differences in VIF values using (faraway) and (car) packages

General
1

Howdy folks,

I am trying to detect multicollinearity using VIF values. I tried to find the values using 2 different packages.

library(car); library(carData)
car::vif(glm(median_splitB ~ Education + EmploymentStatus + MasterDataR$VitB12..pmol.L. + MasterDataR$Folate.nmol.L., family=binomial(link='logit'),data=MasterDataR))

                                         GVIF Df GVIF^(1/(2*Df))

Education 2.166349 3 1.137509
EmploymentStatus 1.391349 1 1.179554
MasterDataR$VitB12..pmol.L. 1.471561 1 1.213079
MasterDataR$Folate.nmol.L. 1.225044 1 1.106817

Though, I got different values when I used faraway package.

EducationPreparatory EducationSecondary completed EducationUniversity EmploymentStatusEmployed MasterDataR$VitB12..pmol.L.
11.898977 10.316796 12.880224 8.855899 9.080728
MasterDataR$Folate.nmol.L.
5.783821

which one should be used?!

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.