Most people who have taken a personality MBTI test walked away feeling understood. The four letter code felt right. It explained why Monday morning meetings drain you, or why you keep rewriting emails that your colleague fires off in thirty seconds. The Myers Briggs Type Indicator sorts you into one of sixteen types and tells you something meaningful about who you are. At least, that is the assumption. And roughly two million people take the test every year, convinced this assumption is true.
The challenge is that when researchers outside the MBTI's own publishing ecosystem have examined the evidence, the picture looks very different. A review in the Consulting Psychology Journal noted that a National Academy of Sciences committee concluded there was not enough well designed research to justify using the MBTI in career counselling programmes. A systematic review covering studies from 2017 to 2025 confirmed that somewhere between 39 and 76 percent of people receive a completely different personality type when they retake the test just five weeks later. A comprehensive validity evaluation published in the Review of Educational Research found insufficient evidence to support the claims made about the test's utility, including its ability to predict job performance or career outcomes.
So why do 88 percent of Fortune 500 companies still use it? And what happens when the most popular personality framework in the world collides with the evidence that tests its claims?
The Comfortable Myth of Personality MBTI Types
The MBTI's appeal is easy to understand. It takes something impossibly complicated, the full range of human personality, and makes it neat. You are either an Extravert or an Introvert, a Thinker or a Feeler. Organizations love this because it gives them a shared vocabulary. Teams can talk about their differences using a common language. Workshops feel productive. People feel seen.
But that neatness is exactly the problem. The MBTI assumes that personality falls into distinct categories, the way blood types do. You are either type A or type B. You cannot be half of each. The instrument was built on Carl Jung's theory of psychological types, which Jung himself described as a framework for observation rather than a rigid classification system. Katharine Cook Briggs and her daughter Isabel Briggs Myers transformed Jung's exploratory ideas into a fixed typology during the 1940s. Neither had formal training in psychometrics or psychology. They created a tool that felt deeply insightful, and that feeling has carried it forward for decades, regardless of what scientific evidence says.
The MBTI industry generates enormous revenue. The Myers Briggs Company calls it the world's most widely used personality assessment, and it has been estimated that the broader personality testing market will reach 6.5 billion dollars by 2027. When that much money depends on a product continuing to sell, the incentive to question its foundations weakens.
What Decades of Personality MBTI Research Actually Show
The scientific case against the MBTI rests on three pillars: reliability, validity, and theoretical structure. Each one matters, and the evidence on each is concerning.
The Reliability Problem: Your Type Keeps Changing
If a personality test is measuring something real and stable, it should give you roughly the same answer every time you take it. Psychologists call this test retest reliability, and it is one of the most basic requirements for any assessment. The MBTI's performance on this measure is genuinely troubling.
A systematic review published in the International Journal of Social Science Research, covering studies from 2017 to 2025, found that roughly half of people who retake the MBTI within five weeks receive a different four letter type. The Thinking/Feeling and Judging/Perceiving scales were particularly unstable. The Extraversion/Introversion scale performed somewhat better, but even there, the results fell short of what psychologists expect from a stable personality measure.
A meta analytic reliability study published in Educational and Psychological Measurement, synthesising data across dozens of studies, found that while the MBTI's continuous scores showed acceptable average reliability, individual studies ranged wildly. One study reported a test retest coefficient as low as 0.48 on the Thinking/Feeling scale, a number most psychologists would consider unacceptable for making any meaningful decision about a person.
To put this in context, the Big Five personality model typically achieves test retest reliability coefficients between 0.80 and 0.90, even over intervals of years. The MBTI's continuous preference scores sometimes reach that range. But most users never see continuous scores. They see the four letter type, and that is where the instability is most obvious. When your scores sit near the middle of any scale, which is where most people's scores fall, a small shift in your answers flips you from one type to another entirely.
The Validity Question: Can Personality MBTI Types Predict Anything?
Even a somewhat unreliable test might be worth using if it predicted something important. Does knowing that someone is an ENTJ tell you anything useful about how they will perform at work, lead a team, or navigate their career?
The evidence says: barely. The 1991 National Academy of Sciences review found that only the Extraversion/Introversion scale showed adequate construct validity, meaning it correlated well with similar scales from other established tests. The Sensing/Intuition and Thinking/Feeling scales showed relatively weak validity. The committee's conclusion was direct: the popularity of the instrument in the absence of proven scientific worth is troublesome.
More recent research has not substantially improved the picture. A 2023 study of 529 participants published in Frontiers in Psychology examined whether MBTI personality types predicted leadership behaviours. Of twenty hypothesised relationships between MBTI dichotomies and specific leadership practices, only seven reached statistical significance. The researchers encouraged caution in drawing causal conclusions from MBTI scores.
A 25 year psychometric review of the MBTI Form M, published in 2025 in the Journal of Counseling and Development, aggregated data from 178 articles involving more than 57,000 participants. The review found some support for convergent validity, with roughly 43 percent of comparisons with other personality instruments reaching acceptable thresholds. But it also noted that structural validity and test retest studies were entirely absent from the 25 year literature it sampled. The authors warned that clinicians should not stretch their interpretations beyond what the current research findings support.
Researchers Stein and Swan published a detailed evaluation of MBTI theory in Social and Personality Psychology Compass, identifying several foundational problems. They pointed out that the MBTI's dichotomies, such as Thinking versus Feeling, are presented as opposites when research on human cognition suggests they operate independently. They also noted that the MBTI's definitions of its dimensions differ drastically from personality dimensions that have been refined through decades of empirical research, leaving it in what they called a parallel universe governed mostly by commerce rather than peer review.
The Structural Problem: Boxes Where There Should Be Spectrums
Beyond reliability and validity, there is a deeper issue with how the MBTI categorises people. The instrument assumes personality is typological: you are one thing or its opposite. But virtually all personality research conducted over the past several decades shows that personality traits are distributed along continuous spectrums, not clustered into distinct types.
If personality types were real, you would expect to see a bimodal distribution when you plot people's scores, two clear humps at opposite ends of each scale. Instead, scores on the MBTI consistently follow a normal bell curve distribution, with most people clustering in the middle. The MBTI draws an arbitrary line through the centre of that curve and assigns everyone on one side to one category and everyone on the other side to the opposite category. Two people whose scores are almost identical can receive different type labels simply because they fall on opposite sides of that dividing line.
This matters enormously in practice. When an organisation sorts its employees into MBTI types and builds team exercises or career development pathways around those types, it is treating a continuous, blurred reality as though it were a clean binary. The consequence, as the systematic review in the International Journal of Social Science Research concluded, is that the MBTI's binary structure oversimplifies personality and ignores the diverse continuum of personality traits.
Related: Personality MBTI: Why the World's Most Popular Personality Test May Be Its Least Scientific
What Actually Works: The Trait Based Alternative
If the MBTI is this problematic, what should organisations use instead? The answer, according to the weight of evidence, is trait based personality models, particularly the Big Five (also known as the Five Factor Model). The Big Five measures five broad dimensions: Openness to Experience, Conscientiousness, Extraversion, Agreeableness, and Emotional Stability (the inverse of Neuroticism). Critically, it treats each dimension as a continuous spectrum, not a binary type.
The evidence base behind the Big Five is enormous. A landmark meta analysis by Barrick and Mount, published in Personnel Psychology, examined the relationship between the Big Five dimensions and job performance across five occupational groups. They found that Conscientiousness predicted job performance consistently across all occupational groups and all types of performance criteria. The remaining dimensions showed meaningful relationships that varied by occupation and context. This was not a vague finding buried in a single study. It was a pattern confirmed across 117 validation studies.
A more recent synthesis of over 50 meta analyses strengthened the picture further. Conscientiousness remained the strongest predictor of job performance overall, while Extraversion and Emotional Stability proved particularly important in specific contexts. The associations varied meaningfully across different types of performance, which is exactly what you would expect from a model that captures genuine complexity rather than forcing it into boxes.
Research on work engagement and personality has added further depth. A meta analysis published in the Journal of Occupational Health found that the Big Five model explained roughly 30 percent of the variance in work engagement. Conscientiousness showed the strongest association, followed by Extraversion and Openness to Experience. These are meaningful, replicated findings, and they come from a model that does not force people into types they may not actually belong to.
The difference in personality assessment across different employment positions has also been illuminated by Big Five research. Studies comparing entrepreneurs, managers, supervisors, and employees found meaningful personality differences that align with what the attraction, selection, and attrition model would predict: people gravitate toward roles that match their personality profiles, and roles exert selection pressure in return. The MBTI has struggled to produce comparable findings with the same consistency or specificity.
Related: What the Big Five Personality Traits Actually Tell You (And What They Do Not)
What This Means For Your Organisation and Career
If you have taken the MBTI and felt a flash of recognition when you read your type description, that feeling is not entirely meaningless. The MBTI does measure some real aspects of personality. Its Extraversion/Introversion scale, for instance, correlates at around 0.74 with the Big Five's Extraversion dimension. The problem is not that the MBTI captures nothing real. The problem is that it captures real personality differences less precisely, less reliably, and less usefully than alternatives that have been subjected to far more rigorous testing.
If you are an HR professional deciding which personality framework to use for hiring, team development, or career counselling, the evidence points clearly away from the MBTI for any high stakes decision. Using it to decide who gets hired, who gets promoted, or who belongs in which role means relying on an instrument that a significant proportion of people would score differently on if they took it again next month. The Myers Briggs Foundation itself states that it is unethical to use the MBTI for hiring. When the test's own publisher tells you not to use it for selection, that should give pause.
Where the MBTI may retain some value is in low stakes contexts: team discussions about working styles, personal reflection, or as a conversation starter about personality differences. Used this way, with full awareness of its limitations, it becomes a loose framework for self exploration rather than a precise measurement tool. The danger arises when organisations treat it as the latter.
Key Takeaways
- The MBTI is the world's most popular personality assessment, taken by over two million people annually and used in the vast majority of Fortune 500 companies. Its popularity, however, has consistently outpaced its scientific support.
- Test retest reliability is a serious problem. Research consistently finds that between 39 and 76 percent of people receive a different MBTI type when retaking the assessment after just five weeks, with certain scales such as Thinking/Feeling performing particularly poorly.
- The MBTI's ability to predict real world outcomes such as job performance, career satisfaction, or leadership effectiveness is weak. The 1991 National Academy of Sciences review and subsequent studies have found insufficient evidence for its predictive validity.
- The MBTI forces a continuous, bell curve distribution of personality scores into binary categories, creating artificial distinctions between people whose actual scores may be nearly identical.
- Trait based models, particularly the Big Five, consistently demonstrate stronger reliability, better predictive validity for job performance and engagement, and a more accurate representation of how personality actually works.
- The MBTI still has a role in low stakes settings such as self reflection and team dialogue, but should never be used for hiring decisions, promotions, or any consequential organisational choice about individuals.
Implications for Practice
If your organisation currently uses the MBTI for recruitment screening or selection, stop. The test's own publisher warns against this, and the research provides no basis for making employment decisions from MBTI results. Replace it with a validated Big Five instrument that provides continuous scores and has a robust evidence base for predicting work outcomes.
If you use the MBTI for team development workshops, reframe how it is introduced. Be explicit that the types are conversation starters, not fixed identities. Discourage employees from labelling themselves or others in ways that limit expectations. Pair the MBTI framework with a clear explanation of its limitations so that participants can engage with it thoughtfully rather than uncritically.
When investing in personality assessment for leadership development, choose instruments grounded in the Big Five framework. These assessments provide dimensional profiles rather than categorical labels, which means they capture the nuance that matters when developing leaders. A manager who scores moderately on Extraversion needs different development support than one who scores very high or very low. Binary type labels erase that distinction.
For career counselling, use interest inventories and work values assessments alongside validated personality measures rather than relying on the MBTI alone. The research consistently shows that career fit depends on multiple factors, and no single four letter code captures enough of the relevant variation to guide consequential career decisions.
Train HR teams and managers to recognise the difference between personality frameworks that feel accurate and those that are empirically accurate. The Barnum effect, the tendency for people to accept vague, generally positive descriptions as uniquely applicable to themselves, explains much of the MBTI's felt validity. Feeling seen by a test result is not the same as being accurately measured by it.
Related Reading on The Human Capital Hub
For more on the role of psychometric testing in hiring decisions, see 18 Reasons Why Psychometric Tests Are Crucial in Hiring. For a practical guide to understanding how psychometric assessments work, explore Psychometric Tests: A Survival Guide. And for more on how personality factors relate to recruitment, see Why Personality Assessment Matters in Recruitment.
