Home / Tools / t Table

t Table: Critical t Values, Made Interactive

A t table turns your degrees of freedom and significance level into the critical t value a test statistic must beat to count as significant. Set the three inputs below to light up the exact cell, read the value off a shaded curve, or flip to reverse mode to turn a t statistic back into a p-value. Every number is generated live and matches R's qt().

I want to find the critical t value

Two-tailed splits α across both tails, so the tool looks up qt(1 − α/2, df). One-tailed puts it all in one side: qt(1 − α, df).

two-tailed · α = 0.05

t* = 2.086

Critical value for df = 20.

critical t*
2.086
df
20
tail area
0.025
confidence
95%

Full t distribution table

The complete critical-value table below is rendered as plain HTML, so it prints, copies and is fully indexable. Each row is a value of degrees of freedom; each column is an upper-tail area (with the matching two-tailed α and confidence level under it). Your current selection highlights the exact cell.

df0.10.050.0250.010.0050.0010.0005
two-tailed α0.20.10.050.020.010.0020.001
confidence80%90%95%98%99%99.8%99.9%
13.0786.31412.70631.82163.657318.309636.619
21.8862.9204.3036.9659.92522.32731.599
31.6382.3533.1824.5415.84110.21512.924
41.5332.1322.7763.7474.6047.1738.610
51.4762.0152.5713.3654.0325.8936.869
61.4401.9432.4473.1433.7075.2085.959
71.4151.8952.3652.9983.4994.7855.408
81.3971.8602.3062.8963.3554.5015.041
91.3831.8332.2622.8213.2504.2974.781
101.3721.8122.2282.7643.1694.1444.587
111.3631.7962.2012.7183.1064.0254.437
121.3561.7822.1792.6813.0553.9304.318
131.3501.7712.1602.6503.0123.8524.221
141.3451.7612.1452.6242.9773.7874.140
151.3411.7532.1312.6022.9473.7334.073
161.3371.7462.1202.5832.9213.6864.015
171.3331.7402.1102.5672.8983.6463.965
181.3301.7342.1012.5522.8783.6103.922
191.3281.7292.0932.5392.8613.5793.883
201.3251.7252.0862.5282.8453.5523.850
211.3231.7212.0802.5182.8313.5273.819
221.3211.7172.0742.5082.8193.5053.792
231.3191.7142.0692.5002.8073.4853.768
241.3181.7112.0642.4922.7973.4673.745
251.3161.7082.0602.4852.7873.4503.725
261.3151.7062.0562.4792.7793.4353.707
271.3141.7032.0522.4732.7713.4213.690
281.3131.7012.0482.4672.7633.4083.674
291.3111.6992.0452.4622.7563.3963.659
301.3101.6972.0422.4572.7503.3853.646
351.3061.6902.0302.4382.7243.3403.591
401.3031.6842.0212.4232.7043.3073.551
451.3011.6792.0142.4122.6903.2813.520
501.2991.6762.0092.4032.6783.2613.496
601.2961.6712.0002.3902.6603.2323.460
701.2941.6671.9942.3812.6483.2113.435
801.2921.6641.9902.3742.6393.1953.416
901.2911.6621.9872.3682.6323.1833.402
1001.2901.6601.9842.3642.6263.1743.390
1201.2891.6581.9802.3582.6173.1603.373
1501.2871.6551.9762.3512.6093.1453.357
2001.2861.6531.9722.3452.6013.1313.340
(z)1.2821.6451.9602.3262.5763.0903.291

The last row (∞) is the normal limit: with infinite df the critical t value equals the z value. R: qt(p, df) reproduces any cell.

How this value is computed
1
2
3
4The critical value is the quantile of Student's t distribution: the point with exactly the target area in the tail. It is found by inverting the t cumulative distribution function, the same routine R uses.
The same thing in R

  

No data leaves your browser Verified against R's qt() and pt() Free, no sign-up

When you reach for a t table

Three everyday jobs, all answered by the same table of critical values.

Build a confidence interval

A 95% interval for a mean is the estimate plus or minus t* × SE, where t* is the two-tailed 0.05 critical value for your df. Smaller samples get a wider t*.

Judge a t test

Compare your test statistic to the critical value for your α and tails. If |t| exceeds it, reject the null hypothesis. The reverse mode turns the statistic straight into a p-value.

See why df matters

With few degrees of freedom the tails are heavy, so the critical value is large. As df climbs, the t distribution tightens toward the normal and t* settles near the familiar z cutoffs.

One-tailed vs two-tailed, and how the columns line up

You haveTail area to look upR calldf = 20
95% CI / two-tailed α = .050.025 in each tailqt(0.975, 20)2.086
99% CI / two-tailed α = .010.005 in each tailqt(0.995, 20)2.845
one-tailed α = .050.05 in one tailqt(0.95, 20)1.725
one-tailed α = .010.01 in one tailqt(0.99, 20)2.528

A two-tailed test asks only whether two things differ, so it guards both directions and needs a larger cutoff. A one-tailed test commits to a direction in advance and spends all of α on that side, which lowers the bar. Pick the tail count before you see the data, not after.

t table questions

How do I read a t table?

Find the row for your degrees of freedom and the column for your significance level, then read the cell where they meet: that is the critical t value. Use the two-tailed α header for a two-sided test and the one-tailed area for a one-sided test. If your observed statistic is more extreme than the cell, the result is significant at that α.

What are degrees of freedom?

Degrees of freedom count the independent pieces of information behind your variance estimate. A one-sample or paired t test has df = n − 1; a Welch two-sample test has a fractional df from its variance-weighting formula. This tool accepts any df, whole or decimal, so a Welch df like 17.4 works directly.

What is the critical t value for a 95% confidence interval?

A 95% interval leaves 0.05 split across the two tails, so you want the value with 0.025 above it: qt(0.975, df). It is 2.086 at df = 20 and 2.009 at df = 50, and it approaches the z value 1.960 as the sample grows.

What are degrees of freedom in a t test?

Degrees of freedom (df) are the number of independent pieces of information used to estimate variability. For a one-sample or paired t test, df equals the sample size minus one. For a two-sample test, R uses the Welch approximation, which can be fractional. This calculator accepts any df, including decimals.

What is the difference between one-tailed and two-tailed critical values?

A two-tailed test splits alpha across both tails, so it uses qt(1 - alpha/2, df). A one-tailed test puts all of alpha in one tail, so it uses qt(1 - alpha, df) and gives a smaller critical value. Choose one-tailed only when your hypothesis specifies a direction in advance.

Does this t table match R?

Yes. Every printed value and every interactive result is generated by the same routine and checked against R's qt() and pt() functions across more than 2500 cases to better than seven significant figures.

Why is the critical value bigger for small samples?

With few degrees of freedom the t distribution has heavier tails than the normal, reflecting the extra uncertainty in estimating the standard deviation from a small sample. A larger critical value is the price of that uncertainty; it shrinks toward the z value as df grows.

Does this match R and printed textbook tables?

Yes. The printed table and the live result share one routine, checked against R's qt() and pt() across more than 2500 cases to better than seven significant figures. Small last-digit gaps versus an old textbook table are rounding in the book, not here.

Keep going