False Discovery Rate (Benjamini–Hochberg) Calculator
Paste a list of p-values and get Benjamini–Hochberg adjusted q-values, the BH cutoff and how many tests survive at your FDR level — with Bonferroni for comparison.
How it works
Sort p₍₁₎ ≤ … ≤ p₍ₘ₎; keep all tests up to the largest i with p₍ᵢ₎ ≤ (i/m)·q; q-value₍ᵢ₎ = min over j ≥ i of p₍ⱼ₎·m/j
When you run many tests, controlling the family-wise error (Bonferroni) becomes so strict that real effects are missed. Benjamini–Hochberg instead controls the expected proportion of false positives among the results you call significant. Ranking p-values and comparing each to a rising threshold keeps far more true discoveries while promising that, on average, no more than q of what you keep is noise — the standard in genomics, A/B testing at scale and any screening study.
Worked example
The 25 p-values above at q = 5%: the BH step-up keeps the 5 smallest (cutoff p = 0.042), whereas Bonferroni (0.05/25 = 0.002) keeps only one.
Frequently asked questions
What is a q-value?
The smallest FDR at which that test would be called significant — the FDR analogue of an adjusted p-value.
Does BH need independent tests?
It holds under independence or positive dependence. For arbitrary dependence use Benjamini–Yekutieli, which divides q by Σ1/i.
Related calculators
For education — check assumptions (sample size, distribution, independence) before using a test result in a real decision.