Information Resources Management

 
 
 
 
Workshop #4
 
An Overview of Partial Least Squares Path Modeling

Wynne Chin

Department of Decision and Information Sciences
C.T. Bauer College of Business
University of Houston
 
 
            Structural Equations Modeling (i.e., causal modeling) is rapidly becoming a methodological technique of choice among social scientists.  Among those researchers using the causal modeling approaches, the covariance based technique as exemplified by LISREL is the more widely known and employed.  It is argued that the lesser known technique of Partial Least Squares (PLS) can often be more appropriate for researchers.
              PLS path modeling is well-suited if your research involves multiple indicators of latent variables or constructs, if you wish to account for measurement error among the indicators, and if you wish to estimate the reliability and validity of these measurements within the context of your theoretical model.  Finally, PLS comes to the fore if you are faced with very complex models with large number of constructs and indicators, data conditions that preclude the necessary conditions for using LISREL (i.e., non-normality and smaller sample size), and would like to create construct scores for predictive purposes.  
            The objective of this seminar is to present a state of the art overview of the Partial Least Squares method and position it among other analytic techniques as well as within IS research.  Furthermore, this presentation will highlight the philosophical and operational issues that should be considered when employing this technique.                        
Attendees of this session should come away with the following:
 
·        an understanding of where PLS stands in relation to other multivariate techniques with an initial brief introduction to regression and factor analysis.  
 
·         an understanding of the conditions when PLS is appropriate for analyses.  
 
·         an understanding of causal diagrams.  
 
·         the general algorithm by which PLS generates its results and implications for sample size, data distributions and epistemological relationships between measures and concepts.  
 
·         how PLS and LISREL compare and compliment one another.  
 
·         how to assess and report PLS analysis including statistical re-sampling using jackknifing, bootstrapping, and blindfolding.  
 
·         a demonstration of the PLS-Graph software package for interactive PLS analyses focusing on all features.    
                                                                                  ==========================================
 
B I O   O F   D R   W Y N N E   C H I N
==========================================  
Wynne W. Chin is the C. T. Bauer Professor of MIS in the department of Decision and Information Sciences in the C.T. Bauer College of Business at the University of Houston and recently the Faculty Senate President. He received his A.B. in Biophysics from U.C. Berkeley, MS in Biomedical/Chemical Engineering from Northwestern University, and an MBA and Ph.D. in Computers and Information Systems from the University of Michigan. Wynne has also taught previously at the University of Calgary, Wayne State University, and the University of Michigan and has been a visiting fellow at the University of Canterbury, Queens University, City University of Hong Kong, and the University of New South Wales. 
 
 
Wynne's research focuses on structural equation models related to Information technology adoption, sales force automation and electronic meeting support systems where he has developed measures for group cohesion, satisfaction, and consensus. Wynne has published in journals such as Information Systems Research, Data Base, Journal of Management Information Systems, MIS Quarterly, and Decision Sciences. Wynne is on the editorial board of Structural Equation Modeling journal, Journal of Information Technology, IEEE Transaction of Management, and previously co-editor of Data Base and on the boards of Information Systems Research, Journal of AIS and MIS Quarterly.
 
 
Wynne has received best paper awards from the Journal of Personal Selling and Sales Management in 2003, the Administrative Sciences Association of Canada (IT division) in 1993 and 1998, a MIS Quarterly Reviewer of the Year in 1996, a Management Science Outstanding Reviewer award in 1996, and the First Biennial Award for Outstanding Ph.D. dissertation from the International Communication Association's Communications and Technology Division. He is also the developer of PLS-Graph, a graphical based software to perform Partial Least Squares analysis and, according to Google Scholar, his 1998 PLS book chapter already has approximately 1450 citations to date.