P-Value Calculator
Turn any test statistic into a p-value in one step. Enter a z-score, t-score, chi-square, or F statistic, pick your tail and significance level, and get the exact p-value, a clear reject / fail-to-reject decision, and a shaded distribution curve — with every step shown.
Textbook-accurate distribution math · Reviewed July 2026 · Follows the NIST e-Handbook on critical values and p-values
Enter Your Test Statistic
The z-score you obtained from your test. Negative values are fine.
Two-tailed is the usual default when you’re testing for any difference.
Your P-Value
Distribution Curve
The shaded area is your p-value
- P-value area (as extreme or more)
- Your statistic
What Is a P-Value?
A p-value is the probability of getting a result at least as extreme as the one you observed, assuming the null hypothesis is true. In plain terms, it measures how surprising your data would be if nothing were really going on. A small p-value means your result would rarely happen by chance alone, which counts as evidence against the null hypothesis. A large p-value means your result is unremarkable — the kind of thing that happens often even when there’s no real effect.
Every hypothesis test starts with a null hypothesis (H₀), the default claim that there’s no effect or no difference — the statistical equivalent of “innocent until proven guilty.” The p-value is the tool that weighs the evidence against that claim. It doesn’t prove anything on its own; it tells you how strong the evidence is, so you can decide whether to reject the null hypothesis or not.
How to Calculate a P-Value from a Test Statistic
Once you have a test statistic, the p-value is the area in the tail (or tails) of its distribution beyond that value. The exact formula depends on the distribution and the direction of your test:
From a z-score (normal distribution)
For a two-tailed z-test, the p-value doubles the area beyond the absolute value of z:
p = 2 × [1 − Φ(|z|)]
For a one-tailed test, you use just one tail: the right-tail area 1 − Φ(z), or the left-tail area Φ(z). For example, a z-score of 2.00 gives a two-tailed p-value of about 0.0455.
From a t-score
The t-distribution works the same way but depends on the degrees of freedom (df = n − 1 for a one-sample test). Its heavier tails give slightly larger p-values than the z-distribution for the same statistic, which corrects for the extra uncertainty of small samples.
From chi-square (χ²) or F
Chi-square and F tests are almost always right-tailed, because only large values indicate a departure from the null. The p-value is the upper-tail area beyond your statistic, using the appropriate degrees of freedom. Chi-square is common for tests of independence and goodness-of-fit; F appears in ANOVA and regression.
How to Interpret a P-Value
You interpret a p-value by comparing it to a pre-chosen significance level, alpha (α) — most often 0.05:
- p ≤ α: the result is statistically significant. There’s strong enough evidence to reject the null hypothesis.
- p > α: the result is not statistically significant. You fail to reject the null hypothesis — which is not the same as proving it true.
A p-value of 0.03 with α = 0.05, for instance, is significant: if the null hypothesis were true, you’d see data this extreme only 3% of the time, which is unusual enough to doubt the null. The calculator above makes this decision for you and shades the exact tail area on the curve.
What does p < 0.05 actually mean?
It means that if the null hypothesis were true, you would observe data at least this extreme less than 5% of the time. The 0.05 threshold is a long-standing convention, not a law of nature — some fields use 0.01 or stricter. What matters is choosing your alpha before you run the test, not after seeing the p-value.
The Big P-Value Misconception
Here’s the single most common error, made even by experienced researchers: a p-value is not the probability that the null hypothesis is true. A p-value of 0.03 does not mean there’s a 3% chance the null is correct, nor a 3% chance your result is a fluke. It’s calculated assuming the null is true, so it can’t also tell you the probability of that assumption.
Two more traps worth avoiding: a large p-value doesn’t prove there’s no effect (your study may simply lack the power to detect one), and a small p-value doesn’t measure the size or importance of an effect. Statistical significance and practical significance are different things — with a big enough sample, even a trivial effect can produce a tiny p-value.
| What a p-value IS | What a p-value is NOT |
|---|---|
| The chance of data this extreme if H₀ is true | The probability that H₀ is true |
| A measure of evidence against the null | A measure of the effect’s size or importance |
| Compared to a pre-set alpha to decide significance | Proof that the alternative hypothesis is true |
One-Tailed vs. Two-Tailed Tests
The tail direction should match your alternative hypothesis. Use a two-tailed test when you’re looking for any difference in either direction (the mean is simply “not equal” to the null value) — this is the safe default. Use a one-tailed test (left or right) only when you have a specific directional prediction decided in advance, such as “the new method is faster.” A two-tailed p-value is exactly double the corresponding one-tailed p-value for symmetric distributions like z and t, which is why one-tailed tests are more likely to reach significance — and why switching to one-tailed after seeing your data is considered a form of p-hacking.
Common Significance Thresholds
| Alpha (α) | Meaning | Typical use |
|---|---|---|
| 0.10 | Lenient — 10% false-positive risk | Exploratory / pilot studies |
| 0.05 | The conventional standard | Most social & business research |
| 0.01 | Strict — strong evidence required | Clinical & medical research |
| 0.001 | Very strict | High-stakes / physics |
Where P-Values Come From
A p-value is only the final step of a hypothesis test — it depends entirely on the test statistic you feed it. That statistic is usually a standardized distance from the null value: a z-score or t-score built from your sample mean, standard deviation, and sample size. If you need to compute that first, our Z-Score Calculator converts a raw value into the z you enter here, and it works from the mean and standard deviation you can get from the Standard Deviation Calculator. P-values and confidence intervals are also two sides of the same coin: if a 95% interval excludes the null value, the two-tailed p-value is below 0.05 — you can build that interval with our Confidence Interval Calculator.
Frequently Asked Questions
A p-value is the probability of getting a result at least as extreme as the one you observed, assuming the null hypothesis (no effect) is true. It measures how surprising your data would be if nothing real were going on. A small p-value means your result would rarely happen by chance, which is evidence against the null hypothesis. A large p-value means the result is unremarkable and consistent with the null.
A p-value below 0.05 means that if the null hypothesis were true, you would observe data at least this extreme less than 5% of the time. By the common convention, that’s considered statistically significant, so you reject the null hypothesis. The 0.05 threshold is a widely used standard, not a universal rule — some fields use 0.01 or stricter. The key is to choose your significance level before running the test.
No — this is the most common misconception. The p-value is calculated assuming the null hypothesis is true, so it cannot also tell you the probability that the null is true. A p-value of 0.03 does not mean there’s a 3% chance the null is correct or a 3% chance your result is a fluke. It only tells you how likely data this extreme would be under the null hypothesis.
Find the tail area of the standard normal distribution beyond your z-score. For a two-tailed test, p = 2 × [1 − Φ(|z|)], where Φ is the normal cumulative distribution function. For a right-tailed test, p = 1 − Φ(z); for a left-tailed test, p = Φ(z). For example, a z-score of 1.96 gives a two-tailed p-value of about 0.05. This calculator does it automatically and shades the corresponding area on the curve.
A two-tailed p-value tests for any difference in either direction and is the standard default. A one-tailed p-value tests for a difference in a single, pre-specified direction and is exactly half the two-tailed value for symmetric distributions like z and t. Use one-tailed only when you have a directional hypothesis chosen in advance. Switching to one-tailed after seeing your data to get a smaller p-value is considered p-hacking.
No. A smaller p-value means stronger evidence against the null hypothesis, but it says nothing about the size or importance of the effect. With a large enough sample, even a tiny, practically meaningless effect can produce a very small p-value. Always look at the effect size and confidence interval alongside the p-value to judge whether a statistically significant result actually matters in practice.
Methodology & formulas used
This calculator computes exact tail probabilities from the relevant distribution, reviewed July 2026. All computation runs locally in your browser; your data is never uploaded.
- Z (normal): two-tailed p = 2[1 − Φ(|z|)]; right p = 1 − Φ(z); left p = Φ(z). Φ is evaluated with the Abramowitz & Stegun 7.1.26 error-function approximation (accurate to about 1×10⁻⁷).
- T (Student): tail areas of the t-distribution with your degrees of freedom, via the regularized incomplete beta function.
- Chi-square (χ²): right-tail p = 1 − CDF, using the lower incomplete gamma function (series and continued-fraction expansions).
- F: right-tail p = 1 − CDF, via the regularized incomplete beta function with numerator and denominator degrees of freedom.
- Decision: the result is statistically significant when p ≤ α; otherwise you fail to reject the null hypothesis.
Chi-square and F tests are treated as right-tailed, the standard convention. Very small p-values are displayed as “p < 0.0001.” Results are computed at full precision and rounded for display.
References
- NIST/SEMATECH e-Handbook of Statistical Methods. Critical Values and P-Values. itl.nist.gov
- Abramowitz, M. & Stegun, I.A. (1964). Handbook of Mathematical Functions, formula 7.1.26. Error function overview
- Wikipedia. p-value — definition, calculation, and interpretation. en.wikipedia.org