A Monte Carlo Investigation of Partial Least Squares, With Implications for Both Structural and Measurement Models

Partial Least Squares (PLS) is a popular technique with extensive adoption within the Information Systems research community. However, the statistical performance of PLS has not been extensively studied, and recent research has questioned some of its purported advantages. The simulation study reported here analyzed the performance of PLS with regards to the recovery and estimation accuracy of both structural and measurement parameters. Somewhat surprisingly, the effects of estimation bias on the latter and their implications for the evaluation of measurement models have not been the focus of past research. Results show the existence of an important degree of bias in both sets of estimates, and the conflicting effect of increased sample size with additional indicators per composite variable.

[1]  Jacob Cohen.  Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[2]  Herman Wold,et al.  Soft modelling: The Basic Design and Some Extensions , 1982 .

[3]  David F. Larcker,et al.  Structural Equation Models with Unobservable Variables and Measurement Error: Algebra and Statistics: , 1981 .

[4]  William Lewis,et al.  PLS, Small Sample Size, and Statistical Power in MIS Research , 2006, Proceedings of the 39th Annual Hawaii International Conference on System Sciences (HICSS'06).

[5]  Detmar W. Straub,et al.  A Practical Guide To Factorial Validity Using PLS-Graph: Tutorial And Annotated Example , 2005, Commun. Assoc. Inf. Syst..

[6]  Wynne W. Chin,et al.  A Partial Least Squares Latent Variable Modeling Approach for Measuring Interaction Effects: Results from a Monte Carlo Simulation Study and an Electronic - Mail Emotion/Adoption Study , 2003, Inf. Syst. Res..

[7]  C. Saunders,et al.  Editor's comments: PLS: a silver bullet? , 2006 .

[8]  Jan-Bernd Lohmöller,et al.  Latent Variable Path Modeling with Partial Least Squares , 1989 .

[9]  Wynne W. Chin.  The partial least squares approach for structural equation modeling. , 1998 .

[10]  Detmar W. Straub,et al.  Validation Guidelines for IS Positivist Research , 2004, Commun. Assoc. Inf. Syst..

[11]  R. P. McDonald,et al.  Path Analysis with Composite Variables. , 1996, Multivariate behavioral research.

[12]  James C. Anderson,et al.  STRUCTURAL EQUATION MODELING IN PRACTICE: A REVIEW AND RECOMMENDED TWO-STEP APPROACH , 1988 .

[13]  William Lewis,et al.  Research Note - Statistical Power in Analyzing Interaction Effects: Questioning the Advantage of PLS with Product Indicators , 2007, Inf. Syst. Res..

[14]  Detmar W. Straub,et al.  Structural Equation Modeling and Regression: Guidelines for Research Practice , 2000, Commun. Assoc. Inf. Syst..