Hi,

I wonder whether anybody can help me. I am trying to run a small sample (n = 45) binary logistic regression, with a mixture of continuous and categorical predictors. One categorical predictor (2x2) has an empty cell, and leads to an incredibly large standard error. I would like to follow Firth (1993) and add .5 to the cells, but there seems to be no way to do it in SPSS (apart from running a multinomial regression) - any suggestions?

Many thanks,

Seb

Topic