The coefficient of variation (CV), also called the relative standard deviation, measures variability relative to the size of the values. A standard deviation of 5 g is huge for a 20 g sample but negligible for a 5 kg bag; dividing by the mean puts both on the same footing (25% versus 0.1%). Because the units cancel, the CV also lets you compare the consistency of quantities measured in different units, such as height in centimetres and weight in kilograms.
Laboratories use it to report the precision of assays, investors compare the risk per unit of return of different assets, and manufacturers track it as a measure of process consistency. A CV below about 10% usually indicates low variability, while above 30% the data are very spread out relative to their mean.
The CV only makes sense for ratio-scale data with a true zero and positive values: it is meaningless for temperatures in °C or for data that mix positive and negative values, where the mean can be close to zero. For small samples the ordinary CV slightly underestimates the population value; the corrected version multiplies it by (1 + 1/(4n)).
How to use the Coefficient of Variation Calculator
- Calculate from
- Data set
- Data are a
- Mean (x̄ or μ)
- Standard deviation (s or σ)
- Sample size (n)
Results update as you type, and everything is calculated in your browser.
How it works
From a data list the calculator finds the mean and the sample (n − 1) or population (n) standard deviation; from summary values it uses the mean and SD you enter. The CV is SD ÷ |mean|, shown as a percentage and as a ratio.
Worked example
Six bag weights in grams: 102, 98, 105, 97, 101, 99. The mean is 602 ÷ 6 = 100.33 g and the sample standard deviation is 2.944 g, so CV = 2.944 ÷ 100.33 = 2.93%, which is very consistent filling. The corrected CV is 2.93% × (1 + 1/24) = 3.06%. From summary values, a mean of 120 with SD 9 gives CV = 7.5%, or 7.875% corrected for n = 5.
Comparing variability on different scales
| Data set | Mean | SD | CV |
|---|---|---|---|
| Bag weights (g) | 100.33 | 2.94 | 2.9% |
| Adult heights (cm) | 170 | 10 | 5.9% |
| Monthly sales (units) | 400 | 120 | 30% |
Although the sales SD (120) is much larger than the height SD (10), the CV shows that sales are also far more variable relative to their size.
Inputs explained
| Field | What to enter |
|---|---|
| Calculate from | Choose one: A list of data values, A known mean and standard deviation |
| Data set | A list of numbers separated by commas, spaces or new lines |
| Data are a | Choose one: Sample (divide by n − 1), Population (divide by n) |
| Mean (x̄ or μ) | A number |
| Standard deviation (s or σ) | A number |
| Sample size (n) | A number |
What each result means
- Coefficient of variation = [Standard deviation]÷abs([Mean])×100
- CV as a ratio = [Standard deviation]÷abs([Mean])
- Small-sample corrected CV = (1 + 1÷(4×[Number of values]))×[Coefficient of variation]
- Mean ÷ standard deviation (signal-to-noise ratio) = abs([Mean])÷[Standard deviation]
Formula
CV = s ÷ x̄ × 100% (sample) or σ ÷ μ × 100% (population) · corrected CV* = (1 + 1/(4n)) × CV
Worked example
Inputs: Calculate from: A list of data values; Data set: 102, 98, 105, 97, 101, 99; Data are a: Sample (divide by n − 1).
Results: Mean = 100.3333; Standard deviation = 2.9439; Number of values = 6; Coefficient of variation = 2.934 %; CV as a ratio = 0.02934; Small-sample corrected CV = 3.056 %; Mean ÷ standard deviation (signal-to-noise ratio) = 34.082; Variability = Low (CV under 10%).
Key terms
- Coefficient of variation
- Standard deviation divided by the mean.
- Relative standard deviation (RSD)
- The CV expressed as a percentage.
- Ratio scale
- A scale with a true zero, on which ratios are meaningful.
- Signal-to-noise ratio
- Mean ÷ standard deviation, the reciprocal of the CV.
Frequently asked questions
How do I calculate the coefficient of variation?
Divide the standard deviation by the mean and multiply by 100. For 102, 98, 105, 97, 101, 99: s = 2.944, mean = 100.33, CV = 2.93%.
What is a good coefficient of variation?
It depends on the field. In laboratory assays a CV under 10% (often under 5%) is considered precise; in finance or biology 20–30% can be normal.
What is the difference between CV and standard deviation?
The SD is in the units of the data; the CV is a unitless percentage of the mean, so it can compare data on different scales.
When should I not use the CV?
When the mean is close to zero, when the data can be negative, or for interval scales without a true zero such as °C or °F.
What is the corrected coefficient of variation?
For a sample of size n, CV* = (1 + 1/(4n)) × CV reduces the small-sample bias. With mean 120, SD 9 and n = 5: 7.5% becomes 7.875%.
Is relative standard deviation the same as CV?
Yes. RSD is the same quantity, usually quoted as a percentage.
How do I compute the CV in Excel?
=STDEV.S(range)/AVERAGE(range), formatted as a percentage.
Sources
Guides on this topic
- Standard Deviation Explained: How to Calculate It and What It Means: Standard deviation is the typical distance from the mean. Learn to calculate it by hand, when to divide by n − 1, and how to read it with the 68–95–99.7 rule.