RSD Calculator Online
Calculate relative standard deviation from raw observations or from a known mean and standard deviation. The result includes variance, the decimal ratio, every step, and a reconstruction check.
Written and checked by TheToolNet Editorial Team · Last reviewed: September 12, 2026 · How we check statistics tools
RSD measures spread relative to the magnitude of the mean. It is undefined when the mean is zero.
Calculate RSD online
Use raw data when you want the calculator to find the mean and SD, or enter summary statistics directly.
How to calculate relative standard deviation
- Find the arithmetic mean.
- Calculate the appropriate sample or population standard deviation.
- Divide SD by the absolute value of the mean.
- Multiply by 100 to express the ratio as a percentage.
RSD is unitless, which makes it useful for comparing relative spread across measurements with different magnitudes.
RSD calculation example
For sample values 10, 12, 11, 9, and 8, the mean is 10 and sample SD is approximately 1.5811. RSD = 1.5811 ÷ 10 × 100 = 15.811%.
Sample or population standard deviation?
| Data represents | Divisor | Choose |
|---|---|---|
| A sample from a larger population | n − 1 | Sample SD |
| The entire population of interest | n | Population SD |
Choose from the study design, not from whichever version produces the preferred percentage.
RSD versus coefficient of variation
For a positive mean, RSD and percentage coefficient of variation have the same numerical value. RSD is conventionally non-negative because it divides by |mean|. A signed CV can be negative, although ordinary relative-spread interpretation becomes unreliable for negative or near-zero means.
For the terminology and ratio-scale limitations, see the coefficient of variation calculator. To use Excel, open the RSD in Excel guide.
Frequently asked questions
What is a good RSD?
There is no universal cutoff. Acceptable RSD depends on the measurement process, field, and decision standard.
Why use the absolute mean?
It keeps RSD as a non-negative magnitude. Even so, negative means often indicate that ordinary relative-spread interpretation is not suitable.
What if the mean is close to zero?
RSD becomes unstable and can be extremely large. Report SD and the mean alongside it and reconsider whether a relative measure is meaningful.