IQ Scores and Differences Between Groups

Dr. Russell T. Warne

Sep 10, 2023

Since the early 20th century, psychologists administering IQ tests have noticed that average differences in IQ between groups. These IQ differences are real, and no one argues about their existence. It is important to remember, though, that these are average differences. There is a lot of overlap between groups, and people from every demographic group can be found at all IQ levels.

Sex Differences in IQ

On most IQ tests, the average difference between males and females is very small, but when it does occur, it usually favors males. On the Reasoning and Intelligence Online Test (RIOT), the average sex difference in IQ was 0.69 points favoring males. That is a trivial difference. But the mean IQ difference isn’t the whole story. On specific subtests, the differences are sometimes noticeably larger. For example, on spatial ability, males outperform females by d = .251 (which is 3.77 points on the IQ scale), but females have reaction times (d = -.197, or 2.96 points on the IQ scale) that are about 5-6% faster. Other sex differences are smaller, and when aggregated, most cancel out.

That is just average differences, though. The differences in variability are larger and have important implications. On the RIOT, the standard deviation of males’ IQ scores was 14.8% larger than the standard deviation for female IQ. That means that males’ scores are more spread out, while females’ scores cluster more towards the middle. You can see this diagrammed in the image below. On the RIOT, there are more females with IQs between 81 and 115. Outside of this range, there are more males. This finding is very common in intelligence research and is not confined to IQ scores.

Source: Warne, R. T. (2025). Technical manual for the Reasoning and Intelligence Online Test, version 1.0. Riot IQ. Used with permission.

The implication of this difference in score spread is that there are more males at high and low levels of IQ. As a result, males are overrepresented in situations where low IQ is more common (like special education classes or group homes for people with intellectual disabilities), but they are also found in environments that require higher to enter (e.g., graduate school).

Age Difference in IQ Scores

Usually, IQ scores are calculated by comparing people to others in their age group. This masks any differences that exist between age groups. However, groups can be compared on their raw scores on specific subtests. These generally show that tests of learned knowledge (like a Vocabulary subtest) increase gradually through adulthood and do not start to decline until late in life. Subtests that require fluid reasoning to discern rules and apply them to answer questions peak much earlier in life and decline more quickly. Some memory and processing speed tasks also show strong age differences (favoring younger adults).

Education Differences in IQ Scores

Unsurprisingly, IQ differences also emerge across educational groups. People with less education tend to have lower IQ scores. Some of this is because education does seem to raise IQ. But the reverse is also true: smarter students tend to stay in school longer.

Race and Ethnicity Differences in IQ Scores

The most controversial finding related to IQ is racial and ethnic differences in IQ. For over 100 years, psychologists have found that East Asian groups have a higher average IQ than groups descended from Europeans. The latter, in turn, have a higher IQ than groups descended from Africans. The race difference that has been the subject of the most study is the Black-White difference in the United States. This difference is usually 12 to 16 IQ points, depending on the test given and other characteristics. On the RIOT, for example, the Black-White average IQ difference is 12.20 IQ points. This difference (and other differences in race/ethnicity on the RIOT) are graphed in the image below.

Source: Warne, R. T. (2025). Technical manual for the Reasoning and Intelligence Online Test, version 1.0. Riot IQ. Used with permission.

Social class differences in IQ

There are also average IQ differences across social class groups. IQ correlates positively with income, meaning that wealthier people tend to perform better on IQ tests. Some of this relationship seems to be causal, meaning that being smarter causes people to earn more money. This likely happens because some well-paying occupations have a minimum IQ needed to enter that career, either explicitly (through test score minimums to attend training) or implicitly (through educational requirements or the demands of on-the-job training). But this does not mean that everyone smart is wealthy; nothing is stopping a bright person from choosing to work a job that has a lower average IQ and pays less if that’s where their interests are or what their lifestyle favors. Other causal pathways are possible. For example, crime is more common in low-IQ individuals, and being incarcerated would definitely have a negative impact on social class.

International Differences in IQ

Another area of controversy is average IQ across nations. These averages are controversial because wealthier countries tend to have higher IQs than poorer countries, and the differences can be stark.

Comparing scores across nations is much more complicated than comparing scores across groups within a country. If a country has a common language and a common base of knowledge among its people, then creating an intelligence test that is fair to all groups is a manageable task. However, cross-cultural testing is much more complicated. One solution has been to use a nonverbal tests (like the Raven’s Progressive Matrices) to measure international IQ scores. While popular, this strategy has major problems that are probably insurmountable when using these tests in low-income countries.

But some researchers have shown that test scores can be portable across nations with similar cultures. Even vastly different cultures can sometimes use the same IQ test. Even when the non-verbal tests are used, it is clear that they’re measuring something, even if that “something” is not intelligence. These IQs correlate with important national variables measuring average health (e.g., life expectancy), economic development, democracy, and more. Examining this, one expert proposed that these tests measure how well different countries have trained their citizens to solve abstract problems. That’s an important skill in a modern economy, but it’s not the same thing as intelligence.

Are IQ Tests Biased?

Given the differences across groups, it is logical to wonder if something is wrong with the tests. Many critics have accused IQ tests of being biased, and this is a claim that IQ test creators take very seriously. Since the 1960s, IQ tests have been the target of studies of bias, and by the 1980s, it was standard practice to screen tests and their items for any bias before releasing them to the public.

Test creators and other experts universally recognize that the existence of average differences in scores is not, by itself, evidence that a test is biased. (It can be a good reason to investigate the possibility of bias, though.) Investigating test bias is a sophisticated manner, but in short, all methods try to identify whether one group has a systematic advantage (or disadvantage) for reasons that are unrelated to what the test is measuring.

Overwhelmingly, the results show that IQ tests are not biased, as long as they are used on people for whom they are intended. This is especially true in the modern era (since the 1980s), when IQ test creators started screening for bias and dropping any biased content before a test was released to the public. However, when tests are used on examinees for which they were not intended (such as examinees in impoverished nations taking the Raven’s Progressive Matrices), then bias may occur. When using a test for a group it is not designed for, it is the test administrator’s responsibility to determine whether the test is functioning properly and without bias for the new examinees. Unfortunately, too few people using tests in this fashion do so. A lack of bias is common, but it cannot be assumed. This issue is a problem mostly when making international score comparisons with a test that was not designed for that purpose. When comparing individuals’ scores within a country that a test was designed for, bias is rarely a problem.

Parting Thoughts

When studying average IQ differences, it is important to remember that there is a lot of overlap. That means that if Group A has a higher average IQ than Group B, then there are still some people in Group B who are smarter than people in Group A. That is why it is always best to judge people individually whenever possible. None of these findings justify discrimination. For that reason, these findings are nothing to fear, but they still must be understood in their proper context.

We hope you found this information useful. For further questions, please join our Discord server to ask a Riot IQ team member or email us at support@riotiq.com. If you are interested in IQ and Intelligence, we co-moderate a related subreddit forum and have started a Youtube channel. Please feel free to join us.