Skip to main content
Intermediate Technical IVT

false discovery rate

Also written as: FDR — False Discovery Rate

Statistical correction method controlling the expected proportion of false positive findings.

Full Definition

False Discovery Rate (FDR) is a statistical method for controlling the expected proportion of incorrectly rejected null hypotheses among all rejected hypotheses when performing multiple comparisons. In neuroimaging, FDR correction is commonly applied when testing thousands of voxels or connections simultaneously to maintain statistical validity while being less conservative than family-wise error rate corrections. The Benjamini-Hochberg procedure is the most widely used FDR method, offering a good balance between Type I error control and statistical power. FDR correction is particularly useful in exploratory neuroimaging analyses where some false discoveries may be acceptable.

Usage

Usage note: Often abbreviated as 'FDR' in methods sections; distinguish from 'false positive rate' which has different meaning.

In Context

  • "Multiple comparisons were corrected using false discovery rate at q < 0.05." — Statistical methods section
  • "The analysis revealed 450 significantly activated voxels after false discovery rate correction." — fMRI activation study results

Also known as

FDR Benjamini-Hochberg correction

Contrasted with

family-wise error rate uncorrected p-values

Don't confuse with

false positive rate family-wise error rate Bonferroni correction

Editors from these organisations have used our services since 1998

Reuters BBC Oxford University Press Penguin Random House Springer Microsoft Suncor Energy United Nations Fisher Investments IBM The Home Depot KODAK CHEVRON