Coefficient of Variation Calculator

This coefficient of variation calculator turns a list of numbers into mean, standard deviation and relative spread in percent. Example: the values 12, 15, 11, 18, 14, 16, 13 and 17 give a CV of about 16.89 %.

Source: Wikipedia – Variationskoeffizient. Updated: .

Separate with spaces or semicolons; commas also separate values, so use a decimal point (2.5), not a comma.
My ToolboxYour inputs are saved in this browser only.

Result

Result
Coefficient of variation CV = 16.89 % (s = 2.44949, mean = 14.5)
All values
QuantityValue
Count n8
Mean x̄14.5
Standard deviation s (sample)2.44949
CV in percent16.89 %
CV as a decimal0.16893
Calculation steps
  • Mean of 8 values: x̄ = 14.5.
  • Standard deviation: s = √(Σ(x − x̄)² / (n − 1)) = √(42 / 7) = 2.44949.
  • Coefficient of variation CV = s / |x̄| = 2.44949 / 14.5 = 16.89 %.
The coefficient of variation has no unit and lets you compare the spread of series at different scales. It only makes sense for ratio scales with a true zero.

How it is calculated

The coefficient of variation divides the standard deviation by the mean, which makes spread comparable across different scales. A standard deviation of 5 is huge next to a mean of 10 but tiny next to a mean of 1,000; the coefficient of variation shows exactly that.

Formula

CV = s / |x̄|, usually quoted as a percentage. For a sample use the standard deviation with denominator n − 1, for a complete population the one with denominator n. The calculator lists the mean, the squared deviations and both intermediate steps.

Example

For 12, 15, 11, 18, 14, 16, 13, 17 the mean is 14.5 and the squared deviations add up to 42. As a sample, s = √(42 / 7) ≈ 2.449, so CV ≈ 16.89 %. Treating the same numbers as a population gives σ = √(42 / 8) ≈ 2.291 and CV ≈ 15.80 %.

Frequently asked questions

When is the coefficient of variation better than the standard deviation?

When you compare series at different levels or in different units, such as the weights of mice and elephants. The standard deviation depends on the magnitude, the coefficient of variation does not.

What is a good coefficient of variation?

It depends on the field. Measurement technology often expects a few percent, while incomes or share prices vary much more. What matters is the comparison with similar series.

Should I choose sample or population?

If the values are a sample, choose n − 1. If you really have every element, such as all employees of one firm, n is correct. With large data sets the difference is small.

Sources and legal basis

As of: · Editorial policy

Related tools