I used JMP to perform a mixed model logistic regression. The outcome is "NO" (binary outcome for career choice), the fixed effects are a few individual level factors and the random effect is the program someone is at.

If I'm not mistaken, the fixed effects I would report the estimate (beta) and standard error. But how do I report the random effect?

I'm performing this test to try and better understand how much the random effect is impacting my data. I note the that within the model there is 1.36 variance, of which program (RE) comprises 0.35 and the other factors (FE) make up the residual 1.0. Is that correct? Can one say the RE accounts for ~26% (0.35/1.36) of the variance and the FEs account for the 74% in my results?

More Alexander R Cortez's questions See All
Similar questions and discussions