Coefficient of Variation Calculator
Coefficient of Variation Calculator
Enter a list of numbers to calculate its coefficient of variation (CV) — the standard deviation expressed as a percentage of the mean. Because it's unitless, CV lets you compare the relative variability of datasets measured in completely different units or scales.
A standard deviation of 10 means very different things depending on context: for a dataset with a mean of 20 that's enormous relative variability, while for a dataset with a mean of 10,000 it's negligible. CV normalizes for this by expressing spread as a percentage of the mean, which is why it's the standard tool for comparing consistency across financial returns, manufacturing tolerances, or lab measurements taken on entirely different scales — a lower CV always means more consistency relative to the average, regardless of the underlying units.
- Formula: Coefficient of Variation = (Standard Deviation ÷ Mean) × 100 — using the sample standard deviation.
- Why it's useful: a standard deviation of 10 means very different things for a dataset with a mean of 20 versus one with a mean of 10,000 — CV normalizes for this, making comparisons meaningful.
- Rule-of-thumb bands: under 15% is generally considered low variability, 15-35% moderate, and above 35% high — though what counts as "high" varies by field.
When is CV not a useful measure?
CV becomes unreliable or meaningless when the mean is close to zero (since you'd be dividing by a near-zero number) or when the data includes both positive and negative values that could make the mean itself close to zero despite real variability.
Can I compare CV values across completely different datasets, like heights in cm versus weights in kg?
Yes — that's exactly what CV is designed for, since it strips away the original units, letting you meaningfully compare relative spread even across unrelated measurement scales.
Coefficient of Variation Calculator
Enter a list of numbers to calculate its coefficient of variation (CV) — the standard deviation expressed as a percentage of the mean. Because it's unitless, CV lets you compare the relative variability of datasets measured in completely different units or scales.
A standard deviation of 10 means very different things depending on context: for a dataset with a mean of 20 that's enormous relative variability, while for a dataset with a mean of 10,000 it's negligible. CV normalizes for this by expressing spread as a percentage of the mean, which is why it's the standard tool for comparing consistency across financial returns, manufacturing tolerances, or lab measurements taken on entirely different scales — a lower CV always means more consistency relative to the average, regardless of the underlying units.
- Formula: Coefficient of Variation = (Standard Deviation ÷ Mean) × 100 — using the sample standard deviation.
- Why it's useful: a standard deviation of 10 means very different things for a dataset with a mean of 20 versus one with a mean of 10,000 — CV normalizes for this, making comparisons meaningful.
- Rule-of-thumb bands: under 15% is generally considered low variability, 15-35% moderate, and above 35% high — though what counts as "high" varies by field.
When is CV not a useful measure?
CV becomes unreliable or meaningless when the mean is close to zero (since you'd be dividing by a near-zero number) or when the data includes both positive and negative values that could make the mean itself close to zero despite real variability.
Can I compare CV values across completely different datasets, like heights in cm versus weights in kg?
Yes — that's exactly what CV is designed for, since it strips away the original units, letting you meaningfully compare relative spread even across unrelated measurement scales.