Coefficient of variation: what it means and how to read it

In the report: cv — std / mean; empty when the mean is 0.

How to read it

The spread relative to the level: 0.1 means the values vary by about 10% of the mean. It compares variability across units and scales, and makes sense only for data with a true zero (weights, prices), not for rating scales.

In plain words

A farmer weighs her hens and her eggs. Hens weigh 2000 grams on average and usually differ by about 200 grams, so the standard deviation is 200. Eggs weigh 60 grams and usually differ by 12. Which of them differ more? 200 grams looks bigger than 12, but that is unfair, since a hen is much heavier. The cv came out 0.1 for hens and 0.2 for eggs. Now she can say that eggs differ by about a fifth of their weight and hens only by a tenth. It works only where zero means nothing at all, like weight or price, not for a 1 to 5 rating.

How it is computed

cv = std / mean

Where you see it

STATS_DESCRIPTIVE — Column-wise descriptive statistics with flexible metric selection and optional column filtering.

Relative dispersion — std / median; empty when the median is 0.
Skewness — The asymmetry of the distribution, with the sample correction as in Excel and SPSS; empty for fewer than 3 values or a column of equal values.
Kurtosis — Excess kurtosis: how heavy the tails are compared with the normal distribution (0), with the sample correction as in Excel and SPSS; empty for fewer than 4 values or a column of equal values.
First quartile — The first quartile: 25% of the values are below it; computed as in SPSS, so it can differ from the Sheets QUARTILE and PERCENTILE functions.
Third quartile — The third quartile: 75% of the values are below it; computed as in SPSS, so it can differ from the Sheets QUARTILE and PERCENTILE functions.

Get it in your sheet

One formula in a cell gives the whole report, Coefficient of variation included. ALS Stats is a free add-on for Google Sheets™.

Install ALS Stats from Google Workspace Marketplace (opens in a new tab)