Spurious Correlations

May 28, 2014

An important rule of research is; correlation does not equal causation. Just because two events track each other over time does not mean that one caused the other. An example of this is ice cream sales and murder rates rise in US cities during the summer. Despite the two data being correlated, it is not difficult to understand why one doesn’t cause the other.

To help us understand this point and at the same time have some fun, Tyler Vigen, of the Harvard Law School, wrote a computer program to mine data for such correlations. His most amusing searches are on his web site, Spurious Correlations.