Chapter 7: Key Takeaways
Core Concepts
-
The MBTI has poor test-retest reliability (~50% type change on retest), no evidence for discrete types in the data (scores form continuous distributions), poor predictive validity for job performance, and doesn't measure neuroticism. The National Research Council found insufficient evidence to justify its use.
-
The Enneagram has even less evidence — no consistent factor structure, poor psychometric properties, no predictive validity. Its origins are spiritual, not scientific.
-
The Big Five (OCEAN) is the consensus personality model among researchers. It uses continuous dimensions rather than types, has been replicated across 50+ cultures, and predicts real-world outcomes including job performance, mental health, and relationship quality.
-
The MBTI persists because of the Barnum effect, identity value, institutional momentum, the $2–4B certification industry, and high participant satisfaction (Kirkpatrick Level 1) — none of which require the assessment to be psychometrically valid.
-
The Big Five is culturally unpopular precisely because of the features that make it scientifically valid: continuous dimensions (no catchy labels), inclusion of neuroticism (unflattering), and lack of narrative (no community-building).
-
Carl Jung never intended his types as a classification system. The MBTI took his fluid, nuanced descriptions and hardened them into a 16-type system — exactly what Jung warned against.
-
The virality-accuracy trade-off applies directly to personality assessment: the most shareable frameworks (types, labels, identities) are the least valid, and the most valid framework (Big Five) is the least shareable.
Evidence Ratings in This Chapter
| Claim | Rating | Summary |
|---|---|---|
| "The MBTI is scientifically valid" | ❌ DEBUNKED | Poor reliability, no type structure in data, no predictive validity |
| "The Enneagram has a scientific basis" | ❌ DEBUNKED | No factor structure, no psychometric support, no predictive validity |
| "The Big Five is a valid personality model" | ✅ SUPPORTED | Replicated 50+ cultures, predicts outcomes, good psychometrics |
| "If 80% of companies use it, it must work" | ❌ DEBUNKED | Corporate adoption reflects marketing, not validity |
Key Terms Introduced
- Test-retest reliability: Consistency of results when the same person takes the same test twice
- Predictive validity: The ability of an assessment to predict real-world outcomes
- Factor structure: The pattern of underlying dimensions revealed when assessment data is statistically analyzed
- Construct validity: Whether an assessment measures the psychological construct it claims to measure
- Binary categorization problem: Forcing continuous personality dimensions into either/or categories, losing information and creating instability
One Sentence to Remember
The most popular personality assessment in the world (MBTI) is scientifically unsupported, while the most scientifically supported personality model (Big Five) is virtually unknown to the public — a perfect demonstration of the gap between what sells and what works.