Exact paired sign-flip test
Enumerate sign assignments for a prespecified mean paired-difference statistic.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Use 2–18 paired differences, preserving one independent difference per unit. Enumeration is exact over all 2^n sign assignments, including zero differences.
The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.
Calculation results
- Mean paired difference
- 1.4
- Enumerated sign assignments
- 32
- Two-sided exact p
- 0.1875
How to interpret the output
The two-sided p-value counts absolute permuted means at least as large as the observed absolute mean. Sign exchangeability requires an appropriate randomization null or symmetry under the null; merely being paired is insufficient. This is not the binomial sign test and no Monte Carlo correction is needed for complete enumeration.
Method referenceSource checked 2026-11-06. Read the reproducible calculator validation report.