I want to fit logistic regression model for a binary outcome variable to see the impact of other explanatory variable on it. What are selection criteria for selecting variables in the adjusted logistic regression model? What is the basic difference between unadjusted and adjusted logistic regression model?


1 Answer 1


From what I can tell, there is no difference between unadjusted and adjusted logistic regression; to tell what happens when you add a variable to a logistic regression model, you simply add that variable to the model and see what happens.

You have to be careful about overfitting an colinearity, but that is the basic idea.

  • $\begingroup$ Thank for your comment. But how can I find model which fits the data set best? Is there rule of thumb for fitting best model?@Peter Flom $\endgroup$
    – JRK
    Mar 5, 2014 at 11:23
  • $\begingroup$ The model with the most variables will fit best, but that may be overfit. There are various ways to balance fit to the data and complexity, look up AIC, AICc, BIC, etc. $\endgroup$
    – Peter Flom
    Mar 5, 2014 at 11:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.