Enneagram of Personality

Status: Documented, not shipped · Evidence: C · Family: Self and team awareness · Verdict: reject (2026-06-11)

What it is

The Enneagram of Personality maps a person to one of nine interconnected types, each defined by a core motivation, a characteristic fear, and a besetting “passion” or fixation (Type 1 the Reformer, Type 2 the Helper, Type 3 the Achiever, Type 4 the Individualist, Type 5 the Investigator, Type 6 the Loyalist, Type 7 the Enthusiast, Type 8 the Challenger, Type 9 the Peacemaker). The nine points sit on a geometric figure - a circle with an inner hexad and triangle - and the system layers on secondary structure: each type has two adjacent “wings” that flavour it, and “arrows” or lines connecting each type to two others, said to describe how the person shifts under stress (disintegration) and growth (integration). A person learns their type by self-report, often via a questionnaire (the Riso-Hudson RHETI, the WEPSS, the iEQ9) or by guided self-study.

The honest description of the move is this: it is a self-typing instrument that hands you a rich, motivation-centred vocabulary for talking about your own habitual pattern - what you chase, what you fear, where you go when you are stretched. Its appeal is depth and narrative: unlike a trait score, a type is a story about why you do what you do. That is also where the trouble starts, because the structure that gives it its narrative power - exactly nine types, the wings, the integration and disintegration arrows - is the part the evidence does not support.

When it helps / when it misleads

It helps, modestly, as a vocabulary for self-reflection and conversation in low-stakes, voluntary, developmental settings. Several studies in the review literature report that participants found the Enneagram useful for personal and spiritual growth, and the mechanism is plausible: a motivation-centred frame (“you may be chasing significance / avoiding conflict / managing your anger”) can prompt a person to notice a pattern they had not named. In that use the cost of a wrong label is low and the value lives in the conversation the typology starts, not in the accuracy of the assignment. As an icebreaker, a coaching prompt, or a shared language for a team that has opted in, it does real work.

It misleads, sometimes seriously, the moment the type is treated as a measured fact rather than a conversational frame:

• The type is not stable. Categorical type agreement on retest is materially lower than the continuous-scale reliability the publishers quote, because the nine boundaries amplify error near the cutoffs; a meaningful minority of people are assigned a different type on a second sitting. A label that can flip on retest is not a basis for a decision about a person.
• It is not a selection or staffing tool. There is no validity evidence supporting Enneagram type as a predictor of job performance or fit, and using it to hire, place, or evaluate is both unsupported and, depending on jurisdiction, legally exposed. Treat any vendor claim of predictive validity for selection as a vendor claim, not a finding.
• The wings and arrows have essentially no empirical support. The secondary theory that gives the system most of its explanatory swagger - that you predictably move to Type X under stress and Type Y in growth - is the part the systematic review found no studies supporting. Reasoning from a person’s “stress arrow” is reasoning from decoration.
• It invites the Barnum read. The type descriptions are evocative and broad enough that most people can find themselves in several, and the act of self-selecting a type then makes the description feel confirmed (the same self-fulfilling loop that makes any typology feel uncannily accurate).

The clean boundary: useful as a self-reflection vocabulary you opt into; misleading the instant it becomes a measurement you act on about yourself or, worse, about someone else.

What the evidence says

The honest grade is C (conceptually plausible but under-tested), which confirms the registry’s preliminary tag. The reason it is C and not the X assigned to its cousin MBTI in this family is a real distinction: the Enneagram is not contradicted on a central claim the way the learning-styles meshing hypothesis is. It is, rather, an under-built structure with a partially-validated kernel and an unvalidated superstructure. And it is not the solid practitioner P of a tool with uncontested adoption for its actual function, because the function it advertises - typing a person across nine empirically-distinct categories - is the part the data do not deliver.

The systematic review is the anchor. Hook, Hall, Davis, Van Tongeren and Conner (2021, Journal of Clinical Psychology) reviewed 104 independent samples and reported mixed evidence of reliability and validity. Three findings from it govern the grade. First, on structure: factor-analytic studies “typically found fewer than nine factors,” and no study used clustering to derive the nine types - the nine categories were reasoned out of the system’s theory and a priori symbolism, not recovered from data. Second, on the secondary theory: the review found no studies supporting the wings or the intertype-movement (stress and growth arrow) concepts. Third, on the partial-validity kernel: Enneagram subscales did show theory-consistent relationships with the Big Five, and the authors note genuine clinical appeal for personal and spiritual growth. Their overall posture is that the empirical foundation needs strengthening before broader validation claims hold - a textbook C profile.

The questionnaire-level evidence is mixed, leaning weak on the categorical claim. Newgent, Parr, Newman and Higgins (2004, Measurement and Evaluation in Counseling and Development), analysing the RHETI on 287 participants, found adequate internal consistency (alpha roughly .70 to .82 on several type scales) and “mixed support for construct validity” by correlational and canonical analysis - but their factor analysis did not support the nine-type construct, returning a two-factor solution per type rather than nine clean type factors. The authors found the strongest support for the instrument’s “heuristic value,” which is precisely the self-reflection-vocabulary use this dossier endorses and not the measurement use it warns against.

The most-cited practitioner review reaches the same hedge. Anna Sutton’s “But is it real? A review of research on the Enneagram” (2012, Enneagram Journal) and the associated empirical work (Sutton, Allinson and Williams, 2013, in Personality and Individual Differences, which found meaningful but partial Enneagram-to-Five-Factor correlations) conclude that the questionnaires (RHETI, WEPSS) are reasonably reliable on their continuous scales while urging caution about validity. A sympathetic insider review landing on “reliable-ish scales, be cautious on validity” is, read honestly, a C verdict from a friendly witness.

What is excluded on the evidence rule. Vendor and advocacy pages (for example claims that a proprietary instrument like the iEQ9 demonstrates strong reliability and validity, or popular “the research says yes” explainers) are V-tier marketing and do not lift the grade; their figures are not independent. No specific effect size or hit-rate from those sources is carried into this dossier as fact. The honest net is: a partially-validated kernel (the type scales do correlate with established traits), a nine-category structure that factor analysis does not recover, and a secondary theory (wings, arrows) with no supporting studies. That is C - under-tested and partly contradicted on its defining structural claim, not flatly refuted on a single hypothesis the way an X method is.

Why it is / is not a skill here

Verdict: Reject (status: excl). This confirms the registry’s preliminary verdict, and the stated reasons are two, either of which is sufficient.

Ground one - it is an instrument, not a cognitive move, and the library does not ship instruments. Every shipped skill here takes a problem, decision, or claim as input and emits a structured artifact that operates on it. The Enneagram takes a person as input and emits a type label. That is the genre line this family was created to police: the self-and-team-awareness candidates were queued precisely because they profile a person or team rather than work a problem, so the shippable fraction is small. The same library logic already routed MBTI to reject/X and the proprietary instruments (CliftonStrengths V, DISC V, Strong Interest Inventory M) to reject; the Enneagram sits with them. A “find your nine-type” skill would emit a label, not a thinking artifact, and would fail the “artifact, not prose” commitment in spirit - the label is the output, and the label is the part the evidence does not support.

Ground two - the evidence gate. The move it advertises (type a person across nine empirically-distinct categories, then reason from their wings and arrows) is graded C, and the specific structural claim is the under-supported part: factor analysis does not recover nine types, and the wings/arrows have no supporting studies. The library’s identity is honest evidence grading. Shipping a nine-type self-typing skill would mean either (a) shipping the structure the data do not support, or (b) quietly discarding the nine-type structure and keeping only the motivation-vocabulary - at which point it is no longer the Enneagram, it is a generic “examine your core motivation and fear” prompt that overlaps the reflective skills the library already has.

Why excl rather than flag, fold, or pm. Not flag: flag means “include with caveats,” which implies we ship something, and we are shipping nothing here - the honest landing is excluded-on-the-merits, the same place MBTI lands. Not a single-target fold: the one defensible kernel (motivation-and-fear self-reflection, and the trait-correlated signal) does not map cleanly onto one shipped skill’s mechanism. Its trait-correlated signal is the generic counterpart already captured by the family’s trait-lens-perspective (Big Five / HEXACO) candidate, and its self-questioning use overlaps the shipped reflective skills, but neither is the same move as Enneagram typing - the overlap is diffuse, so a clean fold would misrepresent it. Not pm: it is not a product-management method that belongs in the sibling library; it is a personality instrument, out of this library’s scope on the merits. Excl records the honest decision: documented in full, not shipped, with the defensible generic signal pointed at trait-lens-perspective.

The learning value of this decision: the Enneagram is one of the most beloved typologies in coaching and team-building, and it has a genuinely useful self-reflection vocabulary buried in it. But “beloved and generative” is not “validated,” and a famous instrument can clear neither gate at once - it profiles a person rather than working a problem, and the nine-category structure that makes it the Enneagram is the part the research does not support. Documenting it as a deliberate exclusion, with an honest C and a pointer to the better-evidenced trait lens, keeps the catalog honest and keeps a famous-but-weak instrument from being laundered into a shipped skill.

Lineage and who to read

The Enneagram symbol is old and contested - a nine-pointed figure popularised in the West by George Gurdjieff in the early twentieth century for his sacred-movement and esoteric teaching, with claimed (and unverifiable) roots traced back through various mystical traditions. The personality system is modern and well-attributed. The Bolivian-born philosopher Oscar Ichazo, at his Arica School (founded 1968), was the first to map psychological meaning onto the nine points, teaching a system he called Protoanalysis that linked each point to a passion, virtue, and ego-fixation. The Chilean psychiatrist Claudio Naranjo learned it from Ichazo around 1970, developed the detailed psychological type descriptions, and taught it in the United States - principally at the Esalen Institute and in his SAT (Seekers After Truth) groups - using live interview panels to illustrate each type. From Naranjo’s students the system spread into the popular and Christian-spirituality streams associated with later authors and the Riso-Hudson “Enneagram Institute.” There is no single trademark holder: multiple schools (Riso-Hudson, Palmer-Daniels narrative tradition, Integrative9, and others) teach competing versions, which is itself part of why the instrument’s claims do not converge. The entry is therefore documented descriptively and is not flagged as branded; specific test names (RHETI, iEQ9) are the trademarks of their respective publishers.

Named sources

• Joshua N. Hook, Todd W. Hall, Don E. Davis, Daryl R. Van Tongeren & Mackenzie Conner, “The Enneagram: A systematic review of the literature and directions for future research,” Journal of Clinical Psychology 77(4) (2021): 865-883. Reviewed 104 independent samples; mixed reliability and validity, factor analyses typically recovering fewer than nine factors, no support for the wings or intertype-movement concepts, theory-consistent Big Five relationships. The governing source. (Systematic review)
• Rebecca A. Newgent, Patricia A. Parr, Isadore Newman & Karen K. Higgins, “The Riso-Hudson Enneagram Type Indicator: Estimates of Reliability and Validity,” Measurement and Evaluation in Counseling and Development 36(4) (2004): 226-237. N=287; adequate internal consistency on several type scales but factor analysis not supporting the nine-type construct (a two-factor solution per type); strongest support for “heuristic value.” (Empirical, instrument-level)
• Anna Sutton, “But is it real? A review of research on the Enneagram,” The Enneagram Journal 5(1) (2012). A sympathetic-insider review concluding that the questionnaires are reasonably reliable on their continuous scales while urging caution about validity. (Practitioner/review)
• Anna Sutton, Christopher W. Allinson & Helen Williams, “Personality type and work-related outcomes: An exploratory application of the Enneagram model,” reported with related work in Personality and Individual Differences (2013). Found meaningful but partial correlations between Enneagram types and the Five-Factor Model - the partially-validated kernel. (Empirical)
• Oscar Ichazo (Arica School, from 1968) and Claudio Naranjo (Esalen / SAT groups, from circa 1970): the originators of the modern personality system. Primary lineage rather than evidence; read for what the system claims, not for whether it holds. (Primary/foundational)

Excluded on the evidence rule: vendor and advocacy validity claims (for example proprietary-instrument reliability and validity figures from publishers such as Integrative9/iEQ9, and popular “the research says yes” explainers) are not independent and do not influence the grade. No effect size or hit-rate from those sources is carried into this dossier as fact.