Nonparametric Tests

Wilcoxon Signed-Rank Test Table: Critical Values of T

Exact critical values for the Wilcoxon signed-rank test, the nonparametric alternative to the paired t-test.

  • n = 5 to 50
  • Exact values
  • Exact p-value calculator

Wilcoxon Signed-Rank Table

Rows: number of non-zero differences n · Columns: α · Cells: largest T that is significant. Reject H₀ if T ≤ the cell.

Wilcoxon Signed-Rank Test Table (Critical Values)
One-tailed α →0.050.0250.010.005
Two-tailed α →
n ↓
0.10.050.020.01
50———
620——
7320—
85310
98531
1010853
11131075
12171397
132117129
1425211512
1530251915
1635292319
1741342723
1847403227
1953463732
2060524337
2167584942
2275655548
2383736254
2491816961
25100897668
26110988475
271191079283
2813011610191
29140126110100
30151137120109
31163147130118
32175159140128
33187170151138
34200182162148
35213195173159
36227208185171
37241221198182
38256235211194
39271249224207
40286264238220
41302279252233
42319294266247
43336310281261
44353327296276
45371343312291
46389361328307
47407378345322
48426396362339
49446415379355
50466434397373

Tip: click any value to highlight its row and column.

How to read this table

  1. Compute each paired difference and drop any that are zero. n is the number of differences left.
  2. Rank the absolute differences, add up the ranks of the positive and of the negative differences, and let T be the smaller sum (for a two-sided test).
  3. Reject H₀ if T is less than or equal to the table value.
  4. A dash (—) means no result can be significant at that α with so few pairs.

Worked example

Twelve patients have blood pressure measured before and after treatment, with no zero differences, and the smaller rank sum is T = 10. Row n = 12, two-tailed α = 0.05 gives 13. Because 10 ≤ 13, the change is significant at the 5% level.

Click the underlined links to highlight the value in the table.

Calculator

For values between the rows and columns of the table. Results update as you type.

Formula and how the values were computed

Reject H₀ if T ≤ Tα, n, where P(T ≤ Tα, n) ≤ α

Critical values come from the exact null distribution of T, obtained by counting all 2ⁿ sign assignments. Each value is the largest T with P(T ≤ t) ≤ α (one-tailed). The table assumes no tied absolute differences.

Frequently asked questions

Some published tables differ by one. Why?

Tables differ in whether the listed value is the largest T with P(T ≤ t) ≤ α (used here, so the test never exceeds α) or the nearest value. This table always keeps the true significance level at or below α.

What about ties and zeros?

Zero differences are dropped before ranking; tied absolute differences get the average of their ranks. With many ties the exact table is only approximate, and a normal approximation with a tie correction is often used.

What if n is larger than 50?

The calculator above handles n up to 100 exactly. For large n, T is approximately normal with mean n(n + 1)/4 and variance n(n + 1)(2n + 1)/24.

Related calculators and tutorials