VIF cannot be used on categorical data. ... If you want to check independence between 2 categorical variables you can however run a Chi-square test.

How do you deal with multicollinearity in categorical variables?

Conclusion: In this article, we have discussed how to avoid multicollinearity in categorical data. Lasso regularization can also be used to avoid the effect of multicollinearity. By choosing the correct value of alpha for lasso regularization the weight of the extra features will be equated to zero.

Does multicollinearity apply to categorical variables?

Multicollinearity means "Independent variables are highly correlated to each other". For categorical variables, multicollinearity can be detected with Spearman rank correlation coefficient (ordinal variables) and chi-square test (nominal variables).