e-Learning Support
Seminar: 08-29-FEM-2 Structural Equation Modeling / Blockseminar - Details

Seminar: 08-29-FEM-2 Structural Equation Modeling / Blockseminar - Details

Nachrichtenbox schließen
Sie sind nicht in Stud.IP angemeldet.

Allgemeine Informationen

Veranstaltungsname Seminar: 08-29-FEM-2 Structural Equation Modeling / Blockseminar
Untertitel
Veranstaltungsnummer 08-29-FEM-2
Semester SoSe 2026
Aktuelle Anzahl der Teilnehmenden 11
erwartete Teilnehmendenanzahl 30
Heimat-Einrichtung Soziologie
Veranstaltungstyp Seminar in der Kategorie Lehre
Nächster Termin Donnerstag, 04.06.2026 09:00 - 13:00, Ort: (UNICOM 7.1050)
Art/Form
Englischsprachige Veranstaltung Ja
ECTS-Punkte 6

Räume und Zeiten

(UNICOM 7.1050)
Donnerstag, 21.05.2026 - Freitag, 22.05.2026, Donnerstag, 04.06.2026 - Freitag, 05.06.2026, Freitag, 19.06.2026 09:00 - 13:00

Modulzuordnungen

Kommentar/Beschreibung

This course introduces you to the wondrous world of Structural Equation Modelling (SEM). Structural Equation Models are a very versatile tool for social scientists, as they combine measurement of complex concepts (e.g., values or attitudes) with the estimation of structural effects. The course focuses on the statistical foundation of this approach, the estimation of SEMs (using R-package lavaan), as well as the interpretation of SEM-results. We cover topics such as latent variable modelling, direct and indirect effect estimation, multigroup analysis and model fit evaluation. Basics background knowledge in statistics (e.g., descriptive statistics, null-hypothesis testing, and regression analysis) is required.

By the end of this course you will…
• …be able to specify, estimate and interpret a CFA and SEM models, ranging from simple models to more complex ones (including multigroup models)
• …understand a variety of fit indices used to evaluate SEM models
• …have a better understanding of different approaches for measuring latent variables
• …have the necessary knowledge to separate direct from indirect effects in explanation models

Key topics covered
Over the course of five modules (one per lecture day), a variety of topics that are fundamental to SEM are discussed.

Module 1 covers the very basics of SEM. We start from questions such as: What is the difference between SEM and more conventional multivariate techniques, such as regression analysis? What are the advantages and specificities of the SEM approach? Module 1 introduces important terminology and notation systems, and provides an overview of different models in the SEM tradition.

In Module 2, we highlight a particular branch within the SEM family, namely latent variable models / measurement models / confirmatory factor analysis. This application allows us to discuss the issues of model specification, parameter estimation, model restrictions and identification in greater depth.

Module 3 provides three different approaches to SEM estimation that each imply a different epistemological position. The wide variety of procedures to evaluate model fit is reviewed and the assumptions of the SEM model are discussed.

Module 4 focuses on a second branch of the SEM approach: structural models. We cover examples such as path analysis, structural regression models and MIMIC models.

Finally, Module 5 introduces the concept of multigroup SEM, and illustrates how this approach can be used in comparative research (including tests for measurement equivalence).

Course dates: 21 & 22 May; 4 & 5 June; 19 June

Anmeldemodus

Die Auswahl der Teilnehmenden wird nach der Eintragung manuell vorgenommen.

Nutzer/-innen, die sich für diese Veranstaltung eintragen möchten, erhalten nähere Hinweise und können sich dann noch gegen eine Teilnahme entscheiden.