Berkson's Paradox
Berkson's Bias · Berkson's Fallacy
A statistical illusion where two positive traits appear negatively correlated because you're only observing a filtered subset. The missing data you never see is what creates the false pattern.
EverydayConcepts.io
Origin
Named after statistician Joseph Berkson, who described the bias in a 1946 paper on hospital admission rates. Berkson showed that studying only hospitalized patients could create spurious correlations between diseases — because the act of selecting for one condition changes the apparent relationship with another. The classic restaurant example: places with good burgers seem to have bad fries, but only because you'd never eat at a place where both were bad.
Updated July 11, 2018