Can anyone explain why we can't use odds ratio in cohort study using simple logical approach rather than statistical one?
I read that the investigator can change the number of control and experiment to his favor, and that is why we can't use risk ratio. It does not make sense to me.
Thanx sooo much. It makes sense. Risk referes to the future so cohort study. It may not academic, but at least makes sense
You can in fact speak of a causal odds ratio in a cohort study.It is just the ratio of the risk of being a case to the risk of not being a case in the exposed compared to the same ratio in the non exposed..
tTe relations between rate ratio, risk ratio and odds ratio are very well treated in Rothman and Greenland, Modern Epidemiology 2nd ed p 48-50 and 244-247
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis