What Is a Good IQ Score? (Is 120 Good? Is 130?) It Depends What For

Few numbers spark as much curiosity as an IQ score. You finish a test, see “118” and immediately wonder: is that good? The honest answer has two parts. First, how your score compares with other people, which is simple maths. Second, what you want the score to be good for, because a score that is plenty for one purpose may not meet a specific cut-off for another.

How IQ scores work, in one minute

Modern IQ tests don’t divide “mental age” by real age any more — that formula from the early 1900s broke down for adults. Instead they use a deviation IQ: your performance is compared with a large reference group of people your age, and the result is placed on a scale where:

• the average is 100,
• one standard deviation is 15 points,
• scores follow a bell curve (a normal distribution).

The OpenStax Psychology 2e textbook summarises it plainly: on most IQ tests the average score is 100, and in modern IQ testing one standard deviation is 15 points. Because the scale follows a normal curve, the share of people in each band is fixed. The NIST/SEMATECH e-Handbook of Statistical Methods gives the rule for any normal distribution: 68.27% of values fall within one standard deviation of the mean, 95.45% within two and 99.73% within three. For IQ, that means about 68% of people score 85–115, about 95% score 70–130 and about 99.7% score 55–145.

Test publishers attach slightly different labels to these ranges, but the common pattern is: 90–109 average, 110–119 high average, 120–129 superior (or “very high”), 130 and above very superior (or “extremely high”). The full chart, every classification and a percentile calculator for any score live on our IQ scale and IQ chart page. If you want the statistics behind the curve itself, our guide to the IQ bell curve and standard deviations walks through it.

Is 115, 120, 130 or 145 a good IQ score?

These are the four scores people ask about most. Each one is a point on the same curve, so here is where it sits and how rare it is:

• [An IQ of 115](/en/iq-score/115) is one standard deviation above average. It beats about 84% of people (roughly 1 in 6 score this high or higher) and sits in the high-average band. A clearly good result.
• [An IQ of 120](/en/iq-score/120) beats about 91% of people, roughly 1 in 11. It is where most publishers start the “superior” or “very high” band.
• [An IQ of 130](/en/iq-score/130) is two standard deviations up, about the 98th percentile and roughly 1 in 44 people. It is the score most often linked with giftedness and with Mensa.
• [An IQ of 145](/en/iq-score/145) is three standard deviations up, about the 99.9th percentile and roughly 1 in 741 people. At this level, small differences between tests and measurement error matter a lot.

Notice how quickly rarity grows at the top. Moving from 100 to 115 lifts you above a third of the population; moving from 130 to 145 lifts you above only about 2% more — but makes the score about 17 times rarer. For any other number, the calculator on the IQ scale page gives the exact percentile.

Good for what? It depends on the question

“Good” is not a property of the number alone. Here is what the research says about the situations people usually have in mind.

Good compared with other people

Because 100 is the middle by definition, any score above 100 is above average, and most people would call 110 or more a good result. A score just below 100 is not “bad” either: half of all people score below average, which is simply how the scale is designed.

Good for school and university

IQ is one of the better single predictors of grades, but far from the only one. A 2015 meta-analysis in Intelligence by Roth and colleagues pooled 240 samples with 105,185 participants and found a correlation of 0.54 between intelligence test scores and school grades. That is a strong relationship, yet it accounts for only about 29% of the differences in grades. Study habits, motivation, teaching and health make up much of the rest. A score of 115 or 120 is a real advantage for learning complex material; it does not guarantee results, and a 100 does not rule them out.

Good for work

Reasoning ability helps with learning a job, especially a complex one, but the link to job performance is more modest than often claimed. A 2022 re-analysis in the Journal of Applied Psychology by Sackett and colleagues concluded that earlier meta-analyses had overestimated the validity of many selection tools, cognitive tests included, by 0.10 to 0.20 points, and ranked structured interviews at the top. In practice, no employer can tell from a single number whether you will be good at a job.

Good enough for Mensa

Here “good” has a hard definition. Mensa International admits people who score in the top 2% of the population. American Mensa’s list of qualifying test scores translates that into a score of 130 on the Wechsler scales or the Stanford-Binet 5, and it states that unsupervised testing, including internet-based tests, is not accepted as proof of eligibility. So an online result is a way to see whether trying a supervised test is worthwhile, not a ticket in.

Good enough to be called gifted

OpenStax notes that a score of 130 or above is considered a superior level and that people in this range may be classified as gifted in educational systems. Actual rules vary by school system and usually involve more than one test, so treat 130 as a common reference point rather than a universal cut-off. Where the label “genius” begins is a separate debate, covered in what IQ is considered genius.

What is the average IQ?

The average IQ is 100 at every age, because each age group is compared only with itself. A 16-year-old who scores 100 performs like a typical 16-year-old; a 60-year-old who scores 100 performs like a typical 60-year-old. Our guide to average IQ by age explains what does change as people get older.

You may have seen rankings of “average IQ by country”. Treat them with caution: they combine very different tests, samples and years, and are strongly influenced by schooling, health and test familiarity. They say little about any individual.

Averages also drift over time. A 2014 meta-analysis in Psychological Bulletin (Trahan et al.) pooled 285 studies and found that scores rose by an average of 2.31 points per decade, and by about 2.93 points per decade on modern Stanford-Binet and Wechsler tests normed since 1972. This rise is known as the Flynn effect, and it is why tests are periodically re-normed so that 100 stays the middle.

Why your IQ score can vary by a few points

No test measures anything perfectly, so professional reports give a confidence interval around every score. In Pearson’s sample WAIS-IV score report, for example, a Full Scale IQ of 102 comes with a 95% confidence interval of 97 to 107, about five points either side. Things that commonly move a result:

1. Measurement error. A different set of questions samples your ability slightly differently.
2. Practice effects. A meta-analysis of 50 studies and more than 134,000 people by Hausknecht and colleagues found that scores on cognitive ability tests rise on retesting, with larger gains when the same test form is used or coaching is involved.
3. Your state on the day. Sleep, stress, time pressure and distractions all matter.
4. The type of test. A test heavy on vocabulary measures something different from one built on visual patterns.

That is why a single score is best read as a range: 118 really means “somewhere around the high-average-to-superior boundary”.

How online IQ tests compare with clinical tests

Clinical tests such as the Wechsler scales are administered one-to-one by a trained psychologist, take over an hour and are normed on carefully selected national samples. They are used for diagnosis and educational decisions.

An online test is a different tool: faster, cheaper and useful for self-discovery and practice, but not a diagnosis. On IQtonics you answer 40 questions across five areas — patterns, spatial reasoning, numbers, logic and memory — with a 20-minute time limit. Harder questions carry more points, and your weighted score is reported on the same mean-100, standard-deviation-15 scale described above, together with a percentile. Our methodology page explains how the test is built and scored, and how accurate online IQ tests are covers the fine print for any online test. If you want to know what the questions look like, read our guide to the five types of IQ test questions.

What an IQ score doesn’t tell you

IQ tests measure reasoning, pattern recognition, working memory and processing speed — abilities that help with learning complex material. They do not measure creativity, motivation, emotional skills, expertise or character, all of which shape real-world outcomes at least as much. A good IQ score is a strength, not a destiny. For a fuller picture of what the number means, start with what IQ is and how it is calculated.