Hi mates
I've read the handbook and the first step with PLS is to check collinearity problems. If they exists:
a) Remove predictors
b) Join predictors
c) Build HOC constructs
In my model I've predictors with VIF > 5 or 10 (inner model) and also items (outer).
I've removed the two predictors with collinearity problems and now my model seems to be ok. However, some of predictors I retain have items with VIF higher than 5 or 10. Given that the inner model is ok, should I concern about these items? Could I proceed with my model? Only reflective relationships.
Thank you in adavance.
EDIT:
After reading the following post, I realized that for my reflective model I only have to check the VIF of the inner model:
viewtopic.php?f=3&t=3479&p=11524&hilit= ... ity#p11524
How to deal with collinearity problem?
-
- SmartPLS Developer
- Posts: 1287
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: How to deal with collinearity problem?
Yes, you are right: VIF is only a concern for formative outer models. For reflective models, you only have to check inner model VIFs.
Rather than just deleting you should think about the possible reasons for collinearity (are the predictors conceptually very similar?) and may rather combine the constructs or build HOC models. If you simply delete you will like loose information along the way.
Rather than just deleting you should think about the possible reasons for collinearity (are the predictors conceptually very similar?) and may rather combine the constructs or build HOC models. If you simply delete you will like loose information along the way.
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
Re: How to deal with collinearity problem?
Thank you Prof. Becker