Case 1
- Mean (μ)
- 50
- Standard deviation (σ)
- 5
10.00%
Open with these valuesCoefficient of Variation Calculator
10.00%
Result: 10.00 %The coefficient of variation states the standard deviation as a percentage of the mean, so the units drop out. That makes it the one spread measure you can carry between datasets that share nothing — kilograms against seconds, euros against counts. A lower value means the data sit more tightly around their average.
The numbers at a glance
Held fixed: Mean (μ) 50.0000.
| Standard deviation (σ) | Result (%) |
|---|---|
| 0.0000 | 0.00 |
| 2.0000 | 4.00 |
| 4.0000 | 8.00 |
| 5.0000Your value | 10.00 |
| 6.0000 | 12.00 |
| 8.0000 | 16.00 |
| 10.0000 | 20.00 |
Worked examples
Case 2
- Mean (μ)
- 100
- Standard deviation (σ)
- 15
15.00%
Open with these valuesCase 3
- Mean (μ)
- 20
- Standard deviation (σ)
- 5
25.00%
Open with these valuesHow it's calculated
CV = (σ ÷ μ) × 100
- StepEnter the mean of your dataset. It has to be greater than zero.
- StepEnter the standard deviation of the same data.
- ResultRead the spread as a percentage of the mean.
What this number means
Unit-free, so it travels between datasets
Dividing the standard deviation by the mean cancels the units, so what is left is a plain percentage. That is why a coefficient of variation can compare kilograms with seconds, which a raw standard deviation cannot.
Breaks down as the mean approaches zero
The formula divides by the mean, so the value grows without limit as the mean nears zero, and at exactly zero it has no value at all. That is why the mean entered here has to stay above zero.
Meant for positive ratio data
The measure is reserved for ratio data with a true zero and positive values — prices, weights, counts. Data that can turn negative falls outside what it is built for.
No universal high or low
As a rough guide, under about 10 percent is often called low variability, 10 to 30 percent moderate, and above 30 percent high. Judge the value against what is typical for your own kind of data rather than a fixed threshold.
Commonly misread
A coefficient of variation of 25 means a standard deviation of 25.
It means the standard deviation is 25 percent of the mean. A mean of 80 with a standard deviation of 20 gives 25, and so does a mean of 20 with a standard deviation of 5.
The dataset with the larger standard deviation is the more variable one.
Only when both share the same mean and units. A standard deviation of 10 on a mean of 200 gives 5, while 5 on a mean of 20 gives 25 — the smaller spread belongs to the more variable data.
The output is a ratio, so 10 means ten times the mean.
The ratio is multiplied by 100 and reported as a percentage. A value of 10 means the standard deviation is one tenth of the mean.
Reference table
| Mean, SD | Reading | CV in % |
|---|---|---|
| 200, 10 | Very consistent data | 5 |
| 50, 5 | Spread is one tenth of the mean | 10 |
| 100, 15 | Moderate spread | 15 |
| 80, 20 | A quarter of the mean | 25 |
| 20, 5 | Same ratio, much smaller numbers | 25 |
Questions
How do I calculate the coefficient of variation?
Divide the standard deviation by the mean, then multiply by 100: CV = (σ ÷ μ) × 100. For a mean of 50 and a standard deviation of 5 that is (5 ÷ 50) × 100 = 10 percent — the standard deviation is one tenth of the mean.
What does the coefficient of variation tell me?
It measures relative variability: how large the spread is compared with the average. Because it is a percentage with no units, it lets you compare the consistency of datasets that use different scales, which a raw standard deviation cannot do. A lower value means more consistent data.
What counts as a high or low coefficient of variation?
There is no universal cut-off — it depends on the field. As a rough guide, a value under about 10 percent is often called low variability, 10 to 30 percent moderate, and above 30 percent high. Always judge it against typical values for your own kind of data rather than a fixed threshold.
Why is the coefficient undefined when the mean is zero?
The formula divides by the mean, and division by zero has no value. A mean of zero also makes the ratio meaningless, which is why the measure is reserved for ratio data with a true zero and positive values — prices, weights, counts.
How is it different from the standard deviation?
The standard deviation reports spread in the original units, so 5 kg and 5 cm are not comparable. The coefficient of variation rescales that spread relative to the mean and reports a unit-free percentage, so variability can be compared across datasets of completely different magnitudes.
Sources and last check
Information, not professional advice.