Conceptual
Browse Instruct Curate
Login

Multiple Hypothesis Testing Correction in Statistics

Methods for controlling the inflated rate of false positives that arises when many statistical hypotheses are tested at once, such as family-wise error control (Bonferroni, Holm) and false-discovery-rate procedures (Benjamini-Hochberg). Adjusting significance thresholds or p-values across a family of tests keeps the overall error rate at the intended level.

Questions this Concept answers

  • Why does testing many hypotheses simultaneously require a stricter significance threshold than a single test?