Fun with Statistics: Correlation and Causation

Whenever you look at statistics, there are two things to keep in mind:

1. Correlation is not causation.
2. There are three types of lies: lies, damn lies, and statistics.

The second maxim is pretty easy to remember, especially if you disagree with the numbers in question. But the first one—distinguishing between correlation and causation—is somewhat more challenging.

To illustrate that point, let's look at some cases where correlation most likely does not imply causation. (Thanks to www.tylervigen.com for the examples.)

Stop eating cheese now—people are dying!

Per capita consumption of cheese and the number of people who died by becoming tangled in their bedsheets

Correlation: 94.7 percent

Hydrogenated fats are bad for marriage.

The divorce rate in Maine and per capita consumption of margarine

Correlation: 99.25 percent

Be careful in the sauna and kitchen—Miss America is almost 23.

The age of Miss America and murders by steam, hot vapors, and hot objects

Correlation: 87 percent

But where are the birds?

Honey-producing bee colonies and the marriage rate in Vermont

Correlation: 93.7 percent