Page 2 of 2

Posted: Wed Jun 30, 2010 5:14 am
by jarhead01e
I followed the Dr Bido's instructions above, along with the utube video, and ran PLS to assess for CMB, but I think something is wrong. Almost all of my AVE and communality scores are 1.00000 The METHOD AVE and communality is .52

Does that seem logical or do you think I built the model incorrectly?

Posted: Tue Dec 07, 2010 1:37 pm
by mronkko
This is bit of a bumping an old thread, by we have published a new version of our PLS CMV paper. Also analyzed the Liang et al method and found out that it does not work the way that the paper argues it does. See this thread and the links there.


Posted: Mon Dec 27, 2010 2:20 pm
by christian.nitzl
Thank you very much for sharing the document!

Best regards,


Posted: Wed Jun 27, 2012 8:35 pm
by stefanperras
Just to clarify:
If the model includes constructs that are measured through formative indicators, a statistical ex-post identification of common method bias is not possible with PLS.

For explanation see page 900 of
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903


Posted: Sat Aug 04, 2012 9:58 pm
by Hengkov
See Chin et al. 2012 article International Symposium USA for test CMV with Latent Merker Variable Approach and in MIS Quarterly fortchoming.

Re: How to test for Common Method Variance in PLS?

Posted: Mon Sep 26, 2016 1:54 pm
by mak

I know it is quite an old threat but I am struggling with the same problem and I was wondering whether I could get an access to the youtube video as for now it is private...

It would be super helpful!

Thank you in advance!



Posted: Tue Sep 10, 2019 5:33 am
by shikhab
christian.nitzl wrote:
Sat Jun 12, 2010 8:57 pm
Hey Criston,

It is easy to produce a LV with "repeated indicator”. You have only to take a second time the indicators which you already used (drag and drop).

How it works you can see here:

Furthermore there is an working paper that could be interesting for the discussion her:

Ylitalo (2009): Controlling for Common Method Variance with Partial Least
Squares Path modeling: A Monte Carlo Study; ... yli09b.pdf

Best regards

Dear Prof. Christian,

Thanks a lot for clarification. Can you please reshare these links as they are not accessible anymore.

Also, please confirm in order to test CMB, do we take take square of path coefficients of method and substantive constructs.

Thanks a lot,