The IQ Bell Curve: How the Normal Distribution Works

What is the IQ bell curve?

The IQ bell curve is a graph showing how IQ scores (intelligence quotient — a score comparing your reasoning ability to other people your age) are spread across a population. Picture a smooth hill: tall in the middle, sloping gently down on both sides, symmetrical left to right. Most people cluster near the middle, and fewer and fewer people appear as you move toward either end.

This shape has a proper name — the normal distribution, sometimes called a Gaussian curve after the mathematician who described it. It shows up all over nature and human traits: height, blood pressure, reaction times, and yes, IQ scores. It's not a coincidence and it's not something test designers force onto the data by hand — it's the natural shape you get when a trait is influenced by many small, mostly independent factors (genes, environment, education, and plain luck all nudging a score up or down).

Why does IQ follow a normal distribution?

IQ tests are actually built to produce this curve, through a process called norming. When a test is developed, it's given to a large, carefully chosen sample of people. Their raw scores (the number of items answered correctly) almost always come out roughly bell-shaped already, because reasoning ability — like most complex human traits — results from the combined effect of countless small variables rather than one single switch.

Test designers then convert those raw scores into a standardised scale where the average is fixed at 100. That conversion process is explained in detail in how IQ is calculated, but the short version is this: your raw score gets compared against the norming sample, and that comparison is what turns "23 out of 30 correct" into a number like 112.

Mean, median, and mode: why they're all 100

In a perfectly normal distribution, three different measures of "the middle" all land in the same spot:

• Mean — the mathematical average of every score in the population.
• Median — the middle score if you lined everyone up from lowest to highest.
• Mode — the single most common score.

On the IQ bell curve, all three equal 100. That's not a natural law — it's a deliberate design choice. Test publishers set the scale so that the average performance in the norming sample equals exactly 100, then everyone else's score is measured relative to that anchor. For more on why 100 was chosen and how it holds steady across different tests, see what the average IQ actually means.

What is standard deviation, and why does it matter for IQ?

Standard deviation (SD) is a measure of how spread out scores are around the average. A small SD means most people score close to the mean; a large SD means scores are more scattered. For IQ, the standard deviation is set at 15 points on most modern tests (some older or specialised tests use 16 or 24, which is worth checking if you're comparing scores across different tools).

This number does the real work of the bell curve. It tells you exactly how rare or common a given score is, because the normal distribution follows a predictable rule:

Range from the mean	IQ range (mean 100, SD 15)	Share of population
±1 SD	85–115	~68%
±2 SD	70–130	~95%
±3 SD	55–145	~99.7%

This is often called the "68-95-99.7 rule." It means roughly two-thirds of everyone falls in a fairly narrow 30-point band (85 to 115), and scores beyond 130 or below 70 are genuinely uncommon — not because of some arbitrary cutoff, but because that's simply how far out on the curve's tail they sit. For a fuller breakdown of what each band is typically labelled, the IQ scale explained walks through every range from 55 to 145 and beyond, and the IQ levels chart gives plain-language descriptions for each one.

How the bell curve connects to percentiles

A percentile tells you what share of people a given score outranks — for example, being in the 90th percentile means you scored higher than 90% of the comparison group. Because IQ follows a normal distribution, every IQ score maps onto an exact percentile, and the two move together in a very non-linear way.

Here's the part that surprises people: moving from 100 to 115 (one standard deviation) jumps you from the 50th to roughly the 84th percentile. But moving from 130 to 145 (also one standard deviation) only takes you from about the 98th to the 99.9th percentile. Equal point gaps near the edges of the curve represent far smaller percentile jumps than equal point gaps near the middle, simply because there are fewer people out there to overtake. The full mechanics, plus a complete chart, are covered in IQ percentile: what your ranking really means.

Common misconceptions about the bell curve

"A score of 150 means you're 50% smarter than average." No — IQ is not a ratio scale like weight or distance, where 150 is literally 1.5x of 100. It's a rank on a curve. A 150 tells you how rare that score is relative to the population, not a multiplier on ability.

"The curve is a fixed, universal law of intelligence." The curve's shape is a real statistical pattern, but the number 100 and the SD of 15 are conventions tied to a specific norming sample and a specific test edition. Norms get updated periodically (a phenomenon related to the Flynn effect, where average raw performance shifts over generations), so a "100" today isn't measured against the exact same yardstick as a "100" from decades ago.

"Everyone's score sits exactly where the curve predicts." The bell curve describes the population, not any one test session. Your actual score can shift a few points between sittings due to fatigue, practice, or mood — which is one reason IQ is best treated as an estimate of reasoning ability, not a fixed label. Related traits like fluid vs crystallized intelligence can also shift the profile behind a single overall number.

Why this matters for reading your own score

Understanding the bell curve changes how you should read any IQ result, including your own. A score isn't a verdict — it's a position on a well-mapped curve, and that position comes with a percentile, a typical range of measurement error, and a context (which test, which norming sample, which age group).

If you're curious what a "good" score even means in practice, what is a good IQ score unpacks that question directly, using the same distribution logic covered here.

Key takeaways

• The IQ bell curve is a normal distribution: symmetrical, with most scores clustered near the mean.
• Mean, median, and mode all equal 100 by design, not coincidence.
• Standard deviation (typically 15 points) determines exactly how common or rare any score is.
• About 68% of people score 85–115, about 95% score 70–130, and about 99.7% score 55–145.
• Equal point differences represent very different percentile jumps depending on where they fall on the curve.
• IQ scores shift slightly between test sittings, so treat any single number as an estimate, not a fixed label.

None of this replaces professional psychological assessment, and iqmetria's test — like the bell curve concept itself — is meant as an orientative tool for self-knowledge, not a clinical or medical diagnosis. If you have real concerns about cognitive functioning, a qualified professional is the right person to talk to.