Is the 16 personalities test scientifically valid?

Partially. Its underlying questionnaire measures continuous traits similar to the well-validated Big Five model, and those trait percentages are reasonably informative. The weak link is the type label: cutting continuous scores into 16 boxes discards information, which is why psychometricians treat four-letter types as a self-reflection vocabulary rather than a scientific classification.

Why did my 16 personalities type change when I retook it?

Most people score near the middle on at least one dimension, so a small mood- or context-driven shift in answers flips the letter. The Myers-Briggs Company — the publisher of the MBTI, the older instrument this format comes from — reports that only about 50% of people get the same four-letter type on a retest, while about 90% get the same result on three or four of the four scales. Your underlying scores usually move only slightly; it's the binary cut that flips.

Why does my personality profile feel so accurate?

Partly because it may be, and partly because of the Barnum effect — the tendency to accept vague, broadly applicable descriptions as uniquely personal. In Bertram Forer's 1949 demonstration, students were each handed what they were told was an individual personality sketch based on their answers. Every sketch was identical, and they rated it about 4.3 out of 5 for accuracy. A profile feeling true is weak evidence that it is precise.

Are personality tests accurate in general?

It depends what you mean by accurate. Well-built questionnaires measure self-reported traits consistently, and that part is reasonably solid across most reputable tests. What they do not do well is sort people into discrete categories, or predict specific outcomes like job performance or relationship success. Judge a test by whether it reports where you sit on a continuum, tells you how confident it is, and avoids claiming to predict your future.

Is 16 personalities the same as the Myers-Briggs test?

No. 16Personalities is operated by NERIS Analytics and uses its own questionnaire and model (including a fifth Assertive/Turbulent axis), presented with four-letter codes similar to the Myers-Briggs tradition. The official MBTI is a separate, paid instrument from The Myers-Briggs Company.

Which personality test is the most accurate?

For scientific purposes, trait-based Big Five assessments have the strongest evidence base: their scores are continuous, stable and predict real outcomes modestly but reliably. Type-based tests trade some accuracy for memorability — a four-letter label is easier to use than five percentile scores.

Is the 16 personalities test free?

The basic test and type result are free; the company sells premium profiles and courses on top. Free type tests are useful for self-reflection — just treat any type label, theirs or ours, as a starting point rather than a verdict.

Is 16 Personalities Accurate? An Honest Review

The 16 personalities test is accurate as a snapshot of your self-reported traits, but the four-letter type it assigns you is a simplification that psychometric research treats with caution. Whether that counts as "accurate" depends on what you want from it: a vocabulary for self-reflection (it does that well) or a scientific measurement with predictive power (no type test does that well).

IQ Revealed is not affiliated with, endorsed by or connected to 16Personalities, NERIS Analytics Limited, or The Myers-Briggs Company. 16Personalities® and MBTI® are the trademarks of their respective owners, referred to here only to discuss them. Our own personality assessment is independently developed. This page is published by IQ Revealed, which offers a competing personality test.

What the test actually measures

16Personalities — operated by NERIS Analytics — asks Likert-scale questions ("agree" to "disagree") and scores you on five continuous dimensions, closely paralleling the academically dominant Big Five trait model. It then converts those percentages into a four-letter code plus an Assertive/Turbulent suffix. The measurement layer is conventional and reasonably solid; the conversion layer is where accuracy leaks out.

The dichotomy problem

Human traits are distributed continuously, and most people cluster near the middle. A person scoring 52% extraverted and one scoring 95% get the same E — while the 52% person is far more similar to a 48% introvert than to their fellow E. Cutting a bell curve in half throws away most of the information the questionnaire collected. This is the core academic criticism of every four-letter system, and it is why your letters can flip on a retake while your underlying scores barely move.

The clearest evidence for this comes from the publishers themselves. The Myers-Briggs Company, which sells the MBTI — the older instrument the four-letter format descends from — publishes its own reliability data. It reports that the individual preference scales are highly stable: internal consistency around .90, and one-month test-retest correlations of .94 to .97, falling to .57 to .81 over four years. But on the four-letter type as a whole, its own figure is that about 50% of people get the same type on a retest, while about 90% get the same result on three or four of the four scales.

Read those two numbers together, because they are the whole argument. The measurement is stable; the label built on top of it is not. Nine in ten people are consistent on most of the underlying scales, yet only half keep the same code — the difference is entirely the act of cutting each scale at its midpoint. In fairness to the publisher, it points out that 50% is roughly eight times what chance alone would give you, since there are 16 possible types. That is true, and it is also not the standard most people have in mind when they ask whether a result is accurate.

One caveat worth stating plainly, because most pages on this question blur it: those figures describe the MBTI, not 16Personalities. They are separate instruments from separate companies. NERIS Analytics publishes reliability claims for its own assessment, but we were unable to retrieve that page to quote it here, so nothing above should be read as a measurement of the 16Personalities test specifically. It is the closest published evidence about the format they share.

Is 16 Personalities the same as the MBTI?

No, and the distinction matters if you are trying to judge accuracy. The MBTI is a paid, proctored instrument from The Myers-Briggs Company, historically administered by certified practitioners. 16Personalities is a free online assessment operated by NERIS Analytics that borrows the four-letter output but builds it on a different questionnaire and a Big Five-style trait model, plus a fifth Assertive/Turbulent dimension the MBTI does not have. So "is the MBTI accurate?" and "is 16 Personalities accurate?" are genuinely different questions, even though the answers rhyme: both measure self-reported preferences reasonably consistently, and both compress those measurements into a category that is less stable than the measurement underneath it.

The Barnum effect: why almost any profile feels accurate

There is a second reason a type description rings true, and it has nothing to do with the questionnaire. In 1949 the psychologist Bertram Forer gave students what each was told was a personality sketch derived from their own answers. Every student received the identical text, assembled from a newsstand astrology book. They rated it about 4.3 out of 5 for accuracy.

That result — the Barnum effect — is why "this describes me perfectly" is weak evidence about any personality test, ours included. Descriptions written to apply broadly will feel personal to almost everyone. The useful test is not whether a profile feels right, but whether it says something specific enough that it could have been wrong: a profile that predicts you sometimes need time alone has told you nothing; one that says you decide by consistency rather than by effect on people has made a claim you can actually disagree with.

How to judge any personality test

The question "are personality tests accurate?" has a better answer than yes or no. Four things separate a defensible assessment from an entertaining one:

Does it show you a continuum, not just a category? A percentage on a scale carries the information a letter throws away.
Does it tell you when you are near a boundary? A borderline score is the single most useful thing a type test can disclose, and the thing most of them hide.
Does it avoid predicting your future? Claims about job performance, compatibility or ability are where type tests stop having evidence behind them.
Does it publish how it was built? Reliability figures and an honest account of the limits are a better signal than a large accuracy percentage on a landing page.

What type tests are genuinely good for

Self-reflection: a type profile is a structured mirror — reading one makes you articulate how you actually prefer to work, decide and recharge.
Shared vocabulary: teams and couples get a neutral language for real differences ("you need closure, I need open options") without anyone being wrong.
Memorability: "INFJ" sticks in a way "73rd percentile openness, 41st percentile extraversion…" never will. That's a feature, and also exactly where the precision goes.

What they can't do

No type test — 16Personalities, ours, or the original paper-and-pencil instruments — reliably predicts job performance, relationship success or ability. Personality type and measured intelligence are different things entirely: every type spans the full ability range, as we cover in the smartest personality type. And as with type compatibility, the honest answer is that types describe preferences, not destinies.

So should you trust your result?

Trust it the way you'd trust a well-made mirror, not a medical scan. If the profile reads true, use it; if a letter feels borderline, you probably are borderline on that dimension — read both profiles and keep what fits. That's also how we built our own 16 type profiles: original descriptions, preference language only, no ability claims, and an explicit note that borderline scores are normal. If you want to see where you land, our 24-question personality test gives you a full type profile — and if you want measurement rather than description, that's what an IQ test is for.

Is 16 Personalities Accurate? What Psychometrics Says