MindsMetric exists to make credible psychometric insight accessible to anyone, anywhere. Self-understanding shouldn't require a referral, a waiting list, or a large budget — but it should always rest on legitimate science.
◈ Science first. We don't invent constructs. Every assessment maps to established, peer-reviewed psychometric frameworks with published reliability and validity data. Our role is to make those frameworks more accessible — not to replace them.
◇ AI as an instrument, not an authority. Artificial intelligence generates assessment items aligned to each framework's constructs, interprets response patterns against empirically defined scoring models, and writes personalized reports. The AI operates within the boundaries of each framework — it doesn't freelance.
◆ Transparency over mystique. We tell you exactly what we measure, how we measure it, and what the results do and don't mean. MindsMetric is a self-discovery tool. It is not a clinical diagnosis, and we say so clearly.
MindsMetric's two core assessments — IQ and Personality — are each anchored to established, peer-reviewed frameworks from cognitive science and personality psychology. Career pathways are generated by AI from your combined profile.
Grounded in the Cattell–Horn–Carroll (CHC) theory of cognitive abilities. The IQ section uses Computer Adaptive Testing (CAT) powered by a 3-Parameter Logistic (3PL) Item Response Theory model. Each item has calibrated difficulty, discrimination, and guessing parameters. The engine selects the most informative item based on your current ability estimate (theta), updates the estimate after each response, and ensures balanced coverage across all five cognitive domains: Fluid Reasoning, Logical Reasoning, Verbal Reasoning, Working Memory, and Processing Speed. The test administers exactly 60 items within a 30-minute time limit. Your IQ score is derived from the final theta estimate, converted to the standard scale (mean 100, SD 15) with a 95% confidence interval.
Built on the Big Five / OCEAN model — the most robustly validated framework in personality psychology. Measures Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism across both broad trait dimensions and 30 narrower facets (6 per trait). Each user receives 80 items sampled from a 150-item bank, with randomised facet coverage ensuring every facet is represented. Items use a 5-point Likert scale with a mix of forward and reverse-scored statements. Scores are normalised to 0–100 per trait.
Career recommendations are generated by AI (Claude by Anthropic) using your combined IQ domain scores and Big Five personality profile. The AI identifies 6 career pathways where your specific cognitive and personality strengths align, grounded in your actual numbers rather than generic advice. Each career is categorised as Strong Fit, Good Fit, or Growth Opportunity. Only careers with strong long-term prospects are recommended.
AI is involved at every stage of a MindsMetric assessment — but always operating within the guardrails of the underlying scientific framework.
The AI generates assessment questions mapped to specific constructs within each framework — for example, a fluid reasoning item for IQ or a self-regulation scenario. Items measure what the framework says should be measured, not arbitrary traits.
As you respond, the AI adjusts item difficulty and focus — similar to computerized adaptive testing (CAT) in standardized assessments. This aims to produce more reliable measurement in fewer questions.
The AI scores and interprets your response pattern against the framework's empirically-defined constructs. Your Big Five profile, for example, is computed from how your answers load onto each trait dimension and its subfacets.
Rather than returning a generic template, the AI writes an individualized narrative — what your scores mean, how they relate to each other, and practical implications for self-understanding. Every report is written fresh.
◇ A note on AI limitations. Large language models are powerful tools for natural language understanding and generation, but they are not infallible. AI-generated assessment items may not carry the same psychometric properties (reliability, validity, item discrimination) as items developed through traditional test construction with large norming samples. We are transparent about this because honesty serves you better than false precision.
Credibility requires clarity about scope. Here's where we draw the line.
A self-discovery platform that uses established psychometric frameworks and AI to give you structured, science-informed insight into your cognitive ability (IQ) and personality (Big Five), with AI-generated career pathway recommendations.
A clinical diagnostic tool. MindsMetric results are not medical diagnoses, should not replace professional evaluation, and are not administered or supervised by licensed clinicians. If you need clinical assessment, we encourage you to consult a qualified professional.
Every framework we use is published, peer-reviewed, and open to scrutiny. We name them explicitly so you can verify. We don't hide behind proprietary "secret algorithms" — we want you to understand exactly how your results are derived.
We are actively working toward independent validation studies, clinical advisory partnerships, and norming our AI-generated items against established benchmarks. This page will be updated as those milestones are reached.
MindsMetric was built by Tech-Domain — a product studio based in Africa, with a deep interest in human development, psychometrics, and the biological and psychological factors behind how people think, feel, and grow. Tech-Domain is not a clinical practice, and MindsMetric does not claim clinical authority. What it does claim is a sincere commitment to making credible psychometric insight accessible to people who otherwise wouldn't have it — built on real science, delivered through thoughtful technology.
MindsMetric is actively collecting validation data. All 140 assessment items — 60 IQ questions (adaptively selected by CAT from a 75-item pool) and 80 personality questions (sampled from a 150-item bank covering 30 facets) — are being offered to research participants free of charge. The anonymised data will be analysed by our psychology advisory board to compute reliability statistics, item quality metrics, and scoring calibration. We will publish the results on this page when the validation study is complete.