Pooled Variance Calculator

Data Input

Separate values with commas, spaces, tabs, or new lines
0 values entered

Enter data values to see results

What Is Pooled Variance?

Pooled variance combines the variances of two or more independent groups into a single estimate, on the assumption that the groups come from populations with the same variance. Each group's variance is weighted by its degrees of freedom (its size minus one), so larger groups count for more. It is the variance used in the classic two-sample t-test and the "mean square within" in one-way ANOVA.

Formula

Each group's sample variance is weighted by its degrees of freedom:

Pooled Variance (Two Groups)

sp² = [(n₁−1)s₁² + (n₂−1)s₂²] / (n₁ + n₂ − 2)
  • sp²Pooled variance
  • n₁, n₂Sizes of group 1 and group 2
  • s₁², s₂²Sample variances of group 1 and group 2

Pooled Variance (k Groups)

sp² = Σ[(nᵢ−1)sᵢ²] / Σ(nᵢ−1)
  • kNumber of groups
  • nᵢSize of group i
  • sᵢ²Sample variance of group i

How to Calculate Pooled Variance Step by Step

  1. Find each group's sample variance.Calculate s² for every group separately, dividing by n − 1.
  2. Note each group's size.Record n for each group.
  3. Weight each variance.Multiply each s² by its group's n − 1 and add the results.
  4. Add the degrees of freedom.Add up all the (n − 1) values.
  5. Divide.sp² = weighted total / total degrees of freedom.

Pooled Variance Examples

1Comparing Two Teaching Methods

Data:Method A (n=12, s²=18) | Method B (n=15, s²=22)
Solution:sp² = [(11)(18) + (14)(22)] / (11 + 14) = (198 + 308) / 25 = 506 / 25.
Answer:Pooled variance = 20.24

2Drug Trial: Two Dosage Groups

Data:Low dose (n=20, s²=9.5) | High dose (n=25, s²=11.2)
Solution:sp² = [(19)(9.5) + (24)(11.2)] / (19 + 24) = (180.5 + 268.8) / 43 = 449.3 / 43.
Answer:Pooled variance ≈ 10.45

3Three Manufacturing Shifts

Data:Morning (n=30, s²=4.2) | Afternoon (n=28, s²=5.1) | Night (n=25, s²=4.8)
Solution:sp² = [(29)(4.2) + (27)(5.1) + (24)(4.8)] / (29 + 27 + 24) = (121.8 + 137.7 + 115.2) / 80 = 374.7 / 80.
Answer:Pooled variance ≈ 4.68

4Two Class Sections

Data:Section 1 (n=18, s²=30) | Section 2 (n=22, s²=25)
Solution:sp² = [(17)(30) + (21)(25)] / (17 + 21) = (510 + 525) / 38 = 1035 / 38.
Answer:Pooled variance ≈ 27.24

How to Use This Calculator for Pooled Variance

The calculator above works on one list of numbers at a time. To pool several groups:

  1. Set the calculator to Sample.
  2. Enter the first group's raw values and note its sample variance (s²) and its count (n).
  3. Clear the input, enter the next group and note its s² and n. Repeat for every group.
  4. Put the figures into the formula above: multiply each s² by its n − 1, add them, and divide by the total of the (n − 1) values.

Why Pool Variances?

  • A pooled estimate uses all the data, so it is more stable than any single group's variance.
  • It is the variance the equal-variance (Student's) two-sample t-test is built on.
  • It generalises directly to three or more groups as the within-group mean square in ANOVA.

Where Pooled Variance Is Used

Two-sample t-test

The equal-variance t-test divides the difference in means by a standard error built from sp².

ANOVA

The mean square within groups (MSW) is the pooled variance of all the groups.

Effect sizes

Cohen's d divides the difference in means by the pooled standard deviation √sp².

Clinical trials

Comparing treatment and control groups when their variances are similar.

Education research

Comparing outcomes between classes taught different ways.

Assumptions and Limitations of Pooled Variance

Pooling is only appropriate when its assumptions hold:

Equal population variances

If the groups' variances clearly differ, the pooled figure describes none of them well. Levene's test is a common check.

Independent groups

If the same people appear in more than one group (before-and-after measurements, for example), use a paired analysis instead.

Unequal group sizes

A much larger group dominates the pooled estimate. Unequal sizes combined with unequal variances are the case where the pooled t-test is least reliable.

Pooled is not the same as combined

Pooled variance ignores differences between group means. The variance of all the data lumped together includes them and is larger whenever the means differ.

The alternative

Welch's t-test does not assume equal variances and does not pool. Many statisticians use it by default.

Frequently Asked Questions: Pooled Variance Calculator

Everything you need to know about variance.

What is pooled variance?

Pooled variance is a weighted average of the variances of two or more independent samples, used when you assume the populations share the same variance. For two groups: sp² = [(n₁ − 1)s₁² + (n₂ − 1)s₂²] / (n₁ + n₂ − 2). Larger samples get more weight.

When do you use pooled variance?

In the two-sample t-test with equal variances, in one-way ANOVA (where it is the mean square within groups), and whenever you want a single variance estimate from several groups believed to have the same spread. Check that assumption first, for example with Levene's test.

How do you calculate pooled variance step by step?

Group 1: n = 10, s² = 8. Group 2: n = 15, s² = 12. sp² = [(9)(8) + (14)(12)] / (10 + 15 − 2) = (72 + 168) / 23 = 240 / 23 ≈ 10.43.

What is the difference between pooled variance and combined variance?

Pooled variance averages the within-group variances and ignores differences between the group means. The combined variance of all the data put together also includes the spread between group means, so it is larger whenever the means differ.

What assumption does pooled variance require?

Equal population variances (homoscedasticity), plus independent groups. If the variances clearly differ, use Welch's t-test, which does not pool.

Can I pool variances from more than two groups?

Yes. For k groups: sp² = Σ[(nᵢ − 1)sᵢ²] / Σ(nᵢ − 1). This is the mean square within (MSW) in one-way ANOVA.