Why your result can change

This is the most common reason by a wide margin. Suppose your Thinking/Feeling split came out at 52% Thinking. That does not mean you are a Thinking type with a small Feeling streak — it means the two are close enough that a couple of answers landing differently flips the letter entirely.

A single question you hesitated over, or read slightly differently the second time, can move a 52/48 to a 48/52 without anything about you having changed at all. This is why we show the strength of each preference next to the letter. Two people with the same four letters can be in completely different situations: one is a clear INTJ at 90/10 across the board, the other is barely on the J side and one thoughtful answer away from INTP.

Tests are not administered in a vacuum. If you took the first one after a good week and the second during a difficult month, you were not answering about the same person. Someone under sustained pressure answers the Extraversion questions differently than someone on holiday — and the same is true of Feeling and Judging items, which are the ones most sensitive to how much energy you have for other people and for structure.

There is a specific version of this that catches people out on the Enneagram: it asks about your core motivation, not your behaviour, and people under stress often describe the behaviour they have adopted rather than the motivation underneath it.

This one is real, and it is the answer people least want to hear because it is the least convenient. Preferences do shift, particularly in the twenties and thirties, and particularly after a major life event — a demanding job, a long relationship ending, becoming a parent, moving country, a bereavement. What a test measures is partly habit, and habits are learned.

It is also entirely normal for someone to spend years behaving as one type because a role demanded it, and to test as a different type once that role ends. The test did not get more accurate; you stopped answering on behalf of the role.

This is the least comfortable one and everyone does it. Test questions are transparent: it is usually obvious which answer corresponds to being organised, or empathetic, or creative. If you have just started a job that rewards decisiveness, you may answer the Thinking questions as the person you are trying to become.

The second quiz also arrives with knowledge the first one did not have: you already know your previous result. That prior knowledge is surprisingly sticky. Some people unconsciously answer to confirm the type they were given; others answer to escape it.

If the two results came from different websites, this may be the whole explanation. There is no single canonical MBTI test. Different implementations weight the dichotomies differently, use different question counts, and some quietly fold the Assertive/Turbulent dimension into the main four letters. Two tests can both honestly call themselves MBTI and disagree on a borderline case.

The same applies to the Enneagram, where competing schools disagree about wings, subtypes and which line a type moves along under stress.

Usually neither, in the sense you are hoping for. A four-letter code is a compressed summary of a set of answers you gave on one particular day. It is a useful summary — that is why the vocabulary is worth having — but it is not a fact about you that a retest can confirm or falsify.

The more useful question is not “which type am I?” but “what did I just learn?” If a test showed you that you are closer to the middle on Thinking/Feeling than you assumed, that is real information regardless of which letter you ended up with. If two results disagree, the disagreement itself — that you sit near a boundary — is the finding.