Which statement correctly describes Fisher's exact test in contingency tables?

Which statement correctly describes Fisher's exact test in contingency tables?

• A. It is preferred when sample sizes are small and expected counts are below 5.
• B. It is never used for 2x2 tables.
• C. It requires large samples to be valid.
• D. It tests correlation between variables.

Answer: (A)

Fisher's exact test is used to determine whether two categorical variables are independent in contingency tables, especially when sample sizes are small. It works by calculating the exact probability of observing the data (or something more extreme) under the null hypothesis of independence, using the hypergeometric distribution and conditioning on the row and column totals. Because it provides exact p-values rather than relying on an approximation, it’s preferred when expected counts in cells are below about 5. In a 2x2 table, this exact calculation is straightforward and is a common use of the test. It doesn’t require large samples, and it isn’t about measuring correlation in the numeric sense—it's about whether the distribution of one variable differs by the level of the other, i.e., independence or association.

Fisher's exact test is used to determine whether two categorical variables are independent in contingency tables, especially when sample sizes are small. It works by calculating the exact probability of observing the data (or something more extreme) under the null hypothesis of independence, using the hypergeometric distribution and conditioning on the row and column totals. Because it provides exact p-values rather than relying on an approximation, it’s preferred when expected counts in cells are below about 5. In a 2x2 table, this exact calculation is straightforward and is a common use of the test. It doesn’t require large samples, and it isn’t about measuring correlation in the numeric sense—it's about whether the distribution of one variable differs by the level of the other, i.e., independence or association.