How is the average IQ score determined?

The average IQ score is **defined as 100** through the norming process. Test publishers administer the test to a large, representative sample (typically 2,000-8,000 people) and use statistical methods to set the mean at 100 and the standard deviation at 15. This is recalibrated every 15-20 years to account for the Flynn effect -- the documented rise of approximately 3 IQ points per decade. So the "average" is not discovered; it is ***constructed*** by design, ensuring that 100 always represents the population midpoint for any given norming period.

What does a standard deviation of 15 mean in IQ testing?

A standard deviation of 15 means that the "natural spread" of IQ scores is measured in 15-point units. **68% of the population** falls within one standard deviation of the mean (IQ 85-115), **95%** falls within two standard deviations (IQ 70-130), and **99.7%** falls within three (IQ 55-145). Practically speaking, each 15-point jump represents a significant cognitive difference: moving from 100 to 115 takes you from the 50th to the 84th percentile, meaning you now outperform 84% of the population instead of 50%. The Wechsler and modern Stanford-Binet tests both use SD = 15, but older tests like the Cattell use SD = 24, which is why you must always check which scale is being used before comparing scores.

Can online IQ tests provide accurate scores?

Most online IQ tests **lack proper norming and standardization**, meaning their scores are unreliable. A scientifically valid IQ test requires: (1) a large, representative norming sample; (2) standardized administration conditions; (3) demonstrated test-retest reliability above 0.90; and (4) validated correlation with established tests. Many free online quizzes meet none of these criteria and may inflate scores to encourage sharing. However, well-designed online assessments that follow psychometric principles can provide **reasonable estimates** -- not clinical-grade precision, but useful approximations. Our [full IQ test](/en/full-iq-test) and [practice test](/en/practice-iq-test) are developed with sound psychometric methods. For a definitive clinical score, seek testing from a licensed psychologist using instruments like the WAIS-IV or Stanford-Binet 5.

Why do IQ scores sometimes change over time?

IQ scores can change for several legitimate reasons: (1) **Education** -- each additional year of schooling is associated with 1-5 IQ point gains (Ritchie & Tucker-Drob, 2018); (2) **Health changes** -- illness, medication, head injury, or recovery can shift scores; (3) **Test conditions** -- fatigue, anxiety, motivation, and environmental distractions affect performance; (4) **Measurement error** -- even highly reliable tests (r = 0.96) have a confidence interval of about 5-8 points; (5) **Genuine cognitive development or decline** -- the Ramsden et al. (2011) study found IQ shifts of up to 20 points in teenagers between ages 14 and 18, accompanied by measurable brain structure changes. However, *relative rankings* tend to be more stable -- a person who scores in the top 20% at age 15 is likely to score in the top 20-30% at age 45.

What is the Flynn effect in IQ testing?

The Flynn effect, named after researcher **James Flynn**, is the well-documented phenomenon of rising average IQ scores over generations -- approximately **3 points per decade** throughout the 20th century. This has been observed in over 30 countries. The causes are believed to include improved nutrition, better healthcare, more cognitively stimulating environments, increased access to education, and reduced exposure to environmental toxins like lead. The Flynn effect has major implications for IQ scoring: tests must be **re-normed regularly** because scores on outdated norms will be inflated. Notably, recent research by Bratsberg and Rogeberg (2018) using Norwegian military data found evidence that the Flynn effect has **reversed** in some developed countries since the 1990s, with average scores declining by about 0.2 points per year -- a finding that remains actively debated.

Are IQ scores affected by age?

IQ tests are ***age-normed***, meaning your score reflects performance compared to others in your same age group, not the general population as a whole. This is critical because different cognitive abilities follow different age trajectories: **fluid intelligence** (abstract reasoning, processing speed) typically peaks in the mid-20s and declines gradually, while **crystallized intelligence** (vocabulary, general knowledge) continues to increase into the 60s or 70s. Because of age-norming, a 70-year-old who scores 100 has performed at the average level *for 70-year-olds*, even though their raw performance on some subtests might be lower than a 25-year-old who also scored 100. This design ensures that IQ scores remain meaningful and comparable across the lifespan.

How Are IQ Tests Scored? (Standard Deviation Made Simple)

Quick Answer: IQ test scoring converts your raw number of correct answers into a standardized score where 100 is the average and 15 points equals one standard deviation. This means about 68% of people score between 85 and 115, about 95% score between 70 and 130, and a score of 130 puts you in the top 2% of the population. The scoring system is based on comparing your performance to a large, carefully selected reference group of the same age.

Why Understanding IQ Scoring Matters

IQ scores are among the most widely used metrics in psychology, education, and workplace assessment -- yet most people who receive a score have little idea how it was actually calculated. The number on your report is not simply a count of correct answers. It is the product of a sophisticated statistical process designed to make your performance comparable to millions of other test-takers across different ages, time periods, and test versions.

"An IQ score is not a measure of worth or potential. It is a statistical summary of how someone performed on a specific set of cognitive tasks, compared to a reference population."
-- Alan Kaufman, clinical psychologist and co-author of the Kaufman Assessment Battery for Children

Understanding this process matters for several reasons:

It helps you interpret your own scores accurately rather than over- or under-estimating what they mean
It reveals why two different tests might give you slightly different scores
It explains why IQ scores are not fixed -- and how the scoring system itself changes over time
It protects you from misleading claims by unvalidated online quizzes

This article walks through the entire scoring process, from raw answers to final IQ number, in plain language.

Step 1: From Questions to Raw Scores

Every IQ test begins the same way: you answer a series of questions designed to assess different cognitive abilities. Each correct response earns points, and these points are tallied into a raw score.

What IQ Tests Actually Measure

Modern IQ tests are not a single exam. They are batteries of subtests, each targeting a specific cognitive skill:

Subtest Category	Example Tasks	What It Measures
Verbal Comprehension	Vocabulary definitions, similarities between concepts	Language knowledge, abstract verbal reasoning
Perceptual Reasoning	Matrix puzzles, block design, visual patterns	Nonverbal reasoning, spatial ability
Working Memory	Repeating number sequences, mental arithmetic	Short-term memory, attention, mental manipulation
Processing Speed	Symbol matching, rapid coding tasks	Speed of simple cognitive operations

The two most widely used professional IQ tests in the world are:

Test	Publisher	Age Range	Standard Deviation	Current Version
Wechsler Adult Intelligence Scale (WAIS)	Pearson	16-90 years	15	WAIS-IV (2008)
Stanford-Binet Intelligence Scales	Riverside	2-85+ years	15 (previously 16)	SB5 (2003)
Kaufman Assessment Battery (KABC)	Pearson	3-18 years	15	KABC-II (2004)
Raven's Progressive Matrices	Pearson	5-65 years	15	RPM (various editions)

Important: A raw score alone tells you almost nothing. Answering 45 out of 60 questions correctly on one test version might represent average performance, while 45 out of 60 on a harder version might represent exceptional ability. The raw score must be transformed before it has meaning.

Step 2: The Norming Process -- Building the Reference Group

The most critical step in IQ scoring is norming -- the process of establishing what "average" actually means for a given population.

How Norming Works

Test publishers administer the test to a large standardization sample -- typically between 2,000 and 8,000 people -- carefully selected to represent the broader population. This sample must mirror the population in:

Age distribution
Gender balance
Racial and ethnic composition
Geographic region
Education level
Socioeconomic status

"The quality of an IQ test is only as good as the quality of its norms. Without a representative standardization sample, the scores are meaningless numbers."
-- David Wechsler, creator of the Wechsler Intelligence Scales

The performance of this norming group becomes the reference standard against which all future test-takers are measured.

Norming Sample Sizes for Major IQ Tests

Test	Standardization Sample Size	Countries/Regions
WAIS-IV	2,200 adults	United States
WISC-V	2,200 children	United States
Stanford-Binet 5	4,800 individuals	United States
Raven's SPM Plus	6,000+	Multiple countries

Step 3: Converting Raw Scores to Standardized Scores

Once norms are established, individual raw scores are converted into standardized scores using a mathematical process.

The Z-Score Formula

The first step is calculating a z-score, which expresses how many standard deviations a raw score falls from the mean:

z = (Individual Raw Score - Mean Raw Score) / Standard Deviation of Raw Scores

For example, if the mean raw score on a subtest is 30 with a standard deviation of 5, and you scored 40:

z = (40 - 30) / 5 = +2.0

This z-score of +2.0 means your performance was two standard deviations above the mean.

Converting Z-Scores to IQ Scores

The z-score is then mapped onto the IQ scale using this formula:

IQ Score = 100 + (z x 15)

Using the example above:

IQ = 100 + (2.0 x 15) = 130

Your z-Score	Meaning	Resulting IQ Score
-3.0	3 SDs below mean	55
-2.0	2 SDs below mean	70
-1.0	1 SD below mean	85
0.0	Exactly at the mean	100
+1.0	1 SD above mean	115
+2.0	2 SDs above mean	130
+3.0	3 SDs above mean	145

This conversion is why IQ scores are called deviation IQ scores -- they are defined entirely by how far (how many standard deviations) your performance deviates from average.

"The deviation IQ replaced the old ratio IQ precisely because it provides a consistent, interpretable metric at every age. A score of 130 means the same thing whether you are 10 or 50."
-- John B. Carroll, psychometrician and author of Human Cognitive Abilities

Step 4: The Bell Curve -- Understanding the Normal Distribution

IQ scores follow a normal distribution (bell curve), which is the statistical pattern that emerges when you measure any trait influenced by many small, independent factors.

Properties of the IQ Bell Curve

The bell curve has several mathematically precise properties:

Range	Standard Deviations	Percentage of Population	Approximate Number (per 1,000 people)
Below 55	Below -3 SD	0.1%	1
55-69	-3 to -2 SD	2.1%	21
70-84	-2 to -1 SD	13.6%	136
85-99	-1 SD to mean	34.1%	341
100-114	Mean to +1 SD	34.1%	341
115-129	+1 to +2 SD	13.6%	136
130-144	+2 to +3 SD	2.1%	21
Above 145	Above +3 SD	0.1%	1

Key facts derived from this distribution:

50% of people score between 90 and 110
68% score between 85 and 115 (within 1 SD)
95% score between 70 and 130 (within 2 SDs)
99.7% score between 55 and 145 (within 3 SDs)
Only about 1 in 750 people scores above 145
Only about 1 in 30,000 people scores above 160

IQ Score Classifications: What Each Range Means

Different IQ score ranges are associated with specific classifications. While terminology varies slightly between test publishers, the most commonly used system is:

Wechsler Classification System

IQ Score Range	Classification	Percentile Range	Population Proportion	Real-World Context
160+	Exceptionally Gifted	99.99th+	~1 in 30,000	Theoretical physicists, top mathematicians
145-159	Highly Gifted	99.9th	~1 in 750	Many research scientists, elite academic achievers
130-144	Very Superior / Gifted	98th-99.9th	~2 in 100	Typical threshold for gifted education programs
120-129	Superior	91st-97th	~7 in 100	Many professionals in demanding fields
110-119	High Average	75th-90th	~16 in 100	Successful college graduates on average
90-109	Average	25th-74th	~50 in 100	The majority of the adult population
80-89	Low Average	9th-24th	~16 in 100	May benefit from additional academic support
70-79	Borderline	2nd-8th	~7 in 100	May qualify for learning disability services
Below 70	Extremely Low	Below 2nd	~2 in 100	Clinical criterion for intellectual disability diagnosis

Important context: These classifications describe statistical position relative to the population, not human worth or potential. Many highly successful people have average IQ scores, and a high IQ does not guarantee achievement in any particular domain.

"IQ tells you something about a person's cognitive abilities. It tells you very little about their character, their determination, their creativity, or their capacity for kindness."
-- Howard Gardner, psychologist, Harvard University, creator of the theory of multiple intelligences

Understanding Percentiles

Percentiles tell you what percentage of the population scored lower than you. They are often more intuitive than IQ scores for understanding where you stand.

Complete IQ-to-Percentile Conversion Table

IQ Score	Percentile	Meaning
70	2nd	Scored higher than 2% of the population
75	5th	Scored higher than 5%
80	9th	Scored higher than 9%
85	16th	Scored higher than 16%
90	25th	Scored higher than 25%
95	37th	Scored higher than 37%
100	50th	Scored higher than 50% (exact average)
105	63rd	Scored higher than 63%
110	75th	Scored higher than 75%
115	84th	Scored higher than 84%
120	91st	Scored higher than 91%
125	95th	Scored higher than 95%
130	98th	Scored higher than 98%
135	99th	Scored higher than 99%
140	99.6th	Scored higher than 99.6%
145	99.9th	Scored higher than 99.9%

Notice a critical feature of this table: the percentile gaps between IQ points are not uniform. Moving from an IQ of 100 to 105 jumps you from the 50th to the 63rd percentile (13 percentage points), while moving from 130 to 135 only shifts you from the 98th to the 99th percentile (1 percentage point). This is because the bell curve is steepest near the center and flattens at the extremes.

Composite Scores vs. Subtest Scores

A full-scale IQ score is a composite -- a weighted combination of performance across multiple subtests. Understanding the difference between composite and subtest scores provides a much richer picture of cognitive ability.

WAIS-IV Score Structure

Level	Score Type	What It Tells You
Full Scale IQ (FSIQ)	Overall composite	General cognitive ability across all domains
Index Scores	Domain composites	Strength in specific cognitive areas (verbal, perceptual, memory, speed)
Subtest Scores	Individual task scores	Performance on specific tasks (vocabulary, block design, digit span, etc.)

Why Score Profiles Matter

Two people with the same Full Scale IQ of 110 might have very different cognitive profiles:

Score Component	Person A	Person B
Verbal Comprehension	125	95
Perceptual Reasoning	95	125
Working Memory	110	110
Processing Speed	105	105
Full Scale IQ	110	110

Person A has a clear verbal strength, suggesting they may excel in language-heavy tasks like writing, law, or teaching. Person B has a perceptual reasoning strength, suggesting talent in visual-spatial domains like engineering, architecture, or surgery.

"The Full Scale IQ is useful as a summary, but it can hide as much as it reveals. The pattern of scores -- the peaks and valleys -- often tells us more about a person's cognitive functioning than the single number."
-- Alan Kaufman, clinical psychologist

This is why professional IQ assessments always report both composite and subtest scores. If you want to explore your own cognitive profile, take our full IQ test which provides scores across multiple domains.

How Different IQ Tests Compare

Not all IQ tests use exactly the same scoring system, which can lead to confusion when comparing results across tests.

Standard Deviation Differences

Test	Mean	Standard Deviation	Score of 145 Equals...
Wechsler Scales (WAIS, WISC)	100	15	3 SDs above mean
Stanford-Binet 5	100	15	3 SDs above mean
Stanford-Binet (older editions)	100	16	2.8 SDs above mean
Cattell Culture Fair	100	24	1.9 SDs above mean

This means a score of 145 on the Cattell scale is not equivalent to 145 on the Wechsler. On the Cattell (SD = 24), a score of 145 is only 1.9 standard deviations above the mean, placing you around the 97th percentile. On the Wechsler (SD = 15), a score of 145 is 3 standard deviations above the mean, placing you at the 99.9th percentile.

Always check which test was used before comparing IQ scores.

Score Equivalence Across Tests

Wechsler (SD=15)	Old Stanford-Binet (SD=16)	Cattell (SD=24)	Percentile
85	84	76	16th
100	100	100	50th
115	116	124	84th
130	132	148	98th
145	148	172	99.9th

Reliability and Validity: Can You Trust the Score?

Test-Retest Reliability

A good IQ test produces consistent scores when administered to the same person multiple times. The major professional tests achieve high reliability:

Test	Test-Retest Reliability (FSIQ)	Meaning
WAIS-IV	0.96	Extremely consistent
WISC-V	0.95	Extremely consistent
Stanford-Binet 5	0.95	Extremely consistent

A reliability coefficient of 0.96 means that 96% of the score variation reflects true cognitive differences between people, and only 4% is random measurement error.

However, even with high reliability, individual scores carry a confidence interval. A reported IQ of 115 on the WAIS-IV has a 95% confidence interval of approximately 111 to 119, meaning the person's "true" IQ is very likely somewhere in that 8-point range.

Validity: Does It Measure What It Claims?

IQ test validity is supported by decades of research showing correlations with real-world outcomes:

Outcome	Correlation with IQ	Interpretation
Academic achievement	0.50-0.70	Strong predictor
Job performance (complex jobs)	0.40-0.55	Moderate to strong predictor
Income	0.30-0.40	Moderate predictor
Health and longevity	0.15-0.25	Weak to moderate predictor
Creativity	0.15-0.30	Weak predictor (above threshold of ~120, correlation diminishes)

"IQ tests are the most reliable and valid psychological tests that exist. That does not mean they are perfect -- it means that, compared to every other measurement tool in psychology, they have the most evidence supporting their consistency and accuracy."
-- Ian Deary, psychologist, University of Edinburgh

The Flynn Effect: Why Scores Must Be Re-Normed

The Flynn effect, discovered by James Flynn, is one of the most important phenomena in IQ testing. It refers to the well-documented rise in average IQ scores over time -- approximately 3 points per decade throughout the 20th century.

Implications for Scoring

Because the population gets higher raw scores over time, test norms become outdated. If a test normed in 2000 is still being used in 2020, test-takers will appear to score higher than they would on a freshly normed test -- not because they are smarter, but because the reference group is 20 years old.

Test Normed In	Score If Tested in 2020 (Unadjusted)	Score on 2020 Norms
2000	~106	100
1990	~109	100
1980	~112	100

This is why major test publishers re-norm their tests every 15-20 years. When you take a test, check when it was last normed. Scores on outdated norms may be inflated by several points.

Recent development: Some researchers have found evidence that the Flynn effect has slowed or reversed in several developed countries since the 1990s, a phenomenon sometimes called the "negative Flynn effect." The causes remain debated, with hypotheses ranging from changes in educational practices to dysgenic effects.

Common Misconceptions About IQ Scoring

Misconception 1: "IQ Scores Are Just Correct Answers Divided by Total Questions"

Reality: Raw scores undergo statistical transformation through norming and standardization. The same raw score can produce different IQ values depending on the test version, the norming sample, and the test-taker's age.

Misconception 2: "An IQ of 100 Means You Got 100% Correct"

Reality: An IQ of 100 means you performed at the population average. On many tests, scoring 100% correct would produce an IQ well above 140.

Misconception 3: "My IQ Is Exactly 115"

Reality: Every IQ score has a confidence interval. A reported score of 115 likely means your true score falls somewhere between approximately 111 and 119. Treat IQ as a range, not a precise number.

Misconception 4: "All IQ Tests Give the Same Score"

Reality: Different tests use different standard deviations, measure slightly different cognitive constructs, and were normed on different populations. Variations of 5-10 points between tests are normal and expected.

Misconception 5: "Online IQ Tests Are as Valid as Professional Ones"

Reality: Most online IQ tests lack proper norming, standardization, and validation. Only assessments developed using psychometric principles provide meaningful scores. Our practice test and full IQ test are designed with these principles in mind, though they should not be considered equivalent to a professionally administered clinical assessment.

Practical Applications of IQ Scores

Education

Application	How IQ Scores Are Used
Gifted program identification	Typically requires IQ of 130+ (98th percentile)
Learning disability diagnosis	Discrepancy between IQ and academic achievement may indicate specific learning disabilities
Intellectual disability classification	IQ below 70 combined with adaptive behavior deficits
Educational planning	Score profiles guide individualized instruction strategies

Famous Historical IQ Estimates

While these are estimates (not actual test scores in most cases), they illustrate the range of scores among well-known figures:

Individual	Estimated IQ	Known For
Albert Einstein	~160	Theoretical physics, general relativity
Marie Curie	~180	Radioactivity research, two Nobel Prizes
Bobby Fischer	~187	Chess world champion
Stephen Hawking	~160	Theoretical physics, cosmology
Marilyn vos Savant	~190 (disputed)	Listed in Guinness Book of World Records for highest IQ

Caveat: Historical IQ estimates are reconstructions and should be treated with considerable skepticism. They are included here for illustrative purposes only.

How to Get the Most Accurate IQ Score

If you want to understand your cognitive abilities as accurately as possible, follow these guidelines:

Choose a validated test: Use assessments with proper norming and standardization. Start with our full IQ test for a comprehensive online evaluation, or seek a professional assessment for clinical-grade accuracy.

Test under optimal conditions: Take the test when you are well-rested, alert, and free from distractions. Avoid testing when ill, stressed, or sleep-deprived.

Avoid "studying" for the test: IQ tests are designed to measure cognitive ability, not knowledge. Practicing specific test items inflates your score without reflecting genuine ability.

Understand the confidence interval: Your score is a range, not a precise number. A 5-point difference between two testing sessions is well within normal variation.

Consider multiple assessments: Take our quick IQ test for a fast estimate, our timed IQ test for processing speed emphasis, and a practice test to familiarize yourself with the format before a more formal assessment.

Conclusion: Putting IQ Scores in Perspective

IQ test scoring is a rigorous statistical process that transforms raw cognitive performance into a standardized, interpretable number. The system is built on three pillars: a representative norming sample, the normal distribution, and the standard deviation as a unit of measurement.

Understanding this process empowers you to interpret your own scores accurately, compare results across tests meaningfully, and avoid the common misconceptions that surround IQ testing. A score of 100 is not mediocre -- it is the statistical center of human cognitive ability. A score of 130 is not just "a bit above average" -- it places you ahead of 98% of the population.

But no single number captures the full scope of human intelligence. IQ scores measure specific cognitive abilities -- reasoning, memory, processing speed -- and they do so with impressive reliability. What they do not measure is equally important: creativity, emotional wisdom, perseverance, practical intelligence, and the countless other qualities that contribute to a meaningful and successful life.

"Intelligence tests measure important things, but they do not measure everything that matters."
-- Robert Sternberg, psychologist, Cornell University

Explore your own cognitive abilities with our full IQ test, get a quick snapshot with our quick IQ test, or prepare with our practice test.

References

Wechsler, D. (2008). Wechsler Adult Intelligence Scale -- Fourth Edition (WAIS-IV). Pearson.

Roid, G. H. (2003). Stanford-Binet Intelligence Scales, Fifth Edition (SB5). Riverside Publishing.

Flynn, J. R. (2007). What Is Intelligence? Beyond the Flynn Effect. Cambridge University Press.

Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge University Press.

Kaufman, A. S. (2009). IQ Testing 101. Springer Publishing.

Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and educational achievement. Intelligence, 35(1), 13-21.

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262-274.

Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Ceci, S. J., ... & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51(2), 77-101.

Trahan, L. H., Stuebing, K. K., Fletcher, J. M., & Hiscock, M. (2014). The Flynn effect: A meta-analysis. Psychological Bulletin, 140(5), 1332-1360.

Bratsberg, B., & Rogeberg, O. (2018). Flynn effect and its reversal are both environmentally caused. Proceedings of the National Academy of Sciences, 115(26), 6674-6678.