Is the MBTI Personality Test Scientifically Valid? What Researchers Found About 16 Types and the Big Five
The MBTI is a useful self-reflection tool but is not a scientifically valid measure of personality in the way researchers use that term. Its four-letter types overlap with the Big Five dimensions—McCrae and Costa’s 1989 analysis placed the Extraversion scale at r ≈ 0.74—yet the model forces continuous traits into 16 fixed boxes. It works best for sparking self-awareness, not for predicting job fit or relationship success.
What the MBTI actually measures
The Myers-Briggs Type Indicator, often called the 16 personality types or simply MBTI, sorts people into one of 16 categories using four either-or preferences: Extraversion or Introversion, Sensing or Intuition, Thinking or Feeling, and Judging or Perceiving. The APA Dictionary of Psychology describes it as a self-report inventory built on Carl Jung’s theory of psychological types, adapted by Katharine Briggs and Isabel Myers in the mid-20th century.
Each preference is treated as a switch, not a dial. You are either E or I, never “mostly E with a dash of I.” That forced choice is the feature users love—it gives a tidy label—and the limitation researchers critique.
The MBTI’s core design choice is to turn dimensions into categories, which is exactly where its scientific problems begin.
What the Big Five measures instead
The Big Five, also known as the Five Factor Model or the OCEAN model, describes personality as five continuous dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Rather than placing you in a box, it scores you on a sliding scale for each trait, so most people land somewhere in the middle.
These five personality dimensions emerged from decades of factor-analytic studies of how people actually describe one another, not from a single theorist’s framework. That bottom-up origin is a big part of why it became the default language of personality psychology.
The Big Five treats traits as points on a spectrum, which matches how personality is actually distributed in real populations.
MBTI vs Big Five: where the two frameworks disagree
The two systems are not enemies—the MBTI’s four scales do track the Big Five to a degree. But they diverge in structure, coverage, and what you can safely conclude from a result.
| Aspect | MBTI (16 personality types) | Big Five (Five Factor Model) |
|---|---|---|
| Underlying model | Four dichotomies → 16 types | Five continuous dimensions (OCEAN) |
| How you are scored | Forced choice into E/I, S/N, T/F, J/P | Rated on a sliding scale for each trait |
| Neuroticism covered? | No equivalent dimension | Yes, as a core factor |
| Stability on retest | ~35% receive a different type after 4 weeks | Median test–retest ≈ 0.82 over two months |
| What researchers use it for | Self-reflection, team ice-breakers | Predicting outcomes, clinical screening |
A forced binary label loses the information that sits between the two poles—and that missing middle is where most people actually live.
How well does each model hold up to scientific scrutiny?
The evidence is lopsided. According to McCrae and Costa’s 1989 analysis, the MBTI Extraversion–Introversion scale correlates with the Big Five Extraversion factor at r ≈ 0.74, while the Thinking–Feeling scale links to Agreeableness at only r ≈ 0.44, showing the types do not map cleanly onto the traits researchers measure.
Reliability is the sharper problem. According to a 2014 meta-analysis in the Journal of Research in Personality that combined 74 studies and 14,923 participants, the Big Five shows a median test–retest reliability of 0.816 across intervals from one day to two months. The MBTI’s own manuals have long acknowledged that roughly 35 percent of test-takers receive a different four-letter type when retested after four weeks.
For predictive power, the Big Five wins clearly. According to Barrick and Mount’s 1991 meta-analysis of 117 studies, Conscientiousness was the single strongest personality predictor of job performance across every occupation examined. And according to Roberts and colleagues’ 2007 review in Perspectives on Psychological Science, personality traits predicted mortality, divorce, and career attainment about as strongly as socioeconomic status and IQ did.
The MBTI correlates with real traits but is less reliable and far less predictive than the Five Factor Model researchers rely on.
Why the MBTI stays popular anyway
If the science is lukewarm, why do millions still take it? Because a four-letter code is memorable, shareable, and conversation-friendly in a way a vector of five scores is not. In workplaces and friend groups, “I’m an ENFP” opens a discussion; “I score high on Extraversion and Openness, moderate on the rest” does not.
My own experience facilitating personality workshops is that attendees remember the four-letter type long after the dimension scores fade, which is precisely why the MBTI is sticky even where it is weak. Used as a doorway into self-reflection, that stickiness is a feature.
If you want to feel the difference directly, sites like personalitree.com let you take a Big Five assessment and a 16-type assessment back to back, which is the fastest way to experience a dimension versus a label.
The MBTI’s value is as an on-ramp to curiosity, not as a verdict about who you are.
Frequently Asked Questions
Is the MBTI completely useless?
No. The MBTI is a weak measurement instrument but a decent reflection prompt. It helps people name preferences and talk about differences, which has real value in teams and relationships. The mistake is treating the four-letter code as a fixed, predictive fact rather than a loose self-description.
Why did my personality type change when I retook the test?
That instability is expected. Because the MBTI forces a binary at every scale, a small shift near the 50 percent midpoint can flip your letter and therefore your whole type. People who score close to the middle are the most likely to get a different result on a second attempt.
Which personality test do psychologists actually use?
Most personality researchers use the Big Five or its formal version, the NEO Personality Inventory, and sometimes the six-factor HEXACO model. These are built for continuous scoring and have far more published evidence behind them than the MBTI does.
Can someone be two personality types at once?
Not in a meaningful sense. Traits are continuous, so you are better described as a point on several spectra than as a member of one of 16 boxes. Someone who scores 52 percent on Thinking and 48 percent on Feeling is far closer to the middle than to either pure type.
Is a 16-type test still worth taking?
Yes, as a starting point. A 16-type assessment like the one on personalitree.com is a fine entry point into personality exploration—just treat the four-letter code as a conversation starter, not a verdict. Pair it with a Big Five score if you want the version researchers trust.
Written by: the Personalitree personality-science desk · Last updated: 2026-11-08 · Sources: published trait research, APA Dictionary of Psychology