I was trying to calculate the VIF scores and understand the rationale beneath it.
Am using a reflective model so am only calculating VIF between latent variables. Now am used to VIF values per variable but PLS provides VIF values for each path. What is the formula thats being used? I tried using 1/1-R square and it didnt get me the right answer.
Also, if VIF scores are below 3.3 (Kock, 2015) then I shall assume no multicollinearity?
The VIF scores
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: The VIF scores
1) VIF values should be reported for each independent variable in all regressions on each dependent variable. As a variable can be exogenous to multiple dependent variables in PLS we provide several VIFs values for each regression equation.
2) We use the standard formula and you should get the same results. You need to take the standardized latent variables scores and be careful with missing values. If you calculate outer model VIFs: be carefull to also take standardized indicators as there may be a difference between unstandardized and standardized estimates.
A hint: The R² value is the regression of all other independent variables on the focal variable for a particular regression.
2) We use the standard formula and you should get the same results. You need to take the standardized latent variables scores and be careful with missing values. If you calculate outer model VIFs: be carefull to also take standardized indicators as there may be a difference between unstandardized and standardized estimates.
A hint: The R² value is the regression of all other independent variables on the focal variable for a particular regression.
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
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de