📊

Percentile Calculator (Percentile Rank or Value at a Percentile)

Find the percentile rank of a value in a data set, or the value at any percentile, with the Excel, nearest-rank and (n + 1) methods side by side, a full percentile table, quartiles and the worked steps.

What do you want

Data values

Value or percentile

Percentile method

Percentile rank definition

Percentile table step

Decimal separator

Loading calculator…

How it works

Percentile rank = (values below + ½ × values equal) ÷ n × 100 · Value at p (Excel INC): rank r = p ÷ 100 × (n − 1) + 1, then interpolate · (n + 1) method: r = p ÷ 100 × (n + 1) · nearest rank: r = ⌈p ÷ 100 × n⌉

A percentile rank says what share of a data set a value beats: the count of smaller values, plus half the ties, divided by the total. Other tools count differently — “at or below” (used by many test reports), “strictly below”, or Excel’s PERCENTRANK.INC, which divides by n − 1 — so the calculator shows every definition side by side and lets you choose which one is the headline. Going the other way, the value at a percentile is found by sorting the data, working out a rank position and, when the position falls between two values, interpolating. The default is the linear (n − 1) method used by Excel’s PERCENTILE.INC, Google Sheets and CalculatorSoup; the (n + 1) method matches Excel’s PERCENTILE.EXC and many textbooks, and nearest rank always returns one of your data values. The percentile table lists all three methods for every 1st, 5th, 10th or 25th percentile, and the 50th percentile is the median while the 25th and 75th are the quartiles.

Worked example

In 3, 5, 8, 10, 12, 15, 18, 20 the value 15 has five values below it and one equal, so its percentile rank is (5 + 0.5) ÷ 8 × 100 = 68.75% (or 75% counting “at or below”, and 71.43% with Excel’s PERCENTRANK.INC = 5 ÷ 7). Switching modes, the 50th percentile has rank 0.5 × 7 + 1 = 4.5, halfway between 10 and 12, which gives 11. The 25th percentile has rank 0.25 × 7 + 1 = 2.75, so it is 5 + 0.75 × (8 − 5) = 7.25; the (n + 1) method gives rank 2.25 and 5.75 instead.

Sources: NIST/SEMATECH e-Handbook — percentiles · Hyndman & Fan (1996), Sample quantiles in statistical packages

Frequently asked questions

Why do different tools give slightly different percentiles?

There are several accepted definitions — nearest rank, linear interpolation on (n − 1) or on (n + 1), and different ways of counting ties for a percentile rank. The results agree for large data sets but can differ for small ones. Pick the method that matches your textbook or spreadsheet under More options; the percentile table shows all three methods at once.

Is the 90th percentile the same as the top 10%?

Yes. Being at the 90th percentile means you scored higher than about 90% of the group, which is the same as being in the top 10%. Percentiles describe position within a group, not a raw score.

What is the difference between a percentile and a percentage?

A percentage is a fraction of a total; a percentile is a rank relative to others. Scoring 80% on a test is your share of points, while the 80th percentile means you beat 80% of test-takers — very different ideas.

How do I calculate a percentile in Excel or Google Sheets?

Use PERCENTILE.INC(range, 0.9) (or PERCENTILE) for the 90th percentile with the default method here, and PERCENTILE.EXC for the (n + 1) method. PERCENTRANK.INC(range, x) gives the percentile rank as a fraction; multiply by 100 for a percent.

What is the normal-approximation percentile?

It turns the value into a z-score with the sample mean and standard deviation and reads the percentile off the bell curve. It is only meaningful when the data is roughly normal; otherwise rely on the counted percentile rank.