Wednesday, December 22, 2004

Correlation v. Cause

I know that most people don't have a firm grasp of statistics and probability and quite possibly don't understand exactly what statistical modeling is. But here is one thing that pisses me off.

A correlation does not prove a cause.

That's right. For all those people that don't understand, I'll give you an example. There is a correlation between how much I work and how much sleep I get. That does not mean that my lack of sleep is caused by my being employed. As a matter of fact my lack of sleep is caused by the fact that I stay up and watch moronic TV shows, PodCast with Adam and drink to many alcoholic beverages. My being employed is what causes me to be able to afford to do these things. If you want to prove that one thing causes another you need more than just a correlation.



Blogger Claire said...

dude, that's deep. later dude, Claire

12/23/2004 1:39 PM  

Post a Comment

<< Home

View My Stats