Embedded Files
Nature of Research
Nature of Research
Exploratory Quantitative Research
Exploratory Quantitative Research
Research into the unknown.
Predicting, developing, & exploring theory or theories.
Not supported or only partially supported by theory & literature.
What is going on here, how we explain it?
Confirmatory Quantitative Research
Confirmatory Quantitative Research
When we have a pretty good idea about what's going on.
Testing, comparing, & conforming a theory or theories.
Must be supported by strong theory & solid literature.
Is it true what we believe based on the existing theory & literature?
Types of Analysis
Types of Analysis
Source: Hair et al., (2017).
VB-SEM vs CB-SEM
VB-SEM vs CB-SEM
The Choice
The Choice
In case of theory is less developed, researchers should consider VB-SEM as an alternative to CB-SEM.
Once VB-SEM is employed, the study is considered exploratory.
Reasons of Using PLS
Reasons of Using PLS
Chin (1998); Chin & Newsted (1999); Ramayah et al. (2018)
Chin (1998); Chin & Newsted (1999); Ramayah et al. (2018)
The goal is to predict key target constructs or to identify key "driver" constructs.
The research is exploratory or an extension of an existing structural theory.
The phenomenon to be investigated is relatively new & measurement models need to be newly developed.
The structural equation model is complex with a large number of latent variables & manifest variables.
Relationships between the indicators and latent variables have to be modeled in different modes (i.e., formative & reflective).
The conditions relating to sample size & normal distribution are not met.
PLS is especially useful for prediction.
In situations where prior theory is strong and further testing and development are the goal, CB-SEM is more appropriate.
However, for application & prediction, when the phenomenon under research is relatively new or changing, or when the theoretical model or measures are not well formed, a PLS approach is often more suitable (Chin& Newsted, 1999)
Urbach & Ahleman (2010)
Urbach & Ahleman (2010)
PLS makes fewer demands regarding sample size than other methods.
PLS does not require normal-distributed input data.
PLS can be applied to complex structural equation models with a large number of constructs.
PLS is able to handle both reflective and formative constructs.
PLS is better suited for theory development than for theory testing.
PLS is especially useful for prediction.
Page updated
Google Sites
Report abuse