Coefficient of variation calculator

FAQ

What is the coefficient of variation?

The coefficient of variation (CV) is the standard deviation divided by the mean, usually expressed as a percent. It lets you compare relative precision across measurements that have different units or concentration levels.

Is this the sample or the population standard deviation?

Sample. The standard deviation divides by n−1 (Bessel’s correction). That is the usual choice for a finite set of assay replicates rather than an entire population.

Does a CV% replace a precision study such as CLSI EP05?

No. Repeatability and within-laboratory or multi-site reproducibility still need a designed experiment. CLSI EP05 is a commonly cited evaluation-of-precision protocol for quantitative measurement procedures. This page is the first arithmetic check on a paste of replicates.

How does CV relate to LoD and clinical claims?

Precision often worsens near the limit of detection or quantitation. A CV that looks acceptable at a mid-range control level may not support claims near LoD. Pair precision work with LoD studies (for example CLSI EP17) when those claims matter to the intended use.

Should I pool replicates across days or instruments?

Only when the protocol says so. Mixing days, lots, operators or instruments without a design confounds repeatability with intermediate precision or reproducibility. Keep the paste homogeneous with the precision component you are estimating.

What if the mean is near zero?

CV is undefined or unstable when the mean approaches zero, because you divide by the mean. Use another precision metric or a different concentration level for those data.