Combine standard deviations from multiple independent samples into a single pooled estimate of variation.
Last updated: March 2026
Pooled standard deviation is a weighted average of standard deviations from two or more independent samples. It provides a single estimate of the common standard deviation when you have reason to believe that multiple groups share the same underlying variability, even if their means differ.
This statistic is commonly used in hypothesis testing, particularly in two-sample t-tests and ANOVA. The pooling process combines the sum of squares from each group and divides by the total degrees of freedom, resulting in a more stable and reliable estimate than any single sample's standard deviation, especially when sample sizes are small.
The pooled standard deviation is particularly useful when comparing groups that are expected to have similar variability but different means—such as comparing test scores across different teaching methods, or measuring product quality across different production lines.
Comparing test scores from two classes:
Use pooled SD when conducting two-sample t-tests with equal variances assumption, comparing multiple groups in ANOVA, or when you need a more stable estimate of common variability across groups. Don't use it if groups have very different variances.
Pooled SD automatically accounts for different sample sizes by weighting each group's contribution by its degrees of freedom (n-1). Larger samples contribute more to the pooled estimate, which is statistically appropriate.
Regular SD is calculated from a single sample. Pooled SD combines multiple samples, assuming they share a common variance. It's essentially a weighted average that gives you more data points for a more reliable estimate.
Degrees of freedom for a sample is n-1, where n is the sample size. It represents the number of independent values available to vary. Total df is the sum of df from all groups, used to divide the total sum of squares.
Yes! You can pool any number of samples (groups). This calculator supports up to 6 groups. The formula extends naturally: sum all the sum of squares and divide by the total degrees of freedom across all groups.
If groups have substantially different variances (heteroscedasticity), pooling is inappropriate and can give misleading results. Use Welch's t-test or separate variance estimates instead. Test for equal variances before pooling.
The pooled SD represents the typical spread of data points around their group mean, assuming all groups have the same underlying variability. Larger values indicate more spread; smaller values indicate data clustered closely around means.
Pooled variance is the intermediate step: it's the weighted average of sample variances. Standard deviation is simply the square root of variance, expressed in the same units as the original data for easier interpretation.
Related Tools