Calculate RSD: Formula, Example, and Interpretation

To calculate relative standard deviation (RSD), divide the sample standard deviation by the mean and multiply by 100. Enter your measurements in the list below to get the RSD as a percentage, along with the mean and sample standard deviation.

Formula

RSD = (sample standard deviation ÷ arithmetic mean) × 100%. In symbols, RSD = (s / x̄) × 100%, where s is the sample standard deviation in the same units as the measurements, and x̄ is their arithmetic mean in those units. The result is a percentage, so it has no measurement unit.

For n measurements, the mean is x̄ = Σxᵢ / n. The sample standard deviation is s = √[Σ(xᵢ − x̄)² / (n − 1)]. Here, xᵢ is each observed value, n is the number of observations, and Σ means to add across all observations. The calculator uses sample standard deviation, which divides the sum of squared deviations by n − 1, rather than by n.

RSD is also called the coefficient of variation (CV) when expressed as a percentage. Some contexts write the coefficient of variation as a decimal instead: s / x̄. Check the requested convention before comparing results.

Worked Example

Using the default measurements 10, 12, 11, 13, and 14:

1. Calculate the mean: (10 + 12 + 11 + 13 + 14) / 5 = 60 / 5 = 12.
2. Find each deviation from 12 and square it: 4, 0, 1, 1, and 4. Their sum is 10.
3. Calculate the sample standard deviation: s = √(10 / (5 − 1)) = √2.5 ≈ 1.5811.
4. Divide by the mean and multiply by 100: (1.5811 / 12) × 100 ≈ 13.18%.

The calculator therefore reports an RSD of 13.18%, a mean of 12.00 units, and a sample standard deviation of 1.58 units. The units could be grams, milliliters, seconds, or another consistent measurement unit; they cancel in the RSD calculation.

When to Use RSD

RSD describes variation relative to the scale of the average. It is useful when you want to compare the spread of measurements whose means differ substantially, provided the measurements use comparable methods and have meaningful positive means. For example, a standard deviation of 2 units is small relative to a mean of 100, but large relative to a mean of 4. RSD expresses that contrast as a percentage.

• Repeatability: Compare repeated measurements of the same quantity to see how tightly they cluster around their average.
• Relative variability: Compare variation across datasets measured on different scales, while checking that their means are suitable for this comparison.
• Reporting: Summarize dispersion alongside the mean when a percentage is more interpretable than a standard deviation alone.

RSD is not a measure of accuracy. A set of readings can have a low RSD because the readings agree closely, yet still be consistently far from a correct reference value. Accuracy and precision describe different properties.

How to Interpret the Result

A smaller RSD means the standard deviation is smaller in proportion to the mean; a larger RSD means greater relative spread. There is no universal threshold that makes an RSD “good” or “bad.” Appropriate values depend on the kind of measurement, the purpose of the comparison, and any applicable method or field conventions.

RSD is unchanged if every measurement is converted by multiplying by the same positive factor. For example, converting all lengths from centimeters to meters scales both the mean and standard deviation equally, so their ratio stays the same. Adding a constant to every value is different: it changes the mean but not the standard deviation, and therefore changes the RSD.

Common Mistakes and Limitations

• Using the wrong standard deviation: This calculator uses the sample standard deviation with denominator n − 1. If your data are the complete population of interest and your convention calls for population standard deviation, use the population formula with denominator n instead. The resulting RSD may differ.
• Mixing units: Convert measurements to a consistent unit before calculating. Combining, for example, inches and centimeters as if their numerical values were directly comparable gives a misleading mean and spread.
• Ignoring a zero or near-zero mean: If the mean is zero, RSD is undefined because the formula divides by zero. If the mean is close to zero, the percentage can become extremely large and unstable, even when the absolute spread is modest. For data that can be negative or cross zero, RSD may not be meaningful.
• Rounding too early: Keep full precision during the calculations and round only the reported result. Early rounding can slightly change the final percentage.
• Assuming it describes every distribution well: RSD reduces a dataset to its mean and standard deviation. It does not show outliers, skew, trends, or whether the observations form distinct groups. Inspect the data as well when those features matter.

At least two observations are needed for a sample standard deviation. With one value, there is no sample spread to estimate. The formula also treats each entered observation as an individual measurement; repeated values should not be removed simply because they are duplicates.

Frequently Asked Questions

How do I calculate RSD by hand?
Find the arithmetic mean, calculate the sample standard deviation, divide the standard deviation by the mean, and multiply by 100. Keep both calculations in the same measurement units.

Is RSD the same as coefficient of variation?
Usually, yes: both describe standard deviation relative to the mean. RSD commonly refers to the value expressed as a percentage, while coefficient of variation may be shown either as a decimal or a percentage. Follow the convention used in your instructions or report.

Should I use sample or population standard deviation?
Use the sample version when your observations are a sample used to represent a broader process or population. Use the population version only when the data comprise the entire population of interest and that is the specified convention. State which one you used when the distinction matters.

Can RSD be negative?
With a positive mean, standard deviation is nonnegative, so RSD is nonnegative. If the mean is negative, the signed ratio can be negative; many applications instead regard RSD as unsuitable for such data or use an absolute mean by explicit convention. Do not silently change the formula—check the reporting standard for your context.

What does an RSD of 10% mean?
It means the sample standard deviation is one-tenth of the mean in magnitude, when the mean is positive. It does not by itself establish whether the measurements are acceptable or accurate; that depends on the task and its criteria.

Related

• Aluminum Weight Calculator: Formula and Examples
• 7th Chord Inversion Calculator: Notes and Bass
• 6061 Aluminum Weight Calculator and Formula

list values | Measurements | 10, 12, 11, 13, 14
out rsd | Relative standard deviation | % | 100 * sd(values) / mean(values) | 2
out mean | Mean | units | mean(values) | 2
out s | Sample standard deviation | units | sd(values) | 2