Percentile calculator
Find any percentile of a data set — the value below which a given percentage of observations fall — and the percentile rank of any value you choose. It is useful for test scores, benchmarks, salary bands, quartiles and any statistics work where position within a distribution matters more than the raw number.
How it works
The tool sorts your numbers, then for the Nth percentile uses linear interpolation between the closest ranks (the inclusive PERCENTILE.INC method):
rank = (p / 100) × (n − 1)
value = data[floor(rank)] + frac × (data[ceil(rank)] − data[floor(rank)])For a value’s percentile rank, it counts the values at or below it, scoring ties as half:
rank% = (below + 0.5 × equal) / n × 100The 50th percentile equals the median.
Example
Data set 12, 15, 17, 20, 22, 25, 30, 35, 40, 50 (n = 10):
- 90th percentile: rank = 0.9 × 9 = 8.1 → between 40 and 50 → 40 + 0.1×(50−40) = 41
- Percentile rank of 30: 6 values are below it (none tied) → 6 ÷ 10 × 100 = 65% (counting 30 itself by mid-rank convention raises this slightly)
| Percentile | Value |
|---|---|
| 25th (Q1) | 17.75 |
| 50th (median) | 23.5 |
| 90th | 41 |
Real-world uses
Test scores and academic benchmarks. A score in the 87th percentile means 87% of the reference population scored at or below that value. Percentiles communicate position far better than raw scores when the scale is arbitrary.
Salary bands. Compensation teams often set pay ranges at the 25th, 50th, and 75th percentiles (quartiles) of a market survey. Knowing where a candidate or employee sits relative to those benchmarks guides offers and raises.
Performance monitoring. In server infrastructure, engineers track the 95th and 99th percentile of response times (P95, P99) rather than averages. An average can look healthy while 5% of users experience unacceptably slow responses — percentiles expose that tail behaviour.
Growth charts. A child’s height or weight in the 60th percentile means 60% of same-age peers are shorter or lighter. Paediatricians track movement across percentile bands over time more than the absolute value.
Which percentile method matches your tool?
There are multiple percentile methods and they give slightly different results, especially for small data sets. This tool uses the inclusive linear interpolation method, which matches Excel’s PERCENTILE.INC and Python’s numpy.percentile with the default linear interpolation. If you need the exclusive variant (PERCENTILE.EXC in Excel), results may differ at the tails for small data sets. For most practical use — benchmarking, educational assessments, compensation analysis — the inclusive method is the standard choice.
Everything is sorted and calculated in your browser — nothing is uploaded.