Get PJ Media on your Apple

PJM Lifestyle

11 Curious Spurious Correlations

Proof that statistics can be fun!

Paula Bolyard


May 14, 2014 - 8:00 am
Page 1 of 11  Next ->   View as Single Page

Tyler Vigen, a Harvard Law student, has created a website that appears to demonstrate the truism: “53% of all statistics are made up.” That’s not precisely true of the “research” Vigen presents at Spurious Correlations (his correlations rely on actual data sets) but it does give some insight into the tools the data crunchers have at their disposal for spinning actual facts into what they want us to hear (often to the exclusion of more relevant information). Vigen says, “Empirical research is interesting, and I love to wonder about how variables work together. The charts on this site aren’t meant to imply causation nor are they meant to create a distrust for research or even correlative data. Rather, I hope this projects gets people interested in statistics and research.”

Here are some of Vigen’s best Spurious Correlations:

1. Number of people who tripped over their own two feet and died correlates with Civil engineering doctorates awarded (US)



The more pressing question: Are the civil engineers tripping over their own feet or are they designing things that cause others to trip? 

Comments are closed.

All Comments   (5)
All Comments   (5)
Sort: Newest Oldest Top Rated
Is there a similar site for geographic correlations? This one just has time series.
23 weeks ago
23 weeks ago Link To Comment
Typo: The image for #11 is the same image as the one for #10.
24 weeks ago
24 weeks ago Link To Comment
Thanks, sirspider. We'll get it fixed! :)
24 weeks ago
24 weeks ago Link To Comment
The world is full of things that change, so things that (as far as we know) are independent of each other will still move at the same time.
The charts all seem to track positive correlation: things that move up or down together. There are also negative correlations: things that move in opposite directions, less of one tracking with more of the other. Of course they can be just as coincidental as positive correlation.
I might have chosen the word "coincidental" rather than spurious; to me, spurious brings up the idea of deception, claiming a correlation where none really exists. Like the last 17 years of CO2-based "warming."
24 weeks ago
24 weeks ago Link To Comment
I'm afraid it isn't really a "choice" of words. Unfortunately, "spurious" is the nomenclature of professional and academic statistics regarding observed correlation without causation.

Try and think of it more as "potentially misleading" as humans tend to draw causal inferences about observed correlation. It is just the way our brains are wired.
24 weeks ago
24 weeks ago Link To Comment
View All