Not sure which lesson to do first? Find your level and get a plan →

Statistics · Grade 9-12

Percentiles

Percentiles are values that divide a ranked distribution into 100 equal parts.


Quick Answer

Percentiles are values that divide a ranked distribution into 100 equal parts. The nth percentile has about n% of the data below it, so a percentile describes rank, not score. Use percentiles when you need to say where one value stands among many: a baby's weight among babies of the same age, a test score among everyone who sat the test, a delivery time among all deliveries. They work for any shape of data. Quartiles are special percentiles: P25=Q1, P50 is the median and P75=Q3.

Why This Matters

A raw score means little on its own: 38 out of 50 could top the class on a hard test or sit below average on an easy one. Percentiles turn any value into a rank against a named group. That lets a doctor notice a child sliding from the 50th to the 10th percentile for weight, and lets a delivery company promise that 95% of parcels arrive within two days.

Intuitive Explanation

Line up 100 students by height, shortest on the left. The student in place 91 has 90 people shorter than them, so their height is at about the 90th percentile. A percentile is just 'what share of the line stands behind me'. Put that same student in a line of professional basketball players and their percentile drops sharply, though their height has not changed. Percentiles are always relative to a group.

That is why a percentile is not a percentage score. On a very hard contest, 30 points out of 100 might beat 85% of the entrants: a score of 30% and a rank at the 85th percentile. One number answers 'how much of the test did I get right?', the other answers 'how many people did I beat?'.

To find a percentile in a data set, sort the values and count. In 20 sorted scores, the 75th percentile sits about three-quarters of the way along, around the 15th value. Textbooks and calculators interpolate between neighbouring values in slightly different ways, so a small data set can give answers that differ a little; with large data sets the methods agree closely.

Quartiles and the median are percentiles you already know. The 50th percentile is the median, the middle position, not the mean. For skewed data such as incomes, the two can be far apart.

Think of it like this. A percentile is your place in a queue sorted by score, read as the share of people standing behind you. Swap in a different crowd and your place moves even though your score did not. The picture is sharp for large groups; with ties or only a handful of people, 'your place' gets fuzzy.

Core idea. A percentile says what share of a group falls below a value, so it measures standing relative to that group, and it changes when the group changes even if the value does not.


When to Use

Reach for percentiles when a question asks where one value ranks among a group: 'better than what percent', growth charts, top 10% cut-offs, or the 95th percentile of wait times. They need no bell shape. If the question asks how many standard deviations a value is from the mean, use a z-score; if it wants a typical value, use the median or mean; if it asks what fraction of a test was answered correctly, that is a plain percentage.

Tip. Ask: Do you need the share of the group that falls below a value, stated as a rank out of 100?

How to Recognize It

Before answering, separate the score from the rank: decide whether the question wants a value's standing in its group, a cut point, or a distance in standard deviations.

  1. Does the question ask what percent of the group lies below a value, or which value has a given percent below it?

    Yes is a percentile. If it asks what percent of the questions were answered correctly, that is a percentage score, which says nothing about rank.

  2. Is the cut at a share other than 25%, 50% or 75%, such as the 90th or the 10th?

    Yes needs a general percentile. The three special cuts at 25, 50 and 75 are the Quartiles.

  3. Are the data themselves given, or a table of how many fall below each value, rather than only a mean and standard deviation?

    Yes means you can count positions to get the percentile directly. With only a mean and standard deviation of bell-shaped data, find a Z-Score (Standard Score) first and convert it through the normal curve.

  4. Must the answer work for lopsided data, such as incomes or waiting times?

    Yes favours percentiles, which rely only on order. A z-score converted through the bell curve gives the wrong rank when the shape is far from the Normal Distribution.

Percentiles vs Quartiles vs Z-Score (Standard Score) vs Median

All four say where a value stands, so they get swapped, and none of them is the percent of questions answered correctly. Ask whether you need a rank out of 100, a quarter mark, a distance in standard deviations, or the halfway point.

Percentiles

Meaning
The pth percentile is the value with about p% of the group below it.
Key test
A rank out of 100 that depends on who else is in the group.
Example
A runner at the 80th percentile in a race finished ahead of about 80% of the field.

Quartiles

Meaning
Quartiles are the three percentiles at 25%, 50% and 75% that split data into four groups of equal count.
Key test
Only three fixed cuts; the 25th percentile is Q1.
Example
Exam marks with Q1=58 and Q3=81: a mark of 60 is just inside the middle half.

Z-Score (Standard Score)

Meaning
A z-score counts how many standard deviations a value sits above or below the mean.
Key test
A distance in spread units; it becomes a rank only through the bell curve.
Formula
z=x−μσ
Example
With mean 76 and SD 4, a score of 82 has z=1.5, about the 93rd percentile if normal.

Median

Meaning
The median is the 50th percentile, the value with half of the data below it.
Key test
Exactly one cut, at 50%.
Formula
median position=n+12
Example
If the median household owns 2 cars, at least half of households own 2 or fewer.

Formula & Notation

How to read it: Percentiles are written as Pn or simply stated as 'the nth percentile.' P50 is the median, P25=Q1, and P75=Q3.

Worked Examples

Example 1 — Percentile of a score, and a score at a percentile

Easy

Twenty test scores, sorted: 55, 58, 61, 64, 66, 68, 70, 71, 73, 74, 76, 77, 79, 80, 82, 85, 87, 90, 93, 97. (a) At what percentile is a score of 82? (b) Find the 90th percentile.
Show the steps and the answer
  1. Count the scores below 82: it is the 15th value, so 14 scores are lower.

    A percentile counts how much of the group lies below the value.

  2. Convert to a percent: 1420=0.70, so 82 is at the 70th percentile.

    Dividing by the group size turns the count into a share of the group.

  3. For (b), find the position: 0.90×20=18.

    18 scores should lie below the 90th percentile, so it falls between the 18th and 19th values.

  4. Average those neighbours: 90+932=91.5.

    When the position is a whole number, a common convention averages it with the next value; some calculators interpolate a little differently.

Answer: (a) 82 is at the 70th percentile. (b) The 90th percentile is about 91.5.

Takeaway: Percentile of a value: count below and divide by the total. Value at a percentile: find the position, then read the sorted list.

Example 2 — Percent correct versus percentile

Standard

On a hard math contest Leo answers 18 of 40 questions correctly. Of the 300 students who took it, 246 scored lower than Leo. Leo says he 'got 45%, so he did badly'. Find his percentile and decide whether he is right.
Show the steps and the answer
  1. Compute his percent correct: 1840=45%.

    This compares Leo with the test, not with other students.

  2. Compute his percentile: 246300=0.82, the 82nd percentile.

    This compares Leo with everyone who sat the same hard test.

  3. Interpret: Leo beat about 82% of the entrants.

    A hard test pushes everyone's raw scores down, so a low percent correct can still be a high rank.

Answer: Leo is at the 82nd percentile; he did well relative to the group, even though he answered only 45% correctly.

Takeaway: Percent correct measures you against the test; a percentile measures you against the other test-takers.

Example 3 — Find the mistake: the 50th percentile is not the mean

Application

Yearly incomes of 9 households (thousands of dollars): 32, 35, 38, 41, 45, 48, 52, 60, 250. Sam writes: 'The mean is 66.8, and the 50th percentile is the mean, so half of these households earn more than $66,800.' Find the mistake and correct it.
Show the steps and the answer
  1. Check the mean: 6019≈66.8 thousand.

    Sam's arithmetic is fine; the mistake is calling the mean the 50th percentile.

  2. Find the 50th percentile: it is the median, the 5th of 9 sorted values, 45.

    The 50th percentile is defined by position in the ordered list, not by averaging.

  3. Count who earns more than 66.8: only the household at 250, one of nine.

    The single 250 pulls the mean far above the middle, so 'half above the mean' is false for skewed data.

Answer: The 50th percentile is the median, $45,000; four of the other eight households earn more than that, and only one earns more than the $66,800 mean.

Takeaway: The 50th percentile is always the median; it matches the mean only when the data are symmetric.

Where You'll See It

  • Standardized test score report

    Scoring in the 90th percentile means you did better than about 90% of test-takers; it describes your rank among everyone, not how many questions you got right.

  • Baby growth percentiles

    A doctor saying a baby is in the 60th percentile for weight means the baby weighs more than about 60% of babies the same age, placing them in the distribution of peers.

  • Class rank on a project

    If your project is at the 75th percentile, three-quarters of the class scored at or below you - it locates your position relative to classmates regardless of the raw points.

Common Mistakes

  1. The slip: Scoring in the 90th percentile means you got 90% of the questions correct.

    Why it's tempting: Both use the number 90 and a percent sign, so they feel identical, but one is a rank and the other is a raw score.

    Instead: The 90th percentile means you scored higher than about 90% of people; it says nothing about how many questions you answered correctly.

  2. The slip: A higher raw score always means a higher percentile.

    Why it's tempting: More points feels like a better rank, and usually it is, but percentile depends on how everyone else did, not just your score.

    Instead: Percentile is relative: the same raw score can be a high or low percentile depending on the rest of the distribution.

  3. The slip: The 50th percentile is the mean.

    Why it's tempting: Both sound like 'the middle', so they get conflated, but the 50th percentile is defined by position, not by averaging values.

    Instead: The 50th percentile is the median (the middle position), which equals the mean for symmetric data but can be far from it when the data is skewed.


Mini Practice

Try these on your own. Tap Reveal when you want to check.

  1. In a class of 40 students, Maya scored higher than 30 of them. What is her percentile?

    Hint: What percent of the class scored below her?

  2. 1,200 students took a test, and Juan is at the 85th percentile. About how many students scored below him?

    Hint: Find 85% of 1,200.

  3. Ten ordered scores are 51, 55, 60, 62, 68, 70, 74, 79, 83, 90. Using the percent of values below a score, what percentile is the score 79?

    Hint: Count how many scores are below 79.

  4. On a test, the 20th percentile is 62 points and the 80th percentile is 88 points. What percent of students scored between 62 and 88?

    Hint: Subtract the two percentiles.

  5. Statement A: "Kai answered 90% of the questions correctly." Statement B: "Kai scored at the 90th percentile." Which one compares Kai with other students?

    Hint: One counts Kai's own right answers; the other ranks him.

  6. In a class of 50, Sam has the 5th highest score. A student says: "Sam is at the 5th percentile." What is Sam's percentile?

    Hint: How many students scored below Sam?

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Frequently Asked Questions

Can anyone be at the 100th percentile?

It depends on the definition. If a percentile counts the share of the group strictly below you, the top scorer among 200 people has 199 of 200, or 99.5%, below, never 100%. Definitions that count "at or below" can reach 100. Many score reports avoid the problem by running only from 1 to 99. The same edge appears at the bottom, where the lowest score sits at or near the 0th percentile.

Moving from the 50th to the 60th percentile and from the 90th to the 99th are both gains of about 10. Do they take the same number of extra points?

No. In bell-shaped data many people are packed near the middle, so a small gain in score passes a lot of them, while the tails are thinly populated. Going from the 50th to the 60th percentile takes about 0.25 standard deviations; going from the 90th to the 99th takes about 1.04. The second climb needs roughly four times as many extra points.

Can I average two percentile ranks?

Not safely, because percentile ranks are unevenly spaced in score. Suppose test scores are normal with mean 500 and SD 100. A score of 500 is the 50th percentile and 700 is about the 98th, so the two ranks average to about 74. But the average score, 600, is about the 84th percentile. Average the scores first, then convert the result to a percentile.

Learning Path

← Before

QuartilesMedian
Percentiles

You are here

Build on the Median, which splits sorted data in half, and Quartiles, which split it into four. Percentiles refine the same counting idea into 100 slices, so any value can be given a rank. Next, Z-Scores measure position in standard deviations, and the Normal Distribution links the two: for bell-shaped data a z-score converts directly into a percentile.

See Also