0
$\begingroup$

I have a quasibinomial glm and I'd like to get odds ratios for each of the predictors. Is it acceptable to use the 'oddsratio' package in R in the same method that you would for a binomial model (e.g. https://pat-s.github.io/post/oddsratio/)?

If not, is there a better approach to doing this?

$\endgroup$
0
$\begingroup$

You can get the odds ratio from the exponential of the coefficients given from the glm, so you'd have:

model1 <- x ~ y, data = data, family = quasibinomial(link = 'logit')
exp(coef(model1))
$\endgroup$

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.