Percentile Calculator

StatisticsLast updated: August 17, 2026

A percentile tells you where a given proportion of a data set falls: the 90th percentile, for instance, is the value below which 90% of the data lies. Read the other way round, the percentile position of a score tells you what share of the group that score beats — one of the most widely used measures in exam rankings and growth charts.

This calculator works in both directions: it returns the value corresponding to the percentile p you choose (matching Excel's PERCENTILE.INC), or it locates the percentile position of a value you enter. It suits students comparing an exam result with their class, researchers reporting percentiles, and anyone evaluating measurements.

Both decimal points and decimal commas are accepted (e.g. 3.14 or 3,14). At least 2 values are required.
%
Enter a value between 0 and 100. E.g. 25 → Q1, 50 → the median, 90 → P90.

Formula

Rank = (p / 100) × (n − 1)   [data sorted ascending, index starts at 0]
P(p) = x[⌊Rank⌋] + fraction × (x[⌊Rank⌋ + 1] − x[⌊Rank⌋])
Percentile position = (number of values below the score / n) × 100

This calculator uses the R-7 method with linear interpolation (the same one as Excel's PERCENTILE.INC). Because different sources define percentiles differently, results on small data sets can vary slightly from one method to another; that is not an error.

How to Calculate

  1. Paste your data values into the box — separate them with commas, spaces, or line breaks.
  2. Choose the direction: are you looking for the value at percentile p, or for the percentile position of a value?
  3. For the first option enter p (0–100); for the second, enter the value whose position you want.
  4. Read the percentile value or position in the results section, together with the quartiles.
  5. The percentile curve shows how your data is distributed from P0 to P100.

Worked Examples

The score at the 90th percentile

For the scores of a class of 10, the 90th percentile falls between 85.00 and 90.00 in the sorted data and is found by linear interpolation: P90 = 85.50. So 90% of the scores are at or below 85.50, while the median is 72.00.

P90 — the value at percentile 90: 85.50 · Minimum (P0): 55.00 · First quartile (Q1 = P25): 63.50

Where an exam score sits in the class

A student scoring 77 has 6 of the 10 scores below them: the percentile position is 60%, meaning the score beats 60% of the class. Counting the 1 tied score and the 3 higher ones, the student falls in the top 40% band.

Percentile position (proportion below): 60% · Top band: Top 40% · Number of values below: 6

Finding the first quartile with P25

For the data 10, 20, 30, 40, 50 the 25th percentile lands exactly on an observation: P25 = 20.00, which is also the first quartile (Q1). The median is 30.00 and the third quartile is 40.00.

P25 — the value at percentile 25: 20.00 · Minimum (P0): 10.00 · First quartile (Q1 = P25): 20.00

Frequently Asked Questions

What does a percentile actually mean?
The pth percentile is the value below which roughly p% of the data falls. For example, a child at the 90th percentile for height is taller than 90% of their peers. P25, P50, and P75 are also known as the first quartile, the median, and the third quartile.
Is this the same as an exam ranking band?
It is close, but counted in the opposite direction. Exam bands are counted from the top: being in the top 1% means being among the 1% with the highest scores. The "Top band" row in this calculator follows exactly that logic, giving the share of values equal to or above the score. The "Percentile position" row counts from the bottom, showing what share of the group you beat.
Which percentile method does this calculator use?
It uses the R-7 method with linear interpolation, which returns exactly the same result as Excel's PERCENTILE.INC and Google Sheets' PERCENTILE. It is also NumPy's default. Some software, such as SPSS, uses a different method (R-6), so small differences can appear.
Can I find the position of a value that is not in the data set?
Yes. The percentile position is the number of values below the score divided by the total number of values; the score itself does not have to appear in the data. If your data contains 70 and 74, for example, the proportion below 72 is still computed. For values entirely outside the data range the calculator warns you separately.
Why is P50 the same as the median?
By definition the 50th percentile is the median: half of the data falls below it and half above. In the same way P25 = Q1 and P75 = Q3. Percentiles are therefore a generalization of quartiles, and they form the basis of box plots.
Is it meaningful to compute percentiles on small data sets?
Yes, as long as you interpret them carefully. In a set of 5 to 10 observations, extreme percentiles such as P90 rest largely on interpolation and shift with the method used. To report extreme percentiles (P5, P95, P99) reliably, at least 20 to 30 observations are recommended, and preferably more than 100.