GESIS Training Courses
user_jsdisabled
Suche

Wiss. Koordination

Prof. Dr.
Reinhard Schunck
Tel: +49 221 47694-160

Administrative Koordination

Angelika Ruf
Tel: +49 221 47694-162

Week 2: Structural Equation Models (SEMs)

Dozent(en):
Prof. Dr. Kenneth A. Bollen, , Zachary Fisher

Datum: 13.03 - 17.03.2017 ics-Datei

Veranstaltungsort: Cologne

Referenteninformationen - Prof. Dr. Kenneth A. Bollen

Referenteninformationen - Zachary Fisher

Seminarinhalt

This workshop is about Structural Equation Models (SEMs) and the statistical software to estimate such models. The course provides an overview of and experience in constructing and estimating SEMs. The topics treated include: path analysis, confirmatory factor analysis, simultaneous equation models, the incorporation of multiple indicators and measurement error into structural equations, alternative estimation procedures, and the assessment of model identification, fit, and modification.
Keywords
Latent variables, path analysis, multiple indicators, measurement error, factor analysis, simultaneous equation models, identification


Zielgruppe

  • Deal with multiple indicators of constructs
  • Want estimates of relations between variables that control for measurement error
  • Desire to learn new tests of overall model fit.


Lernziel

By the end of the course participants will:
  • Be familiar with the primary types of structural equation models (SEMs)
  • Learn rules of identification for SEMs
  • Be able to evaluate the fit of a SEM
  • Know about different ways to estimate a SEM
  • Understand the relation between the general SEM and special cases such simultaneous equations and factor analysis. 
  • Gain experience with estimating a number of SEMs with real empirical data.


Voraussetzungen

  • Participants should have a good grasp of multiple regression and befamiliar with basic matrix notation and matrix operations. Background in factoranalysis or path analysis is helpful, but is not required.
  • Basic knowledge of Mpuls.


Zeitplan

Literaturempfehlungen