Best Enneagram Tests in 2026: An Honest, Evidence-Based Comparison

Comparison at a glance

How we compared

We ranked nothing on accuracy, and we published our own product’s shortfalls alongside everyone else’s, because the fair basis of comparison in this category is published evidence and transparency, not marketed precision.

Here is why. The single most important fact about Enneagram testing is that the model itself is only partially supported. Hook and colleagues (2021), reviewing 104 independent samples, found “mixed evidence of reliability and validity”; factor analyses “typically found fewer than nine factors,” and secondary ideas such as wings and arrows have little empirical support. In that landscape, no test can honestly be called “the accurate one.” So instead of an accuracy score, we used seven checkable criteria — free results, wings, a confidence reading, published methodology, item count, price, and fit — and reported each from the sources.

A word on the two claim-kinds below. Research evidence is what an independent or published study actually measured. A marketing claim is what a maker advertises. These are different things, and the recurring problem in this field is that they get blurred — a vendor-run study on internal consistency becomes a billboard about “accuracy.” We keep them separate, and where a marketed figure has no study behind it, we say exactly that and stop there.

The tests, one by one

The Crossing (ours)

The Crossing’s methodology: The Crossing scores all nine types, identifies your likely type and wing, and reports how confident the result is with a four-level band — Clear lead, Moderate lead, Close call, or Split — beneath which it always states that the band reflects how clearly one pattern separated in your answers, not a probability that the result is correct. Its 54 motivation-focused Likert items plus 12 either-or tie-breaker rounds (a design in which all 36 possible type-pairs get exactly one head-to-head test) and its full scoring engine are published in a downloadable technical readout. Research status, stated exactly as our own science page states it: The Crossing is pre-pilot. Its items are in the cognitive-interview (think-aloud) stage, but it has not yet run a beta reliability study or a criterion-validity study. It publishes no accuracy figure and will not until it has run its own validation study; its confidence bands are separations in your answers, not accuracy claims. We include it here because fairness requires disclosing it, not because its evidence outranks the documented incumbents — on published validation, it does not yet.

WEPSS — Wagner Enneagram Personality Style Scales

Research evidence: the WEPSS is the only Enneagram inventory with a formal technical manual (Western Psychological Services) and the only one reviewed in Buros’s Mental Measurements Yearbook, which called it “a praiseworthy effort” while noting that further empirical confirmation was still needed. Reported internal consistency runs α ≈ .78–.88 and test–retest r ≈ .62–.91 at six weeks, though much of this comes from the author’s manual and dissertations rather than independent peer review. It reports percentile scores across all nine styles, with a distinctive healthy/unhealthy split.

iEQ9 — Integrative Enneagram Questionnaire (Integrative9)

Research evidence: a vendor-commissioned 2019 study (n=5,910) reported Cronbach’s alpha of .73–.84 across the nine type scales and a nine-factor structure — genuinely the strongest factor result in the field. Marketing vs. demonstrated: iEQ9 markets the test as “95% accurate” and “the most reliable and accurate Enneagram test currently available.” That figure does not appear in its published study, and independent reviewers note it conflates accuracy, validity, and confidence interval, which are not interchangeable; treat it as an unverified marketing claim, not a demonstrated per-person hit-rate. The adaptive 175-item report is the richest in the category.

RHETI v2.5 — Riso-Hudson (The Enneagram Institute)

Research evidence: independent, dissertation-level work — Newgent and colleagues (2004, n=287), corroborated by Giordano (2008) — reported internal consistency of α ≈ .56–.82 and theory-consistent NEO correlations on small convenience samples. Marketing vs. demonstrated: the RHETI is marketed as “scientifically validated,” which is supported at the level of acceptable internal consistency in small, dated samples, not a current manual or a criterion hit-rate; its forced-choice (ipsative) format also complicates standard reliability metrics. The 144-item test returns a full nine-type profile.

Essential Enneagram — Daniels & Price

Research evidence: this is the one instrument with a published criterion study — 970 US adults unfamiliar with the Enneagram, compared against a blind expert interview and post-course self-reassessment, giving roughly two-thirds per-type agreement (published examples: Type 1 ≈ 66%, Type 9 ≈ 68%; overall Cohen’s κ ≈ .53). That design is rare and valuable, though the sample is old (circa 2000) and author-produced. Marketing vs. demonstrated: its “first and only scientifically determined Enneagram test” superlative overstates modest per-type figures, but the method behind those figures is real. It uses nine best-fit paragraphs and an explicit look-alike disambiguation step.

Truity

Research evidence: Truity publishes a public technical report (January 2026 sample N=23,033) with facet-level alphas of roughly .70–.87 and reported discriminant validity — unusually transparent for a free consumer test, though the analysis validates 21 facets rather than the nine-type structure or a typing rate. Marketing vs. demonstrated: its claim to be “the only online Enneagram assessment that offers publicly available data on reliability and validity” is roughly defensible; its “most accurate Enneagram test available” claim is not independently substantiated. Its 105-item test shows free all-nine basic results with a $19 full report.

Crystal Knows

Research evidence: Crystal publishes descriptive data (type frequencies from its dataset) but no reliability or validity study for its Enneagram scale, and it candidly cites Hook et al. (2021) on the Enneagram’s limited empirical validation. Marketing vs. demonstrated: Crystal’s help materials state its personality predictions are “on average 80% accurate” and its personality tests “on average 97% accurate” — self-reported figures for its broader profiling platform, not an independently verified figure for the Enneagram quiz specifically. It is a DISC-native workplace tool with a free Enneagram test.

Eclectic Energies

Research evidence: none published. What sets it apart is honesty of framing: it carries an explicit disclaimer that “no personality test is completely accurate,” shows percentage scores across all nine types, and tells takers to consider their close scores. Its free tests report type, wing, and instinctual subtype.

Cloverleaf

Research evidence: none specific to Cloverleaf. It states that it is built on the RHETI model and cites the RHETI’s construct-validity sources rather than presenting its own manual, so it borrows the RHETI’s evidence. Marketing vs. demonstrated: it is marketed as “one of the most accurate and widely trusted Enneagram assessments available” — a claim with no Cloverleaf-specific validation behind it. Its free individual test reports type, wings, triads, and arrows and is built for team use.

Personality Path

Research evidence: none published. Personality Path offers a short free rate-the-statements test with an optional paid deeper report (about $27), reporting type and wings. We include it for completeness and fit; it makes no validation claim we could source, and we assert none for it.

The honest bottom line

Pick an Enneagram test for its transparency and its fit to what you want, not for a marketed accuracy number — because no such number in this category is independently substantiated, and the model underneath every one of these tests is only partially supported.

Hook and colleagues (2021) are unambiguous that the nine-factor structure is not yet empirically established and that wings and arrows have thin support. That is the honest floor beneath the entire category, ours included. Within it, the useful questions are practical: want depth and a rich report? The iEQ9 and WEPSS document themselves the most. Want a good free result? Truity is the best-documented, with Crystal, Cloverleaf, and Eclectic Energies as free alternatives that publish no instrument-specific validation. Want the most method-honest typing approach on record? The Essential Enneagram’s criterion study is still the only one of its kind. And if you specifically want a test that tells you how confident it is, that reporting is the documented gap almost no one fills — the feature The Crossing is built around, though as a pre-pilot instrument it has evidence still to earn.

Frequently asked questions

Which Enneagram test is most accurate?

No Enneagram test has published an independent, per-person correct-typing rate, so “most accurate” cannot be answered with a verified number. The only criterion-referenced figure in the field is the Essential Enneagram's roughly 66–68% per-type agreement against a blind expert interview (overall κ≈.53, n=970). Marketed figures like “95%” (iEQ9) and “97%” (Crystal) are vendor-run and not independently demonstrated.

What's the best free Enneagram test?

On published documentation, Truity is the best-documented free option: it shows free all-nine basic results and publishes a technical report. Crystal Knows, Cloverleaf, and Eclectic Energies are also free and show a basic result, but none publishes an instrument-specific reliability or validity study.

Is there a scientifically validated Enneagram test?

Partly, and not fully. Some tests publish reliability evidence — WEPSS (manual plus Buros review), iEQ9 (a large vendor study), Truity (a public technical report) — but Hook et al. (2021) found mixed evidence across all instruments, and no test has an independent criterion-validity study establishing per-person accuracy.

Does any test tell you how confident it is in your result?

Almost none report a per-person confidence band — the reviewed literature calls this an unfilled gap. WEPSS and RHETI show scores across all nine types, and Eclectic Energies warns about close scores. The Crossing (ours) reports an explicit Clear lead, Moderate lead, Close call, or Split band; we disclose that this is our own design and that the instrument is still pre-pilot.

Sources

• Hook et al. (2021) — “The Enneagram: A systematic review of the literature and directions for future research.” Journal of Clinical Psychology, 77(4), 865–883. doi:10.1002/jclp.23097
• Daniels & Price — The Essential Enneagram — validation study (n=970) against blind expert interview and post-course reassessment; overall κ ≈ .53; published per-type examples (Type 1 ≈ 66%, Type 9 ≈ 68%).
• Newgent et al. (2004) — Peer-reviewed evaluation of the 144-item forced-choice RHETI: internal consistency α ≈ .56–.82; mixed construct validity; NEO correlations.
• Wagner, J. — WEPSS technical manual (Western Psychological Services); reviewed in Buros’s Fifteenth Mental Measurements Yearbook (2003).
• Sutton, Allinson & Williams (2013) — “Personality type and work-related outcomes.” European Management Journal, 31(3), 234–249.
• MyDailyEnneagram.com — The Crossing’s validation status & gate tracker and methodology.