How a raw answer becomes an IQ.
Every psychometric instrument compresses messy human performance into a single number. Form 7C does it in four transparent steps — the same pipeline used, in far more elaborate form, by supervised clinical instruments. Below is each step written out in full, with worked examples, followed by the questions we get asked most.
Raw responses
Fifteen items span five faculties — pattern recognition, logical inference, numeric reasoning, verbal analogy and spatial rotation. Each item has exactly one correct option. A correct answer earns the item's points in full; a wrong answer or an expired timer earns zero. There is no penalty for guessing, so an uncertain answer is always better than no answer.
Example: you answer twelve items, get nine right, and let one item time out. The timed-out item is scored exactly like a wrong answer — zero points, no deduction — and your remaining nine correct answers carry forward to the next step at their own weights.
Difficulty weighting
Not every item counts the same. Each carries a weight d ∈ {1, 2, 3}, calibrated from pilot pass-rates: items most people solve are worth 1, mid-range items 2, and items only a minority solve 3. Your raw total is the sum of the weights of the items you answered correctly, out of the sum of all fifteen weights.
Example: solving five easy items (5 × 1 = 5 points) moves your estimate less than solving two of the hardest ones (2 × 3 = 6 points). This is why two people with the same number correct can finish with different scores — the mix matters more than the count.
Tempo coefficient
Your median response latency across the fifteen items applies a small adjustment, scaled by accuracy and capped at ±4 points. Answering quickly while answering correctly nudges the estimate up slightly; answering quickly while answering wrongly buys nothing at all, because the adjustment is multiplied by your accuracy.
Example: two candidates both earn 24 weighted points. One took a median of 9 seconds per item, the other 26 seconds. The faster candidate finishes a couple of points higher — never more than four — because sustained speed at the same accuracy is weak but real evidence of fluency.
Norm mapping
The weighted, tempo-adjusted total is mapped onto a normal distribution with μ = 100 and σ = 15 — the same scale clinical instruments report on. That mapping yields two numbers: your IQ estimate, and your percentile, which is the share of the norm group your score sits above.
Example: a score of 115 sits one standard deviation above the mean, which places you at roughly the 84th percentile — higher than about 84 of every 100 people in the norm group. A score of 130 sits two deviations up, at roughly the 98th percentile.
Frequently asked
It is a real screening test, but not a clinical one. Form 7C uses the same scoring logic as supervised instruments — difficulty weighting, a tempo adjustment and mapping onto a μ = 100, σ = 15 curve — with fifteen items instead of the hundreds an administered battery uses. Treat the number as an informed estimate for entertainment and self-insight. A defensible IQ figure can only come from a proctored instrument administered by a qualified psychologist.
Pattern recognition (spotting the rule inside a visual or symbolic sequence), logical inference (drawing what must follow from given premises), numeric reasoning (relationships between quantities and series), verbal analogy (mapping a relationship between words onto another pair) and spatial rotation (holding and turning a shape mentally). Each faculty contributes items at mixed difficulty, and your result panel shows how many points you earned in each.
Your IQ estimate is converted to a z-score — z = (IQ − 100) / 15 — and the percentile is the area under the standard normal curve to the left of that z. So IQ 100 is the 50th percentile, IQ 115 is about the 84th, and IQ 130 is about the 98th. The shaded region on your results chart is exactly that area: the share of the norm group scoring below you.
We take the median seconds you spent per item, compare it against the 30-second allowance, and turn the difference into an adjustment of at most ±4 points. That adjustment is then multiplied by your accuracy, so it only rewards speed that came with correct answers. Rushing through items you got wrong leaves the coefficient at effectively zero, and a slow, accurate run loses at most a few points.