Saturday, March 7, 2009

Correlation doesn't equal causation.

I saw this excellent joke on the Crossfit NYC blog.


This is of course worth remembering when reading scientific studies. Just because x goes up if y goes up it doesn't mean that y causes x to go up. There are lots of examples especially in the world of diet.....

5 comments:

John Hope said...

Just because he definitely deserves the credit for a great comic... This is an xkcd strip - http://xkcd.com/552/

Chris said...

Thanks John.

I've now linked the cartoon to its source

theorytopractice said...

Excellent!

Allison Bojarski, CFNYC said...

Glad you enjoyed it. I thought it very relevant to the crossfit "black box" approach of inputs and outputs--we might not know exactly how and why our specific training protocols are improving our fitness, but if we see that something is working, we go with it. We're not so concerned with what causes what--we care about the results.

helium said...

When reading XKCD comics allways read the image's title tag as well. In this case: "Correlation doesn't imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there'."