Page 1 of 1

second order reflective-reflective model for PLS-SEM

Posted: Thu Apr 11, 2019 9:22 am
by momoko
Dear all,

I have a question about my model testing. I have a second order reflective-reflective model in my paper and my sample very small (less than 200) so I decided to use PLS. However, when I tell my supervisor she disagrees with that because she said she has not seen any article or paper use PLS to test a second order reflective-reflective model before and PLS is not suitable for mature theory testing.
So I really want to know can I use this, any drawback with it and are there any paper can support it?
Thank you for all your helping.

Best Regards,
Angel

Re: second order reflective-reflective model for PLS-SEM

Posted: Sat Apr 13, 2019 3:37 pm
by jmbecker
There are certainly many paper that use PLS to estimate reflective-reflective higher-order models. You might also find some in your field.

Whether to use or not to use PLS in a purely reflective model is an ongoing debate. You will find some arguments and literature here:
https://www.smartpls.com/documentation/ ... with-cbsem
https://www.smartpls.com/documentation/ ... and-cb-sem

and in this recent publication:
Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. European Business Review. doi:10.1108/EBR-11-2018-0203