Page 1 of 1
Moderating Effects
Posted: Wed Jul 03, 2013 10:45 am
by the_interpreter
I have a model with four LVs. For some reason, SmartPLS will only allow me to create a moderating effect for one of them. It doesn't seem to depend on their relationship with the other LVs, as far as I can see. Could anyone explain to me any characteristics that LVs have to fulfill before you add a moderating effect?
Posted: Sat Aug 17, 2013 1:35 pm
by admin
Hi
That's how we implemented things using an auto function. If you would like to create moderating effects differently, manually create the interaction term, import the dataset with the additional indicators and run the analysis.
Check these articles:
Rigdon, E. E., Ringle, C. M., & Sarstedt, M. (2010). Structural Modeling of Heterogeneous Data with Partial Least Squares. In N. K. Malhotra (Ed.), Review of Marketing Research (Vol. 7, pp. 255-296). Armonk: Sharpe.
Henseler, J., & Chin, W. W. (2010). A Comparison of Approaches for the Analysis of Interaction Effects Between Latent Variables Using Partial Least Squares Path Modeling. Structural Equation Modeling, 17(1), 82-109.
Best
CR
Posted: Thu Sep 19, 2013 5:49 am
by bwilson
admin wrote:Hi
That's how we implemented things using an auto function. If you would like to create moderating effects differently, manually create the interaction term, import the dataset with the additional indicators and run the analysis
A nice SPSS add in by James Jaccard called Zumastat is worth looking at.
Cheap and good for interaction terms and other things.
best regards.