1

Problem: I need to figure out how many times are women more likely to continue attending a class after a certain period of time.

Data:enter image description here

Solution: I used odds ratio (see picture).

Is it correct to assume that women are 2.8 times more likely to continue with the class compare to men?

enter image description here

Cohort 1: 0.59 / 0.34 = 1.7

Cohort 2: 0.40 / 0.65 = 0.61

Odds Ratio = 1.7 / 0.61 = 2.8

Petra
  • 11
  • Please see the beginning of my post at https://stats.stackexchange.com/a/133633/919 for formulas relating odds to probability. – whuber Mar 30 '24 at 13:16

1 Answers1

1

No. Odds is not the same as "times more likely".

About 59% of women stayed, while about 35% of men did. 59 is not 2.8*35.

What you seem to want is relative risk. This is not the same as odds ratio. See e.g. Ranganathan et al (2001) Common Pitfalls in Statistical Analysis: Odds vs. Risk or many other articles.

Peter Flom
  • 119,535
  • 36
  • 175
  • 383