Part of the Math & Statistics suite · 235 calculators

Hypothesis Testing Calculator

Turn a test statistic into a p-value and a decision — for z, t, chi-square or F — with the critical value and a plain reading of what it means.

Every test reduces to the same act: locate your statistic on a reference distribution and measure the tail beyond it.

Results update as you type
Results
p-value
0.03572871
Decision
Critical value
Strength of evidence
One result in how many, under the null
What the p-value is not
Reviewed September 2026. Pure mathematics: the result does not depend on where you are. Terminology follows US usage (math, parentheses, decimal point).
No account required · Google Analytics off unless allowedCalculator arithmetic runs in your browserResults update as you type
All calculations run 100% in your browser. The calculator code does not submit your figures to GlobalCalc to obtain a result.
About hypothesis testing

How the hypothesis testing calculator works

Every test reduces to the same act: locate your statistic on a reference distribution and measure the tail beyond it. That tail is the p-value — the probability of a result at least this extreme if the null hypothesis were true.

What it is not: the probability that the null is true, the probability the result was a fluke, or a measure of how big the effect is. Those confusions are so common that this page states them explicitly.

Formula: p = P(statistic at least this extreme | H₀ true)

Worked examples

Inputsp-valueNote
z = 2.1, two-tailed0.03572871p = 0.036 — significant at 5%
t = 2.1 on 25 df0.04598428p = 0.046 — larger than the z version
A statistic of zero1p = 1

Frequently asked questions

What is a p-value?

The probability of observing a result at least this extreme if the null hypothesis were true.

What is it not?

It is not the probability the null is true, not the probability the finding is a fluke, and it says nothing about the size or importance of an effect.

Is p = 0.049 meaningfully different from p = 0.051?

No. The 0.05 threshold is a convention, and treating it as a boundary between real and unreal is the single biggest misuse of the whole framework.

What if I ran many tests?

Correct for it. Twenty independent tests at 5% give about a 64% chance of at least one false positive. Bonferroni is crude but safe; Benjamini–Hochberg is usually better.

Should I report anything else?

Always: the effect size and a confidence interval. A p-value alone answers only the least interesting question you can ask of your data.

Where these figures come from

Last checked: September 2026. Formulas are fixed by mathematics and do not change with tax years or regulations.