Yeah, I was assuming tongue in cheek with the last paragraph. I think the correlation-causation example that really drove it home for me was exercise and health. Apparently healthy people are better at exercise, enjoy it more, and thus do it more. So for the majority of exercise and health studies the health was 'pre-existing' and the exercise is just a marker. Now, of course there's the dose response of exercise and health, but by and large health makes people exercise, and less so vice-versa.
Comments 2
Reply
Reply
Leave a comment