application of structural equation models in experiments: testing causality with multi-group models
Applications in the R environment
Afternoon session(s)
Multilevel structural equation models: dealing with nested data structures 
Including survey-weights in structural equation models: when to weight the indicators?
Applications in the R environment

Beginn

19. Februar 2020, 09:00 – Enddatum: 20. Februar 2020, 16:00

3GOed-Webmahusterxqi (3GO@uol.duk2exfw) (Stand: 10.12.2019)