forum.smartpls.com

Hi,

I am testing moderation where the independent variable(IV), moderating variable(M) and dependent variable (DV) are higher order constructs. The IV is formative-formative construct and the M is reflective-reflective construct and the DV is reflective-reflective construct. Can someone please explain how to test moderation in such a scenario. I have read about 2 stage process and feel might be the right approach but it will be great if someone can explain the step wise process or point towards literature that explains the entire process.

Thanks,
Arjun

As far as I know there is no literature on using higher-order components in moderation analysis. We covered this topic briefly in the future research section of our LRP paper (Becker, J. M., Klein, K., & Wetzels, M. (2012). Hierarchical latent variable models in PLS-SEM: guidelines for using reflective-formative type models. Long Range Planning, 45(5-6), 359-394.), but and I'm not aware of a study that has actually looked at these problems.

I would say that it would make sense to combine the two two-stage approaches into basically a three stage approach or use the two-stage moderator approach with a repeated indicator higher-order construct. Both should work ok (without having data to back this up).
I would not use the orthogonalizing or product-indicator approach with higher-order components, especilly if you use repeated-indicators. That would like mess things up (but I might be wrong). In addition, most current recommendations anyway suggest to use only the two-stage moderator approach (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, 1-21.)

Thank you so much for clarifying. Can I also request you to suggest on how to report moderation in such cases?

In just the normal way.