Coefficient of Variation Calculator
Written by Thierno Sadou Diallo, formula verified per our methodology • Last checked on 9/6/2026
The coefficient of variation is calculated with CV = (standard deviation ÷ mean) × 100. For a mean of 50 and a standard deviation of 15, the CV is 30%: the relative dispersion of the values around the mean is moderate.
Explanation
The coefficient of variation (CV) expresses the dispersion of a data series as a proportion of its mean, rather than in a raw unit the way the standard deviation alone does. Its main purpose is comparing the variability of two series that don't share the same mean or the same unit: a standard deviation of 15 means something very different depending on whether the mean is 50 or 5,000. By expressing it as a percentage of the mean, the CV allows that comparison directly. One important point, often overlooked: the CV only makes sense for ratio data, meaning data whose zero represents a true absence of the measured quantity (a weight, a duration, a price). It loses all meaning on an interval scale where zero is arbitrary, such as a temperature in degrees Celsius (0°C doesn't mean "no temperature", and the same temperature gap would give a different CV in Fahrenheit). This calculator applies the formula directly from an already-known mean and standard deviation; it doesn't calculate them itself from a raw data sample.
Example: mean 50, standard deviation 15
Inputs
Mean: 50. Standard deviation: 15.
Calculation
Coefficient of variation = (15 ÷ 50) × 100 = 0.3 × 100 = 30%.
Result
The relative dispersion of the values around the mean is 30%.
Frequently asked questions
Why use the coefficient of variation instead of the standard deviation alone?
The standard deviation is expressed in the same unit as the data, which makes it hard to compare two series with very different means. The coefficient of variation, expressed as a percentage of the mean, allows a direct comparison of the relative dispersion of series that share neither the same scale nor the same unit.
Does the CV make sense for any type of data?
No: it assumes a ratio scale, where zero corresponds to a true absence of the measured quantity (weight, duration, income, for example). It becomes misleading on an interval scale whose zero is arbitrary, such as a temperature in Celsius or Fahrenheit, or as soon as the mean is zero or close to zero.
What counts as a "high" or "low" CV?
There is no universal threshold: the interpretation depends entirely on the field. In industrial quality control, a CV of a few percent can already be considered high, whereas in social sciences or biology, CVs of 20 to 30% are common and considered acceptable. Always compare the CV against benchmarks specific to the field being studied rather than a generic threshold.
What happens if the mean is negative?
The formula then returns a negative coefficient of variation, which makes comparison with other series unintuitive. Some conventions use the absolute value of the mean in the denominator; this calculator applies the raw formula, without that convention, so a negative CV here should be interpreted with caution.