You're looking at Less Wrong's discussion board. This includes all posts, including those that haven't been promoted to the front page yet. For more information, see About Less Wrong.

Covariance in your sample vs covariance in the general population

27 RomeoStevens 16 May 2012 12:17AM

A popular-media take on a subtle problem in sampling.  I found the graph quite illustrative.

http://www.theatlantic.com/business/archive/2012/05/when-correlation-is-not-causation-but-something-much-more-screwy/256918/

Comments (3)
Powered by Reddit Powered by Reddit