Introduction to Partial Least Squares / Structural Equation Modelling (PLSSEM) using Stata

PLS-SEM, also referred to as partial least squares path modelling, is a type of SEM, which is being increasing used in social sciences, psychology, business administration and marketing. In a nutshell, PLS-SEM can be viewed as a component-based SEM alternative to the covariancebased structural equation modelling (CB-SEM) which can be described as a factor-based SEM technique. As such, the PLS-SEM approach provides researchers with a multivariate statistical technique that can readily be used to estimate exploratory or/and complex SEM models. Although there are several stand-alone specialized PLSSEM software packages available, this course introduces participants to the PLS-SEM methodology, through the user-written Stata-package, plssem, developed by the course instructors themselves.

In common with TStat’s workshop philosophy, throughout the workshop, theoretical sessions are reinforced by case study examples, in which the course tutor discusses current research issues, highlighting potential pitfalls and the advantages of individual techniques. In this manner, course leaders are able to bridge the “often difficult” gap between abstract theoretical methodologies, and the practical issues one encounters when dealing with real data.

At the end of the course, participants are expected to be able to autonomously implement the theories and methodologies discussed during the workshop.

The PLS-SEM workshop is of particular interest to researchers and professional working in social sciences, psychology, business administration, marketing and management. Due to its introductory nature however, is it also accessible to individuals, regardless of their respective disciplines or fields, who need to acquire the requisite toolset to apply the PLS-SEM methodology to their own data.

When possible, participants should bring their own datasets to the workshop to work with and discuss with the instructors.

It is assumed that participants have previously followed a basic course in statistics. Previous exposure to Stata or other statistical software packages would also be an advantage.

SESSION I: INTRODUCTION

What is structural equation modeling (SEM)?

Different approaches to SEM

What is PLS-SEM?

PLS-SEM versus CB-SEM

SESSION II: BASIC CONCEPTS

Regression

Principal component analysis

Path analysis

Bootstrapping

Reflective and formative measures

SESSION III: DEVELOPING AND ASSESSING A PLS-SEM MODEL

Developing the model

Specification

Example study and measures

Estimation using plssem package in Stata

Assessing the model

Measurement model

Construct and discriminant validity

Structural model

Goodness of fit

Path coefficients

SESSION IV: ADVANCED PLS-SEM MODELS USING plssem PACKAGE IN STATA

Mediation analysis

Barron and Kenny approach and its alternatives

Mediation analysis with observed variables

Mediation analysis with latent variables

Multiple sample models

Multi-group approach

MIMIC approach

Higher-order factor models

Second-order factor models

Interaction-based models

Product-term approach

PLS-SEM models including categorical variables

SESSION V: HOW TO PUBLISH A PLS-SEM STUDY

Scientific journal criteria

Example studies

SESSION VI: HOW TO USE STORED INFO FROM plssem PACKAGE

Accessing scalars, macros and matrices