My understanding is that the exp(B) produced in multinomial regression in SPSS (OR) are equal to RR is that true
many thanks in advance for your help
Hello Walid A. Al-Qerem. Exp(B) = an odds ratio for both binary and multinomial logistic regression. Some software, though, describes Exp(B) as the relative risk ratio (RRR, not RR). E.g., the output from Stata's -mlogit- command does that. I wrote the attached notes for myself after I finally understood why Stata calls it RRR. Perhaps you will find it helpful. HTH.
I found this distinction, in Bruce Weaver 's document, really useful, and am reproducing it here for anyone perusing the question.
- Risk ratio = risk of an event in one group divided by risk of the same event in another(independent) group
- Relative-risk = risk of some event in one group divided by the risk of a different event(e.g., a non-event) in the same group
This is the key to understanding Stata's use of the relative risk ratio, and would have been really useful to me when fending off reviewer comments!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS