Wednesday, February 21, 2007

Statistics 101: Coincidence Is Not Causality

According to this bit of drivel, being born under the sign of Libra makes you a bad driver.

Ummm...no...I don't think so.

This is one of those wonderful cases where two statistical phenomena happen to coincide, and the person doing the analysis has chosen to claim a link between the two.

I can think of a dozen or so other factors that would have to be lined up to draw the conclusion that this study derives. (Starting with the obvious question about distribution of birthdates in the sample - was there a "bubble" in the September/October time? - for one obvious point)

No comments:

Understanding Alberta Separatist Claims As Gaslighting

The arguments that the Alberta separatists make to justify their positions are carefully crafted to seem almost reasonable, while at the sam...