A relative measure of variation estimated as the ratio of the standard deviation to the sample mean. It affords a valid comparison of small and large things and is independent of the unit of measurement. When the variance is the error variance, it is used to indicate the degree of precision of a given experiment.