Nonparametric calculator

Dunn's post-hoc test calculator

Perform pairwise rank sum post-hoc comparisons following a significant Kruskal-Wallis test using Dunn's test with Bonferroni or Holm p-value adjustments.

Run this test live in StatLab npm i @statlab/core

When to use it

Use after Kruskal-Wallis test indicates significant overall group differences.

Required Inputs

  • Rank sum matrix or sample groups
  • Kruskal-Wallis mean ranks
  • Alpha & Multiplicity correction

Mathematical Formula

z = (R̄_i - R̄_j) / √[ (N(N+1)/12) * (1/n_i + 1/n_j) ]

Reporting Cautions

  • Always apply multiplicity corrections (Holm or Bonferroni) to avoid inflated Type I error rates.
  • Calculates mean rank differences, not median differences.

Code Snippets (Python, R, TypeScript)

Python (SciPy / Statsmodels) 
import scikit_posthocs as sp
res = sp.posthoc_dunn(df, val_col='score', group_col='group', p_adjust='holm')
R Language 
library(FSA)
dunnTest(score ~ group, data = df, method = "holm")
TypeScript (@statlab/core) 
import { dunnTest } from '@statlab/core';
const res = dunnTest(groups);

Developer Use Cases & Production Integrations

  • Pinpointing exact multi-region latency rank shifts in VoxelPulse infrastructure monitoring.
  • Evaluating specific prompt framework rank differences in VoxelAssurance.

VoxelPulse Telemetry Control Plane

Continuous A/B testing, statistical telemetry analysis, and automated feedback loops for live backend services.

Explore VoxelPulse →

VoxelAssurance Release Readiness

Fixed-price AI release-readiness sprints, automated regression testing, and reliability statistical benchmarks.

Explore VoxelAssurance →

Related Statistical Test Calculators

Explore static calculators across the StatLab inference engine suite.