Correlation
Kendall's Tau Table: Exact Critical Values of τ
Exact critical values of Kendall's tau, the rank correlation based on concordant and discordant pairs.
- n = 4 to 100
- Exact values
- Exact calculator
Kendall's tau Table
Rows: sample size n · Columns: α (one-tailed on the first header row, two-tailed on the second) · Cells: critical τ. Reject H₀ if |τ| ≥ the cell.
| One-tailed α → | 0.05 | 0.025 | 0.01 | 0.005 | 0.001 |
|---|---|---|---|---|---|
| Two-tailed α → n ↓ | 0.1 | 0.05 | 0.02 | 0.01 | 0.002 |
| 4 | 1.000 | — | — | — | — |
| 5 | 0.800 | 1.000 | 1.000 | — | — |
| 6 | 0.734 | 0.867 | 0.867 | 1.000 | — |
| 7 | 0.620 | 0.715 | 0.810 | 0.905 | 1.000 |
| 8 | 0.572 | 0.643 | 0.715 | 0.786 | 0.858 |
| 9 | 0.500 | 0.556 | 0.667 | 0.723 | 0.834 |
| 10 | 0.467 | 0.512 | 0.600 | 0.645 | 0.778 |
| 11 | 0.419 | 0.491 | 0.564 | 0.600 | 0.710 |
| 12 | 0.394 | 0.455 | 0.546 | 0.576 | 0.667 |
| 13 | 0.359 | 0.436 | 0.513 | 0.565 | 0.642 |
| 14 | 0.363 | 0.407 | 0.473 | 0.517 | 0.605 |
| 15 | 0.334 | 0.391 | 0.467 | 0.505 | 0.581 |
| 16 | 0.317 | 0.384 | 0.434 | 0.484 | 0.567 |
| 17 | 0.309 | 0.368 | 0.427 | 0.471 | 0.545 |
| 18 | 0.295 | 0.347 | 0.412 | 0.451 | 0.530 |
| 19 | 0.287 | 0.334 | 0.392 | 0.439 | 0.509 |
| 20 | 0.274 | 0.327 | 0.379 | 0.422 | 0.495 |
| 21 | 0.267 | 0.315 | 0.372 | 0.410 | 0.486 |
| 22 | 0.265 | 0.308 | 0.360 | 0.394 | 0.472 |
| 23 | 0.257 | 0.297 | 0.352 | 0.392 | 0.455 |
| 24 | 0.247 | 0.290 | 0.341 | 0.377 | 0.450 |
| 25 | 0.240 | 0.287 | 0.334 | 0.367 | 0.440 |
| 26 | 0.237 | 0.280 | 0.330 | 0.360 | 0.428 |
| 27 | 0.231 | 0.271 | 0.322 | 0.357 | 0.419 |
| 28 | 0.228 | 0.265 | 0.313 | 0.344 | 0.413 |
| 29 | 0.222 | 0.262 | 0.311 | 0.340 | 0.404 |
| 30 | 0.219 | 0.256 | 0.302 | 0.334 | 0.394 |
| 31 | 0.213 | 0.252 | 0.295 | 0.325 | 0.390 |
| 32 | 0.210 | 0.246 | 0.291 | 0.323 | 0.380 |
| 33 | 0.205 | 0.243 | 0.288 | 0.315 | 0.375 |
| 34 | 0.202 | 0.238 | 0.280 | 0.312 | 0.369 |
| 35 | 0.197 | 0.234 | 0.278 | 0.305 | 0.362 |
| 36 | 0.194 | 0.232 | 0.274 | 0.302 | 0.359 |
| 37 | 0.193 | 0.229 | 0.268 | 0.298 | 0.352 |
| 38 | 0.190 | 0.224 | 0.264 | 0.292 | 0.346 |
| 39 | 0.188 | 0.220 | 0.261 | 0.288 | 0.342 |
| 40 | 0.185 | 0.218 | 0.257 | 0.285 | 0.339 |
| 41 | 0.181 | 0.215 | 0.254 | 0.281 | 0.335 |
| 42 | 0.178 | 0.213 | 0.250 | 0.276 | 0.329 |
| 43 | 0.177 | 0.210 | 0.247 | 0.274 | 0.325 |
| 44 | 0.174 | 0.208 | 0.244 | 0.269 | 0.322 |
| 45 | 0.172 | 0.205 | 0.241 | 0.267 | 0.318 |
| 46 | 0.170 | 0.202 | 0.239 | 0.264 | 0.315 |
| 47 | 0.168 | 0.199 | 0.236 | 0.260 | 0.310 |
| 48 | 0.167 | 0.197 | 0.233 | 0.258 | 0.307 |
| 49 | 0.164 | 0.196 | 0.230 | 0.254 | 0.303 |
| 50 | 0.163 | 0.192 | 0.228 | 0.251 | 0.300 |
| 51 | 0.161 | 0.191 | 0.226 | 0.249 | 0.298 |
| 52 | 0.159 | 0.189 | 0.224 | 0.246 | 0.295 |
| 53 | 0.157 | 0.188 | 0.221 | 0.244 | 0.291 |
| 54 | 0.156 | 0.186 | 0.219 | 0.242 | 0.288 |
| 55 | 0.155 | 0.183 | 0.217 | 0.240 | 0.285 |
| 56 | 0.152 | 0.181 | 0.215 | 0.237 | 0.282 |
| 57 | 0.152 | 0.180 | 0.212 | 0.235 | 0.280 |
| 58 | 0.150 | 0.178 | 0.210 | 0.232 | 0.277 |
| 59 | 0.148 | 0.176 | 0.209 | 0.230 | 0.275 |
| 60 | 0.147 | 0.175 | 0.207 | 0.229 | 0.273 |
| 65 | 0.141 | 0.168 | 0.199 | 0.219 | 0.261 |
| 70 | 0.136 | 0.162 | 0.191 | 0.210 | 0.252 |
| 75 | 0.131 | 0.156 | 0.184 | 0.203 | 0.243 |
| 80 | 0.126 | 0.150 | 0.178 | 0.197 | 0.235 |
| 85 | 0.123 | 0.146 | 0.172 | 0.190 | 0.227 |
| 90 | 0.119 | 0.141 | 0.167 | 0.185 | 0.221 |
| 95 | 0.115 | 0.137 | 0.162 | 0.180 | 0.215 |
| 100 | 0.112 | 0.134 | 0.158 | 0.175 | 0.209 |
Tip: click any value to highlight its row and column.
How to read this table
- Count the concordant (C) and discordant (D) pairs. S = C − D and τ = S / (n(n − 1)/2).
- Find the row for n and the column for α.
- The association is significant if |τ| is greater than or equal to the table value. To work with S instead, multiply the table value by n(n − 1)/2.
Worked example
Ten patients are ranked by two raters, giving τ = 0.56 (S = 25). Row 10, two-tailed α = 0.05 gives 0.512, so the agreement 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
Values come from the exact null distribution of the number of discordant pairs, which is the distribution of inversions in a random permutation, computed by recursion for every n. Each entry is the smallest attainable τ with P(τ ≥ t) ≤ α.
Frequently asked questions
Kendall's τ or Spearman's rₛ?
Both test the same hypothesis and usually agree. τ is smaller in magnitude (roughly two-thirds of rₛ), has a direct interpretation as a difference between the probabilities of concordance and discordance, and handles ties more gracefully.
What if there are ties?
Use τ-b, which adjusts the denominator for ties. The exact table assumes no ties, so use the normal approximation with a tie-corrected variance when ties are common.