Pooled SD Calculator
Use this tool to calculate the pooled standard deviation for two independent groups. Optionally add means to estimate Cohen’s d.
What is pooled standard deviation?
The pooled standard deviation combines variability from two groups into one estimate. It is commonly used when comparing group means, especially in t-tests and effect size calculations like Cohen’s d. Instead of looking at each group’s spread separately, pooled SD gives a single weighted value based on both sample sizes.
In practice, pooled SD is most appropriate when the two groups are independent and have reasonably similar variances (the homogeneity of variance assumption).
Pooled SD formula (two independent groups)
spooled = √[((n1-1)s12 + (n2-1)s22) / (n1+n2-2)]
- n1, n2: sample sizes for groups 1 and 2
- s1, s2: sample standard deviations for groups 1 and 2
- n1+n2-2: pooled degrees of freedom
When should you use this calculator?
- Comparing two independent groups using classical parametric methods
- Computing Cohen’s d with pooled variability
- Summarizing spread across two samples in a weighted way
- Performing quick checks while writing reports or papers
Worked example
Suppose Group 1 has n = 25, SD = 4.2, and Group 2 has n = 30, SD = 5.1. The pooled variance is:
((24 × 4.22) + (29 × 5.12)) / 53 = 22.2198
Taking the square root gives a pooled SD of approximately 4.7148. That value becomes the shared spread estimate for both groups.
Pooled SD and Cohen’s d
If you provide both group means, the calculator also reports Cohen’s d:
d = (M1 - M2) / spooled
A common interpretation guide is:
- 0.2 = small effect
- 0.5 = medium effect
- 0.8 = large effect
Common mistakes to avoid
- Using population SD instead of sample SD
- Entering non-integer sample sizes
- Applying pooled SD when variances are wildly different
- Computing Cohen’s d with only one mean entered
FAQ
Can I use this for paired or repeated-measures data?
No. Paired designs require a different approach based on difference scores, not independent-group pooled SD.
What if one group is much larger than the other?
That is fine. The formula is weighted by degrees of freedom, so larger groups naturally contribute more to the pooled estimate.
What if pooled SD is zero?
If both group SDs are zero, there is no variability. In that case, Cohen’s d is undefined unless both means are identical.
Final thoughts
A pooled SD calculator saves time and reduces manual errors when working with two-group statistics. Use it as a quick companion for hypothesis testing, effect size reporting, and data interpretation in research, analytics, and education.