mind & bodyWikipedia
People ignore sample size, so small samples produce extreme results more often than expected
Insensitivity to sample size is a cognitive bias in which people judge the odds of a result without considering how large the sample is. Small samples vary more than large ones, so they produce extreme results far more often. In a famous experiment by Amos Tversky and Daniel Kahneman, a town had a large hospital with about 45 births a day and a small one with about 15. Asked which hospital recorded more days with over 60 percent boys, 56 percent of subjects said the two were about the same. In fact the smaller hospital is far more likely to record such days, because its daily figures swing more widely. Even statistically trained psychologists often made the same error.
The bias helps explain why small studies and rural statistics can produce dramatic headlines that later vanish. Checking the sample size is one of the simplest ways to judge how much to trust a striking number.
From Wikipedia, the free encyclopedia · text CC BY-SA 4.0
Illustrative photo · Oleksii.matiiasevych, CC BY-SA 4.0, via Wikimedia Commons
