Free downloadable dataset to be used with this book:

Dataset: Restaurant.csv

Dataset: Photocopier.csv

Dataset: Cafe100.csv

Erratum:

p108:  It should be "Figure 92 and 93" instead of Figure 94 and 95 at the top of the page.
p108: Endnote 100 should be “See Figure 94 and 95” instead of Figure 12 and 13.
p111: ...To reject hypothesis "H6 but accept hypothesis H7” instead "H7 but accept hypothesis H8"

Overview:

Partial least squares is a new approach in structural equation modeling that can pay dividends when theory is scarce, correct model specifications are uncertain, and predictive accuracy is paramount. 

 

Marketers can use PLS to build models that measure latent variables such as socioeconomic status, perceived quality, satisfaction, brand attitude, buying intention, and customer loyalty. When applied correctly, PLS can be a great alternative to existing covariance-based SEM approaches. 

 

Dr. Ken Kwong-Kay Wong, an award-winning marketing professor who has taught applied research for more than fifteen years, wrote this reference guide with graduate students and marketing practitioners in mind. Filled with business examples and downloadable datasets for practice, the guide includes step-by-step guidelines for advanced PLS-SEM procedures in SmartPLS, including: CTA-PLS, FIMIX-PLS, GoF (SRMR, dULS and dG), HCM, HTMT, IPMA, MICOM, PLS-MGA, PLS-POS, PLSc, and QEM.

 

Filled with useful illustrations to facilitate understanding, you’ll find this guide a go-to tool when conducting marketing research.

 

Praise for

Mastering Partial Least Squares Structural Equation Modeling (PLS-SEM) 

with SmartPLS in 38 Hours

 

 

“PLS-SEM is a very robust and advanced technique that is well suited for prediction in multi-equation econometric models. This easy-to-read book helps researchers apply various statistical procedures in SmartPLS quickly in a step-by-step manner. I would highly recommend it to all PLS-SEM user.”

— Prof. Dipak C. Jain

President (European) and Professor of Marketing

CEIBS, Shanghai

 

“Having supervised to completion twenty-seven doctoral candidates, of which 70% utilized quantitative methodology using PLS, I wish I had Dr. Wong's book earlier. Mastering Partial Least Squares Structural Equation Modeling (PLS-SEM) with SmartPLS in 38 Hoursprovides all the essentials in comprehending, assimilating, applying and explicitly presenting sophisticated structured models in the most simplistic manner for a plethora of Business and Non-Business disciplines.  Since PLS-SEM quantitative analysis has gained prominence with most top tiered academic journals, this book is a necessity for aspiring academics who wish to have prolific publications in highly ranked publications.”

— Prof. Siva Muthaly

Dean, Faculty of Business & Management

Asia Pacific University of Technology & Innovation, Malaysia

 

“In a world filled with fake-news, academic research results get ever more important. For that reason, key methodologies like PLS-SEM must become available and understood beyond an elite scholar group. Dr. Wong’s book does just that and is therefore highly recommended.”

— Prof. dr. Jack AA van der Veen

Professor of Supply Chain Management

Nyenrode Business Universiteit, The Netherlands

 

“We teach PLS-SEM as part of our Marketing Research course at Seneca and Ken was able to turn this difficult subject into an easy one for our students. Researchers at all levels would definitely benefit from this well-organized book to become competent in this multivariate data analysis method.”

Chris McCracken

Academic Chair, School of Marketing

Seneca College, Canada

 

“A must-have edition for academics and practitioners alike.  Dr. Wong brings a refreshing approach to this important topic supporting a wider application across sectors. The clarity of the content will encourage those new to the field to enhance their skill set with step-by-step support.  The comprehensiveness of the edition will allow it to also serve as a valuable reference for even the most advanced researchers.”

Prof. Margaret D. Osborne

Former Academic Chair, School of Marketing

Seneca College, Canada

 

“Ken Wong has created an easy-to-use, all-in-one blueprint for academics and practitioners on PLS-SEM.”

Prof. Seung Hwan (Mark) Lee

Interim Director

Ted Rogers School of Retail Management

Ryerson University, Canada

 

“Finally, a step-by-step guide to one of the most used method in academia. Life would be much easier for many of us. A must for anyone wanting to know it — well.”

Prof. Terence Tse 

Associate Professor of Finance

ESCP Europe Business School, UK

 

“The new book of Dr. Ken Wong on PLS-SEM is a good contribution to help researchers in the application of this important tool in marketing research. His lucid writing style and useful illustrations make life simple for students, researchers and practitioners alike. Strongly recommended!”

— Prof. Kanishka Bedi

Professor, School of Business and Quality Management

Hamdan Bin Mohammed Smart University, UAE

“In real world scenarios, researchers as well as practising managers have always struggled with actual data that does not mimic the properties of a statistically normal distribution. Ken’s graphic attempt proposing PLS-SEM as a possible alternate solution to identify complex causal relationships is indeed noteworthy, more so due to the book’s hands on approach in using software with enough downloadable data sets to aid the familiarisation process without overwhelming the reader.”

— Prof. Chinmoy Sahu

Dean, Manipal GlobalNxt University, Malaysia

Table of Contents

Foreword

Preface

About the Author

Acknowledgements

Chapter 1 – Introduction

The Research Dilemma

A Better Way to Measure Customer Satisfaction

Different Approaches to SEM

CB-SEM:

PLS-SEM:

GSCA & Other Approaches:

Why not LISREL or Amos?

The Birth of PLS-SEM

Growing Acceptance of PLS-SEM

Strengths of PLS-SEM

Weaknesses of PLS-SEM

Evolution of PLS-SEM Software

Chapter 2 – Understanding the PLS-SEM Components

Inner (Structural) and Outer (Measurement) Models

Determination of Sample Size in PLS-SEM

Formative vs. Reflective Measurement Scale

Formative Measurement Scale

Reflective Measurement Scale

Should it be Formative or Reflective?

Guidelines for Correct PLS-SEM Application

Chapter 3 – Using SmartPLS Software for Path Model Estimation

Introduction to the SmartPLS Software Application

Downloading and Installing the Software

Solving Software Installation Problem on Recent Macs

Case Study: Customer Survey in a Restaurant (B2C)

Data Preparation for SmartPLS

Project Creation in SmartPLS

Building the Inner and Outer Models

Running the Path-Modeling Estimation

Chapter 4 – Evaluating PLS-SEM Results in SmartPLS

The Colorful PLS-SEM Estimations Diagram

Initial Assessment Checklist

Model with Reflective Measurement

Model with Formative Measurement

Evaluating PLS-SEM Model with Reflective Measurement

Explanation of Target Endogenous Variable Variance

Inner Model Path Coefficient Sizes and Significance

Outer Model Loadings and Significance

Indicator Reliability

Internal Consistency Reliability

Convergent Validity

Discriminant Validity

Checking Structural Path Significance in Bootstrapping

Multicollinearity Assessment

Model’s f2Effect Size

Predictive Relevance: The Stone-Geisser’s (Q2) Values

Total Effect Value

Managerial Implications - Restaurant Example

Chapter 5 – Evaluating Model with Formative Measurement

Different Things to Check and Report

Outer Model Weight and Significance

Convergent Validity

Collinearity of Indicators

Model Having Both Reflective and Formative Measurements

Chapter 6 – Determining Measurement Model Using Confirmatory Tetrad Analysis (CTA-PLS)

Formative or Reflective? Determining the Measurement Model Quantitatively

Case Study: Customer Survey in a Café (B2C)

CTA-PLS Procedures

Chapter 7 – Handling Non-Linear Relationship Using Quadratic Effect Modeling (QEM)

Non-linear Relationship Explained

QEM Procedures

Chapter 8 – Analysing Segments Using Heterogeneity Modeling

Something is Hiding in the Dataset

Establishing Measurement Invariance (MICOM)

A. Modeling Observed Heterogeneous Data

Permutation Test Procedures

B. Modeling Unobserved Heterogeneous Data

(i) FIMIX-PLS Procedures

(ii) PLS-POS Procedures

(iii) Ex-post Analysis

Chapter 9 – Estimating Complex Models Using Higher Order Construct Modeling (HCM)

Case Study: Customer Survey in a Photocopier Manufacturer (B2B)

Conceptual Framework and Research Hypotheses

Questionnaire Design and Data Collection

Hypotheses Development

PLS-SEM Design Considerations

Sample size

Multiple-item vs. Single-item Indicators

Formative vs. Reflective Hierarchical Components Model:

Data Preparation for SmartPLS

Data Analysis and Results

PLS Path Model Estimation:

Indicator Reliability

Internal Consistency Reliability

Convergent Validity

Discriminant Validity

Collinearity Assessment

Coefficient of Determination (R2)

Path Coefficient

Predictive Relevance (Q2)

The f2and q2Effect Sizes

Chapter 10 – Mediation Analysis

Customer Satisfaction (SATIS) as a Mediator

Magnitude of Mediation

Chapter 11 – Comparing Groups Using Categorical Moderation Analysis (PLS-MGA)

Multi-group Analysis – “Business Type” in the Photocopier Manufacturer Example

Summary of Hypothesis Testing

Managerial Implications for the Photocopier Manufacturer

Chapter 12 – New Techniques in PLS-SEM

Estimating Factor Models Using Consistent PLS (PLSc)

Assessing Discriminant Validity Using Heterotrait-Monotrait Ratio of Correlations (HTMT)

HTMT Procedures

Contrasting Total Effects Using Importance-Performance Matrix Analysis (IPMA)

IPMA Procedures

Testing Goodness of Model Fit (GoF) Using SRMR, dULS, and dG

GoF Procedures

Chapter 13 – Recommended PLS-SEM Resources

Books

Conferences

Discussion Forums

Training workshops

Software

Reference Journal Papers

Conclusion

Epilogue

Life after PLS-SEM?

References

Subject Index