WarpPLS and SmartPLS

[homicro]

Hello there,

I have some questions that I wish you answer them:

1- Why some relationships work with WarpPLS but do not work with SmartPLS (same data)?

2- Some items load well in SPSS, but do not load in SmartPLS?

3- If I have a construct that has three sub-constructs, shall I add only items?


Looking forward to hear from you

[jmbecker]

1) WarpPLS does not only offer PLS algorithms, but also some other variants that are not normal PLS. Some of these other variants are not (yet) published and/or validated and I would be careful in using them. (This does not mean that they are not good, but only that little is known about their behavior).

2) SPSS does not offer PLS-SEM algorithms. If you do an exploratory factor analysis in SPSS, it is a different algorithm with different assumptions. If you would do SPSS and a CFA in AMOS or MPLUS, you would likely also get different results.

3) I'm sorry, but I don't understand the question. It sounds like you have a second-order construct. There is a chapter in the new PLS book on higher-order constructs:
Hair, J. F., Sarstedt, M., Ringle, C. M., & Gudergan, S. P. (2018). Advanced Issues in Partial Least Squares Structural Equation Modeling (PLS-SEM). Thousand Oaks, CA: Sage. https://www.smartpls.com/documentation/advanced

In addition, the following article may also be helpful for you:
Becker, Jan-Michael, Kristina Klein, and Martin Wetzels. "Hierarchical latent variable models in PLS-SEM: guidelines for using reflective-formative type models." Long Range Planning 45.5 (2012): 359-394.

[homicro]

Thanks for your reply.

As for the third question, I have a construct with three sub-constructs. I do not know how to model this?

Looking forward

[jmbecker]

It sounds like a second-order construct and the provided literature might help you.