Big Five — the one model with real research behind it

The bars on this page indicate direction, not measurement. They are derived from the correlations between the MBTI’s continuous scales and the five-factor scales, published by McCrae and Costa in 1989 with 468 adults — not from a sample of INTJs given the NEO-PI-R, which nobody has collected. “Openness 92%” here means “the N pole pulls towards the high end of this ruler”; never “the INTJ sits at the 92nd percentile”. The distance between those two readings is the distance between a map and a measurement.

The Big Five is the only model in this section with a real research base: a five-factor structure that reappears across large samples and several languages, with published norms, reliability and measurement error (Costa & McCrae, 1992). That holds for the model. The bridge between it and your four-letter code is another matter altogether: one study, 1989, N = 468, two self-report questionnaires correlated with each other. Firm enough to steer by, fragile enough not to become a verdict.

Openness to experience

VERY HIGH · r = .72

In the NEO-PI-R, openness is the breadth and permeability of mental life, measured across six facets: fantasy, aesthetics, feelings, actions, ideas and values. It is neither intelligence nor accumulated culture — it is appetite for new inner experience, and that appetite tends to show on all six fronts at once (Costa & McCrae, 1992).

This is the cleanest correspondence between the two models: the S–N scale, scored towards N, correlates .72 with openness. The N in your code and this factor are measuring nearly the same ground — and the ground is Ni: hunger for pattern, for theory, for a system nobody has formulated yet. Be careful, though, with the comfortable myth that intuitives run high on ideas and low on aesthetics. The correlation runs across the facets, aesthetics included; nothing in the data carves beauty out of your profile. What sits outside is not beauty — it is the now, the immediate sensory world that inferior Se registers late and badly. You can listen to Bach for three hours and still forget to have dinner.

Conscientiousness

HIGH · r = −.49

Conscientiousness is impulse control in the service of goals: competence, order, dutifulness, achievement striving, self-discipline and deliberation (Costa & McCrae, 1992). It measures how much you plan, persist and resist the detour — not how well you work, and certainly not how much you are worth.

The J–P scale, scored towards P, correlates −.49 with conscientiousness: the J pole goes with the upper half of the ruler, but with a wide margin. Roughly a quarter of the variance is shared, and the other three quarters are exactly where the INTJ with a chaotic desk and a met deadline lives. In your case conscientiousness is manufactured by Te and pointed inward: achievement striving and self-discipline against a plan you wrote yourself, not dutifulness towards somebody else’s rule. The J promises no order in the room. It promises order in the plan.

Extraversion

LOW · r ≈ −.7

Extraversion, in the NEO-PI-R, is the quantity and intensity of interpersonal engagement plus activity level and positive affect: warmth, gregariousness, assertiveness, activity, excitement seeking and positive emotions (Costa & McCrae, 1992). The low pole describes appetite, not capacity — someone scoring low is not incapable of company, they simply do not go looking for it.

The strongest correspondence of the four: the E–I scale, scored towards I, correlates about −.7 with extraversion (−.69 in women, −.74 in men). The I in your code and this factor are nearly the same measurement under two names. And this is where almost every INTJ description ties itself in a knot: assertiveness is a facet of extraversion, the third of the six. “Low extraversion with high assertiveness” only sounds contradictory if you treat the domain as one thing; it is the average of six, and an average is where disagreement between the parts goes to hide. Assertiveness is the facet through which the quiet INTJ still takes command of a room they never wanted to enter. Note what that is and what it is not: nobody has published a facet profile for INTJs. The structure of the domain dissolves the apparent contradiction; it did not measure it.

Agreeableness

LOW TO MEDIUM · r = .44

Agreeableness is cooperative against antagonistic interpersonal orientation, across six facets: trust, straightforwardness, altruism, compliance, modesty and tender-mindedness (Costa & McCrae, 1992). The low pole is not cruelty — it is scepticism, competitiveness and a willingness to disagree out loud.

The .44 between T–F and agreeableness is the weakest of the four correspondences: less than a fifth of the variance in common. That is thin, and it is precisely this thin correlation that common sense stretches into “thinkers are cold”. It does not stretch. T is a decision criterion — impersonal logic above personal value; agreeableness is a style of dealing with people. You can be low on compliance (never agreeing out of convenience, never retreating to keep the peace) and high on altruism, and the domain returns a lukewarm average that describes neither. Notice too the detail almost every INTJ text gets backwards: straightforwardness is a facet of agreeableness, and it measures sincerity against manipulation, not gentleness. Being blunt does not pull you down the ruler — it makes you hard to flatter.

Neuroticism

NO MBTI ANALOGUE

Neuroticism is the proneness to negative affect and to dysregulation under pressure: anxiety, hostility, depression, self-consciousness, impulsiveness and vulnerability (Costa & McCrae, 1992). It is the factor the literature ties most consistently to psychological distress — and it has no bearing on intelligence, character or competence.

This is the most important fact on the page: no MBTI scale corresponds to neuroticism. In 1989 the four dichotomies were correlated with the five factors; four of them found a partner, and the fifth factor was left standing alone. The direct consequence is brutal: your four letters say absolutely nothing about how much you suffer. Two INTJs with the same Ni–Te–Fi–Se stack, one serene and one corroded by anxiety, are handed the same code — and the instrument is not failing when it does that, because it was built to sort preferences in a non-clinical population, not to measure adjustment (Myers et al., 1998). The bar beside this sits in the middle because there is no direction to indicate. It is the one ruler on this page where your type places you nowhere at all.

Shape

Continuous dimensions, and what that does to a four-letter code

A four-letter code is a claim about kind: you are an I, full stop. The Big Five only makes claims about degree. There is no “high-openness type”; there is a score, a norm to compare it against, and a band of measurement error around it. Swapping one for the other is swapping a map of borders for a map of altitude — and then wondering why the border is nowhere in the landscape.

The uncomfortable detail is that the MBTI itself produces a number before it produces a letter. The manual reports a preference clarity index — how hard the answers pulled to one side — and the letter is what remains once a cut point is applied to that number (Myers et al., 1998). Two people separated by a single answer fall into different categories; two people carrying the same I can sit enormously far apart on the extraversion ruler.

This is exactly where Pittenger’s critique bites. The scores spread out continuously, with most people near the middle, rather than piling into two separate humps at the poles; what does the categorising is the cut point, not nature (Pittenger, 1993; 2005). That does not void type language as vocabulary — it voids treating each letter as a boundary found out in the world.

The letter is the sign of the number, not the number.
A marginal preference and an extreme preference are handed exactly the same letter.
Near the cut point, a retest changes the letter without the person having changed.
In the Big Five there is no cut point to defend: the score already is the result.

Correspondences

What the 1989 numbers actually say

McCrae and Costa gave the MBTI and a five-factor inventory to the same 468 adults and correlated scale with scale (McCrae & Costa, 1989). The MBTI scales are scored towards the second pole — I, N, F, P — and that alone is why two of the four numbers come out negative: the sign is a scoring convention, not a discovery about you.

Correlation is not identity. An r of .72 means roughly half the variance in common; .44 means less than a fifth. Saying “N is openness” is a tolerable overstatement; saying “T is low agreeableness” is already caricature. And both sides of the sum are self-report — the same person answering two questionnaires in the same mood — which tends to inflate this kind of correlation. The finding is solid about questionnaires; the agreement between the theories behind them remains your own inference.

The authors’ own reading was economical and not especially flattering to the MBTI: the instrument appears to measure a subset of the five-factor model, one dimension short and one theoretical layer long, and in the same paper they argued that the continuous scores were more defensible than the categories built on top of them.

E–I × Extraversion: about −.7 (−.69 in women, −.74 in men) — the strongest.
S–N × Openness: .72.
J–P × Conscientiousness: −.49.
T–F × Agreeableness: .44 — the weakest of the four.
Neuroticism: no MBTI scale corresponds to it.

The gap

The -A/-T suffix was born from this hole

The fifth ruler did not slip past the study by accident: it does not exist in the instrument. Myers and Briggs built a sorter of preferences for healthy people, and emotional stability was never a preference — it is a trait, and a trait nobody chooses. The result is a system meticulous about how you decide and utterly silent about what it costs you.

NERIS, which publishes the most widely answered sixteen-personalities test in the world, solved the problem the most direct way available: it added a fifth scale. The company states that it moved away from Jungian types and organised its questionnaire around five-factor traits, keeping the Myers letters as a recognisable shorthand and adding the Identity dimension — Assertive against Turbulent (NERIS, n.d.). The -A and -T you see everywhere are, in substance, the missing ruler glued onto the end of the code.

Two consequences, neither comfortable. First: -A/-T is not a finding about INTJs, it is a design decision. Somebody noticed the absence and filled it, which is a different act from discovering that the type comes in two emotional versions. Second: the NERIS instrument is commercial and proprietary, without the peer-reviewed scrutiny the NEO-PI-R has accumulated. If you want to know where you sit on neuroticism, the suffix is a friendly hint, not a measurement.

Facets

Six facets per domain, and the average that lies

The NEO-PI-R does not measure five things: it measures thirty, grouped into five (Costa & McCrae, 1992). Each domain is the sum of six facets that travel together closely enough to form a factor and separately enough to disagree inside the same person. The domain score is an average — and an average is where information goes to die.

That is why “low extraversion with high assertiveness” is no contradiction at all. Assertiveness measures dominance, initiative and willingness to speak first, and it is under no obligation to travel with gregariousness and excitement seeking. An INTJ can be emptied out by company, indifferent to parties and averse to sensory risk — and still be the person who settles the direction of the meeting before the coffee goes cold.

What cannot be done is pretending any of that was measured. The 1989 study correlated domains, not facets, and no facet profile for INTJs has been published. The structure of the NEO-PI-R explains why the popular description does not contradict itself; it does not confirm it.

Neuroticism: anxiety, hostility, depression, self-consciousness, impulsiveness, vulnerability.
Extraversion: warmth, gregariousness, assertiveness, activity, excitement seeking, positive emotions.
Openness: fantasy, aesthetics, feelings, actions, ideas, values.
Agreeableness: trust, straightforwardness, altruism, compliance, modesty, tender-mindedness.
Conscientiousness: competence, order, dutifulness, achievement striving, self-discipline, deliberation.

A common misreading

Openness includes the aesthetic — what sits outside is the present

It circulates in the typology literature, and it circulated in the summary version of this section too: the idea that the intuitive runs high on ideas and low on aesthetics. There is no support for it. Openness correlates with intuition across the facets, aesthetics included; the factor exists precisely because fantasy, art, feeling and idea rise together in the same person.

What genuinely sits outside your reach is not the beautiful — it is the immediate: the sensory present, the body, the raw datum arriving through the senses before it turns into meaning. And that is a statement about inferior Se, the vocabulary of Jung and Myers, not about the Big Five: the five-factor model has no sensing function to score at all. Two different languages, and for this particular word the translation simply has no equivalent.

Limits

What this model does not give you

In the Big Five there is no INTJ. There are five scores. “INTJ” is a category imported from another instrument, and everything on this page is translation — translations lose things, and what is lost here is the narrative. The model has no dominant function, no motivation, no basic fear. It tells no story about you at all, and that refusal is half of its value.

In exchange, it delivers what no other map in this section delivers: norms to compare a score against, retest stability, declared measurement error, and a factor structure that insists on reappearing across different samples and languages (Costa & McCrae, 1992). Unromantic and checkable, which is precisely the bargain science offers.

The final warning cuts both ways. Pittenger argues that the MBTI should not steer consequential decisions — selection, promotion, career direction — because the predictive validity evidence does not support that use (Pittenger, 1993; 2005). A five-factor profile derived by correlation from four letters inherits that limit whole and adds one of its own: you are reading the estimate of an estimate.

The one model in this section that would win the argument

If you keep one sentence from this page, keep this one: the Big Five is the model with real research behind it, and even it refuses to give you a name. It does not call you an INTJ, does not call you a strategist, promises you nothing — it gives five positions and the error on each. Use it to calibrate the rest of this site: when the enneagram, the tritype and the temperament are all singing in suspicious harmony, this is the model that interrupts to ask how much, compared with whom, measured how. And do not forget the missing ruler. The dimension most consistently tied to psychological distress is exactly the one your four letters say nothing about.

Evidence: high for the model, moderate for the bridge to the MBTI, nil for “the INTJ’s Big Five profile” treated as measured data.

Sources

Not every claim about type carries the same weight. The tag beside each work says where it comes from — and what counts as evidence varies a great deal between them.

Academic psychology Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI) Professional ManualPsychological Assessment Resources, Odessa, FL.
Academic psychology McCrae, R. R., & Costa, P. T. (1989). Reinterpreting the Myers-Briggs Type Indicator from the perspective of the five-factor model of personalityJournal of Personality, 57(1), 17–40. N = 468.
Methodological critique Pittenger, D. J. (1993). The utility of the Myers-Briggs Type IndicatorReview of Educational Research, 63(4), 467–488.
Methodological critique Pittenger, D. J. (2005). Cautionary comments regarding the Myers-Briggs Type IndicatorConsulting Psychology Journal: Practice and Research, 57(3), 210–221.
Jung’s own work Jung, C. G. (1921). Psychologische TypenRascher. Ed. inglesa: Psychological Types, Collected Works 6, Princeton UP, 1971.
Official MBTI Myers, I. B., McCaulley, M. H., Quenk, N. L., & Hammer, A. L. (1998). MBTI Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator3ª ed., Consulting Psychologists Press. Amostra nacional representativa, N = 3.009.
Specific model NERIS Analytics Ltd. (s.d.). Our Framework16personalities.com/articles/our-theory, acesso em 2026. Página antes intitulada “Our Theory”. Instrumento comercial, sem manual técnico, normas ou validação publicada.