Multicollinearity in LGM PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Wendong Li posted on Wednesday, March 17, 2010 - 5:40 am
Dear Dr. Muthen,

I am analyzing a latent growth model with 3 predictors to predict latent growth factors (intercept and slope). I am interested in the interactions between one IV a and the other two IVs, b and c. However, the uncorrected correlations between a and b, and a and c are .58 and .41 respectively. So I am worried about the potential problem of multicollinearity. How can I test that in LGM? Do you think the correlations are too high?

Thank you very much in advance!
 Linda K. Muthen posted on Wednesday, March 17, 2010 - 9:25 am
Those correlations don't sound unusually high to me. I would see a regression book to see what it says about multicollinearity.
 Wendong Li posted on Thursday, March 18, 2010 - 2:00 am
Thanks for the response. Does Mplus provide any diagnosis for this problem?
 Linda K. Muthen posted on Thursday, March 18, 2010 - 7:13 am
No.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: