Skip to content
K Knidox Search…
Statistics · Inference

P-Value Calculator

Find the p-value from a z-score for left-, right-, or two-tailed tests.

p-value
0.0500p < 0.05

Statistically significant at α = 0.05.

Normal distribution
z = 1.96

The p-value is the tail area beyond your test statistic under the null. From a z-score, a two-tailed p is 2 × P(Z ≥ |z|). For z = 1.96, P(Z ≥ 1.96) = 0.0250, so p = 2 × 0.0250 = 0.0500 — exactly the 5% threshold, making the result borderline significant.

What a p-value means

The p-value is the probability of observing a result at least as extreme as yours, assuming the null hypothesis is true. It’s the area in the tail (or tails) of the distribution beyond your test statistic. A small p-value means your data would be surprising if nothing were going on — so it counts as evidence against the null. It is not the probability that the null is true.

two-tailed p = 2 × P(Z ≥ |z|)

right-tailed = P(Z ≥ z); left-tailed = P(Z ≤ z)

Worked example

A z-statistic of 1.96 on a two-tailed test.

  1. 1
    Find the cumulative area to the left. From the normal table, P(Z ≤ 1.96) = 0.9750.
  2. 2
    Take the upper-tail area. P(Z ≥ 1.96) = 1 − 0.9750 = 0.0250.
  3. 3
    Double it for two tails. 2 × 0.0250 = 0.0500 — exactly the 5% threshold, so this is borderline significant.

Common two-tailed critical z-values

The z beyond which the two-tailed p falls below each α.

Significance αTwo-tailed |z|p at that z
0.101.6450.10
0.051.9600.05
0.012.5760.01
0.0013.2910.001

Reading and reporting the result

Choosing a tail. Use a two-tailed test when a difference in either direction matters, and a one-tailed test only when you have a directional prediction made before seeing the data. A one-tailed p is half the two-tailed p in the predicted direction.

p < 0.05 is a convention, not a verdict. The 0.05 cutoff is arbitrary. Report the exact p-value and your chosen α so readers can judge for themselves; a p of 0.049 and 0.051 are practically identical.

A large p-value isn’t proof of no effect. Failing to reject the null means you lack evidence for an effect — often because the sample is too small — not that the effect is zero.

Significance is not size. With a big enough sample, trivial differences become statistically significant. Always pair the p-value with an effect size or a confidence interval. If your statistic comes from raw data rather than a known z-score, compute it first.

What distribution does this use?
The standard normal (z) distribution, evaluated with a high-accuracy error-function approximation.
Is p < 0.05 always significant?
0.05 is a common threshold, not a law. Report the actual p-value and your chosen α, and treat values near the cutoff with care.
I have a t-statistic, not z.
For large samples the normal approximation is close. For small samples, use the T-Test calculator, which evaluates the p-value against the t-distribution with the right degrees of freedom.
What’s the difference between one- and two-tailed?
A two-tailed test splits α across both tails and detects a difference in either direction; a one-tailed test puts all of α in one tail and only detects a difference in the predicted direction.
Does the p-value tell me the effect is large?
No. It measures how compatible the data are with the null, not how big the effect is. Pair it with an effect size or confidence interval.
Does a small p-value mean the null is false?
No. The p-value is the chance of data this extreme if the null were true — not the probability that the null is true. A small p is evidence against the null, not proof.