A pooled standard deviation combines the spreads of two groups into one figure. The pooled standard deviation calculator takes the size, mean and standard deviation of two groups, and gives the pooled standard deviation, the pooled variance, the standard error of the difference, Cohen’s d, Hedges’ g and Glass’s delta.
How the calculation works
- The pooled variance is a weighted average of the two group variances, with weights equal to the degrees of freedom: sp² = ((n₁ − 1) × s₁² + (n₂ − 1) × s₂²) ÷ (n₁ + n₂ − 2).
- The pooled standard deviation is its square root, sp. It assumes the two groups share the same population spread.
- The standard error of the difference between the means is sp × √(1 ÷ n₁ + 1 ÷ n₂).
- Cohen’s d = (m₁ − m₂) ÷ sp. Hedges’ g multiplies d by 1 − 3 ÷ (4 × (n₁ + n₂) − 9) to correct a small bias, and Glass’s delta divides by the control group’s standard deviation alone.
Example
Group 1 has 30 people with a mean of 74.5 and a standard deviation of 8.2. Group 2 has 28 people with a mean of 69.1 and a standard deviation of 9.6. The pooled variance is (29 × 67.24 + 27 × 92.16) ÷ 56 = 79.255, so sp = 8.903. The standard error of the difference is 2.339, Cohen’s d is 5.4 ÷ 8.903 = 0.607, Hedges’ g is 0.598 and Glass’s delta is 0.563.
Why it matters
The pooled standard deviation is the denominator of the two-sample t-test and of the standard effect sizes. When a paper reports only group summaries, this calculator turns them into the numbers you need for a test, a power analysis or a meta-analysis.
Good to know
Pooling assumes equal population variances, so check that the two standard deviations are similar. The effect size labels of small, medium and large are rough conventions from Cohen and should be read in the context of the field. The calculator needs standard deviations and not standard errors, so convert if your source gives the latter.
Related tools
The Cohen’s d calculator computes the effect size from group summaries, and the standard error calculator gives the uncertainty of a mean.