Nonparametric calculator

Rank-biserial correlation and Cliff's delta calculator

Calculate rank-biserial correlation (r_rb) and Cliff's delta (d) non-parametric effect sizes for Mann-Whitney U test outcomes.

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

When to use it

Use alongside Mann-Whitney U test results to communicate non-parametric effect magnitude.

Required Inputs

  • Group A values (or U statistic)
  • Group B values (or sample sizes n_A, n_B)

Mathematical Formula

r_{rb} = 1 - (2U / (n_A * n_B)), Cliff’s δ = (2U / (n_A * n_B)) - 1

Reporting Cautions

  • Ranges strictly from -1.0 to +1.0; zero indicates complete distributional overlap.
  • Do not confuse with Pearson point-biserial correlation.

Code Snippets (Python, R, TypeScript)

Python (SciPy / Statsmodels) 
def cliffs_delta(lst1, lst2):
    m, n = len(lst1), len(lst2)
    more = sum(i > j for i in lst1 for j in lst2)
    less = sum(i < j for i in lst1 for j in lst2)
    return (more - less) / (m * n)
R Language 
library(effsize)
cliff.delta(group_a, group_b)
TypeScript (@statlab/core) 
import { rankBiserial } from '@statlab/core';
const r_rb = rankBiserial(groupA, groupB);

Developer Use Cases & Production Integrations

  • Quantifying rank superiority magnitude between telemetry sample runs in VoxelPulse.
  • Expressing non-parametric shift size in release readiness audit reports.

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.