Methodology

Every Work Fit IQ result rests on a 21-trait scoring model. Each trait is a single bounded dimension of how someone works — not a personality stereotype, not a cluster of preferences. The list is intentionally narrower than the Big Five but broader than the four-letter typing most assessments converge on.

The 21 traits are split across four conceptual buckets:

• Energy axes — Social Energy, Autonomy Need, Emotional Labor Tolerance. How human contact, independence, and emotional load each cost or compound on your day.
• Cognitive style — Analytical Thinking, Creativity, Systems Thinking, Big-Picture Thinking, Detail Tolerance, Ambiguity Tolerance, Learning Drive. The default mode your brain operates in.
• Interpersonal — Leadership Drive, Communication Strength, Persuasion Strength. The output side of working with other people.
• Motivation and life-fit — Mission Orientation, Money Motivation, Ambition Drive, Stability Preference, Risk Tolerance, Structure Preference, Repetition Tolerance, Execution Strength. What pulls you toward work and what kind of life you're willing to trade for it.

Each trait is scored 0–100 from the diagnostic responses, then bucketed into four bands (low, mid, high, very high) with band-specific guidance. The bands are directional, not pathological — a low Social Energy score is a clear signal toward solo-focus roles, not a deficit to be fixed.

Every trait has its own page with a full write-up at /profile/traits.

Six well-established career-psychology frameworks shape the diagnostic. Each one is integrated as original prose drawn from the framework's public-domain concepts — no third-party text is reproduced.

The free diagnostic is 12 questions. Each question is calibrated to load onto two or more traits with known weights, so 12 answers produce signal on all 21 dimensions (with confidence intervals tighter on heavily-loaded traits and wider on lightly-loaded ones).

Scores are computed in three passes:

1. Raw load — each answer contributes a signed weight to one or more traits.
2. Normalize — against the max-possible score given how many questions loaded onto each trait, so under-sampled traits aren't penalized for thin data.
3. Band — bucket into low / mid / high / very high. Bands aren't equal- width; they're calibrated against the same-demographic cohort.

The full 100-question version of the diagnostic (in the paid unlock) tightens the confidence interval to roughly ±5 points per trait. The free 12-question version sits at roughly ±12 — informative for direction, not for comparing yourself to your sibling.

The Cognitive Aptitude is a separate instrument. It measures three reasoning domains — numerical, verbal, and spatial — over 100 items in the full version (12 in the free preview).

Every item is original. Work Fit IQ does not extract questions from copyrighted test banks. Items follow the standard psychometric format (number series, antonyms, mental rotation, etc.) but each specific item is hand- authored or generator-built to a published difficulty calibration.

Scoring produces three section bands plus a composite. The composite is mapped to an approximate IQ-band equivalent (with a wide error margin clearly disclosed in the report). Per-question cohort miss-rates are computed live from anonymized response data; questions with fewer than five attempts use a deterministic estimate until enough real attempts accumulate.

Work Fit IQ surfaces 201 careers across 10 clusters — technology, business, creative, healthcare, finance, operations, education, science, legal, entrepreneurship.

30 entries are fully hand-authored at the catalog page level. The remaining 171 are built by a seed-based generator that fills cluster-typical day-to-day breakdowns from a per-cluster template — but every seed carries hand-written upside and downside copy, so each page reads editorial rather than templated.

Salary bands (entry / median / experienced / top 10%) are calibrated to publicly available 2024–2026 US labor-market signals — Bureau of Labor Statistics Occupational Outlook Handbook, O*NET, and Levels.fyi spot checks. They are indicative, not contractual.

Fit-score derivation for each role uses the user's trait signature against a per-role weight vector. Match confidence is computed independently and shown alongside the headline percentage so a user can see whether a 78% match comes with high or low certainty.

Three honest limits worth knowing.

• Not a clinical instrument. The diagnostic doesn't screen for mental-health conditions, learning differences, or neurodivergence. If you suspect any of those would change your work answers, please work with a qualified clinician.
• Not a guarantee of employment. A 92% fit score with Product Manager doesn't mean you'll get hired as one. Labor markets, credentialing, network effects, and luck all matter and we don't model any of them.
• Not a substitute for self-knowledge. The diagnostic is best read as a structured second opinion. If your result disagrees with what you already know about yourself, weigh both — don't blindly defer to the score.

Work Fit IQ is built and operated by Work Fit IQ. The taxonomy, calibration, and item authoring are in-house. The tooling stack is Next.js + Prisma + Postgres + Anthropic Claude (for the optional resume-review interpretation layer). Privacy: we don't sell data; the only third-party services that touch user data are Clerk (auth), Stripe and PayPal (payments), Resend (email receipts), AWS S3 (resume file storage), and Sentry (error monitoring with PII scrubbed).

Reach the team at hello@workfitiq.com with corrections, citation requests, or methodology questions.