Boneyard Tools

Coefficient of Variation (CV) Calculator

Paste a data set to find its coefficient of variation, the standard deviation relative to the mean. The CV is unitless, so it lets you compare the spread of data sets measured on different scales. The tool shows the CV as a ratio and a percentage, along with the mean and standard deviation it used, and lets you switch between the sample and population formulas.

How to calculate the coefficient of variation

  1. Type or paste your numbers into the Data set box, separated by commas, spaces or new lines.
  2. Choose Sample (n - 1) or Population (n) for the standard deviation.
  3. Read the CV (ratio) and CV (percent) cards.
  4. Check the Mean and Std dev cards to see the underlying figures.
  5. Click Copy result to grab all of the values at once.

Examples

Sample CV of a small data set

2, 4, 4, 4, 5, 5, 7, 9 (Sample)
CV 0.427618 (42.7618%), mean 5, std dev 2.13809

Population CV of the same numbers

2, 4, 4, 4, 5, 5, 7, 9 (Population)
CV 0.4 (40%), mean 5, std dev 2

Evenly spaced values

10, 12, 14, 16, 18 (Sample)
CV 0.225877 (22.5877%), mean 14, std dev 3.162278

Frequently asked questions

What is the coefficient of variation?

The CV is the standard deviation divided by the mean. It expresses spread relative to the average, usually shown as a percentage, and it has no units. A CV of 0.427618 means the standard deviation is about 42.76% of the mean.

Why use the CV instead of the standard deviation?

Because it is unitless, the CV lets you compare variability across data sets with different units or very different averages, which a raw standard deviation cannot. A 10 kg spread means one thing on 50 kg loads and another on 5000 kg loads, but the CV captures both on the same scale.

How do I enter my numbers?

Type or paste them into the Data set box separated by commas, spaces or new lines. The parser splits on any run of those characters, so a messy paste from a spreadsheet or a plain comma list both work.

Should I use the sample or population CV?

Use the sample CV, the default, when your data is a sample drawn from a larger group. Use the population CV when your data covers the entire group. Sample divides the squared deviations by n minus 1, population divides by n, so the sample figure is slightly larger.

Why does the sample CV need at least two values?

The sample standard deviation divides by n minus 1, and with a single value that denominator is zero, which is undefined. The population option divides by n and will work on one value, though the spread of a single point is always zero.

Why can't the mean be zero?

The CV divides the standard deviation by the mean, so when the mean is zero that division is undefined and the tool reports an error. This also makes the CV unreliable for data centered near zero, where a tiny mean inflates the ratio.

Can the coefficient of variation be negative?

For all-positive data it is non-negative. If the mean is negative the ratio comes out negative, so in practice the absolute value is used when interpreting spread. The CV is most meaningful for data measured on a ratio scale with a true zero.

How precise are the results?

The engine rounds each figure to six decimal places, and the percentage is shown to four decimals. That is more than enough precision for reporting, and it avoids the long trailing digits that floating-point arithmetic would otherwise produce.

Is my data private?

Yes. The calculation runs entirely in your browser, so your numbers never leave your device and nothing is uploaded to a server.

Learn more

Related tools