Coefficient of Variation
What you'll learn: How to fairly compare the variability of datasets that have different units or scales using a standardized measure.
The Problem with Direct Comparisons
Imagine you want to compare the consistency of two products: one is measured in grams (weights ranging 5–15g with SD = 2g) and another in kilometers (distances ranging 100–300km with SD = 50km). Which varies more? You can't simply compare 2 vs 50 — the units are completely different!
What is the Coefficient of Variation?
The Coefficient of Variation (CV) solves this by expressing variability as a percentage of the mean. It's calculated as:
CV = (Standard Deviation / Mean) × 100%
By dividing the standard deviation by the mean, you create a unitless ratio that shows how large the spread is relative to the average value.
Why It Matters
The CV lets you answer: "Proportionally speaking, which dataset is more variable?"
Example:
- Product A (grams): Mean = 10g, SD = 2g → CV = (2/10) × 100% = 20%
- Product B (kilometers): Mean = 200km, SD = 50km → CV = (50/200) × 100% = 25%
Despite the larger absolute standard deviation, Product B is actually more variable relative to its scale.
When to Use CV
- Comparing datasets with different units (salary in USD vs height in cm)
- Comparing datasets with very different means (bacteria counts vs human populations)
- Assessing relative risk or consistency in finance, manufacturing, or quality control
Key Takeaway: The Coefficient of Variation standardizes variability as a percentage, enabling fair comparisons across datasets with different units or scales.