Fisher Exact Test Calculator

Evaluate small sample associations with exact table probabilities. Review odds ratios, tails, and expected counts. Download neat reports for audits, research, and presentations today.

Calculator Form

Example Data Table

Group Response No Response Total
New Treatment 12 3 15
Standard Care 5 10 15
Total 17 13 30

This sample highlights a small study where exact testing is preferred over large sample approximations.

How Fisher Exact Testing Helps

Why analysts use it

Fisher Exact Test evaluates whether two categorical variables are associated in a 2x2 contingency table. It is ideal when sample sizes are small, margins are fixed, or expected counts are low. Many medical, laboratory, marketing, and quality studies depend on this exact approach.

What this calculator returns

This calculator produces exact left tailed, right tailed, and two tailed p values. It also reports the observed table probability, odds ratio, confidence interval, relative risk, expected counts, chi square statistics, and the phi coefficient. These outputs support interpretation and documentation.

When to prefer it

Use Fisher Exact Test when at least one expected cell count is small, when data are sparse, or when precision matters more than approximation speed. It is especially useful for pilot studies, randomized trials, A/B tests with low events, and safety signal reviews.

How to read the result

A small p value suggests that the observed pattern is unlikely under the null hypothesis of no association. The odds ratio and relative risk explain effect direction and strength. Expected counts and graph views help verify whether the table is balanced or skewed.

Reporting advice

For clear reporting, include the exact p value, the tail used, the 2x2 counts, and the odds ratio with its interval. If a cell contains zero, mention whether a continuity style correction was applied for ratio estimation. Keep the original counts visible in your report.

Formula Used

For a 2x2 table with cells a, b, c, and d, Fisher Exact Test conditions on fixed row and column totals.

Exact table probability:

P = ((a+b)! (c+d)! (a+c)! (b+d)!) / (a! b! c! d! n!)

where n = a + b + c + d.

Odds ratio: OR = (a × d) / (b × c)

Expected count: Eij = (row total × column total) / n

Two tailed exact p value: sum the probabilities of all valid tables with probability less than or equal to the observed table probability.

How to Use This Calculator

  1. Enter the four observed counts in the 2x2 table.
  2. Rename rows and columns if you want custom report labels.
  3. Select the reporting tail that matches your hypothesis.
  4. Choose the confidence level for interval estimates.
  5. Keep zero cell correction enabled when sparse data appear.
  6. Press the calculate button to show the result section.
  7. Review p values, ratios, totals, and expected counts.
  8. Export the summary using the CSV or PDF buttons.

FAQs

1. When should I choose Fisher Exact Test?

Choose it for 2x2 tables with small samples, low expected counts, or sparse outcomes. It gives exact probabilities instead of relying on approximate large sample assumptions.

2. What does the two tailed p value mean?

It measures how unusual the observed table is under no association. It includes all valid tables that are at least as extreme as the observed one, based on exact probability.

3. Why are left and right tailed values different?

Each one tests a directional hypothesis. Left tailed favors smaller top left counts than expected. Right tailed favors larger top left counts than expected.

4. What if one table cell is zero?

The exact p value remains valid. Ratio measures can become unstable or infinite, so the calculator can apply a small correction for more interpretable interval estimates.

5. Is Fisher Exact Test better than chi square?

It is usually better for very small samples or sparse tables. Chi square is faster for larger samples, but Fisher is more exact when counts are limited.

6. What does the odds ratio show?

It compares the odds of the column one event between row one and row two. Values above one suggest higher odds in row one.

7. Why are expected counts included?

Expected counts show what the table would look like under independence. They help you judge sparsity, imbalance, and whether approximate methods might be questionable.

8. Can I use this for business testing?

Yes. It works for marketing lift checks, response studies, small conversion tests, quality reviews, and any 2x2 categorical comparison with modest sample sizes.

Related Calculators

Mann-Whitney U Test CalculatorKruskal-Wallis H Test CalculatorKendall Tau Correlation CalculatorChi-Square Goodness of Fit CalculatorSign Test CalculatorKolmogorov-Smirnov One Sample Test CalculatorKolmogorov-Smirnov Two Sample Test CalculatorJonckheere-Terpstra Trend Test CalculatorMcNemar Test CalculatorBinomial Sign Test Calculator

Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.