MBTI for Hiring: Why It Doesn’t Work

In the last systematic review we did on the MBTI, one uncomfortable figure stopped us cold: roughly 50% of people get a different type classification when they retake the test just five weeks later (Pittenger, 1993, Review of Educational Research). Now imagine basing a hiring decision — one that shapes the next three years of a person’s life and the next $240,000 USD of your company’s money — on something like that.

This post is the long version of why we don’t use the MBTI at Talen.to, what the psychometric evidence says about its validity, and which tool to use instead when a real hiring decision is on the line.

MBTI: What it is and how it reached the corporate world

The Myers-Briggs Type Indicator was developed in the 1940s by Katharine Cook Briggs and her daughter Isabel Briggs Myers, neither of whom had formal training in psychology. They built it as an operationalization of Carl Jung’s theory of psychological types, but without the methodological infrastructure that 20th-century psychometrics was already starting to demand. The work was initially rejected by the academic community, which wasn’t willing to take seriously the research of two women without graduate degrees — a detail that weighs on the instrument’s history and deserves acknowledgment, even if it doesn’t solve the technical problems.

Today the MBTI is still the most popular personality test in the world: more than two million people take it every year, and 88% of Fortune 500 companies have used it at some point. That popularity — not its scientific robustness — is what brought it into hiring processes across LATAM. If a multinational uses it, we assume it must be validated. That assumption is wrong.

The technical problem: MBTI reliability and validity

When we evaluate a psychometric test, we look at two basic things. The first is test-retest reliability: if I give you the test today and repeat it in five weeks, do you get the same result? The second is predictive validity: does that result predict anything relevant in the real world, like job performance or retention at 18 months?

On reliability, the data is damning. Howes and Carskadon (1979) found that only 47% of participants received the same classification across all four dichotomies when they retook the test five weeks later, with stability per individual dimension ranging between 61% and 78%. Later reviews place the type-change rate between 39% and 76% when the test is retaken after just five weeks (Pittenger, 2005). The meta-analysis by Capraro and Capraro (2002) confirms average test-retest correlations of roughly 0.60 over 12-month intervals.

For comparison: the Big Five model typically shows test-retest reliability coefficients between 0.80 and 0.90, even across intervals of years rather than weeks (Costa & McCrae, 1992). That’s a different order of magnitude.

On predictive validity the picture is similar. The meta-analysis by Barrick and Mount (1991) — the foundational study on personality and job performance — did not include the MBTI because the available evidence was insufficient for meta-analytic synthesis. The MBTI’s predictive validity for job performance hovers around r = 0.10–0.20 in the available studies, while the Big Five ranges between r = 0.30 and 0.45 for the same outcome.

“If between 39% and 76% of people get a different MBTI type when they retake the test five weeks later, we’re not measuring personality: we’re measuring the kind of day the candidate had.”

The root problem is methodological. The MBTI forces continuous traits into binary categories: someone who scores 51% on Introversion and 49% on Extraversion gets classified as “I”, but a tiny measurement variation can flip them to “E” on the next retest. It’s not that the person’s personality changes — it’s that the measurement method pulverizes it at the cutoff. Real traits follow a continuous spectrum, not a clean dichotomy. We go deep on this in our explanation of the OCEAN model.

What happens when you use MBTI for hiring

This is the part that worries us most when we talk to HR Managers who are evaluating whether to adopt the MBTI, or who already have it baked into their process.

First: the false sense of objectivity. A code like “INTJ” looks scientific. It’s four letters, there are reports, there are tables comparing types. But that visual precision doesn’t match the underlying psychometric precision. It’s the same as saying someone is “tall-medium” on a three-bin scale instead of 1.78 m: the category erases real information.

Second: candidates who learn to game it. The descriptors of the 16 types are public and live on every corporate blog. A prepared candidate can answer to deliver the “ENTJ visionary leader” or the “ESFJ team player” depending on what they think the company is looking for. The dichotomy-forced structure makes faking easy.

Third — and this surprises most of the HR Managers we talk to — the very company that publishes the instrument advises against using it for hiring. The Myers-Briggs Company explicitly states that using the MBTI for hiring or selection is inappropriate, given that the assessment lacks predictive validity for job performance and doesn’t meet the legal standards for employment testing. The Society for Industrial and Organizational Psychology holds the same official position: personality-type assessments like the MBTI should not be used for employee selection due to insufficient evidence of validity. And the U.S. National Research Council had already reached the same conclusion back in 1991, recommending the Big Five as an alternative.

When a tool is advised against by its own publisher, by the field’s scientific association, and by the national academy, the case for using it in hiring becomes hard to defend.

The alternative: Big Five and OCEAN+

If the MBTI doesn’t work for hiring, what does? The Big Five model — also called OCEAN, for its dimensions Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism — is the personality model with the broadest scientific consensus in the history of psychometrics.

It’s backed by thousands of peer-reviewed studies accumulated over decades, with cross-cultural validation in more than 50 countries. Its test-retest reliability sits between r = 0.80 and r = 0.90 even over long intervals. And, unlike the MBTI, it doesn’t assign types — it places each person at a continuous point within each dimension. To go deeper on the Big Five’s predictive validity for recruitment, we have a post dedicated to the topic.

MBTI vs Big Five / OCEAN+: psychometric comparison

Criterion	MBTI	Big Five / OCEAN+
Test-retest reliability	~50% change type in 5 weeks; r ≈ 0.60 at 12 months (Capraro & Capraro, 2002)	r = 0.80 – 0.90 over months/years (Costa & McCrae, 1992); r = 0.85 – 0.92 in Talen.to’s OCEAN+
Predictive validity (job performance)	r ≈ 0.10 – 0.20	r = 0.30 – 0.45 (Big Five alone); r > 0.60 combined with skills
Measurement structure	4 dichotomies → 16 categorical types	5 continuous scales (6 in OCEAN+)
Body of evidence	Mostly from the publisher; critical independent reviews	Thousands of peer-reviewed studies; academic consensus
Publisher’s position on use in hiring	”Inappropriate and potentially illegal”	Validated for hiring

At Talen.to we work with our own extension of the model: OCEAN+, which adds a sixth dimension called Structure & Rhythm (ER). It measures preference for defined procedures, tolerance for operational ambiguity, and work rhythm — variables we found to be critical for culture fit in operational roles across LATAM, and which the classic Big Five doesn’t capture with the same precision. OCEAN+‘s internal test-retest reliability sits between r = 0.85 and r = 0.92. Combined with Schwartz’s theory of values (10 universal values validated in 82 countries), the result is a continuous Fit Score, not a four-letter label.

If the underlying question is what culture fit is and how to measure it well, we wrote a complete guide to culture fit that covers the rest of the framework.

Can you still use the MBTI for anything?

It would be unfair to say the MBTI is useless. As a non-predictive self-awareness exercise, as shared vocabulary in a team conversation, or as an informal team-building device where no one is going to make a career decision based on the result — it can have value. The MBTI works for self-reflection and team discussions, but it should never be used for hiring, clinical diagnosis, or high-stakes decisions.

The condition is honesty about what it is: a conversation about self-perceived preferences, not a psychometric measurement. The problem starts when that four-letter code is used to decide who enters the pipeline, who moves to the next round, or who gets promoted. At that point it’s not opinion: the very company that publishes the instrument advises against it.

Keep exploring

• Talen.to’s OCEAN+ assessments
• The OCEAN+ profile library by role
• A structured culture fit assessment process before you hire

Frequently asked questions

Does DiSC have the same problems as the MBTI? Structurally, yes. DiSC has fewer peer-reviewed validation studies than the Big Five, or even than the MBTI; its main value lies in its simplicity and applicability for team building, not in its predictive power. For hiring it suffers from the same methodological problem as the MBTI: it categorizes into discrete styles (D, I, S, C) instead of measuring traits on continuous scales.

Are there better free tests than the MBTI? Yes. The IPIP-NEO (International Personality Item Pool) is a public-domain Big Five inventory, validated with 619,150 participants, with average reliability alpha > 0.80 and a correlation of 0.94 with Costa and McCrae’s commercial NEO PI-R. It doesn’t replace a professional assessment with local norms and an interpreted report, but it’s psychometrically superior to the MBTI for self-awareness.

How long does Talen.to’s OCEAN+ assessment take? The OCEAN+ has 50 questions and takes between 20 and 25 minutes. Added to the Schwartz values inventory, the candidate completes the whole process in 35-45 minutes, with the option to pause and resume. The result is a 0-100% Fit Score and a PDF report generated in 1 click.

Does the MBTI have any legitimate use in a company? Yes, in low-risk contexts: optional self-awareness workshops, team-building conversations with the explicit expectation that it’s a discussion tool, not a measurement tool. What has no legitimate use is making a hiring, promotion, or role-assignment decision based on the type.

How to replace the MBTI in your hiring process

If you’ve made it this far, you’re probably already convinced that the MBTI isn’t the right tool for your hiring pipeline. What comes next is practical: replace it with an assessment that has real psychometric validity — OCEAN+ + Schwartz combined into a single Fit Score per candidate, integrable with your ATS and designed for LATAM.

Try Talen.to free for 14 days → No card, no setup on your plate. If your team is coming off the MBTI, we can help you with the changeover.

---

References

• Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44(1), 1-26.
• Capraro, R. M., & Capraro, M. M. (2002). Myers-Briggs Type Indicator score reliability across studies: A meta-analytic reliability generalization study. Educational & Psychological Measurement, 62(4), 590-602.
• Costa, P. T., & McCrae, R. R. (1992). NEO PI-R Professional Manual. Psychological Assessment Resources.
• Howes, R. J., & Carskadon, T. G. (1979). Test-retest reliabilities of the Myers-Briggs Type Indicator as a function of mood changes. Research in Psychological Type, 2, 67-72.
• Johnson, J. A. (2014). Measuring thirty facets of the Five Factor Model with a 120-item public domain inventory: Development of the IPIP-NEO-120. Journal of Research in Personality, 51, 78-89.
• McCrae, R. R., & Costa, P. T. (1989). Reinterpreting the Myers-Briggs Type Indicator from the perspective of the Five-Factor Model of personality. Journal of Personality, 57(1), 17-40.
• National Research Council. (1991). In the Mind’s Eye: Enhancing Human Performance. National Academies Press.
• Pittenger, D. J. (1993). The utility of the Myers-Briggs Type Indicator. Review of Educational Research, 63(4), 467-488.
• Pittenger, D. J. (2005). Cautionary comments regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal: Practice and Research, 57(3), 210-221.
• Stein, R., & Swan, A. B. (2019). Evaluating the validity of Myers-Briggs Type Indicator theory: A teaching tool and window into intuitive psychology. Social and Personality Psychology Compass, 13(2), e12434.
• The Myers-Briggs Company. (2023). MBTI Facts and Common Misconceptions. The Myers-Briggs Company.