Partial Least Squares Path Modeling: Introduction and Application

Partial least squares regression (PLS regression) is a statistical method that bears some relation to principal components regression; instead of finding hyperplanes of minimum variance between the response and independent variables, it finds a linear regression model by projecting the predicted variables and the observable variables to a new space.

ABSTRACT

IMPORTANT INFORMATION: PARTICIPANTS SHOULD BRING THEIR OWN LAPTOPS. THE SOFTWARE SMARTPLS (free download from www.smartpls.de) MUST BE INSTALLED SOME DAYS IN ADVANCE.

Structural Equation Modeling (SEM) with latent variables has become a quasi-standard in investigating complex causal relationships in business and social research. Recently, PLS Path Modeling has encountered increasing popularity across many scientific disciplines as an easy, yet powerful, estimation technique for structural equation models. It is characterized by robustness, prediction orientation and its capability to estimate formative measurement models.

This seminar provides a profound introduction to PLS Path Modeling. Participants will learn how to apply PLS Path Modeling by means of the SmartPLS software. The seminar contains several SmartPLS examples and exercises. This seminar covers the following topics:

  • An introduction to structural equation modeling (SEM) and PLS Path Modeling
  • Formative vs. reflective measurement models in SEM
  • The PLS algorithm in a nutshell
  • Measurement model evaluation and interpretation of PLS results
  • Significance testing by means of bootstrapping
  • Goodness of Fit in PLS
  • Analyzing mediating effects
  • Analyzing moderating effects

TIME AND LOCATION

Wednesday 4th of July & Thursday 5th of July at Agora (Ag Alfa), University of Jyväskylä (Mattilanniemi 2, Jyväskylä)

SPEAKERS

  • Dr. Jörg Henseler, Institute for Management Research, Radboud University Nijmegen

Organizer: Mikko Siponen, University of Oulu

REGISTRATIONS

This seminar is free-of-charge for INFORTE.fi member organization's staff and their PhD-students. For others participation fee is 700 €. The participation fee includes access to the event and the event materials. Lunch and dinner are not included.

Registrations close on June 29 at 8:00am.

SCHEDULE

Day 1 – 4th of July, 2012

Session 1 09:00 - 10:30

  • Motivation for structural equation modeling
  • Essential characteristics of PLS Path Modeling
  • Formative vs. reflective measurement


Coffee break 10:30 - 10:45


Session 2 10:45 - 12:15

  • Example application
  • Software tutorial: SmartPLS software environment and data entry


Lunch 12:15 - 13:15


Session 3 13:15 - 14:45

  • PLS algorithm essentials
  • Creating valid PLS path models
  • Software tutorial: Creating PLS path models with SmartPLS

 

Snacks and coffee 14:45 - 15:00


Session 4 15:00 - 16:30

  • Reliability and validity assessment
  • Software tutorial: Assessing measurement models with SmartPLS
  • Goodness of Fit

Closing 16:30-17:00


Day 2 – 5th of July, 2012 

Session 5 09:00 - 10:30

  • Interpreting PLS path modeling estimates
  • Bootstrapping
  • Software tutorial: Interpreting SmartPLS output and significance testing


Coffee break 10:30 - 10:45


Session 6 10:45 - 12:15

  • Mediating effects
  • Software tutorial: Modeling mediating effects with SmartPLS
  • Second-order constructs


Lunch 12:15 - 13:15


Session 7 13:15 - 14:45

  • Moderating effects
  • Software tutorial: Modeling moderating effects with SmartPLS
  • Multi-group analysis


Snacks and coffee 14:45 - 15:00


Session 8 15:00 - 16:30

  • Outlook & discussion
  • Wrap up


Closing 16:30-17:00

 

CREDIT POINTS FOR PhD STUDENTS

Doctoral students participating in the seminar can obtain 2 (two) credit points. This requires participating on both days and returning a learning diary after the event (4-5 pages).