T Table: Critical t Values for One & Two Tails (df 1 to ∞)
Why does a 95% confidence interval from 16 measurements use 2.131 instead of the 1.96 from the normal curve? Because the standard deviation was estimated from those same 16 values, and a t-table adds the extra room that estimate needs. The table below gives that number, the critical value of t, for any common confidence level or significance level and for degrees of freedom from 1 to infinity.
Values are rounded to three decimals. For degrees of freedom the table skips, or an exact p-value, use the calculator below. It is set to an example further down this page: t = 2.65 with 15 degrees of freedom.
Use z when the population standard deviation is known, or for a test of proportions.
One-sample or paired t-test: df = n − 1. Two-sample (pooled): df = n₁ + n₂ − 2. Welch's df can be a decimal.
Goodness of fit: df = categories − 1. Test of independence: df = (rows − 1) × (columns − 1).
One-way ANOVA with k groups and N values in total: df₁ = k − 1, df₂ = N − k.
Tests whether a Pearson correlation differs from zero, using t with n − 2 degrees of freedom.
Finding One Value: a 95% Interval From 16 Measurements
Say you test 16 batteries and get a mean life of 41.3 hours with a standard deviation of 3.2 hours. To build a 95% confidence interval for the true mean:
- Degrees of freedom. For one sample, df = n − 1 = 15.
- Column. A 95% interval leaves 5% outside it, split between two tails. That is the column headed 0.05 for two tails, 0.025 for one tail and 95% for confidence. All three labels are the same column.
- Read across. Row 15 in that column gives 2.131.
- Build the interval. The standard error is 3.2 ÷ √16 = 0.8, so the margin of error is 2.131 × 0.8 = 1.705. The interval runs from 39.59 to 43.01 hours.
With the normal curve’s 1.96, the margin would be 1.568, narrower than the data justify. For other designs, such as two samples, the degrees of freedom come from a different formula; the p-value calculator page lists them by test.
The t Table: One-Tailed and Two-Tailed Critical Values
Each cell is the value of t with the stated area beyond it. Find your degrees of freedom on the left, then the column for your tail area or confidence level.
| One tail | 0.10 | 0.05 | 0.025 | 0.01 | 0.005 | 0.001 | 0.0005 |
|---|---|---|---|---|---|---|---|
| Two tails | 0.20 | 0.10 | 0.05 | 0.02 | 0.01 | 0.002 | 0.001 |
| Confidence | 80% | 90% | 95% | 98% | 99% | 99.8% | 99.9% |
| 1 | 3.078 | 6.314 | 12.706 | 31.821 | 63.657 | 318.309 | 636.619 |
| 2 | 1.886 | 2.920 | 4.303 | 6.965 | 9.925 | 22.327 | 31.599 |
| 3 | 1.638 | 2.353 | 3.182 | 4.541 | 5.841 | 10.215 | 12.924 |
| 4 | 1.533 | 2.132 | 2.776 | 3.747 | 4.604 | 7.173 | 8.610 |
| 5 | 1.476 | 2.015 | 2.571 | 3.365 | 4.032 | 5.893 | 6.869 |
| 6 | 1.440 | 1.943 | 2.447 | 3.143 | 3.707 | 5.208 | 5.959 |
| 7 | 1.415 | 1.895 | 2.365 | 2.998 | 3.499 | 4.785 | 5.408 |
| 8 | 1.397 | 1.860 | 2.306 | 2.896 | 3.355 | 4.501 | 5.041 |
| 9 | 1.383 | 1.833 | 2.262 | 2.821 | 3.250 | 4.297 | 4.781 |
| 10 | 1.372 | 1.812 | 2.228 | 2.764 | 3.169 | 4.144 | 4.587 |
| 11 | 1.363 | 1.796 | 2.201 | 2.718 | 3.106 | 4.025 | 4.437 |
| 12 | 1.356 | 1.782 | 2.179 | 2.681 | 3.055 | 3.930 | 4.318 |
| 13 | 1.350 | 1.771 | 2.160 | 2.650 | 3.012 | 3.852 | 4.221 |
| 14 | 1.345 | 1.761 | 2.145 | 2.624 | 2.977 | 3.787 | 4.140 |
| 15 | 1.341 | 1.753 | 2.131 | 2.602 | 2.947 | 3.733 | 4.073 |
| 16 | 1.337 | 1.746 | 2.120 | 2.583 | 2.921 | 3.686 | 4.015 |
| 17 | 1.333 | 1.740 | 2.110 | 2.567 | 2.898 | 3.646 | 3.965 |
| 18 | 1.330 | 1.734 | 2.101 | 2.552 | 2.878 | 3.610 | 3.922 |
| 19 | 1.328 | 1.729 | 2.093 | 2.539 | 2.861 | 3.579 | 3.883 |
| 20 | 1.325 | 1.725 | 2.086 | 2.528 | 2.845 | 3.552 | 3.850 |
| 21 | 1.323 | 1.721 | 2.080 | 2.518 | 2.831 | 3.527 | 3.819 |
| 22 | 1.321 | 1.717 | 2.074 | 2.508 | 2.819 | 3.505 | 3.792 |
| 23 | 1.319 | 1.714 | 2.069 | 2.500 | 2.807 | 3.485 | 3.768 |
| 24 | 1.318 | 1.711 | 2.064 | 2.492 | 2.797 | 3.467 | 3.745 |
| 25 | 1.316 | 1.708 | 2.060 | 2.485 | 2.787 | 3.450 | 3.725 |
| 26 | 1.315 | 1.706 | 2.056 | 2.479 | 2.779 | 3.435 | 3.707 |
| 27 | 1.314 | 1.703 | 2.052 | 2.473 | 2.771 | 3.421 | 3.690 |
| 28 | 1.313 | 1.701 | 2.048 | 2.467 | 2.763 | 3.408 | 3.674 |
| 29 | 1.311 | 1.699 | 2.045 | 2.462 | 2.756 | 3.396 | 3.659 |
| 30 | 1.310 | 1.697 | 2.042 | 2.457 | 2.750 | 3.385 | 3.646 |
| 40 | 1.303 | 1.684 | 2.021 | 2.423 | 2.704 | 3.307 | 3.551 |
| 50 | 1.299 | 1.676 | 2.009 | 2.403 | 2.678 | 3.261 | 3.496 |
| 60 | 1.296 | 1.671 | 2.000 | 2.390 | 2.660 | 3.232 | 3.460 |
| 80 | 1.292 | 1.664 | 1.990 | 2.374 | 2.639 | 3.195 | 3.416 |
| 100 | 1.290 | 1.660 | 1.984 | 2.364 | 2.626 | 3.174 | 3.390 |
| 120 | 1.289 | 1.658 | 1.980 | 2.358 | 2.617 | 3.160 | 3.373 |
| 1000 | 1.282 | 1.646 | 1.962 | 2.330 | 2.581 | 3.098 | 3.300 |
| ∞ | 1.282 | 1.645 | 1.960 | 2.326 | 2.576 | 3.090 | 3.291 |
One Tail, Two Tails or a Confidence Level: Picking the Column
The three header rows label the same columns three ways.
- Two-tailed test (“is it different?”): look up α in the Two tails row. At α = 0.05 with df = 15, the critical value is ±2.131. Reject the null hypothesis if |t| ≥ 2.131.
- Right-tailed test (“is it bigger?”): look up α in the One tail row. At α = 0.05 with df = 15, the column is 0.05 and the value is 1.753. Reject if t ≥ 1.753.
- Left-tailed test (“is it smaller?”): the same column, with a minus sign. The t-distribution is symmetric, so the cut-off is −1.753. Reject if t ≤ −1.753.
- Confidence interval: look up the level in the Confidence row. A 90% interval with df = 15 uses 1.753, the same number as the one-tailed 0.05 test, because 90% in the middle leaves 5% in each tail.
The table lists only positive values. A negative t statistic is not a problem: for a two-tailed test, compare its size, |t|, with the table; for a left-tailed test, compare it with the negative of the table value.
When Your Degrees of Freedom Aren’t in the Table
After df = 30 the table jumps to 40, 50, 60 and beyond, so a value like 45 has no row. There are two ways to handle it.
Round down to the next row. For df = 45, use row 40: 2.021 at two-tailed α = 0.05. A smaller df always has a larger critical value, so rounding down makes the test slightly harder to pass. That is the safe direction: you will never call a result significant that the exact value wouldn’t.
Interpolate between rows. Row 40 gives 2.021 and row 50 gives 2.009. Halfway between is 2.015. The exact value for df = 45 is 2.014, so interpolation lands within 0.001.
The same applies to a Welch t-test, where the degrees of freedom are a decimal such as 23.6. Rounding down to 23 gives 2.069, against an exact 2.066. Rounding up to 24 would give 2.064, which is too small, so round down.
Past the last numbered row, the values change slowly. For df = 200 the table’s nearest row is 120, which gives 1.980; the exact value is 1.972. The calculator above takes any degrees of freedom, including decimals, and returns the exact critical value at your α in its Critical value box.
From a t Statistic to a Range for the p-Value
A t-table gives cut-offs, not p-values, but it can tell you which range the p-value falls in. Say a one-sample test with 16 values gives t = 2.65, two-tailed.
- Go to row 15.
- Find the two values that 2.65 falls between: 2.602 (two tails 0.02) and 2.947 (two tails 0.01).
- The p-value falls between the two column headings: 0.01 < p < 0.02.
So the result is significant at the 5% level but not at 1%. The calculator gives the exact two-tailed p-value, 0.0182, which is better to report than the range.
If t is smaller than the first column, the p-value is above that column’s heading: with df = 15, a t of 1.20 is below 1.341, so the two-tailed p is greater than 0.20. If t is beyond the last column, the p-value is below the smallest heading, p < 0.001.
Why Small Samples Need Bigger Critical Values
Read the 95% column from the top. With 1 degree of freedom the critical value is 12.706. With 2 it drops to 4.303, and with 5 to 2.571. By 30 it is 2.042, and the last row, ∞, is 1.960, exactly the normal-curve value.
| Degrees of freedom | 95% critical value | Above the z value of 1.960 by |
|---|---|---|
| 5 | 2.571 | 31.2% |
| 30 | 2.042 | 4.2% |
| 100 | 1.984 | 1.2% |
| 1,000 | 1.962 | 0.1% |
The ∞ row is the z-table in t-table form: 1.282, 1.645, 1.960, 2.326 and 2.576 are the familiar normal cut-offs. The rule of thumb that “n above 30 means you can use z” leaves an interval about 4% too narrow at df = 30. The t value is correct at any sample size whenever the standard deviation comes from the sample, so there is no reason to switch to z at 30. The z-score table covers the case where the population standard deviation really is known.
The extra width pays for estimating the standard deviation: from a few values it can easily come out too small, and the heavier tails allow for that. The t-distribution was first worked out by William Sealy Gosset, who worked at the Guinness brewery in Dublin and needed methods for small samples. Guinness did not let its researchers publish under their own names, so his 1908 paper in Biometrika, “The probable error of a mean”, appeared under the name “Student”. That is why it is still called Student’s t.
