What is the coefficient of variation?
The coefficient of variation is the standard deviation divided by the mean, usually shown as a percent. Because it doesn’t depend on the units, it lets you compare spread between data sets on different scales or with very different means.
Use it only for measurements with a true zero, like weights, lengths or times. Temperatures in °F or °C give a different CV for the same readings, and a mean near 0 makes it jump about.
Source NIST Dataplot Reference Manual · COEFFICIENT OF VARIATION ↗OpenStax · Introductory Business Statistics 2e, 2.7 Measures of the Spread of the Data (coefficient of variation) ↗
How to calculate the coefficient of variation
Here it is with your numbers:
- The mean is 25 and the sample standard deviation is 6.5955.
- CV = 6.5955 ÷ 25 × 100%
- CV = 26.38%
It means the numbers typically sit about 26.38% of the mean away from it.
What is the standard error of the mean?
The standard error of the mean (SEM) is the standard deviation divided by the square root of the sample size. It describes how far, on average, the mean of a random sample would be from the true mean if you took many samples of the same size. Bigger samples give a smaller standard error.
How to calculate standard error
Here it is with your numbers:
- There are 5 numbers, and √5 = 2.2361.
- Standard error = 6.5955 ÷ 2.2361
- Standard error = 2.9496
It means that over many random samples of 5, the sample mean would typically be about 2.9496 away from the true mean.
Coefficient of variation examples
| Numbers | CV | Standard error |
|---|---|---|
| 27, 31, 15, 30, 22 | 26.38% | 2.9496 |
| 1345, 1301, … (10 numbers) | 2.07% | 8.6849 |
FAQs
How do I calculate a coefficient of variation?
Divide the standard deviation by the mean. For 27, 31, 15, 30 and 22 days, the sample standard deviation is 6.5955 and the mean is 25, so the coefficient of variation is 0.2638.
Source NIST Dataplot Reference Manual · COEFFICIENT OF VARIATION ↗
How do you calculate CV%?
Multiply the coefficient of variation by 100. A standard deviation of 6.5955 with a mean of 25 is a CV of 26.38%. The mean must not be 0.
How do you calculate standard error?
Divide the standard deviation by the square root of the sample size. For 5 numbers with a standard deviation of 6.5955, the standard error of the mean is 6.5955 ÷ √5, which is 2.9496.
Should I use SD or SEM?
They answer different questions. The standard deviation describes how spread out the numbers are; the standard error of the mean describes how far the sample mean is likely to be from the true mean across repeated samples. Report the one that matches your question.
Sources
- NIST Dataplot Reference Manual: COEFFICIENT OF VARIATION
- OpenStax: Introductory Business Statistics 2e, 2.7 Measures of the Spread of the Data (coefficient of variation)
- OpenStax: Introductory Statistics 2e, 7.1 The Central Limit Theorem for Sample Means (standard error)