Use it

Know Your Type

Two questionnaires about the same person. One reports five positions on five lines and refuses to give you a label; the other turns four of them into capital letters and hands you an identity. Both are scored from the same kind of answers, in this browser, and the interesting part is how differently certain they are willing to sound.

What it is

Two instruments, and only one of them will admit to a margin

A personality questionnaire is somebody describing themselves, on one day, in somebody else's words. Everything below follows from that, and both of these are worth about as much as that sentence allows.

The five factors came out of decades of asking which words people use about each other and which of those words move together. Five clusters keep reappearing — across languages, across decades, and in the same person years apart. They predict things: not much, and not about any individual, but genuinely and repeatably. That is a lower bar than most people assume a personality test clears, and it is the highest bar any of them clears.

The four letters come from a different tradition — Jung's psychological types, turned into a questionnaire in the middle of the last century and turned into an away-day by the century after that. The underlying preferences are real enough as things to talk about. The format is where it goes wrong, and it goes wrong in a way this page can demonstrate rather than assert.

The problem with a dichotomy, in one sentence

A four-letter code says you are one of sixteen kinds of person, but the answers underneath it are a position on a line, and almost everybody sits near the middle of at least one of those lines. So a total of 25 and a total of 24 produce different letters, in the same size type, with the same confidence — and the person carries the letter around for years. If the underlying trait were genuinely two-humped you would see two humps in the data; you do not.

That is also why retaking it is so unkind to it. The commonly cited finding is that a large fraction of people — often around half — get at least one different letter a few weeks later. You do not have to take anybody's word for it here: answer the thirty-two statements and the result page tells you which of your letters a single different answer would have flipped, and what code you would be holding if it had.

Both are offered because putting them side by side is the point. The same person, the same kind of self-report, the same five-point scale — and one of them hands back "somewhere in the middle third of four out of five scales" while the other hands back a word you could put on a lanyard.

How it is scored

All of it, because a scoring key you cannot read is one you cannot argue with

Every statement is answered on the same five-point scale, from strongly disagree to strongly agree, with a named midpoint. Eight statements go to each scale, and half of every eight are worded the other way round.

The reversal is the part worth understanding. If all eight statements about extraversion agreed with extraversion, then a person who says yes to things scores high on extraversion — and on everything else. That is a real effect with a real name, and it is why four of every eight here point the other way and have their answer turned round before it is added.

It has a consequence you can check on the page in about forty seconds. Agree with every single statement and you score exactly 50 on everything — all five factors, and a dead heat on all four dichotomies. Answering without reading cannot produce a flattering profile, because it cannot produce a profile at all. scripts/personality_test.js asserts that, because it is the one property of the item set that a later edit could quietly break.

The arithmetic

Eight statements at one to five is a total between 8 and 40. That maps onto 0 to 100, so the midpoint — every answer at "neither" — is exactly 50. On the five factors that number is reported as a position, with the middle third (34 to 66) called the middle third rather than rounded up into a label. On the four dichotomies the same number picks a letter, and the distance from 50 is reported alongside it as the margin.

A dead heat still gets a letter — the second one, so I, N, F or P. That is the convention the published instruments use, it is arbitrary, and the page says so on the result rather than presenting a coin toss as a finding.

The five factors

ScaleToward 100Toward 0
Openness to experience curious, drawn to the unfamiliar, comfortable in the abstract practical, conventional, happier with a settled answer
Conscientiousness organised, deliberate, finishes what it starts flexible, spontaneous, works in bursts
Extraversion outward, talkative, refuelled by company reserved, self-contained, refuelled by quiet
Agreeableness trusting, accommodating, keeps the room together sceptical, blunt, says the unwelcome thing
Emotional stability even, hard to rattle, lets criticism land and pass reactive, quick to worry, feels the setback for longer

Every one of those five is named for its flattering end, and the fifth is the one that shows it. "Emotional stability" is the same axis usually called neuroticism, pointing the other way; "agreeableness" makes bluntness sound like a defect and "conscientiousness" makes flexibility sound like one. That is a fact about English rather than about people, and it is worth saying before somebody reads a 38 as a mark out of a hundred.

The four dichotomies

QuestionOne wayThe other
Where the attention goes E — thinks out loud, refuelled by company I — thinks first, refuelled by quiet
What counts as information S — starts from what is measurably there N — starts from the pattern and what it could become
How a decision gets made T — decides on what is defensible F — decides on who it lands on
How the outside gets handled J — wants it decided and booked in P — wants the options kept alive

Where the answers go

Nowhere. The whole game runs in the page: there is no room, no API and no network call of any kind, which makes it the only game on this site that is solo in the strong sense rather than a room with one seat in it. Your answers are held in this browser's local storage so that a refresh thirty statements in does not cost you thirty statements, and "Forget my answers" on the result clears them out of it. There is nowhere else they could be.

A finished result

Made-up answers, rounded to tens — the shape rather than anybody's numbers

This is what the page hands back. The five-factor half reports five positions; the four-letter half reports a code and then immediately undermines it, which is the job.

The five factors

  • Openness to experience 70

    Toward curious, drawn to the unfamiliar, comfortable in the abstract.

  • Conscientiousness 60

    Between the two — neither notably flexible nor notably organised. This sheet does not put you at either end.

  • Extraversion 30

    Toward reserved, self-contained, refuelled by quiet.

  • Agreeableness 50

    Between the two — neither notably blunt nor notably accommodating. This sheet does not put you at either end.

  • Emotional stability 40

    Between the two — neither notably reactive nor notably even. This sheet does not put you at either end.

3 of 5 came out in the middle third. That is the ordinary result and not a failure to answer properly — most people are near the middle of most of these, which is exactly what a label would hide.

The four letters

INFP

  • I Inward thinks first, refuelled by quiet

    Where the attention goes: E 30, I 70. No single different answer would have changed this letter.

  • N Abstract starts from the pattern and what it could become

    What counts as information: S 40, N 60. Any one of 3 different answers would have made this S.

  • F Values decides on who it lands on

    How a decision gets made: T 50, F 50. A dead heat. The letter is a convention, not a finding.

  • P Open wants the options kept alive

    How the outside gets handled: J 40, P 60. One different answer would have made this J.

Had every close call above gone the other way, this page would be showing you ISTJ instead — with the same confidence, in the same size letters. That is the whole argument about four-letter types, and it is your own sheet making it.

What this is not

There is no paragraph here about what an INFP is like, and that is deliberate. Four answers to four either-ors is not a description of a person, and the version of this that reads like a horoscope is the version that gets quoted at somebody in a stand-up.

Neither questionnaire on this page is the published instrument it resembles, and neither is a selection, hiring or clinical tool.

Why the bars on this page are drawn a class at a time

The live panel sets each bar's width through the CSSOM, which the site's content security policy allows. A written page cannot: the policy is style-src 'self' with no unsafe-inline, so a style attribute on a static page is refused by the browser and the bar silently renders empty. So the worked example above rounds its made-up numbers to tens and picks a width class, which is also why it is a worked example rather than a screenshot of anybody's real result.