Full Collinearity VIF

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
Joewin
PLS Junior User
Posts: 6
Joined: Fri May 29, 2015 11:08 am
Real name and title: Joseph Au

Full Collinearity VIF

Post by Joewin »

I'm thinking of checking the CMV of my research and happened to notice that there is a paper of using "Full Collinearity VIF" to assess the CMV.

Author = Kock, N.
Title = Common method bias in PLS-SEM: A full collinearity assessment approach
Journal = Laredo, TX: ScriptWarp Systems
Year = 2015
Pages = 1-11

The information was kindly provided by one of the expert here --> viewtopic.php?f=7&t=21&p=11792&hilit=cmv#p11792

The question is, can smartpls provide the "Full Collinearity VIF" figures? If yes, how can I get it?

Thanks!
jmbecker
SmartPLS Developer
Posts: 1282
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Full Collinearity VIF

Post by jmbecker »

It is not implemented in SmartPLS. We think that there is not sufficient evidence of the usefulness of the "Full Collinearity VIF" statistics and we do not want to bother researchers with unestablished indices.
However, it is quite easy to calculate. To get these VIFs you need to copy the latent variable correlations (which you get in the PLS output under Latent Variables > second Tab “Latent Variable Correlations”) into a software that can calculate matrix operations (as for example MS Excel). Then simply calculate the matrix inverse of the correlation matrix of all latent variables. The diagonal elements are the "Full Collinearity VIF" values for each LV.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Joewin
PLS Junior User
Posts: 6
Joined: Fri May 29, 2015 11:08 am
Real name and title: Joseph Au

Re: Full Collinearity VIF

Post by Joewin »

jmbecker wrote: Wed Jan 31, 2018 1:16 pm It is not implemented in SmartPLS. We think that there is not sufficient evidence of the usefulness of the "Full Collinearity VIF" statistics and we do not want to bother researchers with unestablished indices.
However, it is quite easy to calculate. To get these VIFs you need to copy the latent variable correlations (which you get in the PLS output under Latent Variables > second Tab “Latent Variable Correlations”) into a software that can calculate matrix operations (as for example MS Excel). Then simply calculate the matrix inverse of the correlation matrix of all latent variables. The diagonal elements are the "Full Collinearity VIF" values for each LV.
Thank you professor jmbecker! I'll follow your suggestion.
Post Reply