Pooled Standard Deviation Calculator
Combine the variability of two or more groups into a single pooled standard deviation — the value behind the two-sample t-test and Cohen’s d. Enter raw data or summary statistics and see the full weighted-variance solution, step by step.
Textbook-accurate formula · Reviewed July 2026 · Degrees-of-freedom weighting per standard pooled-variance methodology
Enter Your Groups
Pooled Result
Group SDs vs Pooled SD
What Is Pooled Standard Deviation?
Pooled standard deviation is a single estimate of variability that combines the spread from two or more independent groups. Instead of relying on one group’s standard deviation alone, it blends them into one number by taking a weighted average of the groups’ variances — with each group weighted by its degrees of freedom. The result is a more stable, reliable measure of the common spread, which is exactly why it sits at the heart of the two-sample t-test and effect-size measures like Cohen’s d.
The key idea is that when you assume two or more groups share the same underlying variability, combining their variances gives you a better estimate than either group provides on its own. A group of 200 people contributes more to the pooled figure than a group of 15 — and that weighting by sample size is what makes the estimate trustworthy.
The Pooled Standard Deviation Formula
For two groups, the pooled standard deviation formula is:
sp = √[ ((n₁ − 1)·s₁² + (n₂ − 1)·s₂²) / (n₁ + n₂ − 2) ]
Here n₁ and n₂ are the two sample sizes, and s₁ and s₂ are the two sample standard deviations. The denominator, n₁ + n₂ − 2, is the total degrees of freedom. For more than two groups, the formula generalizes naturally: sum each group’s degrees of freedom (nᵢ − 1) multiplied by its variance, divide by the total degrees of freedom, then take the square root.
How to calculate pooled standard deviation step by step
- Square each group’s standard deviation to get its variance.
- Multiply each variance by its degrees of freedom (the group’s sample size minus one).
- Add those weighted values together across all groups.
- Divide by the total degrees of freedom (total sample size minus the number of groups). This is the pooled variance.
- Take the square root of the pooled variance to get the pooled standard deviation.
The calculator above runs all five steps for you and shows the working, whether you enter raw data or summary statistics.
When Should You Use Pooled Standard Deviation?
Pooled standard deviation is the right tool whenever you need a single measure of within-group variability drawn from several independent samples — and you have reason to believe those groups share roughly equal variances. Its most common uses are:
- Two-sample t-tests — the pooled SD forms the standard error in the classic Student’s t-test for comparing two means.
- Cohen’s d effect size — the pooled SD is the denominator that standardizes the difference between two group means.
- ANOVA — the pooled variance corresponds to the mean square error (within-group variance).
- Lab sciences — chemists and biologists use it to gauge the repeatability of an experiment across runs.
The equal-variance assumption
Pooling rests on one important assumption: that the groups share a common underlying variance (called homogeneity of variance, or homoscedasticity). A quick sanity check is to compare the individual group standard deviations the calculator reports — if they’re close, pooling is sound. If one group’s SD is dramatically larger than another’s, the pooled estimate can be misleading, and you should consider Welch’s t-test instead, which doesn’t pool. Pooling is also inappropriate for paired or repeated-measures data.
| Aspect | Pooled Standard Deviation | Ordinary Standard Deviation |
|---|---|---|
| Purpose | Combine variability across groups | Measure spread within one dataset |
| Inputs | Two or more groups (n and SD each) | A single set of values |
| Weighting | By degrees of freedom (n − 1) | Not applicable |
| Main use | t-tests, Cohen’s d, ANOVA | Describing one sample’s spread |
| Key assumption | Groups share equal variance | None beyond the sample itself |
If you only need the spread of a single dataset, use our Standard Deviation Calculator instead. For a full summary of one dataset — mean, median, quartiles, and more — try the Descriptive Statistics Calculator.
Frequently Asked Questions
Pooled standard deviation combines the variability of two or more groups into a single estimate. It’s used as the denominator in the two-sample (Student’s) t-test, as the standardizing value in the Cohen’s d effect size, and as the within-group variability in ANOVA (where it equals the square root of the mean square error). It’s also used in lab sciences to assess how repeatable an experiment is across runs.
Square each group’s standard deviation to get its variance, multiply each variance by its degrees of freedom (sample size minus one), add these products together, divide by the total degrees of freedom (total sample size minus the number of groups), and take the square root. In formula form for two groups: s_p = √[((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2)]. This calculator does it automatically from raw data or summary statistics.
Weighting by degrees of freedom (n − 1) rather than raw sample size gives an unbiased estimate of the common population variance, consistent with how sample variance itself is calculated using Bessel’s correction. Practically, larger samples still get more influence — a group of 200 outweighs a group of 15 — but the n − 1 weighting correctly accounts for the reliability of each group’s variance estimate.
Ordinary standard deviation measures the spread within a single dataset. Pooled standard deviation combines the spread from two or more separate groups into one estimate, weighted by each group’s degrees of freedom. You use ordinary SD to describe one sample, and pooled SD when comparing groups — for example in a t-test or when calculating an effect size.
Avoid pooling when the groups have clearly unequal variances — if one group’s standard deviation is much larger than another’s, the pooled value becomes misleading. In that case, use Welch’s t-test, which doesn’t assume equal variances. Pooling is also inappropriate for paired or repeated-measures designs, where the observations aren’t independent between groups.
Yes, for two groups the pooled standard deviation always lies between the two individual standard deviations (or equals them if they’re identical). It sits closer to the SD of the larger group, because that group contributes more degrees of freedom to the weighted average. If your pooled value falls outside the range of the group SDs, there’s a calculation error.
Methodology & formulas used
This calculator uses the standard degrees-of-freedom-weighted pooled variance formula, reviewed July 2026. All computation runs locally in your browser; your data is never uploaded.
- Pooled variance: s²_p = Σ(nᵢ − 1)·sᵢ² / Σ(nᵢ − 1)
- Pooled standard deviation: s_p = √(s²_p)
- Degrees of freedom: df = (Σnᵢ) − k, where k is the number of groups
- Group SD (sample): when you paste raw data, each group’s SD uses Bessel’s correction (÷ nᵢ − 1)
The formula assumes independent groups with approximately equal population variances (homoscedasticity). When that assumption is doubtful, non-pooled methods such as Welch’s t-test are preferred. Results are rounded for display but computed at full precision.
References
- Statistics How To. Pooled Standard Deviation — Definition and Formula. statisticshowto.com
- NIST/SEMATECH. Standard Deviation and Variance — e-Handbook of Statistical Methods. itl.nist.gov
- Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences — pooled SD in effect size. Pooled variance (overview)