Well if they’re going to let women edit the Harvard Law Review…

…then we may as well all go to Mars.

Tyler Vigen wrote a programme that looks for (usually) spurious correlations (via Flowing Data). It’s found over 17,000 correlations so far (and counting). Take a look – you can play around and come up with your own favourites.

The fact that correlation does not mean causation is usually driven home in the first week of any research-related course. But that doesn’t mean that we humans don’t look for correlations, and assign causes. It helps us feel like we’re in control and/or reinforces our prejudices to think that crime is innately related to poverty (rather than both stemming from a third, unknown cause) or that sexual violence in related in some way to female empowerment (they never had rapes before women got cellphones).


As an aside, I love, love, love the copyright notice on Tyler’s site. (Scroll all the way down for the cartoon.)

Tags from the story
,

Leave a Reply

Your email address will not be published. Required fields are marked *