Coefficient of Variation
Compute coefficient of variation CV = σ/μ × 100%.
Coefficient of variation (CV)
The coefficient of variation takes the standard deviation and expresses it as a fraction of the mean: CV = σ / μ · 100%. Because it carries no units (it's just a percentage), you can compare how spread out two populations are even when their scales or units differ. Adult human height comes in around CV ~5%, which is very uniform; salary, by contrast, runs CV > 50% and is all over the place. A sample with μ = 15 and σ = 3 gives you CV = 20%. As a quick guide, CV < 10% means low variability and anything past 30% is high. One caveat: skip the CV when μ sits near zero or the variable can go negative. In those cases the number gets unstable, or stops meaning anything at all.
Applications
Quality control leans on it (Six Sigma chases a low CV). So does finance, where the Sharpe ratio is essentially a normalized risk/return figure. Biologists use it to track variability among individuals or populations. In analytical method validation, the laboratory CV of repeated measurements usually stays under 5%. And engineers reach for it when comparing the precision of instruments that measure on very different scales.
FAQ
CV vs standard deviation, when to use which? Reach for σ when you want absolute dispersion in the same unit as the data. Switch to CV when you're comparing variability across datasets that use different units, or whose means are far apart.
Can CV be greater than 100%? It can. That just tells you σ is bigger than μ, which happens often in skewed distributions like income or species counts.
Sample CV or population CV? If you're working with a sample, use the sample standard deviation (n − 1); if you have the entire population, use the population σ (divided by N). For small n that decision actually moves the result.
Related Tools
Pearson Correlation Coefficient
Compute Pearson correlation coefficient r between X and Y lists.
Partial Correlation
Computes the partial correlation between two variables while controlling for the effect of a third. It's the tool against spurious correlations: ice cream sales correlate with drownings, but only because both rise in hot weather — controlling for temperature, the partial correlation drops to near zero. It removes the influence of the control variable from both and measures the direct association that remains. The tool returns the partial coefficient and its significance test. Enter the three paired series.
Prismatic Coefficient (Cp)
Compute a hull's prismatic coefficient (Cp), Cp = ∇/(Am·L), the ratio of the displaced volume to that of a prism with the midship section area (Am) along the whole length. It indicates how volume is distributed lengthwise: a low Cp concentrates volume amidships (good for low speeds), a high Cp pushes it to the ends (better at high speeds). It is decisive in resistance design. Enter the displaced volume, the midship section area and the length.
Cliff's Delta (Effect Size)
Computes Cliff's delta, a non-parametric effect-size measure between two groups. It answers a direct question: if I draw a value at random from each group, how much more likely is one to be larger than the other? It ranges from −1 to +1; zero means complete overlap. Because it works only with order comparisons, it requires no assumptions about the distribution and is robust to outliers. It's the ideal companion to the Mann-Whitney test. Enter the two groups of values.
Reflection Coefficient
Calculate the reflection coefficient, Γ = (Z_L − Z₀) ÷ (Z_L + Z₀), from the load impedance Z_L and the line's characteristic impedance Z₀ (ohms). The result (dimensionless, between −1 and 1) gives the fraction of the incident wave reflected by the impedance discontinuity: Γ = 0 means a perfect match (no reflection); Γ = ±1 is total reflection (open or short circuit). It is the basis of impedance matching in transmission lines and relates directly to VSWR and return loss. Enter the load impedance and the characteristic impedance.
CI for Difference of Two Means (t)
Computes the confidence interval for the difference between the means of two independent groups, using the t distribution with pooled variances. It's the natural companion to the two-sample t-test: instead of only saying whether the difference is significant, it shows the plausible range for the real size of that difference. If the interval doesn't contain zero, there's a significant difference at the chosen level. Enter the two samples of values and the confidence level.
The results provided by this tool are for general informational and educational purposes only and do not constitute professional, financial, medical, legal, tax or accounting advice. Always confirm important decisions with a qualified professional and official sources.