Worked example
Example: Mean 30, SD 2.5. The inputs, in the units selected above:
| Input | Value |
|---|---|
| Mean | 30.000 - |
| Standard deviation | 2.5000 - |
| Lower value | 26.000 - |
| Upper value | 34.000 - |
Applying the formula below gives:
| Result | Value |
|---|---|
| z-score of x₁ | -1.6000 |
| Probability below x₁ | 5.4799 % |
| Probability above x₂ | 5.4799 % |
| Probability between | 89.040 % |
Open this example in the calculator and change any input.
Formula
\(P(X < x) = \frac{1}{2}\left[1 + \operatorname{erf}\left(\frac{x - \mu}{\sigma\sqrt2}\right)\right]\)
Variables
| Symbol | Meaning | SI unit |
|---|---|---|
| \(\mu\) | Mean | - |
| \(\sigma\) | Standard deviation | - |
| \(x_1\) | Lower value | - |
| \(x_2\) | Upper value | - |
| \(z_1\) | z-score of x₁ | - |
| \(P(X < x_1)\) | Probability below x₁ | - |
| \(P(X > x_2)\) | Probability above x₂ | - |
| \(P(x_1 < X < x_2)\) | Probability between | - |
Method and standards
Normal distribution.
Where it is used
- Quality control limits.
- Strength acceptance checks.
Common mistakes
- Using sample SD of very few results.
- Assuming normality without checking.
Assumptions and limits
- Normal distribution. Tail probabilities are computed from the complementary error function, so very small tails (parts per million) keep their precision.
References
- Normal distribution Φ(z) = ½[1 + erf(z/√2)]
Questions and answers
Mean 30, SD 2.5: below 26?
5.48 %; between 26 and 34 89.0 %.
Characteristic strength?
The value with 5 % below is the mean − 1.645 SD.
Is my data normal?
Check a histogram; strengths and errors often are.
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.