The coefficient of variation (abbreviated “CV”), also known as relative standard deviation (RSD), is a normalised measure of dispersion of a probability or frequency in probability theory and statistics. It is commonly represented as a percentage and is widely employed in analytical chemistry, engineering and physics, and quality assurance in industrial production. Economic, organizational, and financial models are frequently employed in economists and social studies.

Because the CV is a dimensionless number, it is unaffected by the unit of measurement or the data mean.

The following are the main arguments in favour of utilising the coefficient of variation rather than the standard deviation:

• It is unaffected by the scale of the measured variable(s), which allows for easier cross-unit comparisons.
• It enables comparisons between groups with dramatically varying means (relative magnitude of effects)

However, there are several drawbacks to utilising relative standard deviation rather than absolute standard deviation. It can’t be used to create confidence intervals for the mean, for example. The CV approaches infinite when the mean is near to zero, making it extremely sensitive to small changes in the mean. Because the CV is invariant to sample sizes, it is also a bad statistic when the number of observations in the different groups varies. In such instances, the standard error of the mean is usually a better option. Although there is considerable criticism of its usage in this way, the coefficient of variation is frequently employed as a measure of economic inequality.

As with any statistic, there are times when utilising a coefficient of variation calculator is suitable and others when it isn’t.