How does one perform logistic regression when your response and predictor variables are factors with two levels.I have three predictor variables with levels yes or no. My response variable is dead or alive. How do I visualise the data and fit an appropriate model?

  • 1
    $\begingroup$ That's not difficult for logistic regression for any software worth mentioning, so what you want to know about how to do it is unclear. Your data define $2$ levels of response and $2^3 = 8$ combinations of three predictors. So, if you show us the combined $2 \times 8$ frequency table, you are likely to get several suggestions. One graph design for the Titanic data (more complicated than yours) is at statalist.org/forums/forum/general-stata-discussion/general/… $\endgroup$ – Nick Cox Apr 16 '18 at 10:20
  • $\begingroup$ Please note that asking how to code something in your favourite software (whatever it is) would be off-topic here. $\endgroup$ – Nick Cox Apr 16 '18 at 10:21
  • $\begingroup$ The graphs mentioned in my first comment are easier to see, and discussed at greater length, in stata-journal.com/sjpdf.html?articlenum=gr0061 $\endgroup$ – Nick Cox Apr 16 '18 at 10:29

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.