Hello all,

I'm looking to run a logistic regression for a smaller sample size (~45-50), with few positive cases for my DV (12.5%). My research indicated that Firth logistic regressions would be best to handle this, unfortunately, I am unable to run a Firth logistic regression with a nominal DV. Is this the case? If so, are there any suggestions for other logistic regression I can run? Or is my sample size simply too small to produce statistically valid logistic regressions?

Best,

This has worked for me by making it a binary variable with 1 and 0 and calling it scale. For example, I have a variable called "Female" with 1 for yes and 0 for no. I hope you've already found a solution by now, but maybe someone else will be asking the same thing