Evaluating Statistical Claims
How to spot misleading statistics, understand sample sizes, and distinguish correlation from causation.
Numbers Can Mislead
Statistics lend an air of authority to any claim, which is exactly why they are so often misused. A headline that says 'Study finds X doubles your risk of Y' sounds alarming — but if the base risk is 1 in a million, doubling it to 2 in a million is trivially small. This is the difference between relative risk (doubled!) and absolute risk (still negligible).
Common statistical red flags include:
Cherry-picked time frames. A chart showing crime rising from 2019 to 2022 might look alarming, but zoom out to 20 years and the trend might be sharply downward.
Misleading averages. If Jeff Bezos walks into a bar, the average net worth of everyone in the room becomes billions. Mean, median, and mode can tell very different stories.
Correlation vs. causation. Ice cream sales and drowning deaths both rise in summer. That does not mean ice cream causes drowning.
Small or unrepresentative samples. A study of 12 people, or a poll of a website's own readers, cannot be generalized to the population.