If the research is exploratory or an extension of an existing structural theory, select PLS-SEM.РІР‚Сњ.If the goal is theory testing, theory confirmation, or comparison of alternative theories, select CB-SEM.If the goal is predicting key target constructs or identifying key 'driver' constructs, select PLS-SEM.The primary purpose of the PLS approach is to predict the indicators by means of the components expansion (1).РІР‚Сњ JР“В¶reskog and Wold, 1982 p. The most important reason to select CB-SEM or PLS-SEM is the research goal (structure or prediction): РІР‚СљThe primary purpose of the ML approach is to study the structure of the observables. (2009) provide recommendations when to use CB-SEM or PLS-SEM.

(2014), JР“В¶reskog and Wold (1982), and Reinartz et al.

For example, Fornell and Bookstein (1982), Chin and Newsted (1999), Hair et al. But PLS-SEM in its original form (Wold, 1982 LohmР“В¶ller, 1989) has not been created to mimic CB-SEM! PLS-SEM researchers should follow RigdonРІР‚в„ўs (2012) call and begin emancipating the method from its CB-SEM sibling (also see Sarstedt et al., 2014 Rigdon 2014). Instead of distinguishing between common factor models and composite models (Henseler et al., 2014), these discussions focus on PLS-SEMРІР‚в„ўs capabilities to mimic CB-SEM. Even though this issue is well-know (see for example JР“В¶reskog and Wold, 1982), many researchers still focus on comparing the differences of model estimations when using covariance-based structural equation modeling (CB-SEM) and partial least squares structural equation modeling (PLS-SEM). Both methods are complementary rather than competitive.

Smartpls example