Full technical article is currently available in English, Simplified Chinese, Traditional Chinese, Japanese, and Korean. English fallback is shown below.

Literature & Transparency

Scientific Critiques

Examining six major academic objections honestly

Last updated:

Instead of defending every claim historically attached to personality typing, we evaluate each major peer-reviewed critique on its empirical merits.

01

“Type scores do not predict job performance.”

Valid

Morgeson et al. (2007), Pittenger (1993)

The Critique

Large reviews find weak to near-zero correlation between four-letter categories and workplace effectiveness, job satisfaction, or leadership success.

We agree completely. Using four-letter types for hiring, screening, or promotion decisions is both unscientific and unfair.

What OpenJung does: Our user guide warns against workplace screening and positions every score strictly for self-reflection.

Read the User Guide →
02

“Repeated testing can shift a respondent’s type.”

Unvalidated locally

Implementation Evidence Status

What Needs Evidence

Test–retest reliability must be measured directly on the specific questionnaire, scoring code, and interval used—results from proprietary instruments do not verify this adaptation.

OpenJung has not yet published test–retest findings for our web adaptation, our 8-item Quick mode, or our translations.

At the same time, the scoring cutoff explains one obvious boundary hazard: in Full mode, E/I 24 yields E (50% I) while 25 yields I (53% I). One answer step near 50% can flip the badge without claiming stability one way or the other.

What OpenJung does: We preserve assessment mode, version, and percentages on results and link directly to interpretation limits.

Read result change notes →
03

“Personality dimensions are continuous, not binary buckets.”

Valid

McCrae & Costa (1989), Grant (2013)

The Critique

Trait scores cluster along unimodal bell curves rather than splitting into distinct bimodal types.

Research overwhelmingly confirms continuous distributions. Categorical letters are shorthand approximations—like grouping people into “early risers” and “night owls” while knowing sleep timing is continuous.

What OpenJung does: We foreground continuous axis percentages and offer our Big Five Trait Explorer alongside empirical correlation caveats (explicitly noting that types cannot convert into personal Big Five scores).

Open Big Five Trait Explorer →
04

“The four-axis model omits Neuroticism / Emotional Stability.”

Partially valid

McCrae & Costa (1989)

The Critique

Emotional stability predicts meaningful everyday stress responses and well-being outcomes, yet Jungian four-letter models leave it unmeasured.

That omission is structural: OEJTS covers E/I, S/N, T/F, and J/P without adding a fifth emotional-reactivity scale.

Whether omitting it helps or hurts depends on context. Clinical screening requires it; casual style discussions often benefit from avoiding evaluative emotional labels.

What OpenJung does: In our Big Five Trait Explorer and System Notation Lookup, we clarify what the four Jungian dimensions leave out and point toward Big Five (IPIP) trait frameworks.

Explore Big Five Trait Explorer →
05

“Type portraits rely on flattering Barnum statements.”

Partially valid

Forer (1949), Skeptic Literature

The Critique

Broad, upbeat descriptions resonate with almost any reader regardless of actual fit.

Pop-psych summaries often fall squarely into the Barnum trap. However, concrete preference items (“prefers one-on-one talks over group parties” or “relies on checklists vs. improvisation”) remain specific enough to disagree with.

What OpenJung does: We strip vague flattery from profiles, surface blind-spot questions, and explicitly remind users to discard descriptions that do not fit.

06

“Commercial certification incentives distort claims.”

Valid

Grant (2013), Jorgenson (2015) OEJTS 1.2

The Critique

Proprietary licensing fees and corporate consulting create financial pressure to overstate scientific certainty.

Openness does not automatically establish measurement validity, yet removing paywalls and proprietary secrecy eliminates commercial exaggeration.

What OpenJung does: OpenJung is free with open MIT scoring code and clearly attributes the 32-item OEJTS 1.2 instrument under CC BY-NC-SA 4.0.

Read original OEJTS 1.2 development PDF →

Summary of Critiques & OpenJung Position

Critique Assessment OpenJung Position
Does not predict job performance Valid Explicitly warn against hiring or promotion use
Retest stability on this implementation Not yet published Document cutoffs; publish no unverified claims
Continuous traits vs. binary categories Valid Show percentages, sensitivity bands, & Big Five context
Omits Neuroticism dimension Partially valid Acknowledge scope gap; cross-reference OCEAN
Barnum effect in broad type copy Partially valid Use concrete falsifiable habits & reflection prompts
Commercial incentives & closed scoring Valid Public MIT scoring code + OEJTS 1.2 attribution

Primary References

  1. Jung, C.G. (1921 / 1971). Psychological Types (Collected Works, Vol. 6). Princeton University Press.
  2. 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.
  3. Pittenger, D.J. (1993). Measuring the MBTI… And Coming Up Short. Journal of Career Planning and Employment, 54(1), 48–52.
  4. Jorgenson, E. (2015). Open Extended Jungian Type Scales (OEJTS 1.2) Development Report. Open Psychometrics.

Continue Reading