In Brief
MBTI is the most widely used personality tool in the world — and one of the most criticized by researchers. Both facts coexist. Before discarding it or trusting it blindly, it's worth understanding precisely what science says, and what it doesn't.
The Test-Retest Reliability Problem
Test-retest reliability measures whether you get the same result when taking a test at two different points in time. For a valid personality instrument, this number needs to be high.
For MBTI, the data is troubling.
A review by Pittenger (2005) summarizing decades of research found that approximately 50% of people get a different type when retaking the test five weeks later. On some dichotomies — particularly I/E and J/P — variability can exceed 75%.
This is not an isolated anecdote. It is the consistent finding across multiple independent studies.
The question this raises: if your "type" changes between administrations, is it truly a stable characteristic of your personality?
The Distribution Problem
MBTI assumes people fall into one category or another on each dimension: either Introverted or Extraverted, either Thinking or Feeling. The categories are binary.
The problem: the data is not bimodal. It follows a normal distribution — most people cluster near the center, not at the extremes. Cutting someone into "I" or "E" means arbitrarily choosing a cutoff point where reality is a continuum.
Imagine measuring height and deciding: below 5'9" you are "short," above you are "tall." You lose all the information in the actual distribution. This is precisely what MBTI does with personality traits.
The Big Five, by comparison, gives a numerical score on each dimension — preserving the nuances.
The Academic Criticism
Researchers McCrae and Costa — the creators of the Big Five model — published direct critiques of MBTI in the 1980s and 1990s. Their main conclusions:
- The four MBTI dimensions approximately correspond to four of the Big Five factors, but imperfectly and with significant loss of information.
- MBTI does not measure what it claims to measure with the precision required for professional applications.
- Predictive validity — the ability to predict real behaviors — is low for most HR applications for which it is sold.
These critiques did not prevent MBTI from becoming a 2-billion-dollar industry.
What MBTI Actually Does Well
The scientific critique is serious, but it does not mean the tool is useless. MBTI has real strengths:
Shared vocabulary. "I'm more F than T" communicates something useful in seconds. In a team or coaching context, this shortcut has genuine value.
Self-reflection. Taking the test and reading the descriptions prompts many people to observe themselves more precisely. The introspection triggered by the tool has value even if the categories are imperfect.
Accessibility. The Big Five is more scientifically rigorous, but its presentation — five abstract numerical scores — is less memorable and less engaging for most people. MBTI tells a story. Humans remember stories.
Jungian cognitive functions. The underlying system of 8 functions (Se, Si, Ne, Ni, Te, Ti, Fe, Fi) is more defensible than the 4 surface letters. It offers granularity and internal logic that research criticism targets less directly.
What MBTI Gets Wrong
Rigid boxes. Being "INTJ" creates a fixed identity. "I'm INTJ so I can't be warm" is an example of how the tool can constrain rather than liberate.
Predictive validity. MBTI does not reliably predict job performance, relationship compatibility, or training outcomes — despite the claims of many consultants who sell it for these purposes.
Recruitment use. Several large companies have abandoned MBTI as a selection tool after finding its predictive validity lacking in this context. Using it to decide whether someone gets a job is particularly problematic.
Why It Persists Despite Criticism
If academic criticism is so serious, why is MBTI still everywhere?
Several factors:
- The Barnum / Forer effect. MBTI descriptions are worded to seem precisely true for almost everyone. They are flattering and avoid negative framings.
- Certified practitioners' investment. Thousands of consultants and coaches have invested in MBTI certification. They have a stake in maintaining its reputation.
- Emotional accessibility. It meets a deep human need: to understand oneself and feel understood.
- Absence of a popular alternative. The Big Five is more valid, but less mainstream.
Scientifically Validated Alternatives
If you want a tool with a stronger empirical foundation:
Big Five (OCEAN). The dominant model in personality psychology. Five dimensions (Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism) measured on continuums. High reliability, documented predictive validity. Less memorable, but more rigorous.
HEXACO. Extension of the Big Five with a sixth dimension (Honesty-Humility). Particularly useful for predicting ethical behaviors and group dynamics.
NEO PI-R. Clinical version of the Big Five, used in therapeutic and research contexts. Longer to complete, but highly precise.
The Shinkofa Approach
On Shinkofa, we use MBTI with the "Reflection" validation badge — neither "Scientific" nor "Traditional." This means:
- The tool is presented as a mirror, not a truth.
- Results are invitations to exploration, not definitive labels.
- We encourage triangulation: MBTI + Big Five + Enneagram + Human Design together give a richer picture than any single tool.
- We regularly remind users of the test-retest limitation: if you retake in a few weeks and get a different type, that's normal — not an error.
MBTI used honestly is useful. MBTI used as absolute truth is a cage.
What the Cognitive Functions Change
Where 4-letter MBTI is most fragile, the Jungian cognitive functions are relatively more defensible.
Understanding your function stack (dominant, auxiliary, tertiary, inferior) offers a more nuanced behavioral analysis framework than "ENFP vs INFP." Research criticism targets the 4-letter categorization system more than the underlying functional theory.
If MBTI interests you, the path toward depth runs through the functions.
In Practice: How to Use MBTI with Integrity
- Take it as a starting point, not a conclusion.
- Read the two types closest to yours — often one will resonate more than the other.
- Retake in 4 to 6 weeks without looking at your previous result. Compare.
- Compare with the Big Five to see if patterns converge.
- Refuse the boxes. Your type is a central tendency, not a prison.
- Explore the cognitive functions if you want to go deeper.
The tool serves self-knowledge. The moment it starts limiting it, set it down.