How to use the p-value calculator
Choose the statistic your test gave you — z, t, chi-square, F, or a correlation r — and type its value. t, chi-square, F and r also need degrees of freedom (or the sample size for r); the note under each field says where to find them. Type your significance level α (0.05 is used if you leave it empty) and choose the type of p-value: two-tailed for any difference, left-tailed for “lower than”, right-tailed for “higher than”.
The result shows the p-value for that tail, whether it is at or below α, and a curve of the distribution with your statistic marked and the p-value area shaded. The table under it gives the left, right and two-tailed p-values at once; click a row to switch tails. Go advanced to turn a p-value (or an area between −x and x, or from the centre to x, for z, t and r) back into a statistic, or to work out t or z from sample means and standard deviations.
Which statistic do I have?
| Your test | Statistic | Degrees of freedom |
|---|---|---|
| z-test (known population standard deviation, or a large sample) | z | none |
| t-test: one sample, paired, or two samples | t | n − 1, or n₁ + n₂ − 2 |
| Chi-square test: goodness of fit, independence | χ² | categories − 1, or (rows − 1) × (columns − 1) |
| ANOVA, comparing two variances, regression F | F | numerator and denominator |
| Correlation | r (with n) | n − 2 |
The formulas
For a correlation, t = r × √(df ÷ (1 − r²)) with df = n − 2. From sample numbers: one-sample t = (x̄ − μ₀) ÷ (s ÷ √n) with n − 1 degrees of freedom; two-sample Welch t = (x̄₁ − x̄₂) ÷ √(s₁²/n₁ + s₂²/n₂). Values worked out by this calculator:
| Statistic | Left-tailed | Right-tailed | Two-tailed |
|---|---|---|---|
| t = 2.1, df = 20 | 0.9757 | 0.02431 | 0.04862 |
| z = 1.96 | 0.975 | 0.025 | 0.05 |
| z = −1.5 | 0.06681 | 0.9332 | 0.1336 |
| χ² = 3.84, df = 1 | 0.95 | 0.05004 | 0.1001 |
| F = 4.96, df = 1 and 10 | 0.9499 | 0.05009 | 0.1002 |
How to read the p-value
If the p-value is at or below your significance level α, the result is statistically significant and you reject the null hypothesis; if it is above α you do not have enough evidence to reject it. Rejecting or not rejecting says nothing about how big or useful the effect is, and a p-value above α does not prove the null hypothesis true. Choose α before you look at the data.
Critical values at α = 0.05
A statistic past the critical value gives a p-value below α. The calculator marks these lines on the curve for your α.
| Test (α = 0.05) | Two-tailed critical value | Right-tailed critical value |
|---|---|---|
| z (normal) | ±1.96 | 1.645 |
| t, 5 degrees of freedom | ±2.571 | 2.015 |
| t, 10 degrees of freedom | ±2.228 | 1.812 |
| t, 20 degrees of freedom | ±2.086 | 1.725 |
| t, 30 degrees of freedom | ±2.042 | 1.697 |
| t, 60 degrees of freedom | ±2 | 1.671 |
| t, 120 degrees of freedom | ±1.98 | 1.658 |
| chi-square, 1 degree of freedom | 0.001 and 5.024 | 3.841 |
| chi-square, 2 degrees of freedom | 0.051 and 7.378 | 5.991 |
| chi-square, 3 degrees of freedom | 0.216 and 9.348 | 7.815 |
| chi-square, 5 degrees of freedom | 0.831 and 12.833 | 11.07 |
| chi-square, 10 degrees of freedom | 3.247 and 20.483 | 18.307 |
Worked example
A t-test gives t = 2.1 with 20 degrees of freedom. Two-tailed: p = 0.04862, below 0.05, so the difference is significant at the 5% level. Right-tailed: p = 0.02431. For r = 0.45 from 25 pairs, t = 2.417 with 23 degrees of freedom and a two-tailed p-value of 0.024.
Related tools
For the spread of your data use the standard deviation calculator, to average a list of numbers the average calculator, and for percentages the percentage calculator.
Sources
- The definition — the p-value is the probability, assuming the null hypothesis is true, of a test statistic at least as extreme as the one observed — and the two-tailed rule p = 2 × min(left tail, right tail) are the standard ones used by Omnicalculator's p-value calculator, GraphPad QuickCalcs, calculator.net, Social Science Statistics and Statology, all opened on 2026-11-10.
- The tails come from the cumulative distribution functions of the normal, t, chi-square and F distributions (regularised incomplete gamma and beta functions). For a correlation, t = r × √(df ÷ (1 − r²)) with df = n − 2. Checked against published values such as t = 2.1 with 20 degrees of freedom → two-tailed p = 0.048618.
- The 0.05 level is a convention, not a law; pick your α before you look at the data. Statistical significance does not measure how large or important an effect is.
For information only. Check your own study design before drawing conclusions.
Common questions
How do I find a p-value from a test statistic?
Pick the statistic (z, t, chi-square, F or r), type its value and, if it has them, the degrees of freedom, then choose the tail. For t = 2.1 with 20 degrees of freedom the two-tailed p-value is 0.04862 and the right-tailed one is 0.02431.
What is a p-value?
The p-value is the probability, if the null hypothesis were true, of getting a test statistic at least as extreme as the one you observed. A small p-value means your result would be unusual under the null hypothesis. It is not the probability that the null hypothesis is true.
What is the difference between a one-tailed and a two-tailed p-value?
A two-tailed test looks for a difference in either direction and counts both tails of the curve; a right-tailed test asks whether the value is higher than the reference, a left-tailed one whether it is lower. For the symmetric z and t distributions the two-tailed p-value is twice the smaller tail: z = 1.96 gives 0.025 in the right tail and 0.05 for two tails.
What p-value is statistically significant?
Compare it with the significance level α you chose before the test. If p ≤ α you reject the null hypothesis. 0.05 is the most common choice, and 0.01 and 0.10 are also used, but none of them is a law of nature.
How do I find the p-value from a t-score?
Choose t-score, type the t value and the degrees of freedom (n − 1 for one sample or a paired test, n₁ + n₂ − 2 for two samples with equal variance). t = 2.3 with 11 degrees of freedom gives a two-tailed p-value of 0.04203.
How do I find the p-value from a z-score?
Choose z-score and type it; no degrees of freedom are needed. z = 1.96 gives 0.05 for a two-tailed test, and z = −1.5 gives 0.06681 in the left tail.
How do I find the p-value for a chi-square test or an F test?
Choose chi-square or F, type the value and the degrees of freedom (F needs two: numerator and denominator). These tests almost always use the right tail: χ² = 3.84 with 1 degree of freedom gives 0.05004, and F = 4.96 with 1 and 10 degrees of freedom gives 0.05009.
Can I get the p-value from a correlation coefficient?
Yes. Choose correlation r and type r and the number of pairs n. The calculator turns r into t = r × √(df ÷ (1 − r²)) with df = n − 2: r = 0.45 with n = 25 gives t = 2.417 and a two-tailed p-value of 0.024.
Can I find the test statistic from a p-value?
Yes. Go advanced and type the p-value in “Find the statistic from a p-value”: you get the z, t, chi-square, F or r that produces it for the tail you chose. A two-tailed p-value of 0.05 with z gives ±1.96.
Can a p-value be exactly 0?
In practice no. Very large statistics give extremely small p-values, which this calculator shows in scientific notation such as 1.524 × 10⁻²³ for z = 10, rather than rounding to 0.