IQ Standard Deviation: Why 15 Points Changes Everything

The IQ standard deviation (a measure of how spread out scores are around the average) is the number that gives any IQ score its actual meaning. On the most common scale, the average IQ (intelligence quotient, a score comparing your reasoning performance to other people your age) is set at 100, and the standard deviation is 15. That single figure — 15 — tells you how rare or common a score like 115, 130, or 70 really is. Change it to 16, and the exact same raw performance would be reported as a slightly different number. Neither is "more correct"; they're just different rulers.

That's the part beginners usually miss: IQ isn't measured in some fixed, universal unit like centimetres. It's a statistical ranking, and standard deviation is the unit of distance used to describe how far you sit from the centre of the pack.

What standard deviation actually means

Standard deviation is a statistics concept that shows how tightly or loosely a group of scores clusters around the average. Imagine measuring the height of every adult in a country. Most people land close to the average height, fewer are noticeably taller or shorter, and almost nobody is extremely tall or extremely short. The standard deviation is the number that describes the "typical" distance from that average.

Applied to IQ, this is exactly the shape you get. Most people score near 100. Fewer score near 85 or 115. Very few score near 70 or 130. This pattern is called the IQ bell curve, because when you plot it, it looks like a symmetric bell — tall in the middle, tapering off on both sides.

Standard deviation is what defines how wide or narrow that bell is. A small SD means scores bunch tightly around 100. A larger SD means they spread out further. On the standard modern scale:

• 1 SD = 15 points
• Roughly 68% of people score within 1 SD of the mean (between 85 and 115)
• Roughly 95% of people score within 2 SD of the mean (between 70 and 130)
• Roughly 99.7% of people score within 3 SD of the mean (between 55 and 145)

This is why a score of 115 isn't just "a bit above average" in a vague sense — it's precisely one standard deviation above the mean, which places you around the 84th percentile (meaning you scored higher than about 84% of people tested).

Why 15 became the standard

Modern IQ tests, including the Wechsler scales (WAIS for adults, WISC for children) and most reputable online tests, use a mean of 100 and a standard deviation of 15. This wasn't an arbitrary pick — it traces back to decisions made by test developers in the mid-20th century who wanted a scale that was easy to interpret and comparable across different age groups and test versions, replacing the older "mental age divided by chronological age" method described in the history of IQ testing.

Once SD 15 became dominant in the field's most widely used and studied tests, it stuck as the de facto convention. Most psychologists, researchers, and serious test publishers report scores this way, which is also why how IQ is calculated almost always references this specific scale.

SD 16: the other scale you'll run into

Not every test uses 15. Some scales, including versions historically associated with the Stanford-Binet test, use a standard deviation of 16 instead. This single-point difference sounds trivial, but it genuinely changes the number you'd see for the exact same underlying ability.

Here's the practical effect: the further a score sits from 100, the more the two scales diverge. Near the average, SD 15 and SD 16 report almost the same number. Out at the tails — very high or very low scores — the gap widens.

Percentile	Score on SD 15	Score on SD 16
50th (average)	100	100
84th	115	116
97.7th	130	132
99.9th	145	148

This is exactly why comparing raw IQ numbers across different tests without checking the scale is a classic mistake — it's a big part of why Mensa's qualifying scores look different depending on which approved test someone took, even though every accepted score represents the same top-2% cutoff. The percentile is the fair comparison point; the raw number alone is not.

Why points don't mean the same thing everywhere on the scale

A common misconception is treating IQ points like a consistent, portable unit — as if 10 points always means the same jump in ability no matter where you start. It doesn't, for two separate reasons.

First, the bell curve itself isn't flat. Moving from 100 to 110 crosses a densely populated part of the curve, so it shifts you past a fair number of people. Moving from 140 to 150 crosses an extremely sparse part of the curve, so even though it's the same 10-point jump, it represents a much bigger leap in rarity. A point near the extremes is statistically worth far more than a point near the middle.

Second, as shown above, the size of a "point" depends on which standard deviation the test uses. Ten points on an SD 15 scale and ten points on an SD 16 scale don't represent identical slices of the population.

How to read your own score correctly

When you get a result from any test — including a quick IQ test or a free IQ test — the number on its own tells you less than the number plus its scale. A trustworthy test will state its standard deviation and mean somewhere in its methodology, which is one of the signals covered in how to tell if an intelligence test is reliable.

Practical steps for interpreting any score you receive:

1. Check whether the test states its standard deviation (15 is the safe assumption if nothing is specified, but verify when you can).
2. Convert the raw score to a percentile in your head, since that's the figure that actually travels well between different tests.
3. Treat scores within a few points of each other as statistically similar, not meaningfully different — a 103 and a 108 aren't really distinguishable results.
4. Be cautious with scores far from 100 reported by casual or unverified tests, since small measurement errors get amplified at the tails.

A quick example

Say two different tests both report someone scoring "132." If Test A uses SD 15, that's roughly the 98th percentile — Mensa-qualifying territory. If Test B uses SD 16, that same 132 sits slightly lower in percentile terms, closer to the upper-90s but not quite the same rarity. The number looks identical; the meaning isn't. This is exactly the kind of nuance that trips people up when comparing results from different platforms, including when researching what counts as a good IQ score or checking where a result lands on the full IQ scale.

Key takeaways

• Standard deviation (SD) measures how spread out scores are from the average of 100.
• Most modern tests use SD 15; some, like certain Stanford-Binet versions, use SD 16.
• The same ability level can produce slightly different numbers depending on the scale used.
• A point near the extremes of the curve represents far more rarity than a point near the middle.
• Percentile is a more reliable way to compare scores across different tests than the raw number alone.

Any test you take, including iqmetria's, gives you an orientative estimate meant for self-knowledge and curiosity — not a clinical or diagnostic measurement. If you ever have real concerns about cognitive functioning, a licensed psychologist using a full validated battery is the right next step.