How To Calculate Fisher’S Exact Test In Excel

Calculate p-values for 2×2 contingency tables with precise statistical analysis

Complete Guide: How to Calculate Fisher’s Exact Test in Excel

Fisher’s Exact Test is a statistical test used to determine if there are nonrandom associations between two categorical variables in a 2×2 contingency table. This guide provides step-by-step instructions for performing Fisher’s Exact Test in Excel, including manual calculations and using Excel functions.

When to Use Fisher’s Exact Test

• When you have a 2×2 contingency table
• When sample sizes are small (typically when expected cell counts are less than 5)
• When you need exact p-values rather than approximations
• For both one-tailed and two-tailed tests

Understanding the 2×2 Contingency Table

A 2×2 contingency table organizes your data into four cells representing different combinations of two categorical variables:

	Variable B (Present)	Variable B (Absent)	Total
Variable A (Present)	a (Cell A)	b (Cell B)	a + b
Variable A (Absent)	c (Cell C)	d (Cell D)	c + d
Total	a + c	b + d	N (a + b + c + d)

Step-by-Step Calculation in Excel

Method 1: Using Excel’s Built-in Function (Excel 2010 and later)

1. Enter your 2×2 table data in Excel (cells A1:B2 for the four values)
2. Click on an empty cell where you want the p-value to appear
3. Type the following formula:

   =FISHERTEST(A1:B2)

4. Press Enter to get the two-tailed p-value
5. For a one-tailed test, you’ll need to divide the result by 2

Method 2: Manual Calculation Using Hypergeometric Distribution

For versions of Excel without the FISHERTEST function, you can calculate it manually:

1. Calculate the row totals (R1 = a+b, R2 = c+d)
2. Calculate the column totals (C1 = a+c, C2 = b+d)
3. Calculate the grand total (N = a+b+c+d)
4. Calculate the probability of the observed table using the hypergeometric formula:

   = (FACT(a+b)*FACT(c+d)*FACT(a+c)*FACT(b+d))/(FACT(a)*FACT(b)*FACT(c)*FACT(d)*FACT(a+b+c+d))

5. Calculate probabilities for all possible tables with the same marginal totals
6. Sum probabilities ≤ observed probability (for two-tailed test, include both tails)

Interpreting the Results

The p-value from Fisher’s Exact Test helps you determine whether to reject the null hypothesis:

• If p ≤ 0.05: Reject the null hypothesis (statistically significant association)
• If p > 0.05: Fail to reject the null hypothesis (no significant association)

Interpretation Guide for Fisher’s Exact Test Results

P-value Range	Interpretation	Strength of Evidence
p > 0.05	Not significant	No evidence against null hypothesis
0.01 < p ≤ 0.05	Significant	Moderate evidence against null hypothesis
0.001 < p ≤ 0.01	Highly significant	Strong evidence against null hypothesis
p ≤ 0.001	Very highly significant	Very strong evidence against null hypothesis

Common Applications of Fisher’s Exact Test

• Medical research (comparing treatment outcomes)
• Genetics (association studies)
• Market research (consumer preference studies)
• Quality control (defect analysis)
• Social sciences (survey data analysis)

Limitations and Considerations

• Only applicable to 2×2 tables (for larger tables, use Chi-square or other tests)
• Can be conservative with large sample sizes
• Assumes fixed marginal totals (conditional test)
• Computationally intensive for large samples

Comparison with Chi-Square Test

Fisher’s Exact Test vs. Chi-Square Test

Feature	Fisher’s Exact Test	Chi-Square Test
Sample Size	Small samples (expected counts < 5)	Large samples (expected counts ≥ 5)
Calculation	Exact probabilities	Approximation
Table Size	Only 2×2 tables	Any size contingency table
Computational Complexity	High for large samples	Low
Assumptions	Fixed marginal totals	Expected frequencies ≥ 5 in most cells

Advanced Tips for Excel Users

1. Use named ranges for your table cells to make formulas more readable
2. Create a data validation dropdown for alternative hypothesis selection
3. Use conditional formatting to highlight significant results automatically
4. Combine with Excel’s ODDS RATIO calculation for more complete analysis
5. For repeated tests, create a template workbook with pre-formatted tables

Authoritative Resources on Fisher’s Exact Test

NIST/SEMATECH e-Handbook of Statistical Methods – Fisher’s Exact Test Laerd Statistics – Fisher’s Exact Test Guide NIH/NLM – Practical application of Fisher’s Exact Test in biomedical research

Frequently Asked Questions

Can I use Fisher’s Exact Test for tables larger than 2×2?

No, Fisher’s Exact Test is specifically designed for 2×2 contingency tables. For larger tables (2×3, 3×3, etc.), you should use the Chi-square test of independence (if sample sizes are adequate) or the Freeman-Halton extension of Fisher’s Exact Test (though this is computationally intensive).

What’s the difference between one-tailed and two-tailed tests?

A one-tailed test looks for an effect in one specific direction (either “greater than” or “less than”), while a two-tailed test looks for an effect in either direction. Two-tailed tests are more conservative and generally preferred unless you have a strong theoretical reason to expect a directional effect.

Why does my p-value differ between Excel and other statistical software?

Small differences in p-values (typically in the 4th or 5th decimal place) can occur due to:

• Different algorithms for calculating factorials
• Different approaches to handling very small probabilities
• Different methods for the two-tailed test calculation

These differences are usually negligible for practical purposes.

Can I use Fisher’s Exact Test for paired data?

Yes, Fisher’s Exact Test can be used for paired data when analyzing 2×2 tables of matched pairs (McNemar’s test is the alternative for paired binary data when you want to test for changes). The same 2×2 table structure applies, but the interpretation focuses on discordant pairs.