Answer twenty questions — drawn fresh each time from a larger set, five per axis — for a playful four-letter type, then explore relationship, work, and growth prompts. This is a private editorial quiz, not an official MBTI® or clinical assessment.
What runs on this page is a twenty-question questionnaire written for this site. It is not the official MBTI® instrument, not a clinical assessment, and not affiliated with anyone who provides one. It does not ask for your name or your birth date; the answers you click are the only input it has.
The questions come from a bank of thirty-two items — eight for each of the four axes, EI, SN, TF and JP. Five are drawn at random from each axis to make twenty, and those twenty are shuffled together once more before you see them. Which axis turns up where is left to that shuffle.
The first sheet is the one exception. Until the page has finished loading it shows a fixed twenty — the first five items of each axis, kept in axis order — which is then swapped for the random draw. If you have JavaScript turned off, the fixed sheet is the one you answer.
Everything is worked out where you are. Nothing is sent anywhere, and no sentence is written on the spot; the questions and every line of the result are read out of tables set down in advance.
How the four letters are decided
The scoring is a straight count. Each of a question's two options is assigned one letter — E or I, S or N, and so on — and the option you pick scores a point for its letter. All thirty-two items in the bank are worth the same single point; there is no heavy question and no light one.
Five questions per axis means every axis lands on 5-0, 4-1 or 3-2. The odd number is there so that nothing can tie, and five is a figure we chose. It does guarantee that each axis has a winner.
Put the other way round: a letter settled 3-2 turns over if you change one answer. The page shows four letters and nothing else, but behind one of them may be a clean sweep and behind another a single question. Two people with the same code did not necessarily arrive by the same margin.
One more consequence worth stating plainly. Answering with the left-hand option throughout gives ESTJ, and the right-hand option throughout gives INFP, whichever items happened to be drawn that time. The options are ordered so that this holds.
Where the element, the match score and the paragraphs come from
The element shown beside your result is a table lookup on the four letters. INTJ, INTP and INFJ map to water; ESTJ and ISTP map to metal; each of the sixteen codes has exactly one assigned to it.
It is not a saju element. The element in a chart is counted from a birth date and time, and this page never receives one. Same word, different origin — and in the code the two are kept deliberately apart so that one cannot be read as the other.
The match score depends only on how many of the four letters two codes share in the same position. Start at 52 and add 8 for each shared position: nothing shared gives 52, all four gives 84. There are only five possible scores — 52, 60, 68, 76 and 84 — and which letters matched is not considered at all. That formula is ours as well.
The short phrase printed next to the score is a band. Four thresholds — 84, 72, 60, and everything below — each carry one phrase, and whichever band the score falls into supplies the words on screen. Because only five scores exist, 68 and 60 land in the same band and read exactly alike.
The paragraphs are assembled as well. The relationships paragraph is keyed on the first and third letters, which is four variants; the work paragraph on the second and fourth, which is another four. So two people with different codes can be handed the same paragraph.
Retaking changes it, and how much weight to give it
Pressing retake draws the questions again. Choosing five of eight per axis means the second set is unlikely to match the first, and any axis that stood at 3-2 can turn over. That is the design rather than a fault.
So this is not a tool that returns the same answer on demand. It keeps no record of the last one either. Retake clears your answers, resets the type you had picked to compare against, and draws a fresh sheet — nothing on this page holds one attempt beside another.
Which means a letter coming out differently is not evidence that something in you moved. The items drawn were different; that is as far as it goes.
This is a count of twenty answers, not a measurement of you. The four letters are the mechanical result of the twenty choices made on this screen. It is not a diagnosis, not a verdict on the work that suits you, and not a real judgement of how you get on with anyone. Take whatever rings true and leave the rest here.