Evidence-appraisal glossary
Family-Wise Error Rate
The probability of making at least one false-positive claim across a whole set, or family, of statistical tests. Methods that control it aim to keep that combined risk at a chosen level.
Also called: FWER, experiment-wise error rate.
When a study runs many tests, the relevant question is not the error rate of any single test but the chance of any spurious finding among them all. The family-wise error rate captures that overall exposure, and procedures like Bonferroni, Holm, or gatekeeping strategies are designed to hold it below a target such as five percent. Controlling it is strict and protects against even one false claim, which suits confirmatory settings with a few key comparisons. It contrasts with the false-discovery-rate, which tolerates some false positives in exchange for more power.
This is a plain-language methodology definition for reading research. It is general education, not medical advice.