Skip to content
Statculator
Descriptive Statistics

Coefficient of Variation Calculator

Free CV calculator: get the coefficient of variation (s / x̄ × 100%) of a sample or population from raw data or a mean and SD, with steps and how to read it.

Free, no sign-up Runs in your browserBy Hira Noor, BS MathematicsUpdated

Data input
8 values

Values should be measurements on a ratio scale with a true zero, such as mass, length, time or concentration.

Your data is a

Results

Coefficient of variation, CV
2.08%
As a ratio, s / x̄
0.0208
Count, n
8
Mean, x̄
50.4375
Sample SD, s
1.0514
Sample variance, s²
1.1055
Population CV, σ / μ (÷ n)
1.95%
Small-sample corrected CV, (1 + 1/(4n)) × CV
2.15%
Mean-to-SD ratio, x̄ / s
47.9697

Interpretation: The standard deviation is 2.08% of the mean, so values typically vary by about 2.08% of their average size, which common rules of thumb would call low relative spread (what counts as acceptable depends on your field). Because the units cancel, you can compare this figure directly with the CV of data measured on another scale.

Step-by-step solution

  1. 1

    Find the mean, x̄

    x̄ = Σx / n = 403.5 / 8 = 50.4375
  2. 2

    Find the sample standard deviation, s

    Σ(x − x̄)² = 7.7388
    s = √[7.7388 / (8 − 1)] = 1.0514
  3. 3

    Divide the standard deviation by the mean

    CV = s / x̄ = 1.0514 / 50.4375 = 0.020846
  4. 4

    Multiply by 100 to express it as a percentage

    CV = 0.020846 × 100% = 2.08%

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

  1. Choose Raw data to paste values, or Mean and SD if you already know both.
  2. For raw data, choose Sample or Population to set whether the standard deviation divides by n − 1 or n.
  3. For summary values, enter the Mean and Standard deviation in the same units.
  4. 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.
ContextCV usually read as lowCV usually read as high
Repeated lab or instrument measurementsUnder 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 sizeUnder 10%Over 30%
Incomes, waiting times and other skewed dataUnder 50%Over 100%

Coefficient of variation formula (CV formula)

Sample coefficient of variation
CV = (s / x̄) × 100%
Population coefficient of variation
CV = (σ / μ) × 100%
Small-sample correction (normal data)
CV* = (1 + 1 / (4n)) × CV
Formula symbols
SymbolMeaning
s, σsample or population standard deviation
x̄, μsample or population mean
nsample 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?

  1. 01Mean: x̄ = 403.5 / 8 = 50.4375 g.
  2. 02Sum of squared deviations: Σ(x − x̄)² = 7.7388, so s = √(7.7388 / 7) = 1.0514 g.
  3. 03CV = s / x̄ = 1.0514 / 50.4375 = 0.020846.
  4. 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.

  1. Mean: μ = 650 / 5 = 130 parts.
  2. Sum of squared deviations: Σ(x − μ)² = 100 + 25 + 4 + 144 + 25 = 298.
  3. Population standard deviation: σ = √(298 / 5) = √59.6 = 7.7201 parts.
  4. CV = σ / μ = 7.7201 / 130 = 0.059385, or 5.94%.
  5. 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.

GroupMeanStandard deviationCV
Adult mice (body mass)25 g2.5 g10%
Adult elephants (body mass)4,000 kg300 kg7.5%

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 deviationCoefficient of variation
UnitsSame as the data (g, mm, $)None: a ratio or percentage
Formulas = √[Σ(x − x̄)² / (n − 1)]CV = s / x̄ × 100%
Compare groups with different means?NoYes
Works for data that can be negative?YesNo: needs a true zero
Changes if you convert g to kg?Yes, it scales with the unitsNo

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%)?

Divide the standard deviation by the mean and multiply by 100 to get a percentage: CV = (s / x̄) × 100%. A dataset with mean 50 and standard deviation 5 has a CV of 10%. Use the sample standard deviation for sample data and the population standard deviation when you have every value.

What is a good coefficient of variation?

It depends on the field. Many analytical labs aim for CVs below 5% on replicate measurements, field experiments in agronomy treat under 10% as low and over 30% as very high, and biological traits often vary by 10–30%. A lower CV always means less variability relative to the mean.

What does a high coefficient of variation mean?

A high CV means the values are widely spread relative to their average, so the mean is a less reliable summary of a typical value. A CV above 100% means the standard deviation is larger than the mean, which is common for skewed data such as incomes but can also mean the mean is close to zero.

Is the coefficient of variation the same as relative standard deviation?

Yes. Relative standard deviation (RSD or %RSD) is the name used in chemistry and quality control for the CV expressed as a percentage: the standard deviation divided by the absolute value of the mean, times 100%.

Should I use the sample or population standard deviation for the CV?

Use the sample standard deviation (n − 1) when your data are a sample from a larger process or population, which is the usual situation. Use the population version only when the values are the entire group. For large n the two CVs are almost identical.

Why can’t I use the CV for temperatures in Celsius?

Celsius has an arbitrary zero, so the mean depends on where zero happens to sit. The same temperatures in Fahrenheit give a different CV even though the spread is physically identical. The CV is only meaningful for ratio-scale data such as kelvin, mass, length or counts.

Can the coefficient of variation be greater than 100%?

Yes, whenever the standard deviation exceeds the mean. This is common for strongly right-skewed positive data such as incomes or waiting times. It also happens when the mean is close to zero, in which case the CV is unstable and should be interpreted with care.

How do I calculate the CV in Excel?

Excel has no CV function. Use =STDEV.S(A2:A20)/AVERAGE(A2:A20) for a sample, or STDEV.P for a population, and format the cell as a percentage. The same formula works in Google Sheets; in R, use sd(x) / mean(x) * 100.

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/