how to do PLS Predict with moderator in SmartPLS4
Posted: Sun Jun 12, 2022 4:32 am
Hi,
I just installed and tried the SmartPLS4. I want to run PLS Predict with moderator interaction terms. I tried with SmartPLS3, kept getting error message. But it worked with SmartPLS4. However, i am not sure if it is the right way to do it. I read some papers that have done PLS Predict in their analysis, none states whether they have included the interaction terms in their model. In my understanding, the model in PLS Predict should be similar as the structural model. In my case, it's disjoint 2-stage approach with interaction terms.
If i am not mistaken, I read in Becker's paper that it is possible to use interaction terms in PLS Predict..
"Similarly, future research should investigate the approaches and data treatment options from an out-of-sample prediction perspective (Shmueli, Ray, Velasquez Estrada, & Chatla, 2016). In doing so, it would be particularly interesting to investigate the impact of different weighting schemes in the context of our research question. By default, the estimation of reflectively specified constructs draws on Mode A, whereas PLS-SEM uses Mode B for formatively specified constructs. However, Becker et al. (2013) show that this reflex-like use of Mode A and Mode B is not optimal in terms of out-of-sample prediction under all conditions."
Becker, J. M., Ringle, C. M., & Sarstedt, M. (2018). Estimating moderating effects in PLS-SEM and PLSc-SEM: Interaction term generation* data treatment. Journal of Applied Structural Equation Modeling, 2(2), 1-21.
I just installed and tried the SmartPLS4. I want to run PLS Predict with moderator interaction terms. I tried with SmartPLS3, kept getting error message. But it worked with SmartPLS4. However, i am not sure if it is the right way to do it. I read some papers that have done PLS Predict in their analysis, none states whether they have included the interaction terms in their model. In my understanding, the model in PLS Predict should be similar as the structural model. In my case, it's disjoint 2-stage approach with interaction terms.
If i am not mistaken, I read in Becker's paper that it is possible to use interaction terms in PLS Predict..
"Similarly, future research should investigate the approaches and data treatment options from an out-of-sample prediction perspective (Shmueli, Ray, Velasquez Estrada, & Chatla, 2016). In doing so, it would be particularly interesting to investigate the impact of different weighting schemes in the context of our research question. By default, the estimation of reflectively specified constructs draws on Mode A, whereas PLS-SEM uses Mode B for formatively specified constructs. However, Becker et al. (2013) show that this reflex-like use of Mode A and Mode B is not optimal in terms of out-of-sample prediction under all conditions."
Becker, J. M., Ringle, C. M., & Sarstedt, M. (2018). Estimating moderating effects in PLS-SEM and PLSc-SEM: Interaction term generation* data treatment. Journal of Applied Structural Equation Modeling, 2(2), 1-21.
- actual third order PLSPREDICT 06062022 with interactions model 1A.png (88.98 KiB) Viewed 1888 times