I'm running a mixed-effect logistic regression in Stata 16.0.

I can't say much about the data because it's confidential. Nevertheless, here is a fictional example: If we sample students within schools, and include a random intercept for "School", why might a key fixed effect predictor in our models (e.g., food security level on GPA) shift majorly, versus when we did not include a random intercept?

I'm trying to work this out conceptually to explore my own data. If anyone can explain this in simple conceptual terms, that would be greatly appreciated.


Your Answer

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

Browse other questions tagged or ask your own question.