What is Fisher's exact test?

To perform Fisher's exact test, choose Stat > Tables > Cross Tabulation and Chi-Square and click Other Stats.

Use Fisher's exact test to analyze a 2x2 contingency table and test whether the row variable and column variable are independent (H0: the row variable and column variable are independent).

The p-value from Fisher's exact test is accurate for all sample sizes, whereas results from the chi-square test that examines the same hypotheses can be inaccurate when cell counts are small.

For example, you can use Fisher's exact test to analyze the following contingency table of election results to determine whether votes are independent of voters' genders.

Gender Candidate A Candidate B
Female 9 26
Male 21 35

For this table, Fisher's exact test produces a p-value of 0.263. Because this p-value is greater than common levels of α, you cannot reject the null hypothesis. Therefore, even though there may seem to be a difference between gender and candidate preference, with this data, there is insufficient evidence to indicate that a voter's gender affects their choice in the election. With a larger sample, you may be able to prove a difference.

Fisher's exact test is based on the hypergeometric distribution. Therefore, the p-value is conditional on the marginal totals of the table.

By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy