IPIP-NEO

Reading your result

How to read your results

You have thirty numbers and five bars. This page explains what they are measured against, which of them deserve your attention, and which of the obvious readings are wrong.

  • A percentile is not a grade
  • The middle is the widest band
  • Facets say more than domains

What your result actually is

The test turns your answers into thirty facet scores and five domain scores, and then compares each one with people of your sex and age band. Everything you see is that comparison. Nothing is an amount, a level, or a diagnosis.

This matters because the raw sums mean nothing on their own. Answering all thirty items of a facet with fours gives you a sum of forty out of fifty, which sounds high until you learn that most people also answer that way, and the score is unremarkable.

So read every number on this site as a position among people, and read the two layers separately: domains give the outline, facets give the picture.

The percentile, in plain words

A percentile answers one question: out of a hundred comparable people, how many scored lower than you? That is all it does, and it is worth pausing on how different that is from a school grade.

A grade compares you with a standard: eighty out of a hundred means you got four fifths of the answers right. A percentile compares you with a queue: eightieth percentile means four fifths of the queue is behind you. Nobody is failing, and there is no maximum to reach.

This is also why percentiles move faster than raw scores near the middle. Half the people are packed into a narrow band of raw sums, so a couple of points there can shift you ten percentiles, while the same couple of points near the edge barely move you at all.

One hundred people from your comparison group
youscored lower than you

At the 72nd percentile, 72 of these hundred scored lower than you and 27 scored higher. That is the whole claim. It says nothing about how much of the trait you have in some absolute sense, because there is no absolute scale to have it on.

Why low and high are thirty percent each

We call the bottom thirty percent low, the top thirty percent high, and the forty percent between them average. The cuts are a convention, not a law of nature, but they are an honest convention.

Cutting at the median would sort everyone into high or low and tell you nothing, since half of any group is above the middle by definition. Leaving a wide band in the centre says the true thing instead: on most traits most people are ordinary, and only the edges carry information worth reading.

low: bottom 30%
average: middle 40%
high: top 30%
Anything between 30 and 70 is the ordinary middle, and most of your facets will land there. A percentile of 55 is not a mild version of high, it is average, and reading it as a leaning in either direction is the most common mistake on this page.

Domains and facets, and why the average domain lies

A domain is the sum of its six facets, so a middling domain has two very different explanations. Either the six facets are all near the middle, and the domain really is unremarkable; or they are scattered from one end to the other and cancel out on the way to the total.

The second case is common and the interesting one. Someone can be highly orderly and highly dutiful yet weak on self discipline, and their Conscientiousness will land at fifty as if nothing were happening.

The practical rule: read the domain first as an orientation, then go straight down to its facets. If they disagree with each other, the disagreement is the finding, and the domain number is just their average.

This is the whole reason the site measures thirty facets rather than five domains. Five numbers are a summary; the picture starts a layer below.

This is one real shape a result can take: a domain sitting quietly at the middle while its facets run from the eighth percentile to the ninety-first. The domain number is honest arithmetic and completely uninformative here. The facets are where this person actually lives.

There are no good and bad profiles

Every trait in this model is a trade rather than a virtue. High agreeableness makes cooperation easy and negotiation hard. Low neuroticism keeps you steady and can leave you late to notice that something is wrong. High openness brings ideas and scatters attention.

So a profile is not a score card. It is a description of where your effort goes cheaply and where it costs you, and the same profile is an advantage in one setting and a burden in another.

The one honest use of a low score is planning around it. The dishonest use is treating it as a verdict about your worth, or handing it to someone else as one.

The mistakes people make reading this

Almost every misreading we see is one of these five:

  • Treating a percentile as a percentage of a trait. Sixty is not sixty percent of anything, it is a position among people.
  • Reading 55 or 62 as a mild high. That is the average band, and the difference from 48 is noise.
  • Comparing your percentile with someone in a different comparison group. Percentiles are relative to sex and age band, so two people can share a raw score and differ by a dozen percentiles.
  • Reading a single facet from the short version as if it were precise. Four questions carry real uncertainty, and a borderline facet can move on a rerun.
  • Taking one sitting as a permanent verdict. A hard month tilts answers, and traits drift slowly over years anyway.

What to do with it next

Start with the two or three facets that sit furthest from the middle, in either direction. They are where the profile is most confident and where the description will feel most like you.

If a domain interests you, read its page: each one explains what the trait is, what the poles look like day to day and how the six facets under it differ. If you want your own facets read together rather than one by one, that is what the extended part of the result does, and it also compares them against each other rather than against a table.

Questions people ask about their numbers

Is a high percentile better?
No. Every pole has its costs, and the model has no good end. A high score tells you where you sit relative to others, not whether that position is an advantage.
My domain says average but the facets are all over the place. Which do I trust?
The facets. An average domain built from scattered facets is arithmetic rather than description, and the scatter is the more useful fact about you.
How much can my result change if I take it again?
Domains move little over months. Individual facets move more, especially on the short version, and a few percentiles either way is ordinary measurement noise rather than change in you.
Why did my percentile change when I gave a different age?
Because the comparison group changed. The same raw score sits differently among twenty year olds and among fifty year olds, and the percentile is always relative to the group named on your result.
Can I compare my result with a friend's?
Only carefully. If you are in different sex or age groups, your percentiles come from different distributions, so compare the shape of the profiles rather than the individual numbers.

Ready to read your own profile

Take either version and the full base result opens straight away: five domains, thirty facets and percentiles for each, with no account needed.

Free, no sign-up, results right away