Worked example
Example: Ten pressure readings. The inputs, in the units selected above:
| Input | Value |
|---|---|
| Reading 1 | 100.20 - |
| Reading 2 | 100.50 - |
| Reading 3 | 99.800 - |
| Reading 4 | 100.10 - |
| Reading 5 | 100.40 - |
| Reading 6 | 100.00 - |
| Reading 7 | 99.900 - |
| Reading 8 | 100.30 - |
| Reading 9 | 100.20 - |
| Reading 10 | 100.10 - |
| Number of readings to use (2 to 10) | 10.000 - |
Applying the formula below gives:
| Result | Value |
|---|---|
| Mean | 100.15 |
| Standard deviation (n − 1) | 0.21731 |
| Standard uncertainty of the mean s/√n | 0.068718 |
| Coefficient of variation | 0.21698 % |
| Range (largest − smallest) | 0.70000 |
Open this example in the calculator and change any input.
Formula
\(\bar x = \frac{1}{n}\sum x_i, \quad s = \sqrt{\frac{\sum (x_i - \bar x)^2}{n - 1}}, \quad u = \frac{s}{\sqrt n}\)
Variables
| Symbol | Meaning | SI unit |
|---|---|---|
| \(x_{1}\) | Reading 1 | - |
| \(x_{2}\) | Reading 2 | - |
| \(x_{3}\) | Reading 3 | - |
| \(x_{4}\) | Reading 4 | - |
| \(x_{5}\) | Reading 5 | - |
| \(x_{6}\) | Reading 6 | - |
| \(x_{7}\) | Reading 7 | - |
| \(x_{8}\) | Reading 8 | - |
| \(x_{9}\) | Reading 9 | - |
| \(x_{10}\) | Reading 10 | - |
| \(n\) | Number of readings to use (2 to 10) | - |
| \(\bar x\) | Mean | - |
| \(s\) | Standard deviation (n − 1) | - |
| \(u\) | Standard uncertainty of the mean s/√n | - |
| \(s/\bar x\) | Coefficient of variation | - |
| \(R\) | Range (largest − smallest) | - |
Method and standards
Type A evaluation of the GUM.
Where it is used
- Instrument repeatability tests.
- Laboratory results.
- Quality control.
Common mistakes
- Using n instead of n − 1.
- Mixing different conditions.
Assumptions and limits
- Independent readings.
References
- JCGM 100:2008 (GUM) §4.2
- Type A evaluation
Questions and answers
What do the ten default readings give?
Mean 100.15, standard deviation 0.21, standard uncertainty of the mean 0.069.
Why n − 1?
The mean is taken from the same data, which uses up one degree of freedom.
What is the coefficient of variation?
The standard deviation as a share of the mean.
Method and article reviewed by FireCAD engineering on 12 November 2026. Every result is checked against an independent calculation on each release: how we verify our calculators.