The word correlations does not equal fun. At least, to me it doesn’t. Maybe you’re a statistics wizard who thinks correlations are the bomb. Power to you. My point is that for most of us, correlations are, well, a necessary evil.
I do know the ground rule of statistics, of course: correlation does not equal causation. Just because two quantities happen to occur at the same time, multiple times, does not mean one is causing the other (or the other way around).
Statisticians call these spurious correlations: a mathematical relationship in which two or more events or variables are not causally related to each other (i.e. they are independent), yet it may be wrongly inferred that they are, due to either coincidence or the presence of a certain third, unseen factor (referred to as a “common response variable”, “confounding factor”, or “lurking variable”). (Source: Wikipedia)
Examples of Spurious Correlations
If I just lost you, let me make it clear with some fun examples of spurious correlations:
Bet you didn’t see that one coming! How about this one?
Another great example:
And my absolute favorite:
These examples are not mine, obviously. Goodness, gracious, I wouldn’t even have the skills to pull this off. Tyler Vigen, a former Harvard Law School student, created these (and many more). He made an entire book with examples of spurious correlations, and I bet they’re all hilarious.
There’s a serious underlying message though. In this age of big data, where we have more access to big data and more tools to analyze it, we need to be careful to jump to conclusions. Just because there’s a 97.8 % correlation between the number of films Jennifer Lawrence appears in yearly and the gross domestic product of Australia, doesn’t mean she’s to blame. Or praise. We can laugh about these examples, but nowadays spurious correlations are really just another version of alternative facts…