IQ Percentile Calculator: What Percentile Is Your Score?

A raw IQ number means very little on its own. The percentile tells you how many people you actually scored above.

What a Percentile Actually Tells You

An IQ score is not a count of anything. It is a ranking, and the percentile is that ranking stated plainly.

A percentile of 84 means you scored higher than 84 out of every 100 people in the comparison group. That is a concrete statement. A score of 115 is not, until you know what scale produced it and how that scale spreads people out.

This is also why the percentile is the only figure worth quoting when two scores come from different tests. The numbers will disagree. The percentiles will not.

Percentile and Percentage Are Not the Same Thing

These two get confused constantly, and the confusion changes how people read their own result.

A percentage is how much of the test you got right. Answer 27 of 40 items correctly and you scored 67 percent.

A percentile is how many people you outperformed. It says nothing directly about how many questions you answered.

Those can differ enormously. On a deliberately difficult test where the average person answers half the items correctly, getting 67 percent right might place you around the 85th percentile. On an easy test where most people score 90 percent, the same 67 percent would place you well below average.

This is why a properly normed test never reports your percentage as though it meant something. It converts your raw performance against the reference group first, and reports where that puts you. A test that shows you a percentage and calls it your result has skipped the step that makes the number meaningful.

IQ to Percentile Table

Reference points on the standard SD 15 scale.

Notice how fast the top thins out. Going from 115 to 130 is 15 points, and it takes you from 1 in 6 to 1 in 44. The next 15 points take you to roughly 1 in 741.

Classification Bands Explained

Test publishers attach labels to score ranges. Those labels have changed over the years and they vary between publishers, so treat them as rough descriptions rather than official categories.

The Wechsler scales long used words like Superior and Very Superior. Recent editions moved toward neutral wording such as High and Extremely High, because the older labels carried judgments that a percentile rank does not.

The band boundaries are conventions too. Nothing changes in a person between 129 and 130. The line sits there because two standard deviations is a familiar statistical cutoff, not because anything happens in cognition at that point. For the full breakdown of every band, see the IQ score chart.

Three Results Read Properly

The same percentile figure supports very different conclusions depending on where it sits. Three worked examples make that concrete.

A score of 104

That is roughly the 61st percentile, comfortably inside the Average band, which most manuals define as 90 to 109. The honest reading is that this person performed like a typical adult. Given a margin of error of about five points either way, a 104 and a 99 are not meaningfully different results, and treating one as better than the other is reading noise as signal.

A score of 128

Roughly the 97th percentile, or about 1 person in 34. This one sits close to a threshold, which is where the error band starts to matter practically rather than theoretically. The realistic range runs from about 123 to 133, which straddles the 130 line used for gifted programmes and Mensa. A supervised retest could land on either side, so this is a case where the band is more informative than the point.

A score of 148

Roughly the 99.9th percentile on an SD 15 scale, about 1 in 1,000. Out here the precision of the instrument becomes the limiting factor rather than the arithmetic. Standardisation samples contain very few people at this level, and most tests run out of sufficiently difficult items above roughly 145. The percentile is directionally right and the exact figure should not be quoted to decimal places.

How Classification Bands Are Decided

The label attached to a score is a publisher convention rather than a fact about cognition, and the conventions have shifted over time.

The Wechsler scales long used terms like Superior and Very Superior. Recent editions moved toward neutral wording such as High and Extremely High, because the older labels carried a judgment that a percentile rank does not make. Different publishers draw the boundaries in slightly different places, which is why you will see 120 described as Superior in one source and High Average in another.

The boundaries themselves are round numbers chosen for convenience. Two standard deviations is a familiar statistical cutoff, so 130 became the gifted threshold. Nothing changes in a person between 129 and 130, and no research identifies a discontinuity there. Our guide to the gifted IQ score threshold explains how that line came to be drawn where it is.

The practical consequence: read the percentile first and the label second. The percentile is a measurement. The label is a description somebody chose for a range.

Why the Top of the Scale Gets Strange

Percentiles behave predictably in the middle of the distribution and badly at the edges.

Standardisation samples are built to represent the general population, which means very few participants sit far out in either tail. A representative sample of 2,000 adults contains roughly eight people at or above 140. The norms up there are estimated from a handful of cases.

Most mainstream tests also run out of sufficiently difficult items somewhere around 145 to 160, which creates a ceiling. Beyond that point a reported percentile is an extrapolation rather than a measurement.

The practical effect: a gap of 5 points near 100 is meaningful, and a gap of 5 points near 150 is mostly noise. Anyone quoting a percentile to several decimal places above about 145 is claiming precision the instrument does not have.

Why Percentile Beats the Raw Number Every Time

If you take one thing from this page, make it this: the percentile is the durable figure and the IQ number is not.

An IQ score is a position on one particular scale, and that scale was chosen by the publisher. Change the instrument and the number changes while the person does not. A result of 130 from a Wechsler test, 132 from an older Stanford-Binet and 148 from a Cattell all describe the same standing, and someone comparing those three numbers without knowing the scales would reach three different conclusions about the same performance.

The percentile does not have that problem, because it is a statement about people rather than about the test. It says what share of the comparison group you outperformed, and that share is the same however it gets expressed. This is why clinical reports lead with percentile ranks and confidence intervals rather than the headline composite, and why any conversation about scores becomes clearer the moment somebody asks which percentile the number corresponds to.

It is also the reason this calculator reports percentile, rarity and classification band together rather than just converting one number into another. Those three together tell you where you stand. The raw figure on its own tells you almost nothing without the scale that produced it. If you want to move a score between scales directly, the IQ scale converter does that in one step.

Percentiles Outside IQ Testing

The concept is not unique to intelligence testing, and recognising it elsewhere makes it easier to read.

Paediatric growth charts use exactly the same logic. A child on the 75th percentile for height is taller than 75 percent of children the same age, and the number carries no judgment on its own. Standardised school assessments report percentiles for the same reason: the raw mark depends on how hard that particular paper was, while the percentile describes standing against everyone who sat it.

Two habits transfer directly from those contexts. First, the comparison group matters enormously. A percentile against a national sample and a percentile against one school mean different things, just as an IQ percentile depends on which population the test was normed against. Second, movement near the middle of the distribution is much less meaningful than movement at the edges, because that is where most people are clustered.

Four Ways People Misread a Percentile

Each of these comes up constantly, each one changes the conclusion somebody draws about themselves, and each is easy to avoid once you have seen it named. None of them require any statistics background to spot.

Reading it as a mark out of 100. The 84th percentile does not mean you scored 84 percent. It means 84 percent of people scored lower than you.

Assuming equal gaps mean equal differences. Moving from the 50th to the 60th percentile is a much smaller change in raw ability than moving from the 89th to the 99th, because the distribution thins out as you move away from the centre.

Comparing percentiles from different norm groups. A percentile is always relative to a specific comparison population. Two figures from tests normed on different samples, or in different decades, are not directly comparable.

Treating a single figure as fixed. Sleep, illness, stress and simple familiarity with the test format all move results by several points. A percentile is a snapshot taken under particular conditions, not a permanent attribute.

Where This Tool Fits

This page answers one question: what percentile is a given score. Two related questions have their own tools.

If you want to know what a score equals on a different test, use the IQ scale converter. If you want to know whether a score clears the entry requirement for Mensa or another high-IQ society, use the Mensa eligibility checker.

Common Questions