Should PLS become factor-based or should CB-SEM become composite-based? Both!

Abstract

ABSTRACT This paper critically discusses a recent EJIS paper entitled “Will PLS have to become factor-based to survive and thrive?”, which laments that covariance-based structural equation modelling cannot handle composite models, postulates that PLS should become factor-based, and advocates the use of the PLSF-SEM method. We clarify that covariance-based structural equation modelling can handle composite models, e.g. using the so-called Henseler – Ogasawara specification, agree that PLS in its original form cannot consistently estimate the parameters of reflective measurement models, and reiterate that consistent PLS is a viable means to overcome this limitation. Finally, we study the performance of PLSF-SEM through scenario analyses. Based on our results, we conclude that PLSF-SEM is inferior to covariance-based structural equation modelling and consistent PLS in almost all respects. Therefore, we cannot currently recommend the use of PLSF-SEM, and it is up to future research to demonstrate potential advantages over existing structural equation modelling techniques.