forum.smartpls.com

Dear all friends and colleges,

I want to test the effect of AMT on performance. The AMT has 8 dimensions and each dimension has 4 items. I want to analyze the effect of AMT on performance as well as the effect of AMT dimensions on performance. In result of PLS, Some path coefficients are problematic and they are more than 1. In order to solve this, I think I should use two separate PLS models. is it correct? if not could you please guide me what should I do?
Thank you very much
Regards

How do you model AMT? It seems like it is a second-order construct. Do you use the repeated-indicator approach or the two-stage approach?

I good read in this context is the following article:
Becker, Jan-Michael, Kristina Klein and Martin Wetzels (2012): Hierarchical Latent Variable Models in PLS-SEM: Guidelines for Using Reflective-Formative Type Models, Long Range Planning, Vol. 45, Issue 5-6, 359-394. http://dx.doi.org/10.1016/j.lrp.2012.10.001

Dear Dr. Jan-Michael Becker,
Thank you very much for your reply. Yes AMT is a second-order construct and I want to use repeated-indicator approach. The question here is that can I have a path from first-order construct ( one of AMT dimensions) to the performance?

Thank you once again
Best regards

Usually, you would not model direct effects of your first-order dimensions on other endogenous constructs (like performance), but only the effect of the second-order construct (AMT). Otherwise, the model would not make much sense.
You should assess the effects of the first-order dimensions on performance by using the indirect effects.

Dear Dr. Jan-Michael Becker,
Thank you very much for your reply and great guidance.Now my problem is solved and I know what should I do exactly.
Thank you so much once again
Best regards