How IQ Scores Are Calculated
An IQ score looks like a grade, and that is exactly what trips most people up. It is not a percentage of questions answered correctly, and 100 is not a passing mark you scraped or missed. It is a rank — a statement about where your performance sits relative to everyone else your age. Understanding the machinery that turns your answers into that number makes the result far more useful, and far less intimidating.
Why "quotient" no longer means what it used to
The word quotient is a historical artifact. Early 20th-century tests literally divided one number by another: mental age over chronological age, times 100. A ten-year-old reasoning like an average twelve-year-old scored 120. Tidy — and completely broken for adults, because cognitive ability does not keep climbing in neat yearly steps after maturity. A 40-year-old does not have twice the "mental age" of a 20-year-old.
Modern tests abandoned the ratio entirely. What you get today is a deviation IQ: a score built not from a division but from your position in a distribution. The test compares your performance against a large, representative sample of people in your own age band, and reports how far above or below that group's average you landed. The label stuck; the math underneath it is different.
The four-step pipeline
Your final number is the end of an assembly line, not a single calculation:
- Raw score — how many items you got right. On its own this means little, because a hard item and an easy item both count as "1."
- Scaled scores — raw scores are converted, subtest by subtest, onto a common scale that accounts for item difficulty. Now a strong verbal result and a strong spatial result are expressed in the same currency.
- Composite/index scores — related subtests are grouped into indices such as verbal comprehension, perceptual reasoning, working memory, and processing speed.
- Full Scale IQ — the composites are combined and mapped against the age-group norming table to produce the headline number.
Each step strips out a source of noise. The point of all this transformation is comparability: a 125 should mean the same thing whether you earned it through words or through patterns, and whether you took version A or version B of the test.
The bell curve, and why 15 matters
Plot the scores of a large unselected population and you get the normal distribution — the bell curve. Two numbers define it:
- The mean is fixed at 100. This is the peak, the most common result, the definition of "typical."
- The standard deviation is 15. This sets how quickly the curve thins out as you move away from the center.
Those two constants are what let us convert any score into a statement about rarity:
| Score | Distance from mean | Approx. percentile | Roughly 1 in… |
|---|---|---|---|
| 85 | −1 SD | 16th | 6 below it |
| 100 | mean | 50th | the median |
| 115 | +1 SD | 84th | 6 |
| 130 | +2 SD | 98th | 50 |
| 145 | +3 SD | 99.9th | 1,000 |
About 68% of people fall between 85 and 115 — the entire "average" band is one standard deviation either side of the mean.
A point is not a point
The table hides something important: the relationship between score and rarity is not linear. Near the middle, points are cheap — the curve is crowded, so a few points barely move your rank. Out in the tails, points are expensive. Moving from 115 to 125 takes you from the top ~16% to the top ~5%; the same ten-point gap near the center would shuffle you past far fewer people. This is why chasing "just a few more points" near the average is meaningless, while small differences at the extremes describe genuinely different rarities.
Why the goalposts move: re-norming and the Flynn effect
If 100 is always the average, then "average" has to be recalculated as populations change — and they have changed. Across the 20th century, raw performance on IQ tests rose roughly three points per decade, a trend named the Flynn effect after James Flynn, who documented it. Better nutrition, more schooling, and a world saturated with abstract symbols all plausibly contribute.
Because the mean is anchored at 100 by definition, test publishers must periodically re-norm — re-administer to a fresh sample and reset the scale. Without it, scores would inflate and lose meaning. Re-norming is why your 100 is measured against your contemporaries, not your grandparents.
What the number is — and isn't
Properly understood, your IQ score is a confidence-bounded estimate of your standing on general cognitive ability, today, relative to your peers. It is not a fixed quantity stamped into you at birth, not a measure of worth, and not a precise reading — every score carries a margin of error, which is why a responsible report gives a range, not false precision.
See your own breakdown
My Path's IQ assessment runs verbal, numerical, and abstract/spatial reasoning across 40 items, then returns an estimated range with a confidence interval, a percentile, and a subtest profile showing where you are stronger or weaker — the index-level view, not just a single digit.
Once you know where your cognitive strengths sit, pair the result with the Big Five and EQ assessments to see how trait and emotional factors shape what you actually do with that horsepower.