Skip to main content
Advanced Technical IVT

permutation testing

Statistical method using random data shuffling to establish significance thresholds.

Full Definition

Permutation testing is a non-parametric statistical approach that generates null distributions by randomly reshuffling data labels or values to assess the significance of observed statistical effects. In neuroimaging, permutation testing is particularly valuable for controlling family-wise error rates when performing multiple comparisons across thousands of voxels or connections. The method makes minimal assumptions about data distribution and provides exact p-values for complex statistical designs. Permutation testing is often used in conjunction with cluster-based correction methods and is implemented in most major neuroimaging software packages.

Usage

Usage note: Distinguish from 'combination testing' and ensure proper citation of software implementation used.

In Context

  • "Statistical significance was determined using permutation testing with 10,000 iterations." — Statistical analysis methods section
  • "Permutation testing revealed significant group differences in connectivity strength (p < 0.001)." — Functional connectivity study results

Also known as

randomization testing shuffling test

Contrasted with

parametric testing

Don't confuse with

bootstrapping cross-validation Monte Carlo simulation

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