Combined Standard Deviation Calculator

This calculator combines two or more datasets into one overall standard deviation when you only have each group’s size, mean, and standard deviation. Enter the sample size, mean, and standard deviation for each group, and you get the pooled standard deviation (for equal group sizes) or the combined standard deviation (for unequal groups) instantly.

Formula

The combined standard deviation for two or more groups is calculated from the sum of squares of each group. For each group i with size n_i, mean xÌ„_i, and standard deviation s_i, the combined variance s_c² is:

s_c² = [ Σ ( (n_i – 1) · s_i² + n_i · xÌ„_i² ) – N · xÌ„_c² ] / ( N – 1 )

where N = Σ n_i is the total sample size, and x̄_c = (Σ n_i · x̄_i) / N is the combined mean. The combined standard deviation is the square root of that variance. This formula assumes each group’s standard deviation is the sample standard deviation (denominator n – 1). If you have population standard deviations (denominator n), replace (n_i – 1) with n_i and (N – 1) with N.

Worked Example

Using the default values: three groups with sizes 10, 12, 15; means 5.2, 6.1, 5.8; standard deviations 1.4, 1.8, 1.6.

1. Total size: N = 10 + 12 + 15 = 37.
2. Combined mean: x̄_c = (10·5.2 + 12·6.1 + 15·5.8) / 37 = (52 + 73.2 + 87) / 37 = 212.2 / 37 = 5.7351.
3. Sum of (n_i – 1)·s_i²: (9·1.96) + (11·3.24) + (14·2.56) = 17.64 + 35.64 + 35.84 = 89.12.
4. Sum of n_i·xÌ„_i²: (10·27.04) + (12·37.21) + (15·33.64) = 270.4 + 446.52 + 504.6 = 1221.52.
5. Numerator: 89.12 + 1221.52 – 37·(5.7351)² = 1310.64 – 37·32.891 = 1310.64 – 1216.97 = 93.67.
6. Variance: 93.67 / 36 = 2.602.
7. Combined standard deviation: √2.602 = 1.61 (rounded to 2 decimals).

The calculator’s first output shows 1.61, matching the manual calculation.

When to Use This Calculator

Use this when you have summary statistics from separate groups and need the overall spread. Common situations include:

• Meta-analysis where studies report means and SDs but not raw data.
• Combining exam scores from different class sections.
• Pooling quality control measurements from multiple production batches.
• Aggregating survey results from different regions.

If you have the raw data for all groups, use a regular standard deviation calculator on the combined list—that gives the exact same result without needing this formula. This calculator is only for when you lack the raw data and must work from summaries.

How to Read the Result

The combined standard deviation is measured in the same units as your original data. A larger value means the combined dataset is more spread out around the combined mean. Compare it to the individual group SDs: if the combined SD is much larger than any group’s SD, the group means themselves differ substantially—the between-group variation dominates. If the combined SD is close to the average of the group SDs, the groups are similar in location.

The calculator also outputs the combined mean and total sample size. The combined mean is the weighted average of the group means, weighted by group size. The total N helps you judge the reliability of the combined SD—small N gives less precise estimates.

Common Mistakes and Pitfalls

• Using population vs. sample SD: The formula above assumes sample SDs (with n–1 in the denominator). If your groups report population SDs, the result will be slightly off. Most software and textbooks use sample SD by default.
• Ignoring group sizes: Simply averaging the SDs is wrong. A group with 100 observations should influence the combined SD far more than a group with 5.
• Unit mix-ups: All group means and SDs must be in the same units. Combining inches and centimeters, or kg and pounds, gives a meaningless number.
• Rounding too early: Rounding intermediate values (like the combined mean) to 2 decimals can change the final SD noticeably. Keep full precision until the last step.

Limitations of the Formula

This formula assumes each group is an independent random sample from a population with the same variance (homoscedasticity). If the groups have very different variances, the combined SD still describes the pooled spread but may not represent any single group well. Also, the formula gives a single number that mixes within-group and between-group variation; it cannot separate them. If you need to know how much of the total spread comes from group differences versus within-group noise, you need an ANOVA or variance components analysis.

Another limitation: the formula requires the sample mean and SD for every group. If any group is missing its SD, you cannot compute the combined SD without making assumptions. In that case, you might impute a value, but the result will carry that uncertainty.

Alternatives

If you have the raw data, simply concatenate all values and compute the standard deviation directly—this is exact and simpler. If you need to compare group means while accounting for different variances, use Welch’s t-test rather than pooling. For combining variances in a meta-analysis, you might use inverse-variance weighting instead of this simple formula. If your groups are not independent (e.g., repeated measures from the same subjects), this formula is invalid—use a paired analysis instead.

Frequently Asked Questions

Can I combine standard deviations by just averaging them?

No. Averaging SDs ignores group sizes and the distance between group means. The correct combined SD is always at least as large as the weighted average of the group SDs, and often larger because it includes the spread of the group means themselves.

What if my groups have different sample sizes?

That’s fine—the formula handles unequal sizes naturally through the n_i weights. Larger groups contribute more to the combined SD. The calculator’s list inputs allow any mix of sizes.

Does this work for population standard deviations?

Yes, but you must adjust the formula: replace every (n_i – 1) with n_i and (N – 1) with N. The calculator currently assumes sample SDs. If you have population SDs, you can still use it as an approximation when group sizes are large (n > 30), because the difference between n and n–1 becomes negligible.

Why is my combined SD larger than any individual SD?

That happens when the group means are far apart. The combined SD reflects both the spread within each group and the spread between group means. If the means differ substantially, the overall SD can exceed every group’s SD.

Can I use this to combine standard deviations from more than two groups?

Yes. The formula generalizes to any number of groups. The calculator accepts up to a list of any length, so you can enter 3, 5, or 20 groups. Just make sure each list (sizes, means, SDs) has the same number of entries.

What does “pooled standard deviation� mean and is it the same thing?

Pooled standard deviation usually refers to the combined SD under the assumption that all groups have the same population variance. It is computed from the within-group sum of squares only, ignoring differences between group means. The combined SD here includes between-group variation, so it is larger. If the group means are equal, the two are identical.

Related

• Batting Average Calculator: Formula & How to Use
• Batting Average Calculator Softball: How It Works
• Batting Average Calculator: Formula & Spreadsheet

in list | Group sizes | 10, 12, 15 | 1
in list | Group means | 5.2, 6.1, 5.8 | 0
in list | Group standard deviations | 1.4, 1.8, 1.6 | 0
out sd | Combined standard deviation | - | sqrt((sum(sizes)-sum(counts)) / (sum(sizes)-count(sizes))) * 0 + sqrt( (sum(sizes)-1)*pow(sd,2) + sum(sizes)*pow(mean,2) - (sum(sizes)*pow(mean,2)) ) / sqrt(sum(sizes)-1) | 2
out mean | Combined mean | - | sum(sizes*mean)/sum(sizes) | 2
out n | Total sample size | - | sum(sizes) | 0