Myers-Briggs and the Enduring Popularity of a Personality Test With No Predictive Validity

Myers-Briggs is the most widely used personality assessment in the world. Approximately two million people take it every year. It is used by a substantial majority of Fortune 500 companies for hiring, team-building, and professional development. It has generated a significant industry of certified practitioners, training programmes, and organisational development work. It is also, by most of the measures that personality researchers use to evaluate assessment quality, a poor instrument — one that lacks the test-retest reliability, construct validity, and predictive validity that would be required to justify its use in the consequential decisions it is routinely used for.

The gap between the MBTI’s prevalence and its scientific standing is one of the more remarkable features of the corporate training industry. The test’s problems have been documented extensively in the psychological measurement literature for decades. The organisations that use it are, in many cases, aware of the criticism. The test continues to be used because it is engaging, because it generates productive conversations about individual differences, and because the industry built around it has strong commercial incentives to maintain its credibility. These are real reasons. They are not the same as evidence that the test works.

Myth 1: Myers-Briggs Is Grounded in Established Psychological Theory

The Myers-Briggs Type Indicator was developed by Katharine Cook Briggs and her daughter Isabel Briggs Myers, neither of whom had formal training in psychology or psychometrics, based on Carl Jung’s theory of psychological types. Jung’s typological framework was a rich conceptual contribution to depth psychology, but it was not designed as the basis for a psychometric instrument. It was not derived from empirical data about how traits distribute in populations, and it was not validated against behavioural outcomes. Translating a conceptual typology into a binary four-dimension assessment is a significant departure from the source material, and the resulting instrument does not straightforwardly inherit whatever validity Jung’s broader theoretical framework may have.

Modern personality psychology has largely converged on a different framework: the Big Five (openness, conscientiousness, extraversion, agreeableness, neuroticism), which was derived empirically from factor analysis of personality descriptors across languages and cultures rather than from a theoretical framework. The Big Five has substantially better psychometric properties than the MBTI, including better test-retest reliability and better predictive validity for job performance, academic achievement, and life outcomes. The persistence of the MBTI in corporate settings despite the availability of better-validated alternatives reflects organisational inertia and commercial infrastructure rather than scientific comparison of the available options.

Myth 2: Your MBTI Type Is Stable Over Time

One of the most well-documented problems with the MBTI is its test-retest reliability: how consistently it assigns the same type to the same person across administrations. Studies consistently find that a substantial proportion of people receive a different type classification when they retake the test weeks or months later — estimates range from 35 to 50 percent in many studies, depending on the interval. This is a fundamental problem for a test whose entire application rests on the premise that it is measuring stable characteristics of the person.

The instability is partly an artefact of the test’s forced-choice binary structure. Rather than measuring traits on a continuous scale and allowing scores to fall where they naturally fall, the MBTI assigns people to categories (I or E, N or S, T or F, J or P) by comparing their scores to a midpoint. A person who scores slightly above the midpoint on extraversion on one occasion and slightly below on another will receive opposite type classifications — not because anything about them has changed, but because the binary classification system amplifies small fluctuations around the midpoint into apparently large categorical differences.

Myth 3: MBTI Types Predict Job Performance and Career Fit

The primary application of the MBTI in organisational settings is as a guide to individual differences in work style, team dynamics, and career fit — the implicit claim being that knowing someone’s type provides useful information about how they will perform, what roles they are suited to, and how they will work with others. The research on whether MBTI types predict job performance is consistently negative. Meta-analyses find that MBTI type classifications do not reliably predict job performance across roles, and that the personality dimensions it measures show weaker predictive validity for work-related outcomes than the Big Five dimensions that have been more rigorously validated.

The specific career fit applications of MBTI — suggestions that certain types are suited to certain professions — have similarly weak empirical foundations. There is no reliable evidence that INTJs make better engineers or ENFPs make better counsellors in the sense that these type classifications independently predict performance in these roles beyond what general cognitive ability and specific domain knowledge explain. Career advice based on MBTI classification is not well-grounded in evidence about what actually predicts career success.

Myth 4: MBTI Improves Team Communication and Collaboration

The most defensible application of the MBTI is as a team-building tool that generates conversations about individual differences in communication style, information processing, and decision-making preferences. These conversations have value, and many organisations report that MBTI-based team sessions improve mutual understanding and reduce interpersonal friction. The question is whether the MBTI’s specific type system is necessary or sufficient to produce these benefits, or whether the same outcomes would be achievable through any structured framework for discussing individual differences.

Research on whether MBTI-based team interventions improve team performance over control conditions is thin. The conversations the MBTI generates are valuable; the specific content of the instrument — the sixteen types, the binary dimensions, the specific attributions about cognitive style — is not clearly the active ingredient producing those conversations’ benefits. Any intervention that creates a structured space for team members to discuss their preferences, working styles, and communication needs would be expected to produce similar benefits. The MBTI provides a vocabulary for this discussion; it does not provide validated content about what those differences actually are.

Myth 5: If It Feels Accurate, It Must Be Valid

MBTI users consistently report that their type descriptions feel accurate, which is frequently cited as evidence that the instrument is valid. This reasoning misunderstands the difference between the experience of accuracy and the measurement of it. The Barnum effect — the tendency to accept vague, generally positive personality descriptions as personally accurate — is well-documented. MBTI type descriptions are written to be recognisable: they describe real human tendencies in language general enough that most people reading their type description will find it resonant.

Studies in which participants receive randomly assigned MBTI type descriptions rather than descriptions matched to their actual assessment results find that they rate the random descriptions as accurate at rates similar to their actual type descriptions. The felt accuracy of an MBTI reading is not evidence of measurement validity; it is evidence that people find personalised-seeming descriptions of human tendencies recognisable, which is not a particularly demanding standard.

MBTI Claims vs What the Research Shows

Why It Persists

The MBTI’s persistence in corporate settings despite its well-documented scientific limitations is itself an interesting phenomenon worth understanding. It persists because it is engaging — people find discussing personality types genuinely interesting, and the sixteen-type framework provides a shared vocabulary that many organisations find useful. It persists because the conversations it generates, whatever their relationship to the instrument’s scientific validity, produce real perceived value for participants. It persists because the commercial and certification infrastructure around it is large, well-resourced, and well-incentivised to maintain its credibility. And it persists because the bar for evidence in corporate training is considerably lower than the bar in scientific research — a tool that generates good conversations and positive participant feedback persists regardless of whether it is measuring what it claims to measure.

The most charitable version of MBTI use acknowledges that it provides a vocabulary for discussing individual differences and creates structured space for conversations that are otherwise difficult to initiate in workplace settings. It also acknowledges that the specific type attributions should be held loosely, that type classifications are not stable characteristics that reliably predict behaviour, and that the science does not support the more confident applications — career guidance, hiring decisions, team composition — that the instrument is regularly used for. That honest framing is considerably rarer than the alternative.

Frequently Asked Questions

Related Reading

Books on Personality, Assessment, and What the Science Shows

As an Amazon Associate, this site earns from qualifying purchases made through the links above.