IQ Score Chart: What Every Score Range Really Means

IQ scores mean very little without context. A score of 115 sounds high — but is it? A score of 95 sounds mediocre — but where does it actually sit in the distribution used to norm the test? The IQ score chart is meaningful only when you understand the scale behind it: on the commonly used model, the mean is 100 and the standard deviation is 15, with approximately half the population in the conventional 90–109 average band. This article gives you the complete picture — score ranges, common descriptive labels, model-based percentiles, population frequencies, and an honest description of what each score can and cannot mean, building on the fundamentals of what an IQ score actually is.

Want to see which band you fall into? The free DesperateMinds IQ test places your result against population norms across several cognitive domains, in a single short session.

The Complete IQ Score Chart

The table below uses common descriptive labels for orientation. These terms are not a universal taxonomy: test publishers may use different wording and cutoffs, and Pearson describes qualitative score labels as communication aids rather than evidence-based natural categories (Pearson, 2024). The percentile values are theoretical estimates from a normal distribution with a mean of 100 and a standard deviation of 15; a clinical report should use the norm tables for the specific test.

To understand how these scores are actually calculated from raw test performance — including how standard deviations produce these exact percentile bands — the how IQ tests are scored article explains the full norming methodology.

IQ Percentile, IQ Levels and the IQ Scale Explained

Direct answer: An IQ percentile shows where a score falls relative to the test’s norm group. IQ levels are broad descriptive bands used to summarise score ranges, while the IQ scale is the standardised scoring system that converts test performance into comparable scores. On the most commonly used modern scale, the mean is 100 and the standard deviation is 15.

What is an IQ percentile?

An IQ percentile is the percentage of the norm group that scored below a particular result. For example, a score at the 84th percentile is higher than the scores of about 84% of the people in that norm group. A percentile is not the percentage of questions answered correctly, and it is not the same as an IQ score.

How does the IQ scale work?

On the common mean-100, SD-15 IQ scale, each 15-point step represents one standard deviation. IQ 115 is one standard deviation above the mean, IQ 85 is one below it, IQ 130 is two above, and IQ 70 is two below. This structure is what allows an IQ score to be translated into a percentile.

Not every intelligence test uses the same numerical scale. Mensa therefore bases eligibility on the 98th percentile from an approved, properly supervised test rather than on one universal IQ number. When comparing results from different instruments, the test name, norm group and percentile are more informative than the printed score alone.

What are IQ levels?

“IQ levels” usually means broad score bands described with terms such as average, high average, very high, low average or very low. These labels can make a report easier to read, but they vary between tests and publishers. Pearson states that qualitative descriptors are suggested communication tools rather than evidence-based natural categories, so the terms should not be treated as fixed identities or diagnoses.

A responsible interpretation includes the test name, IQ score, percentile, confidence interval and pattern of domain scores. At the lower end of the scale, an IQ score alone cannot diagnose intellectual disability: clinical criteria also require significant limitations in adaptive behaviour and onset during the developmental period.

Each Range Explained in Plain Language

145+ — Exceptionally Gifted (Top 0.1%)

Fewer than 1 in 1,000 adults score in this range on a properly normed test. This is the territory of prodigious mathematical ability, rapid acquisition of complex systems, and the kind of pattern recognition that feels almost involuntary to the person experiencing it. People in this range often describe feeling cognitively isolated — thinking in ways that are difficult to share with most people around them. Notable figures estimated or tested in this range include many prominent mathematicians, physicists, and chess grandmasters.

130–144 — Gifted / Very Superior (Top 2.3%)

This band includes scores around the upper-two-percent region used by Mensa, but qualification depends on the cutoff for an approved, properly supervised test rather than on one universal IQ number. Under the mean-100, SD-15 model, approximately 2.3% score 130 or higher. People with scores in this range may learn some complex material quickly, but education, motivation, opportunity, personality and domain knowledge remain important to academic and occupational outcomes.

A telling historical footnote: Francis Galton, who first proposed measuring human intelligence in the 1880s, estimated his own IQ posthumously at around 200 — far above this range. His work, while foundational, was later substantially revised by researchers who recognised that the relationship between test scores and real-world intellectual output is far more complex than Galton imagined. Even at the gifted level, motivation, persistence, and domain-specific knowledge matter enormously.

120–129 — Superior (Top 6–9%)

A score in this range places you solidly in the top tenth of the adult population. This is the range most commonly associated with strong academic performance, professional success in demanding fields, and the ability to learn complex skills efficiently. The difference between this range and the gifted ranges above is primarily one of speed and automaticity rather than fundamental capacity. People here can do most things the gifted range can — it may simply require more deliberate effort. For a detailed breakdown of what this specific score range means in practice, the IQ 120 explained article covers it thoroughly.

110–119 — High Average (Top 10–25%)

This is a comfortable above-average range representing about one in six adults. People scoring here handle complex information competently, learn new skills reliably, and perform well when they apply sustained effort. This range is probably the most "invisible" in terms of cognitive identity — above average enough to be capable, not exceptional enough to feel distinctly different from peers. Research by Gottfredson (1997) shows that professionals in this range succeed across nearly all mainstream occupational demands.

90–109 — Average (Middle 50%)

This broad band contains half the adult population. The difference between 90 and 109 is substantial in real cognitive terms even though both fall under the "average" label. Someone scoring 109 outperforms 73% of adults. Someone scoring 91 outperforms 27%. Within this band, specific domain strengths can be highly meaningful for career and life outcomes even when the composite sits in the middle.

80–89 — Low Average (Bottom 10–25%)

Scores in this range indicate cognitive performance below the majority of adults but within the normal variation of the population. People in this range can manage most everyday life demands and many occupational roles, particularly those that are procedural or skill-based rather than analytically demanding. This range is not a clinical classification and does not imply any neurological condition.

70–79 — Borderline (Bottom 2–8%)

Scores in this range sit at the lower end of the normal distribution and approach the threshold where clinical assessment becomes relevant. A single test score here is not diagnostic — clinical assessment of intellectual disability requires comprehensive evaluation by a licensed psychologist across multiple domains over time. Test conditions, anxiety, and format familiarity significantly affect scores at this range.

Below 70 — Intellectual Disability Range

A score below 70 on a properly administered clinical instrument, combined with significant deficits in adaptive functioning, meets the criteria for intellectual disability diagnosis. An online screening result below 70 is not a diagnosis and should not be interpreted as one. If you or someone you know scores consistently low on cognitive assessments, a full clinical evaluation by a licensed psychologist is the appropriate next step.

Important Caveats About This Chart

IQ ranges and their labels vary between testing frameworks, publishers and historical periods. This chart uses common descriptive terms for orientation rather than presenting them as a universal classification system. Pearson released the WAIS-5 in 2024 with updated norms, replacing WAIS-IV as the current edition of that instrument; other tests, including the Stanford-Binet, may use different terminology and score reporting.

A single test score is always an estimate, not a precise measurement. Even clinical IQ tests carry a standard error of measurement of approximately 3–5 points. Online tests have higher error. Your "true" IQ score is best understood as a range rather than a single number — and anyone presenting an online result as definitive is overstating the precision of their instrument.

The score ranges above were normed on samples from specific populations and periods. The Flynn Effect — the documented rise of approximately 3.0 IQ points per decade observed across the 20th century — means that norms become outdated over time. A test normed in 1990 would produce inflated scores relative to current population ability. All major clinical tests are periodically renormed to correct for this.

IQ measures a specific cluster of cognitive abilities — primarily abstract reasoning, working memory, processing speed, and verbal comprehension. It does not capture creativity, emotional intelligence, practical wisdom, social skill, or the many other dimensions of human cognitive life. This is exactly why researchers distinguish between different types of intelligence that no single score captures. High IQ is a useful tool. It is not a complete picture of anyone's mind. For the limitations of IQ testing as an instrument, the IQ test accuracy article covers the research in detail.

Where Does the Average Person Actually Score?

The mean of 100 is not just a mathematical convention — it reflects the genuine centre of the adult population on these tasks at the time of norming. "Average" has become culturally loaded in a way that distorts how people interpret their scores. Being average on IQ does not mean being average at life, average at your career, or average as a human being. Some of the most consequential contributions to human culture and knowledge have come from people with thoroughly unremarkable test scores.

What IQ predicts most reliably is performance on tasks that resemble IQ tests — abstract reasoning under time pressure, novel problem solving, academic learning. Gottfredson (1997) identified a threshold around IQ 115 above which individuals can handle most professional-level cognitive demands without accommodations — but this threshold is far lower than most people assume, and the research shows diminishing returns beyond approximately 120 for most real-world outcomes.

Is IQ destiny? The honest answer is: partly. A meta-analysis by Schmidt & Hunter (1998) found that general cognitive ability predicts job performance at r = 0.51 — the strongest single predictor in the employment research literature. But that correlation still leaves nearly 75% of performance variance unexplained. Conscientiousness, motivation, domain knowledge, and interpersonal skill all contribute substantially to outcomes that IQ alone cannot predict.

What IQ Scores Actually Predict

The real-world consequences of IQ scores are more nuanced than either critics or enthusiasts typically acknowledge. Here is what the evidence actually shows.

Educational attainment correlates with IQ at r ≈ 0.55 across large samples (Deary et al., 2007) — strong, but far from deterministic. Job performance correlates at r = 0.51 for complex occupations, falling to r = 0.23 for low-complexity roles (Schmidt & Hunter, 1998). Income correlates with IQ at roughly r = 0.30–0.40, meaning cognitive ability explains somewhere between 9% and 16% of income variance — influential, but dominated by other factors. The IQ and income research article goes deeper on these correlations and what they mean practically.

Health outcomes also correlate with IQ — individuals scoring higher on cognitive ability tests live longer on average, a relationship that researchers attribute primarily to better health literacy and safer decision-making rather than any direct biological link (Gottfredson & Deary, 2004).

What IQ does not predict well: creativity, leadership effectiveness, moral reasoning, emotional competence, or life satisfaction. These dimensions operate largely independently of IQ scores, and in some studies show weak negative correlations with extremely high IQ — a counterintuitive finding that researchers attribute to social isolation and perfectionism at the extreme upper end of the distribution.

How These Scores Are Calculated

Every IQ score is a comparison to a reference population, not a raw count of correct answers. Administered to a large representative sample, the raw scores are transformed so that the distribution has a mean of exactly 100 and a standard deviation of 15. This transformation is called standardisation.

The 15-point standard deviation is the key to understanding the chart. One standard deviation above the mean (IQ 115) sits at roughly the 84th percentile. Two standard deviations above (IQ 130) sits at the 97.7th percentile — the Mensa threshold. Three standard deviations above (IQ 145) sits at the 99.87th percentile. The math is the same in both directions: IQ 70 sits at the 2.3rd percentile, exactly as far below average as IQ 130 sits above it.

One question comes up constantly among non-specialists: why do IQ scores sometimes vary substantially between different tests taken by the same person? The answer lies in what each test measures. Full-scale IQ batteries such as the WAIS-5 aggregate performance across several cognitive domains, each with its own subtest structure. A person who scores 135 on verbal comprehension but 105 on processing speed will receive a full-scale score somewhere in between, potentially masking both a genuine strength and a genuine weakness. The composite number is useful for research and broad comparison; for understanding an individual, the domain breakdown matters more — which is the case for reporting subscores alongside any composite.

For a complete explanation of norming methodology, composite score calculation, and what major cognitive subtests measure, the IQ test scoring methodology article covers the technical details without requiring a statistics background.

Frequently Asked Questions