How to Calculate Confidnece Interval for Fisher's Exact Test

Fisher's exact test is a statistical method used to determine if there are non-random associations between two categorical variables. While the test provides a p-value indicating the strength of the association, calculating a confidence interval gives additional insight into the range of possible effect sizes.

What is Fisher's Exact Test?

Fisher's exact test is a statistical test used to analyze contingency tables with small sample sizes. It calculates the exact probability of observing the given distribution of data, assuming that there is no association between the variables.

The test is particularly useful when sample sizes are small and the assumptions of the chi-square test are violated. It provides a p-value that indicates whether the observed association between the variables is statistically significant.

Why Calculate Confidence Interval?

While the p-value from Fisher's exact test tells us whether an association exists, the confidence interval provides a range of plausible values for the true effect size. This is particularly important in medical research, epidemiology, and other fields where understanding the magnitude of an effect is crucial.

Confidence intervals help researchers determine whether the effect is clinically or statistically meaningful. A wide confidence interval suggests that the true effect size could be very different from the observed value, while a narrow interval indicates more precision in the estimate.

How to Calculate the Confidence Interval

Calculating the confidence interval for Fisher's exact test involves several steps:

Calculate the odds ratio (OR) from the contingency table.
Determine the confidence level (typically 95%).
Use the exact binomial distribution to find the critical values.
Calculate the lower and upper bounds of the confidence interval.

Key Formula

The odds ratio is calculated as:

OR = (a * d) / (b * c)

Where a, b, c, and d are the cell counts in the 2×2 contingency table.

The exact method involves finding the range of odds ratios that would be compatible with the observed data at the chosen confidence level. This is typically done using statistical software or specialized calculators.

Example Calculation

Consider a 2×2 contingency table with the following counts:

	Exposed	Not Exposed
Disease	10	5
No Disease	15	70

The odds ratio is calculated as:

OR = (10 * 70) / (5 * 15) = 700 / 75 ≈ 9.33

Using exact methods, the 95% confidence interval for this odds ratio might be approximately 3.2 to 26.8, indicating a wide range of possible effect sizes.

Interpreting the Results

The confidence interval provides several key pieces of information:

Precision: A narrow interval indicates more precise estimates, while a wide interval suggests greater uncertainty.
Effect Size: The interval shows the range of plausible effect sizes.
Significance: If the interval does not include 1, the effect is statistically significant.

Always consider the context of your study when interpreting confidence intervals. A statistically significant result may not always be clinically meaningful, and vice versa.

FAQ

What is the difference between Fisher's exact test and a chi-square test?

Fisher's exact test is used for small sample sizes where the chi-square test's assumptions may not hold. It provides exact p-values rather than approximate ones.

How do I choose the confidence level?

The most common confidence level is 95%, but you can choose 90% or 99% depending on your study's requirements for precision and significance.

Can I calculate the confidence interval by hand?

For small tables, you can use exact methods, but for larger tables, statistical software is recommended for accuracy.

What does a confidence interval of [0.5, 2.0] mean?

This means there's a 95% chance that the true odds ratio falls between 0.5 and 2.0. If this interval includes 1, the effect is not statistically significant.