📊 Coefficient of Variation Calculator
By ToolNimba Editorial Team · Updated 2026-06-19
CV = (standard deviation / mean) x 100. Sample CV uses the n-1 SD; population CV uses the n SD.
The coefficient of variation (CV), also called relative standard deviation (RSD), measures how spread out your data is relative to its mean. Paste a list of numbers and this calculator returns the count, mean, both the sample and population standard deviation, and the CV as a percentage. Because the CV is unitless, it lets you compare the variability of two data sets even when they have different means or different units.
What is the Coefficient of Variation Calculator?
The coefficient of variation answers a question a plain standard deviation cannot: how big is the spread compared with the size of the values themselves? It is defined as CV = (standard deviation / mean) x 100, expressed as a percentage. A salary data set with a standard deviation of $5,000 sounds volatile, but if the mean salary is $200,000 the CV is only 2.5%, which is very stable. The same $5,000 spread around a mean of $20,000 gives a CV of 25%, far more variable. By dividing the spread by the mean, the CV strips away the scale and leaves pure relative variability.
Because it is a ratio of two quantities in the same units, the CV itself has no units. That is what makes it useful for comparison. You can compare the consistency of a manufacturing process measured in millimetres against one measured in grams, or compare the volatility of a stock priced at $30 against one priced at $3,000, on a single common scale. A lower CV means the data clusters tightly around its mean (more consistent), while a higher CV means the values are more scattered relative to the average.
There is one important condition: the CV is only meaningful for data measured on a ratio scale with a true, meaningful zero and a positive mean. Temperature in Celsius, for example, is an interval scale (its zero is arbitrary), so a CV computed on Celsius readings is misleading. The CV also becomes unstable or undefined when the mean is near or equal to zero, because you are dividing by a tiny or zero denominator. This calculator reports the CV using both the sample standard deviation (dividing by n-1) and the population standard deviation (dividing by n), so you can pick the one that matches your situation: use the sample version when your numbers are a sample drawn from a larger group, and the population version when they are the entire group.
When to use it
- Comparing the variability of two data sets that have different means or different units, where a raw standard deviation would be misleading.
- Quality control in labs and manufacturing, where a low CV (often under 2% to 5%) signals a precise, repeatable process.
- Comparing the risk-per-unit-return of investments, where a lower CV means less volatility for each unit of expected return.
- Checking the reliability of repeated measurements or assay replicates by reporting the relative standard deviation.
How to use the Coefficient of Variation Calculator
- Paste or type your numbers into the box, separated by commas, spaces, or new lines.
- Read off the count, mean, and sum that the calculator computes from your data.
- Note the sample standard deviation (n-1) and population standard deviation (n).
- Use the sample CV if your numbers are a sample, or the population CV if they are the whole group. The CV is shown as a percentage.
Formula & method
Worked examples
Find the population and sample CV of the data set 2, 4, 4, 4, 5, 5, 7, 9.
- Count n = 8 and sum = 40, so mean = 40 / 8 = 5
- Squared deviations from the mean: 9, 1, 1, 1, 0, 0, 4, 16, which sum to 32
- Population variance = 32 / 8 = 4, so population SD = sqrt(4) = 2
- Population CV = (2 / 5) x 100 = 40%
- Sample variance = 32 / (8 - 1) = 4.5714, so sample SD = sqrt(4.5714) = 2.1380
- Sample CV = (2.1380 / 5) x 100 = 42.76%
Result: Mean 5 · Population CV 40% · Sample CV ≈ 42.76%
Find the population and sample CV of the data set 10, 12, 14, 16, 18.
- Count n = 5 and sum = 70, so mean = 70 / 5 = 14
- Squared deviations from the mean: 16, 4, 0, 4, 16, which sum to 40
- Population variance = 40 / 5 = 8, so population SD = sqrt(8) = 2.8284
- Population CV = (2.8284 / 14) x 100 = 20.20%
- Sample variance = 40 / (5 - 1) = 10, so sample SD = sqrt(10) = 3.1623
- Sample CV = (3.1623 / 14) x 100 = 22.59%
Result: Mean 14 · Population CV ≈ 20.20% · Sample CV ≈ 22.59%
Interpreting the coefficient of variation (rough, context-dependent guide)
| CV range | Relative variability | Typical reading |
|---|---|---|
| Under 10% | Low | Data is tightly clustered around the mean (very consistent). |
| 10% to 20% | Moderate | Reasonable spread; common in many real-world measurements. |
| 20% to 30% | High | Noticeably scattered relative to the average. |
| Over 30% | Very high | Highly variable; the mean may not represent the data well. |
Standard deviation versus coefficient of variation
| Property | Standard deviation | Coefficient of variation |
|---|---|---|
| Units | Same as the data | Unitless (a percentage) |
| Measures | Absolute spread | Spread relative to the mean |
| Compare across data sets | Only if same units and scale | Yes, even across units or scales |
| Needs a meaningful zero | No | Yes (ratio scale, positive mean) |
Common mistakes to avoid
- Using the CV on interval-scale data. The CV is only valid on a ratio scale with a true zero, such as length, mass, or count. Applying it to temperature in Celsius or Fahrenheit, or to calendar years, gives a meaningless number because the zero point is arbitrary.
- Computing the CV when the mean is near zero. Dividing by a mean close to zero makes the CV blow up to huge or unstable values, and a mean of exactly zero leaves it undefined. For data centred near zero, report the standard deviation instead.
- Mixing up sample and population standard deviation. Sample SD divides by n-1 and population SD divides by n, so they give slightly different CVs. Use the sample version for a sample of a larger group and the population version only when your data is the entire group.
- Forgetting the percentage multiplier. CV is usually reported as a percentage, so multiply the SD-over-mean ratio by 100. A ratio of 0.4 is a CV of 40%, not 0.4%. Reporting the raw ratio confuses readers expecting a percent.
Glossary
- Coefficient of variation (CV)
- The standard deviation divided by the mean, times 100. A unitless measure of relative variability, expressed as a percentage.
- Relative standard deviation (RSD)
- Another name for the coefficient of variation, used widely in chemistry and lab work.
- Standard deviation
- A measure of how far the values typically fall from the mean, in the same units as the data.
- Mean
- The arithmetic average: the sum of the values divided by how many there are.
- Ratio scale
- A measurement scale with a true, meaningful zero (length, mass, count), where ratios between values make sense.
- Sample vs population
- A sample is a subset drawn from a larger group; the population is the entire group. The sample SD divides by n-1, the population SD by n.
Frequently asked questions
What is the coefficient of variation?
The coefficient of variation (CV) is the standard deviation divided by the mean, multiplied by 100 to give a percentage. It measures how large the spread of a data set is relative to its average, so it is often called relative variability or relative standard deviation.
How do I calculate the coefficient of variation?
Find the mean of your data, find the standard deviation, then compute CV = (standard deviation / mean) x 100. This calculator does it automatically and shows the result using both the sample standard deviation (divide by n-1) and the population standard deviation (divide by n).
Is the coefficient of variation the same as relative standard deviation?
Yes. Relative standard deviation (RSD) is just another name for the coefficient of variation. The two terms are interchangeable, with RSD being more common in laboratory and analytical chemistry contexts.
What is a good coefficient of variation?
It depends on the field. In quality control a CV under about 2% to 5% is considered very precise. More broadly, under 10% is low variability, 10% to 30% is moderate to high, and over 30% means the values are very scattered relative to the mean. Always judge it against norms for your own data.
Why use the CV instead of the standard deviation?
The standard deviation is in the units of your data, so you cannot fairly compare a spread in millimetres with one in grams, or compare two data sets with very different means. The CV is unitless, so it puts relative variability on one common scale and makes such comparisons valid.
What happens when the mean is zero?
The coefficient of variation is undefined when the mean is zero, because the formula divides by the mean. When the mean is merely close to zero, the CV becomes very large and unstable. In those cases report the standard deviation instead, or use a measure suited to data centred near zero.