Numbers can be spun to prove almost anything, something this industry often seems to be guilty of (thanks, percentage change!). And, in an effort to show that correlation is not causation, ‘statistical provocateur’ Tyler Vigen has created a brilliant, data-rich, graph-heavy website.
As the BBC wrote, Vigen, a criminology student at Harvard Law School, created a computer programme to mine datasets for statistical correlations. Then he puts the best ones—like per capita consumption of cheese (US) correlating with the number of people who died by becoming tangled in their bedsheets, or per capita consumption of chicken (US) correlating with
total US crude oil imports—in sexy, easy to grasp but entirely unscientific graph form.
“What’s kind of fun about it is it allows people to be their own scientist for a few minutes, because they get to come up with their own hypothesis,” he said.