Overview
A percentile rank says what share of a group scored below a given value. This works it out for a normal distribution.
Enter the value, the mean and the standard deviation. The target percentile field isn't used here.
Worked examples
| Inputs | Result |
|---|---|
| Observed Value74Distribution Mean62Distribution Standard Deviation9Target Percentile (%)90 | Percentile rank: 90.88%z-score: 1.33 |
Show 2 more examples
| Observed Value50Distribution Mean62Distribution Standard Deviation9Target Percentile (%)90 | Percentile rank: 9.12%z-score: -1.33 |
| Observed Value130Distribution Mean100Distribution Standard Deviation15Target Percentile (%)90 | Percentile rank: 97.72%z-score: 2 |
Percentile rank by observed value
With the calculator's starting values, the percentile rank is 90.88%. The table shows how it changes with the observed value.
| Observed Value | Percentile rank |
|---|---|
| 40 | 0.73% |
| 50 | 9.12% |
| 60 | 41.21% |
| 70 | 81.3% |
| 74starting value | 90.88% |
| 80 | 97.72% |
| 90 | 99.91% |
| 100 | 100% |
| 110 | 100% |
Other inputs: Distribution Mean: 62 · Distribution Standard Deviation: 9 · Target Percentile: 90%. Change them in the calculator above for your own numbers.
How it works
The value is turned into a z-score, (x − μ) ÷ σ, and the percentile rank is the normal curve's area below it. It assumes scores are roughly bell-shaped.
The formula
Practical tips
- 74 with μ = 62 and σ = 9 is about the 91st percentile.
- A z of 0 is the 50th percentile, and 1 is about the 84th.
- For small groups, count the share below directly instead.
Frequently asked questions
Is the 90th percentile 90% correct?
No. It means scoring higher than 90% of the group.
What if my data are skewed?
The normal estimate may be off. Use the real data instead.
Can a percentile rank be 100?
Not with this method, since the normal curve never quite reaches 100%.