Resources · Education
Variance vs Standard Deviation
Variance and standard deviation are two ways of measuring how spread out a set of numbers is. Both are based on the same idea (average squared distance from the mean), but they finish differently. Variance leaves the answer in squared units. Standard deviation takes the square root and lands back in the same units as your data, which is usually easier to read. This guide explains the difference, when to use each one, and how the two numbers relate.
6 min read
What is variance?
Variance is the average of the squared distances between each value in a data set and the mean. To compute it: find the mean, subtract it from every value, square those differences, and average the squares.
Variance is a real, useful number, but it lives in squared units. If your data is in dollars, variance is in squared dollars. If your data is in minutes, variance is in squared minutes. That makes it harder to read on its own.
For a fast answer on any list of numbers, the variance calculator reports both sample and population variance in one step.
What is standard deviation?
Standard deviation is the square root of variance. That single extra step puts the answer back into the same units as your data, which usually makes it easier to read and easier to compare to the mean.
For a longer plain-language overview, see What Is Standard Deviation? For a quick computed answer, the standard deviation calculator reports the mean, both standard deviation variants, and both variances at once.
The main difference between variance and standard deviation
The single difference: units.
- Variance is in squared units.
- Standard deviation is in the same units as your data.
They describe the same underlying spread. Picking one over the other comes down to whether you want a value you can read at a glance (standard deviation) or a value that behaves nicely in further math (variance).
Why variance uses squared units
Variance squares each deviation from the mean before averaging. Squaring does two useful things at once:
- It removes the sign of each deviation, so values above and below the mean both contribute positively to the spread.
- It gives more weight to values that are far from the mean, which makes spread more sensitive to outliers.
The cost is that the answer ends up in squared units. Your data may be in test points, but the variance is in squared test points, which is hard to picture. Standard deviation fixes that by taking the square root at the end.
Why standard deviation is easier to interpret
Standard deviation reads naturally because it shares units with the data and with the mean.
If a class has an average score of 80 and a standard deviation of 5, the typical student is within 5 points of the average. That is easy to picture. The variance for the same class would be 25 squared points, which most people cannot picture at all.
That is why most reports, dashboards, and explanations quote standard deviation rather than variance. The variance is sitting in the background; standard deviation is just easier to talk about.
Variance vs standard deviation formula
Each measure has population and sample versions. The structure is identical; only the denominator changes.
Variance
- Population: σ² = Σ(xᵢ − μ)² / N
- Sample: s² = Σ(xᵢ − x̄)² / (n − 1)
Standard deviation
- Population: σ = √σ²
- Sample: s = √s²
Variance relationship
standard deviation = √variance
The σ² and s² notation is a hint about the relationship: standard deviation is just σ or s, and squaring it gives you variance. For a step-by-step walkthrough of the math, see Standard Deviation Formula. For when to use n − 1 versus N in the denominator, see Sample vs Population Standard Deviation.
Simple example with numbers
Take the same small data set used in the other guides:
2, 4, 4, 4, 5, 5, 7, 9
- Mean = 5
- Sum of squared deviations = 32
From there, both pairs of numbers fall out:
Population
Divide squared deviations by N (8)
σ² = 4 · σ = 2
Sample
Divide squared deviations by n − 1 (7)
s² ≈ 4.5714 · s ≈ 2.1381
Notice that variance is exactly the square of standard deviation in each pair. The variance and standard deviation tell the same story; the variance just lives in squared units.
When to use variance
Variance is the better tool when the math demands it.
- Statistical tests. ANOVA, F-tests, and many regression diagnostics work with variances directly.
- Combining independent quantities. For independent values, the variance of their sum equals the sum of their variances. Standard deviations do not add the same way.
- Theoretical work. Many textbook formulas are written in terms of variance because squared units behave more cleanly under algebra.
When to use standard deviation
Standard deviation is the better tool when humans read the result.
- Reporting and dashboards. Numbers in the same units as the mean are easier for everyone to read.
- Quality control and tolerances. Tolerances are usually written in original units, not squared units.
- Day-to-day analysis. For a quick sense of how variable a data set is, standard deviation pairs naturally with the mean.
In practice, most people who care about spread reach for standard deviation first and only think about variance when the math forces them to.
Quick comparison table
A side-by-side view of how the two numbers differ:
| Variance | Standard deviation | |
|---|---|---|
| Meaning | Average squared deviation from the mean | Square root of variance |
| Symbol | σ² or s² | σ or s |
| Formula | Σ(x − mean)² / n (or n − 1) | √variance |
| Units | Squared units of the data | Original units of the data |
| Interpretation | Hard to read on its own | Easy, in the same units as the mean |
| Best used for | Math, statistical tests, theory | Reporting, dashboards, day-to-day |
| Example value (2, 4, 4, 4, 5, 5, 7, 9) | σ² = 4 · s² ≈ 4.5714 | σ = 2 · s ≈ 2.1381 |
Common mistakes
A few traps that catch people:
- Reporting variance as if it were standard deviation. Variance is in squared units. A variance of 25 in a class with an average of 80 is not 25 points; it is 25 squared points. Standard deviation is the square root of that, so the typical gap from the mean is 5 points.
- Forgetting the square root. Standard deviation is variance with one extra step. Skip that step and you have variance, not standard deviation.
- Treating variance as smaller than standard deviation always. When variance is between 0 and 1, the square root is actually larger. Variance and standard deviation only match at 0 and 1.
- Mixing up sample and population versions. Both variance and standard deviation come in sample (s², s) and population (σ², σ) flavors. Use the same flavor throughout. The full comparison is in Sample vs Population Standard Deviation.
- Comparing standard deviations across data sets in different units. A standard deviation of 5 dollars is not directly comparable to a standard deviation of 5 minutes. Coefficient of variation (standard deviation divided by the mean) is the right cross-context measure.
Quick summary
- Variance and standard deviation both measure spread.
- Variance is the average of squared distances from the mean.
- Standard deviation is the square root of variance.
- Variance is in squared units; standard deviation is in the original units.
- Use variance for math and theory; use standard deviation for reporting and intuition.
- The variance calculator and the standard deviation calculator both report the mean, both variants, and both spread measures at once.
Run the numbers
Three calculators that work directly with variance, standard deviation, and the z-scores derived from them.
Variance Calculator
Population and sample variance, mean, count, and standard deviation in one step.
Standard Deviation Calculator
Mean, population and sample standard deviation, and both variances from a list of numbers.
P Value Calculator
Convert a z-score (in standard deviation units) to a one-tailed or two-tailed p-value.
Frequently asked questions
Variance is the average of the squared distances from the mean. Standard deviation is the square root of variance. They describe the same spread, but variance is in squared units while standard deviation is in the original units of the data.
No. Variance and standard deviation are connected, but they are not the same number. Standard deviation equals the square root of variance, so they only match when both equal 0 or 1.
Variance involves squaring the deviations from the mean, which puts the answer in squared units (like squared dollars or squared minutes). Taking the square root undoes the squaring at the end and brings the result back into the original units, which makes it easier to interpret.
Standard deviation, almost always. Standard deviation is in the same units as your data, so a value of 5 dollars or 5 minutes is intuitive. Variance is in squared units (squared dollars, squared minutes), which most people cannot picture without converting back.
Variance is most useful in mathematical and statistical contexts where squared units are convenient: ANOVA, hypothesis tests, and any work where variances of independent values get added together. It is rarely the right thing to report to a non-technical audience.
Standard deviation is the right choice for everyday reporting, dashboards, and any explanation aimed at people who are not statisticians. It pairs naturally with the mean and reads in the same units as your data.
Yes. When variance is between 0 and 1, the square root is larger than the original number. For example, a variance of 0.25 has a standard deviation of 0.5. For variances above 1, the standard deviation is smaller; for variances below 1, it is larger; and at variance equal to 0 or 1, they match.
Yes for most data sets. The math is simple but tedious. The variance calculator and the standard deviation calculator each take a list of numbers and return the mean, both sample and population variants, and both spread measures in one step.