Part of the Maths & Statistics suite · 235 calculators

Pooled Standard Deviation Calculator

Combine the spread of two groups into a single estimate — the denominator of a two-sample t-test and of Cohen's d.

The pooled standard deviation is a weighted average of the two variances, weighted by degrees of freedom: s²ₚ = ((n₁−1)s₁² + (n₂−1)s₂²) ÷ (n₁+n₂−2).

Results update as you type
Results
Pooled standard deviation
5.988984
Pooled variance
Degrees of freedom
Standard error of the difference
Ratio of the two SDs
Is pooling safe?
Reviewed September 2026. Pure mathematics: the result does not depend on where you are. Terminology follows the Australian Curriculum (maths, brackets, decimal point).
No account required · Google Analytics off unless allowedCalculator arithmetic runs in your browserResults update as you type
All calculations run 100% in your browser. The calculator code does not submit your figures to GlobalCalc to obtain a result.
About pooled standard deviation

How the pooled standard deviation calculator works

The pooled standard deviation is a weighted average of the two variances, weighted by degrees of freedom: s²ₚ = ((n₁−1)s₁² + (n₂−1)s₂²) ÷ (n₁+n₂−2). Take the square root.

It assumes both groups share a common variance. If the two standard deviations differ by more than about a factor of two, that assumption is doing real damage and Welch's test — which does not pool — is the better tool.

Formula: sₚ = √(((n₁−1)s₁² + (n₂−1)s₂²) / (n₁+n₂−2))

Worked examples

InputsPooled standard deviationNote
SDs of 5 and 75.988984pooled ≈ 5.98
Equal SDs6pooled = 6 exactly
Very unequal13.5396pooling is unsafe here

Frequently asked questions

What is a pooled standard deviation?

A single spread estimate built from two groups, weighted by their degrees of freedom.

Why weight by n − 1?

Because each group's variance estimate is worth its degrees of freedom. A group of 100 should count far more than a group of 3.

When should I not pool?

When the two standard deviations differ substantially — more than about a factor of two — or when the group sizes are very unequal. Use Welch's t-test instead.

Is it just the average of the two SDs?

No. It averages the *variances*, then takes the root, so it is always at least as large as the plain average of the SDs.

Where is it used?

The denominator of the classic two-sample t-test, and of Cohen's d for effect size.

Where these figures come from

Last checked: September 2026. Formulas are fixed by mathematics and do not change with tax years or regulations.