I got these results from a logistic regression in R. The data are the proportion of women elected in Uk elections, according to their party.

enter image description here

As you can see, I used exp() to get rid off of the log function, so these are the odds, not the log odds.

How can I interpret these, since I don't see how I can say that for every 1 unit increase in party, there is an increase in the odds. Are these odd ratios comparing with each other, I mean comparing with the odd ratios for all other parties?

  • $\begingroup$ Please type your question as text, do not just post a photograph or screenshot (see here). $\endgroup$ Sep 3 '20 at 19:03

This isn't really a 'discrete' variable, it's a categorical variable. Only the intercept is the odds of a woman (in the reference level party) being elected. The other coefficients are odds ratios. You multiply those odds ratios times the odds in the intercept to get the odds of a woman in the, say, green party being elected (in that case the odds is .3125, or about 24%).

It may help you to read:

  • $\begingroup$ Thanks man. First let me tell you that I consider you a legend haha. Just another silly question: where that 24% comes from? And, just to make sure, R compares every of those values against the reference that is in this case the Conservative Party, which the intercept, right? $\endgroup$ Sep 3 '20 at 22:51
  • $\begingroup$ @RafaelHernándezSalazar, the indicated levels are compared to the reference level. To convert from odds to a probability, you do o/(1+o) (see the thread linked in my answer above). $\endgroup$ Sep 4 '20 at 1:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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