What is a Coefficient of Variation Calculator?
A coefficient of variation calculator (CV calculator) finds the standard deviation as a percentage of the mean: CV = (s / x̄) × 100%. Because the units cancel, the CV compares relative spread across datasets measured on different scales, such as the weight of mice versus elephants or the consistency of two production lines. A lower CV means the values sit more tightly around their average.
What the Coefficient of Variation Calculator does
Paste raw data to get the sample CV (s / x̄) or population CV (σ / μ), or enter a mean and standard deviation you already have. The calculator shows the CV as a percentage and as a ratio, the standard deviation it used, the mean-to-SD ratio and, for samples, a small-sample corrected CV, with every step of the calculation. Use it as a CV calculator (also written C.V.) for quality control, investment returns, survey data, biology or any other measurements on a ratio scale.
Use the CV when you want to know whether variability is large relative to the typical value. A standard deviation of 5 g is huge for a 20 g mouse and trivial for a 4,000 kg elephant; the CV makes that difference explicit. Chemists report the same number as the relative standard deviation; if you are checking lab replicates against an acceptance limit, the relative standard deviation (%RSD) calculator adds a pass/fail check and a confidence bound.
How to use the Coefficient of Variation Calculator
- Choose Raw data to paste values, or Mean and SD if you already know both.
- For raw data, choose Sample or Population to set whether the standard deviation divides by n − 1 or n.
- For summary values, enter the Mean and Standard deviation in the same units.
- Read the CV as a percentage, then use the table below to judge whether it is low or high. Check the notes: the calculator warns you when the mean is close to zero or when the data include zero or negative values.
How to interpret the coefficient of variation
The CV answers one question: how big is the typical deviation compared with the average? A CV of 5% means values usually sit within about 5% of the mean; a CV of 50% means they routinely sit half the mean away from it. A lower CV means more consistent data, and when you compare two groups, the one with the higher CV is relatively more variable, even if its standard deviation is smaller.
There is no universal cut-off for a “good” or “high” CV, because acceptable variability depends on what you measure. The ranges below are common rules of thumb, not standards; always use the limit from your own field, method or specification when there is one.
- CV above 100%: the standard deviation is larger than the mean. This is normal for strongly right-skewed data, but if your mean is close to zero it signals that the CV is unstable.
- Comparing CVs: only compare CVs of variables measured on a ratio scale with a true zero, and preferably with similar sample sizes.
| Context | CV usually read as low | CV usually read as high |
|---|---|---|
| Repeated lab or instrument measurements | Under 5% | Over 10–15% |
| Field experiments in agronomy (Pimentel-Gomes classification) | Under 10% | Over 20% (over 30% is “very high”) |
| Biological traits such as body size | Under 10% | Over 30% |
| Incomes, waiting times and other skewed data | Under 50% | Over 100% |
Coefficient of variation formula (CV formula)
| Symbol | Meaning |
|---|---|
| s, σ | sample or population standard deviation |
| x̄, μ | sample or population mean |
| n | sample size |
The only difference between the two CV formulas is the standard deviation: the sample version divides the sum of squared deviations by n − 1, the population version by n. The mean is calculated the same way in both.
The sample CV slightly underestimates the population CV in small samples. For roughly normal data, multiplying by (1 + 1 / (4n)) removes most of that bias; with n = 8 the factor is 1.03125.
Worked example: sample CV
How consistent is a packing machine?
Problem. A quality inspector weighs eight bags of a snack sold as 50 g and gets 52.1, 49.8, 50.6, 51.4, 48.9, 50.2, 51.0, 49.5 g. The bags are a sample of the day’s output. What is the coefficient of variation?
- 01Mean: x̄ = 403.5 / 8 = 50.4375 g.
- 02Sum of squared deviations: Σ(x − x̄)² = 7.7388, so s = √(7.7388 / 7) = 1.0514 g.
- 03CV = s / x̄ = 1.0514 / 50.4375 = 0.020846.
- 04As a percentage: CV = 0.020846 × 100% ≈ 2.08%.
Answer. The CV is about 2.08%: bag weights typically differ from their mean by about 2% of the mean, which is tight filling. The small-sample corrected CV is 2.15%. These are the calculator’s default values, so you can check each step above.
Worked example: population CV
A small workshop has exactly five machines, and the owner wants the spread of yesterday’s output across all of them. The values are 120, 135, 128, 142, 125 parts. Because these are every machine, not a sample, use the population formula.
- Mean: μ = 650 / 5 = 130 parts.
- Sum of squared deviations: Σ(x − μ)² = 100 + 25 + 4 + 144 + 25 = 298.
- Population standard deviation: σ = √(298 / 5) = √59.6 = 7.7201 parts.
- CV = σ / μ = 7.7201 / 130 = 0.059385, or 5.94%.
- For comparison, the sample formula would give s = √(298 / 4) = 8.6313 and a CV of 6.64%. The difference shrinks as n grows.
Comparing spread across different scales
The standard deviation alone cannot tell you which group is more variable when the groups have very different means. The CV can. In the table below the elephants have a standard deviation 120,000 times larger than the mice, yet relative to their size they are less variable.
The same logic applies to comparing two assays with different concentration ranges, the volatility of a cheap and an expensive stock, or the consistency of small and large production lines.
| Group | Mean | Standard deviation | CV |
|---|---|---|---|
| Adult mice (body mass) | 25 g | 2.5 g | 10% |
| Adult elephants (body mass) | 4,000 kg | 300 kg | 7.5% |
Coefficient of variation vs standard deviation
Both describe spread, but they answer different questions. Use the standard deviation when you need the spread in the original units, for example to build a confidence interval or a control chart. Use the CV when you need to compare spread between groups whose means or units differ.
| Standard deviation | Coefficient of variation | |
|---|---|---|
| Units | Same as the data (g, mm, $) | None: a ratio or percentage |
| Formula | s = √[Σ(x − x̄)² / (n − 1)] | CV = s / x̄ × 100% |
| Compare groups with different means? | No | Yes |
| Works for data that can be negative? | Yes | No: needs a true zero |
| Changes if you convert g to kg? | Yes, it scales with the units | No |
Common use cases
- Quality control and manufacturing: compare the consistency of production lines that make parts or packs of different sizes.
- Analytical chemistry and clinical labs: describe the precision of replicate measurements; for acceptance limits, use the %RSD calculator.
- Biology and ecology: compare variability of traits, such as body size, between species of very different scale.
- Finance: compare risk per unit of average return between investments with different average returns.
- Surveys and social data: compare how spread out incomes, spending or response times are between regions or groups.
Assumptions and limitations
- Ratio scale only. The CV needs a true zero. It is meaningless for temperatures in °C or °F, standardized scores, or any variable that can be negative, because moving the zero point changes the CV.
- Unstable near zero. When the mean is small relative to the standard deviation, tiny changes in the mean swing the CV wildly. A CV above 100% is a warning sign.
- There is no universal threshold for a “good” CV. Acceptable values depend on the field: a few percent can be poor for a reference balance and excellent for a biological assay.
- The CV inherits the standard deviation’s sensitivity to outliers. For skewed data, a robust alternative is the MAD divided by the median.
Frequently asked questions
How do you calculate the coefficient of variation (CV%)?
What is a good coefficient of variation?
What does a high coefficient of variation mean?
Is the coefficient of variation the same as relative standard deviation?
Should I use the sample or population standard deviation for the CV?
Why can’t I use the CV for temperatures in Celsius?
Can the coefficient of variation be greater than 100%?
How do I calculate the CV in Excel?
Cite this page
Statculator. (2026, November 14). Coefficient of Variation Calculator. Retrieved November 14, 2026, from https://statculator.com/descriptive-statistics/coefficient-of-variation-calculator/