Research Workbench

Walsh pseudomedian comparison

Distinguish a paired pseudomedian from the median difference.

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.

Follow the labeled row format. Use synthetic values only. All calculations stay in your browser; no uploads. Finite magnitudes up to 1e9 and at least 1e-12, or zero.

The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.

Calculation results

Walsh pseudomedian
2.75
Ordinary median difference
1
Original differences
3
Dependent Walsh averages
6

How to interpret the output

Median of all Walsh averages (d[i]+d[j])/2 with i<=j, including self pairs. Also reports ordinary median. The averages are dependent and do not increase sample size. No signed-rank test or interval. Population pseudomedian and median coincide under symmetry, not generally. Source checked 18 November 2026.

Method reference

Read the worked guide and interpretation examples →