What is the coefficient of variation?
The coefficient of variation (CV) is a measure of relative variability. It compares the size of the standard deviation to the mean, which makes it useful when you want to compare variability across different datasets that may have different units or very different average values.
Formula: CV = (Standard Deviation / Mean) × 100%
Why use a CV calculator?
Standard deviation alone tells you how spread out values are, but it does not tell you that spread relative to the average level. CV solves that problem by putting variability in context. This is especially useful in:
- Finance: compare risk per unit of expected return.
- Quality control: evaluate process consistency.
- Laboratory measurement: track precision across tests.
- Operations: compare stability of different metrics.
How to use this coefficient of variation calculator
Option 1: Raw data
Select Calculate from raw data, paste your numbers, and choose sample or population standard deviation. The calculator computes mean, SD, and CV automatically.
Option 2: Summary statistics
If you already know your mean and standard deviation, switch to Calculate from mean and standard deviation and enter those values directly.
Interpreting your CV result
In many practical settings, a smaller CV means more consistency and a larger CV means more relative dispersion. A simple rule of thumb:
- Below 10%: very low relative variability
- 10% to 20%: low variability
- 20% to 35%: moderate variability
- 35% to 60%: high variability
- Above 60%: very high variability
These cutoffs are context-dependent. Different fields have different expectations, so always interpret CV with domain knowledge.
Sample vs population standard deviation
When your data is a subset from a larger group, use sample SD (n − 1). When your data is the entire group of interest, use population SD (n). This choice changes SD slightly, and therefore changes CV.
Important limitations
- CV is not defined when the mean is zero.
- CV can be misleading when mean values are near zero.
- It is most meaningful for ratio-scale data (where true zero exists).
- Comparisons are strongest when data are measured under similar conditions.
Quick example
Suppose a process has mean output 50 units and standard deviation 5 units. CV = (5 / 50) × 100% = 10%. Another process has mean 20 and SD 4, so CV = 20%. Even though SD is smaller in the second process, it is less consistent relative to its average.