Odds ratios

Hi John:
I just want to understand the term ORs. In logistic regression, we have ORs as effect size, correct? And also, we can calculate ORs based on a 2x2 table. So, apparently, these two are different in terms of how they are calculated. But we can still use these terms? Thank you.


TS Contributor
Yes, odds ratios are analogous to "effect size" and can be interpreted in the same way for logistic regression and for 2x2 tables.

I believe they're computed the same way in both logistic regression and contingency tables - they're simply the ratio of two odds.