Fisher’s Exact Test Calculator (2×2 Table)
Use this tool to compute one-tailed and two-tailed p-values for a 2×2 contingency table. Enter non-negative whole numbers only.
| Outcome Yes | Outcome No | |
|---|---|---|
| Group 1 | ||
| Group 2 |
Two-tailed p-value is computed using the standard “sum of tables with probability ≤ observed table” approach.
What is Fisher’s Exact Test?
Fisher’s Exact Test is a statistical method used to evaluate whether two categorical variables are associated in a 2×2 table. It is especially useful when sample sizes are small, expected cell counts are low, or when chi-square assumptions may be questionable.
Instead of relying on approximation, Fisher’s test computes exact probabilities under the null hypothesis that there is no association between row and column variables. This gives a precise p-value even for sparse data.
When should you use a Fisher test calculator?
- When your data can be represented as a 2×2 contingency table.
- When one or more expected frequencies are small (often < 5).
- When you need exact p-values rather than asymptotic approximations.
- When analyzing clinical, biological, behavioral, or A/B test outcomes with limited sample size.
How to use this calculator
Step 1: Fill in the 2×2 counts
Enter counts for all four cells. For example, Group 1 might be a treatment group and Group 2 a control group, while columns could represent “success” and “failure.”
Step 2: Set your alpha level
Alpha (α) defines your significance threshold. The default of 0.05 is common, but you can use stricter values like 0.01 if needed.
Step 3: Calculate and interpret
Click Calculate to get left-tailed, right-tailed, and two-tailed p-values, along with odds ratio estimates and a basic significance interpretation.
Understanding the results
- Two-tailed p-value: Tests for any association (in either direction).
- Left-tailed p-value: Tests whether Group 1 tends to have fewer “Yes” outcomes than expected.
- Right-tailed p-value: Tests whether Group 1 tends to have more “Yes” outcomes than expected.
- Odds ratio: Effect size indicator. Values above 1 suggest higher odds in Group 1; below 1 suggest lower odds.
Fisher vs. Chi-square: which is better?
Neither test is universally “better”; it depends on your data. The chi-square test is fast and widely used for larger samples. Fisher’s Exact Test is preferred when precision matters for small samples or sparse tables. If expected counts are low, Fisher’s method is usually the safer choice.
Common mistakes to avoid
- Entering percentages instead of raw counts.
- Using negative values or non-integers.
- Interpreting p-value as effect size (use odds ratio for magnitude).
- Ignoring study design, confounders, or multiple testing considerations.
Practical tip
Statistical significance does not automatically imply practical importance. Always interpret p-values alongside context, effect size, confidence intervals, and domain expertise.