Home›Glossary›Percentile

Percentile

General

Percentile Rank

A value below which a given percentage of observations in a dataset fall. The 50th percentile is the median, splitting the data exactly in half.

Definition

A percentile is a value below which a given percentage of observations in a dataset fall. For example, if a score sits at the 80th percentile, that means 80% of the values in the comparison dataset are lower than that score, and 20% are higher. The 50th percentile is a special case known as the median, which splits the data exactly in half. Percentile rank is calculated using the Percentile Rank Calculator, which takes a raw value and a dataset (or reference distribution) and returns its relative standing.

Percentiles are widely used whenever a raw number needs context. A standardized test score of 720 means little on its own, but knowing it corresponds to the 95th percentile immediately communicates that it outperforms 95% of test-takers. This same logic underlies many health-category percentile calculators — BMI percentile, height percentile, weight percentile, head circumference percentile, and similar growth-tracking tools — which all compare an individual's measurement against a reference population using the identical percentile concept, just applied to different clinical datasets.

Percentiles are closely tied to quartiles, which are simply the 25th, 50th, and 75th percentiles. The spread between the 25th and 75th percentile — the Interquartile Range — is a commonly used measure of how tightly the middle of a dataset is clustered.

Formula

For a sorted dataset with n observations, the rank position for the P-th percentile is:

Rank = (P ÷ 100) × (n + 1)

If the rank is not a whole number, interpolate between the two nearest ranked values in the sorted dataset.

Worked Example

Consider 10 exam scores, sorted: 55, 60, 65, 70, 72, 75, 80, 85, 90, 95. To find the 70th percentile:

Rank = (70 ÷ 100) × (10 + 1) = 0.7 × 11 = 7.7

This falls between the 7th value (80) and the 8th value (85). Interpolating 0.7 of the way between them:

70th Percentile = 80 + 0.7 × (85 − 80) = 80 + 3.5 = 83.5

So a score of 83.5 would sit right at the 70th percentile of this dataset.

Key Things to Know

  • The 50th percentile is the median: both terms describe the exact midpoint of a sorted dataset.
  • Percentile depends entirely on the comparison group: the same raw score can correspond to very different percentiles depending on which dataset or population it's measured against.
  • Quartiles are just specific percentiles: the 25th, 50th, and 75th percentiles define the first, second (median), and third quartiles, and the gap between the 25th and 75th is the interquartile range.
  • Widely used in growth and fitness tracking: BMI, height, weight, and head circumference percentile calculators all apply this same concept against age- and sex-specific reference populations.
  • Not the same as percentage score: a percentile reflects relative rank, while a percentage reflects an absolute proportion — the two numbers can differ substantially for the same result.

Related Calculators

Percentile Rank Calculator →Find what percentile a value falls at within any dataset. Enter your numbers and a target value to instantly see the percentile rank and counts below or above it.

Related Terms

Interquartile RangeInterquartile Range (IQR)

Frequently Asked Questions

Being in the 90th percentile means a value or score is higher than 90% of the observations in the comparison dataset, with only 10% of observations above it. It does not mean the score is 90% correct or 90% of the maximum possible value — percentile rank is purely about relative position within a dataset.
A percentage describes a proportion out of 100 for a single value, such as scoring 85% correct on a test. A percentile describes how that value ranks relative to a distribution of other values, so scoring 85% correct could place a student anywhere from the 40th to the 99th percentile depending on how everyone else performed.
The 50th percentile is mathematically identical to the median — both represent the value that splits a dataset exactly in half, with 50% of observations below and 50% above. They are simply two names for the same measure of central tendency.
One common method sorts the dataset and uses the formula rank = (P ÷ 100) × (n + 1), where P is the desired percentile and n is the number of observations, then interpolates between the nearest ranked values. Statistical software and calculators may use slightly different interpolation methods, which can produce marginally different results for small datasets.
Percentiles let doctors compare an individual child's measurement — such as height, weight, or head circumference — against a large reference population of children the same age and sex, without needing raw comparison numbers. A child at the 75th percentile for height is taller than 75% of children of the same age and sex in the reference data.