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 pass mark you scraped or missed. It is a rank: a statement about where your performance sits relative to a reference group. Knowing how answers become that number makes any result more useful and less intimidating, and it shows why an online estimate and a clinical IQ are different things.

Why "quotient" no longer means what it used to

The word quotient is a historical leftover. Early tests literally divided one number by another. In Lewis Terman's 1916 Stanford–Binet, a child's mental age was divided by their chronological age and multiplied by 100, so a ten-year-old who reasoned like a typical twelve-year-old scored 120 (Terman, 1916). That works tolerably for children and breaks for adults, because reasoning ability does not keep climbing in yearly steps after maturity. A 40-year-old does not have twice the "mental age" of a 20-year-old.

David Wechsler replaced the ratio with a deviation IQ (Wechsler, 1939). Instead of dividing, the test compares your performance with a large sample of people in your own age band and reports how far above or below that group's average you landed. Every major clinical test works this way today. The label stuck; the arithmetic underneath it changed.

How a clinical battery builds the number

On an individually administered test such as the WAIS-5 (adults, published 2024) or the WISC-V (children), the final number is the end of an assembly line:

Raw score — how many points you earned on each subtest.
Scaled scores — each raw score is looked up in the norm table for your age band and converted to a common scale (mean 10, standard deviation 3), so a strong vocabulary result and a strong block-design result are expressed in the same currency.
Index scores — related subtests are combined into indexes on the familiar scale (mean 100, standard deviation 15). The WISC-V and WAIS-5 report five: Verbal Comprehension, Visual Spatial, Fluid Reasoning, Working Memory and Processing Speed.
Full Scale IQ — selected subtests are combined and converted, again through the age-band norms, into the headline number, reported with a confidence interval.

The point of all this is comparability: a 125 should mean the same whether you earned it through words or through patterns. An online estimate does not have most of these steps, as described below.

The bell curve, and why 15 matters

Deviation IQs are scaled so that the norm sample averages 100 with a standard deviation of 15. Because scores in a large unselected population are close to normally distributed, those two constants turn any score into a statement about rarity:

Score	Distance from mean	Percentile	Roughly 1 in…
70	−2 SD	2.3rd	44 score lower
85	−1 SD	16th	6 score lower
100	mean	50th	half score higher
115	+1 SD	84th	6 score higher
130	+2 SD	97.7th	44 score higher
145	+3 SD	99.87th	about 740 higher

About 68% of people fall between 85 and 115, and about 95% between 70 and 130.

A point is not a point

The relationship between score and rarity is not linear. Near the middle each point passes many people; in the tails each point passes few people but makes the score much rarer. Going from 100 to 115 moves you 34 percentile points; from 130 to 145, just over 2. In the other direction, 130 is 1 in 44 while 145 is about 1 in 740. So small differences near the average reorder a lot of people, and at the extremes a few points change rarity a great deal while the test itself becomes less precise, because few people in the norm sample score that high.

Why the goalposts move: re-norming and the Flynn effect

If 100 is always the average, "average" has to be re-measured as populations change, and they have. Through much of the 20th century, average raw performance on IQ tests rose in many countries, by roughly 3 points per decade in James Flynn's estimates (Flynn, 1987). A meta-analysis of data from 1909 to 2013 confirmed the gains and found them slowing in recent decades (Pietschnig & Voracek, 2015).

Because the mean is anchored at 100, publishers periodically re-norm: they test a fresh sample and reset the scale. The WAIS-5, published in 2024, replaced norms the WAIS-IV collected in 2007–2008. Without re-norming, scores on an old test drift upward and lose meaning.

What the number is, and isn't

A clinical IQ is an estimate of your standing on general reasoning ability, today, relative to people your age, with a margin of error that the report states as a confidence interval. It is not a fixed quantity stamped into you at birth, not a measure of worth, and not exact to the point.

How My Path's estimate is calculated

My Path's IQ test has a 12-item short form and a 60-item full form, both free, covering six areas: fluid, verbal, numerical and spatial reasoning, working memory and processing speed. It is not a clinical battery, and the arithmetic is simpler:

Each area gets a percentile: your accuracy compared with people who sat the same form on My Path. Until an area has enough sittings, a provisional table based on published norms is used instead, and the result page says which applies.
The overall figure combines the six areas with weights borrowed from published g-loadings. On the 60-item full form, that weighted combination of your accuracies is ranked against everyone who took the full form, so it is a true percentile rank. On the 12-item short form it is a weighted average of the six area percentiles, which is pulled toward the middle and is not strictly a percentile rank.
The range is a deliberately wide placeholder, ±12 percentile points on the short form and ±7 on the full form, not a calibrated standard error.
Estimated IQ ≈ N is that overall percentile converted to the familiar scale (mean 100, SD 15) through the normal curve and limited to 60–145.

The comparison group is everyone who sat the same form, not your age band: the estimate is not age-normed and comes from no representative norming study. Read it as your position among this site's test-takers, re-expressed on a familiar scale. For more on the conversions themselves, see the bell curve and percentile conversions on our sister site, My IQ Path, which runs the same test.

Take the free IQ test →

Once you know where your reasoning strengths sit, pair the result with the Big Five and EQ assessments to see how traits and emotional skills shape what you actually do with them.