The family-wise error rate (FWER) is the probability of making at least one Type I error (false positive) when conducting multiple statistical tests. As the number of tests increases, the risk of incorrectly rejecting at least one true null hypothesis also rises. Researchers often use correction methods, like the Bonferroni correction, to control the FWER and reduce the chance of false positives across the set of tests.
For example, if a researcher tests the effectiveness of 10 different diets and performs a separate hypothesis test for each one, without correction the chance of making at least one false positive finding becomes quite high. Using a Bonferroni adjustment would help control the family-wise error rate.
Synonyms:
Experiment-wise error rate