Confoundle · a reasoning trap

Correlation ≠ causation

Two things can rise and fall together perfectly and still have nothing to do with each other. Very often a hidden third thing is pulling both strings at once, so it looks like one causes the other when neither does. Before you believe a headline that “X is linked to Y,” ask what else could be driving both.

The rule

Two things moving together doesn't mean one causes the other. Often a third thing is quietly driving both.

What it looks like

More chocolate, more Nobel Prizes. Should your country stock up?It's a real, published finding: across 23 countries, the more chocolate people eat, the more Nobel laureates the country has produced, a strong correlation (r ≈ 0.79). The trend is hard to argue with.

The chocolate isn't doing anything.Wealthier countries can afford more chocolate AND fund more universities, labs and research, which is what actually wins Nobel Prizes. Wealth drives both, so chocolate and Nobels rise together without one causing the other. Hand out free chocolate and you'd get sweeter teeth, not more laureates.

Why it works

A correlation only says two things tend to move together. That can happen for several reasons: one really does cause the other; the causation runs the other way; a hidden third factor drives both (a common cause, like hot weather driving both ice-cream sales and drownings); or it's a coincidence, which gets more likely the more data you sift through. Spotting a correlation is the easy part. Working out which of these is behind it is the real job, and usually needs an experiment, not just a chart.

Source

Messerli FH. Chocolate consumption, cognitive function, and Nobel laureates. N Engl J Med. 2012;367(16):1562-1564. (r ≈ 0.79 across 23 countries; a deliberately tongue-in-cheek demonstration that correlation is not causation.)

See if it fools you

This page gives the answer away. The puzzle version shows you the same figures first and asks you to commit before the reveal.

Play this one