How the Ikigain online IQ test works
This page documents the exact released form score and its limits. Published cognitive-ability research informs the design principles, but evidence from another instrument does not automatically validate these questions or create population norms for them.
Protected keys
Correct options and scoring logic stay outside the browser before completion.
Transparent score
The 0–100 result is percent correct, never an invented percentile or standardized IQ.
Low-stakes use
The online result is never a diagnosis, hiring screen or school-placement decision.
Form and evidence
Seven parts of the released interpretation
Construct and form design
The released English form contains 20 independently written items: five each for inductive, quantitative, verbal-relational and spatial-transformation reasoning. The four labels describe task families sampled by this form; they are not separate clinical subtests or population-normed ability scores.
Item and answer-key controls
Every released item has one keyed answer under the rule stated in its prompt. The form checks exact item and option identifiers, requires one response per item, and rejects duplicate, missing or foreign answers. Correct options, solutions and distractor notes stay outside the browser delivery bundle.
Administration and accessibility
The test presents one current question at a time, provides previous and next controls, preserves keyboard and screen-reader semantics, supplies text alternatives for visual stimuli and has no countdown. The result uses accuracy only; device or reading speed is not added to the score.
Guest, account and server lifecycle
A guest can start without email, resume an eligible attempt and receive the complete free result. The canonical score is calculated and stored on the server with exact instrument, item, design, scoring, interpretation and renderer versions. Signing in may claim an eligible guest result; marketing consent remains a separate choice.
Transparent scoring
Each keyed answer is worth one point. The raw total runs from 0 to 20. The visible form score is raw correct divided by 20 and multiplied by 100, so 16 correct becomes 80/100. Each domain is also reported as a literal count from 0 to 5. No curve, speed bonus, hidden weighting or percentile conversion is applied.
Why the result is not a standardized IQ
A percent-correct form score is not population standing. A defensible IQ-style conversion would require a suitable reference sample, reliability and dimensionality evidence, external validation, fairness analysis and uncertainty for this exact form and intended population. The report therefore does not invent an IQ number, percentile or low/average/high norm band.
Maintenance and future calibration
Completed results remain attached to the exact form and interpretation a participant received. Material item, scoring or wording changes require new versions. If a future representative study supports a standardized estimate, that conversion will have its own norm version, documented uncertainty and eligibility rules rather than rewriting earlier form-score reports.
Research foundation
The design is informed by broad cognitive-ability research and by the International Cognitive Ability Resource (ICAR) research program. Ikigain copies no ICAR item into this form. Current ICAR registration terms restrict publishing its items and scoring key; citing the project documents background principles but does not transfer its evidence or authorize copying its bank.
- Condon, D. M., & Revelle, W. (2014). Development and initial validation of a public-domain cognitive ability resource.
- The International Cognitive Ability Resource publishes its project aims, item families and supporting evidence.
- The project's registration terms are why this form uses independently written questions instead of republishing ICAR items.
Current status
The protected four-domain form, current-item delivery, guest/account server-attempt lifecycle, immutable result route, dashboard registration, accessible runner, strict complete-response scoring and transparent result are released in English. The product reports percent correct and domain counts only. A representative norm study, reliability analysis and external validation remain a separate future project before any standardized IQ or percentile can be considered.