skip to content

Latent Variable Structural Equation Modeling - Using Partial Least Squares

General information

Structural equation modeling depicts an extension of the classical factor analysis model to explain relations among latent conceptual variables. Thus, it enables empirical validation of theoretically established causal models in the various social science disciplines. Partial Least Squares Path Modelling (PLS-PM) is one possible procedure for estimating such models. It proxies latent conceptual variables with linear weighted composites that aim at maximizing explained variances among the set of determinate composite scores. As such, it has a special emphasis on prediction orientation. In contrast to covariance based structural equation modeling (CB-SEM) it trades parameter efficiency for prediction accuracy, simplicity and fewer assumptions. Thus, the method is applicable to large and complex models with relatively few observations. It has gained popularity in various fields such as marketing, management, and information systems.

The objective of this course is to provide an in-depth introduction into the development of structural equation models and the operationalization of latent conceptual variables. It will cover the nature of causal modeling, analytical objectives and some statistics, as well as the evaluation of the estimation results with a focus on the PLS-PM method and an introduction to available software packages (e.g., SmartPLS).

Practical applications and a discussion of the applicability of the method for your own research are an integral part of the course. Students should have a solid foundation in statistics and be familiar with multivariate data analysis. In addition, a basic understanding of factor analytic approaches, regression analysis as well as testing procedures is helpful, but not an essential requirement for understanding the content of the course.

Teaching staff

Dr. Jan-Michael Becker (Department of Marketing and Brand Management)

Time and Location

  • 21.10.2016, 10am – 2pm, Room 327 WiSo-Building
  • 14.11.2016, full day, Room 327 WiSo-Building
  • 25.11.2016, full day, Room 327 WiSo-Building
  • 13.01.2017, full day, Room 327 WiSo-Building

Further information about this CGS-Course here

Registration

Students who want to attend this course need to register. Master students need to register via KLIPS in the first application phase (20.07. – 01.08.2016).

Doctoral students apply via e-mail. The deadline for doctoral student registration is 10.10.2016. Please send an email (in German or English) to Dr. Jan-Michael Becker which should inform about:

  • your name
  • your contact email address
  • whether you are a CGS-student or a doctoral student of the Faculty of Management, Economics and Social Sciences
  • the supervisor of your doctoral thesis and topic
  • your background in statistics/econometrics and empirical research as well as any prior knowledge of PLS or any other SEM method