Part of the Maths & Statistics suite · 235 calculators

Fisher's Exact Test Calculator

The exact p-value for a 2×2 table — no approximation, valid at any sample size, and the correct choice when counts are small.

With the row and column totals fixed, the number of ways to fill the table follows the hypergeometric distribution exactly.

Results update as you type
Results
Two-tailed p-value
0.06977852
One-tailed p-value
Probability of exactly this table
Odds ratio
Decision at 5%
Total observations
Expected count in cell a
Reviewed September 2026. Pure mathematics: the result does not depend on where you are. Terminology follows the Australian Curriculum (maths, brackets, 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 fisher's exact test

How the fisher's exact test calculator works

With the row and column totals fixed, the number of ways to fill the table follows the hypergeometric distribution exactly. Fisher's test sums the probability of the observed table and every table at least as extreme.

Because it is exact rather than asymptotic, it is valid however small the counts are — which is why it is the standard for tables where chi-square's expected-count rule fails.

Formula: P = C(a+b, a) C(c+d, c) / C(n, a+c)

Worked examples

InputsTwo-tailed p-valueNote
8/2 against 3/70.06977852p = 0.070 — not quite significant
The lady tasting tea0.02857143Fisher's original 1935 experiment: p = 0.029
No association1p = 1

Frequently asked questions

When should I use Fisher's exact test?

Any 2×2 table where an expected count falls below 5 — and, since it is exact, you can use it at any size.

Why is it called exact?

Because it computes the p-value from the exact hypergeometric distribution rather than approximating it with a chi-square curve.

What was the lady tasting tea?

Fisher's 1935 example: a colleague claimed she could tell whether milk went into the cup before or after the tea. Eight cups, four each way, all correct — p = 0.029 one-tailed.

Why do one- and two-tailed p-values differ so much?

Because the hypergeometric distribution is usually asymmetric, so the opposite tail rarely mirrors the observed one.

Is it conservative?

Yes, somewhat — conditioning on the margins makes it slightly less powerful than an unconditional exact test. It is nonetheless the standard.

Where these figures come from

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