Variance measures the average squared deviation of observations from their mean. In sample statistics, the sum of squared deviations is divided by n − 1 to obtain an unbiased estimator of population variance under the usual sampling assumptions. Because deviations are squared, variance has squared measurement units and can be difficult to interpret directly. The coefficient of variation (CV), also known as relative standard deviation when expressed as a percentage, compares the standard deviation with the absolute value of the mean. It is useful when evaluating variability between datasets with different means or measurement scales, provided the mean is meaningfully different from zero and the data are on an appropriate ratio scale.
Variance and Coefficient of Variation
Variance and coefficient of variation are statistical measures used to describe the variability of analytical measurements. Variance expresses squared deviations from the mean, while coefficient of variation expresses relative variability as a percentage.
Explanation
Variables
| Symbol | Name | Description |
|---|---|---|
xᵢ | Individual observation | The value of the ith analytical measurement. |
x̄ | Sample mean | The arithmetic average of the sample measurements. |
μ | Population mean | The average value of the population. |
s² | Sample variance | The estimated variance of the population based on sample observations. |
σ² | Population variance | The average squared deviation of population values from the population mean. |
s | Sample standard deviation | The square root of sample variance. |
CV | Coefficient of variation | The ratio of standard deviation to the absolute mean. |
n | Sample size | The number of sample observations. |
Example or application
A worked example has not yet been added.
Definition
Variance is the mean squared deviation of observations from their mean, while the coefficient of variation is the ratio of standard deviation to the absolute mean, commonly expressed as a percentage.
Purpose
To quantify the dispersion of analytical measurements and compare relative variability between datasets or analytical procedures.
Importance
Standard deviation alone may not provide a fair comparison when two analytical datasets have substantially different mean values. Variance provides the mathematical basis for many statistical procedures, while the coefficient of variation expresses dispersion relative to the measurement magnitude.
Key Concepts
Squared deviations
Sample variance
Population variance
Degrees of freedom
Standard deviation
Coefficient of variation
Relative standard deviation
Absolute versus relative variability
Scale-dependent and scale-independent measures
Principle
Variance increases with the squared distance of observations from the mean. The coefficient of variation normalizes the standard deviation by the magnitude of the mean, allowing relative dispersion to be compared across suitable datasets.
Working Process
Collect replicate analytical measurements.
Calculate the arithmetic mean.
Subtract the mean from each observation.
Square each deviation and add the squared values.
Divide by n − 1 to calculate sample variance.
Take the square root of variance to obtain standard deviation.
Divide standard deviation by the absolute mean.
Multiply by 100 when reporting the coefficient of variation as a percentage.
Compare CV values only when the datasets and measurement scales make relative comparison appropriate.
Equations
Sample variance: s² = Σ(xᵢ − x̄)² / (n − 1)
Population variance: σ² = Σ(xᵢ − μ)² / N
Sample standard deviation: s = √s²
Coefficient of variation: CV = s / |x̄|
Coefficient of variation as a percentage: CV (%) = (s / |x̄|) × 100
Relative standard deviation: RSD (%) = CV (%)
Units
Variance: square of the original measurement unit
Standard deviation: same unit as the original measurement
Coefficient of variation: dimensionless
Coefficient of variation: commonly reported as a percentage
Examples
An analytical method with a mean concentration of 100 mg/L and standard deviation of 2 mg/L has a CV of 2%.
A second method with a mean of 20 mg/L and standard deviation of 1 mg/L has a CV of 5%. Although its standard deviation is smaller, its relative variability is greater.
Variance is used in analysis of variance to compare variation within groups with variation between groups.
Applications
Comparing precision across analytical methods
Evaluating instrument repeatability
Monitoring laboratory quality-control samples
Comparing variability at different concentration levels
Analyzing calibration and validation data
Statistical modeling and analysis of variance
Advantages
Variance is a fundamental measure used in many statistical procedures.
The coefficient of variation enables suitable comparisons across different measurement magnitudes.
CV is easy to interpret as a percentage.
Relative variability can be more informative than absolute standard deviation when comparing appropriate datasets.
Limitations
Variance is expressed in squared units and is less intuitive than standard deviation.
The coefficient of variation becomes unstable when the mean is close to zero.
CV is generally inappropriate for measurements on interval scales with arbitrary zero points.
CV comparisons can be misleading for strongly skewed or heterogeneous datasets.
Different analytical error structures may produce different relationships between concentration and CV.
Common Mistakes
Confusing variance with standard deviation.
Reporting variance in the original measurement unit instead of squared units.
Forgetting to multiply by 100 when expressing CV as a percentage.
Calculating CV when the mean is zero or very close to zero.
Assuming a lower absolute standard deviation always indicates better relative precision.
Treating CV as a direct measure of accuracy.
Prerequisites
Arithmetic mean
Median and standard deviation
Basic algebra
Understanding of measurement units
Learning Objectives
Define sample variance and population variance.
Calculate variance from repeated analytical measurements.
Explain the relationship between variance and standard deviation.
Calculate the coefficient of variation.
Compare relative variability between suitable analytical datasets.
Recognize when coefficient of variation is inappropriate.
Practice Questions
How does variance differ from standard deviation?
Why does sample variance use n − 1 in its denominator?
Calculate the CV for a dataset with a mean of 50 mg/L and standard deviation of 2 mg/L.
Why is the coefficient of variation useful when comparing analytical methods?
Under what condition does the coefficient of variation become unreliable?
Can a dataset have a smaller standard deviation but a larger CV than another dataset? Explain.
Key Takeaway
Variance describes absolute squared variability, while the coefficient of variation describes variability relative to the mean. Both are useful for evaluating analytical precision when applied appropriately.
References
| Title | Type |
|---|---|
Miller and Miller, Statistics and Chemometrics for Analytical Chemistry | textbook |
Harris, Quantitative Chemical Analysis | textbook |
Skoog, Holler and Crouch, Principles of Instrumental Analysis | textbook |
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.
Suggest a correction or source · More in Analytical Chemistry