The Science of MBTI: How Reliable Is It?

The Origins of MBTI

The Myers-Briggs Type Indicator (MBTI) was developed in the mid-20th century by Katharine Cook Briggs and her daughter Isabel Briggs Myers, building on Carl Jung’s theory of psychological types. It divides personality into preferences across four dichotomies, which combine into 16 types.

The four dimensions are Extraversion (E)/Introversion (I), Sensing (S)/Intuition (N), Thinking (T)/Feeling (F), and Judging (J)/Perceiving (P). Each person’s four-letter combination forms a type code such as ENFJ, INTP, or ISTJ.

Criticism From the Academic Community

• The dichotomy problem: MBTI sorts people into either-or categories, but psychological research shows that personality traits exist on a continuum rather than as dichotomies. Most people fall somewhere in the middle.
• Low test-retest reliability: studies find that around 50% of people receive a different result when retested after five weeks, which shows the classification is not stable.
• Weak predictive validity: MBTI types predict real-world outcomes such as job performance and marital satisfaction poorly — far less than the Big Five does.
• Commercial incentives: the MBTI publisher CPP Inc earns hundreds of millions of dollars a year from it, which raises suspicions of over-marketing.

Why It Still Has Value

Despite these criticisms, MBTI remains widely popular, and its popularity itself says something about its value. We see MBTI’s worth not as scientific classification but as a language for self-reflection.

MBTI offers an accessible vocabulary that helps ordinary people think about their preferences and tendencies. When someone says “I’m an INTP, I need alone time to recharge,” MBTI gives them a language for their inner needs — and that kind of self-awareness has value in itself.

The four MBTI dimensions are not entirely without basis — the E/I dimension correlates strongly with Big Five Extraversion, and N/S with Openness. The problem is not the dimensions themselves, but forcing a continuous spectrum into rigid dichotomies.

How Astralis Uses MBTI

At Astralis, we treat MBTI as a snapshot of cognitive preferences rather than a verdict on your personality. We label the confidence level of every MBTI result: if you actually completed an MBTI assessment, confidence is high; if the MBTI was inferred by AI from your chart, it is labeled “inferred.”

We recommend using MBTI together with the Big Five — MBTI offers an intuitive type description, while the Big Five provides precise dimension scores. The two complement each other and together give a fuller picture than either alone.

Astralis’s psychological profile module supports both MBTI and the Big Five. When a user completes an actual assessment, the result is saved to their profile, and the AI’s integrated interpretation prioritizes real assessment results over inferred data.