topic Re: collinearity between two variables in Statistical Procedures

collinearity between two variables

sebai — Tue, 04 Sep 2018 21:54:06 GMT

I'm creating a regression model and I want to test if some co-variables are linear or co-dependent? I am thinking, among other things, of '' access to land '' and '' possession of livestock ''.

Which test can I use to show the collinearity between two categorical variables?

Thanks

Re: collinearity between two variables

PaigeMiller — Tue, 04 Sep 2018 22:31:02 GMT

The Chi-squared test in PROC FREQ can test the independence of two categorical variables.

Re: collinearity between two variables

sebai — Tue, 04 Sep 2018 22:37:12 GMT

oh ok, thank u so much for ur answer!

Then if the p-value of the chi-square test is less than 0.05 ==> Can I assume that there is a collinearity between the two variables

and if the p-value is higher than 0.05 ==> then the 2 variables are independent (no collinearity)

Is this right!?

Re: collinearity between two variables

PaigeMiller — Tue, 04 Sep 2018 22:40:52 GMT

Collinearity is not the right word for categorical variables.

p<0.05 indicates lack of independence.