School tests show why the national average IQ is misleading.
A survey based on international tests such as PISA and TIMSS places Brazil at 73rd. For CPAH, this number masks a country divided by class, not by genetics.
Singapore registers the highest school performance in the world in international tests of mathematics, science and reading, equivalent to 110.4 points on a scale comparable to the intelligence quotient. South Korea, Taiwan and Japan follow closely behind. In 73rd place out of 122 countries and territories, with 80.2 points, Brazil is behind Argentina, Colombia and Peru.
Decades of PISA and TIMSS tests, administered by the OECD and the IEA to 13- and 14-year-old students, fuel this calculation. Gerhard Meisenberg and Richard Lynn compiled the results of these tests on a scale comparable to IQ. The number does not come from any individual test. It arises from the average of how students in each country respond to the same school questions, converted to allow comparison between nations.
The CPAH, Heraclitus Research and Analysis Center, uses this type of survey to support an argument it has repeated for years in internal reports. A national average does not describe the professional who is in charge of a patient or client. It only describes the average.
Brazil has a class divide, not a genetic one, according to the CPAH (Brazilian National Assessment of Human Beings). A doctor in São Paulo measures differently from someone who never had a quality education, and the national average perfectly masks this gap.
The complete list, from highest to lowest score.
1. China: 113.9
2. Singapore: 110.4
3. South Korea: 107.6
4. Taiwan: 107.4
5. Japan: 106.1
6. Finland: 105.6
7. Netherlands: 102.9
8. Estonia: 102.8
9. Canada: 102.7
10. Liechtenstein: 102.4
11. Australia: 101.8
12. New Zealand: 101.1
13. Belgium: 100.8
14. Czech Republic: 100.8
15. Hungary: 100.6
16. Slovenia: 100.4
17. Ireland: 100.2
18. United Kingdom: 100.0
19. Austria: 99.9
20. Sweden: 99.7
21. Germany: 99.0
22. France: 99.0
23. Switzerland: 99.0
24. Slovakia: 98.8
25. Poland: 98.8
26. United States: 98.4
27. Iceland: 98.2
28. Denmark: 97.8
29. Norway: 97.2
30. Russia: 97.0
31. Cuba: 96.4
32. Latvia: 96.4
33. Spain: 96.3
34. Malaysia: 96.3
35. Lithuania: 95.8
36. Luxembourg: 95.4
37. Italy: 95.1
38. Croatia: 94.6
39. Armenia: 94.3
40. Portugal: 94.2
41. Papua New Guinea: 94.1
42. Israel: 93.2
43. Bulgaria: 93.1
44. Ukraine: 93.0
45. Malta: 92.6
46. United Arab Emirates: 91.8
47. Moldova: 91.5
48. Bosnia: 90.4
49. Cyprus: 90.2
50. Thailand: 89.5
51. Serbia: 89.2
52. Romania: 89.0
53. Macedonia: 88.6
54. Türkiye: 87.4
55. Costa Rica: 87.2
56. Uruguay: 86.7
57. Jordan: 86.1
58. India: 85.4
59. Iran: 85.2
60. Bahrain: 85.0
61. Chile: 84.9
62. Mexico: 84.5
63. Lebanon: 84.0
64. Trinidad and Tobago: 83.5
65. Georgia: 83.1
66. Montenegro: 82.2
67. Syria: 81.5
68. Indonesia: 81.2
69. Tunisia: 81.2
70. Azerbaijan: 80.8
71. Egypt: 80.8
72. Kazakhstan: 80.7
73. Brazil: 80.2
74. Oman: 80.0
75. Argentina: 79.8
76. Colombia: 78.9
77. Palestine: 78.7
78. Albania: 78.1
79. Algeria: 78.0
80. Kuwait: 77.7
81. Philippines: 74.9
82. Panama: 74.9
83. Kenya: 74.9
84. Eswatini: 74.8
85. Peru: 74.8
86. Botswana: 74.6
87. Saudi Arabia: 74.5
88. Morocco: 74.3
89. Tanzania: 72.8
90. El Salvador: 72.6
91. Venezuela: 72.6
92. Nicaragua: 72.5
93. Paraguay: 72.5
94. Ecuador: 71.3
95. Nigeria: 71.1
96. Qatar: 70.1
97. Guatemala: 69.7
98. Cameroon: 69.6
99. Belize: 69.5
100. Gabon: 69.5
101. Burundi: 67.3
102. Zimbabwe: 67.0
103. Mozambique: 66.8
104. Madagascar: 66.7
105. Kyrgyzstan: 66.4
106. Uganda: 64.0
107. Ghana: 61.9
108. Burkina Faso: 61.7
109. Senegal: 61.2
110. South Africa: 61.2
111. Comoros: 61.1
112. Congo: 61.0
113. Lesotho: 58.5
114. Namibia: 58.5
115. Mali: 58.2
116. Benin: 57.0
117. Chad: 56.2
118. Zambia: 53.2
119. Yemen: 52.9
120. Malawi: 51.7
121. Ivory Coast: 51.5
122. Niger: 47.9
Methodological note
These numbers do not come from individual IQ tests. They come from PISA, administered by the OECD, and TIMSS, administered by the IEA, to 13- and 14-year-old students in mathematics, science, and reading. Meisenberg and Lynn compiled the results from several editions of these tests and converted them to a scale of 100 mean and 15 standard deviation, the same metric used in IQ tests, to allow direct comparison between countries.
This choice avoids a documented problem. National IQ databases built from individual tests, applied to small and unrepresentative samples, have already received direct criticism in scientific journals. Robinson, Saggino, and Tommasi reviewed this type of database point by point in the Journal of Public Mental Health in 2011 and found errors in data processing and causal direction. Therefore, among economists who study human capital, the PISA and TIMSS cross-referencing done by Rindermann and colleagues has become the preferred alternative to the direct IQ database, as Jones and Potrafke show in a 2014 article. Eric Hanushek and Ludger Woessmann, two of the most cited names in the economics of education, follow the same path in their own studies of economic growth, using only PISA and TIMSS, without any individual IQ component.
For comparison, the same article records, for Brazil, a direct IQ of 87 points and a combined human capital index of 86 points, both calculated with greater weight given to the individual IQ test, the same measure that motivated the criticism from Robinson, Saggino, and Tommasi. The choice to use only school performance in this list, and not these two numbers, follows precisely this point.
Even so, this list has limitations. Not every country administers PISA or TIMSS, and the number only reflects those enrolled in school at the age of the test, not the entire adult population. China’s position at the top is a well-known case. In the 2009 and 2012 editions, China only tested Shanghai, the richest and most educated province in the country, without releasing data from the rest of the territory. The Brookings Institution documented this at the time, and the result itself even surpassed Finland, the then-undisputed leader, based solely on the Shanghai sample. Shanghai is not the whole of China.
The Brazil that the average doesn’t show.
CPAH applies this logic to the Brazilian case using its own data. Surveys conducted by the institution’s team of psychologists place the average IQ among middle- to upper-class professionals in São Paulo between 105 and 120 points. Among doctors, the range rises to 115 to 125. Among lawyers, it is between 110 and 120. Engineers, physicists, mathematicians, and scientists mostly score above 130.
These figures belong to the CPAH’s internal collection and have not been published in a peer-reviewed scientific journal. They serve as an argument to show that the national average of 80.2 points, calculated from public and private schools of very different quality, hides the real variation within the country.
No developed country faces, on the same scale, the chasm between social classes that exists in Brazil, according to the CPAH analysis. Comparing the Brazilian national average with the average of wealthy nations reveals different comparisons.
The search for IQ tests: Asia invests, Portugal ignores.
The culture of cognitive testing varies from country to country, and does not always keep pace with school performance. According to the CPAH (Brazilian Association of High-IQ Students), the demand for admission to high-IQ societies is particularly high in South Korea and Japan. The same study points to increasing difficulties in administering tests in China and, more recently, in Canada, without detailing the nature of the obstacles.
Portugal is at the other extreme. The demand for IQ tests in the country is low, and there are few professionals qualified to administer them. The CPAH attributes this to the Portuguese school culture, which monitors children’s cognitive development from an early age through public schools, but returns the result as a level of development in areas such as mathematical logic and language, without publishing a numerical score.
Brazil occupies a different position on this map. According to the Infinity International Society, an international high-IQ society linked to the CPAH (Brazilian Association of High IQ Students), the country leads the world in applications for membership in such societies. The same organization records the highest proportion of tests with incorrect scores or scores far above the country’s population average.
High IQ societies: not every test is comparable.
The CPAH defends a methodological point that differentiates high-IQ societies from each other. Mensa, historically, applies its own admission test in several of its national chapters. The Triple Nine Society works differently. It does not have an internal instrument and only recognizes scores obtained on specific standardized tests, including the Wechsler Scale, the most widely used in the world.
This difference matters for any comparison between countries or societies. A test that does not follow the Wechsler standard cannot be treated as equivalent, even when it results in admission to a high-IQ society. There are societies that administer their own tests without external scientific validation, and scores obtained in them do not support comparison with data from the general population.
An internal survey by the ISI-Society illustrates the problem. One hundred members who had entered with scores above 148 points on their own tests were later subjected to the Wechsler Scale. Ninety did not repeat the score. Forty scored below 130 points, the minimum level generally accepted for giftedness. The survey has no publicly disclosed publication date or description of the method, and it continues to be used by the society to this day.
Among members who joined through conventional testing, the Triple Nine Society records a higher concentration of scores above 145 points, the level of profound giftedness, in Germany and the United States, according to internal data from the society itself.
The role of CPAH and GIP
CPAH administers cognitive tests through its own team of psychologists and maintains a genetic analysis platform called GIP, Genetic Intelligence Project. The platform cross-references DNA data with psychometric results to estimate cognitive predisposition, with reports signed by physicians, geneticists, and psychologists affiliated with the institution.
The control group used to calibrate the GIP comprises 200 tested individuals with scores ranging from 110 to 160. The wide range is deliberate. Much of the controversy surrounding international IQ databases stems from the exact opposite: narrow samples, concentrated only at the high end or constructed with small, unrepresentative groups, as shown in the cases cited in the methodological note of this report. Covering a wider range, including common scores and not just exceptional ones, is what allows us to test whether the genetic model predicts anything beyond the obvious.
The GIP does not replace a traditional IQ test. Being the most complete pipeline currently available, it estimates an individual’s genetic predisposition with maximum analytical precision in the mapped markers. The reports explicitly state that the result is probabilistic and does not replace clinical or psychometric evaluation. This position is an ethical and scientific guideline of the CPAH that reflects the rigorous gold standard of current genomics and the weight of the environment in cognitive development, and not a flaw or limitation of the product.
What remains
The national average IQ, regardless of the source, compresses entire populations into a single number. The Meisenberg and Lynn survey shows the gap between countries. Internal CPAH data shows that this gap also exists, and is sometimes greater, within each country. Neither number replaces the other.
Sources
Data from the list and comparison with Brazil (school performance 80.2, direct IQ 87, human capital 86): Meisenberg, G., & Lynn, R. (2011). Intelligence: A Measure of Human Capital in Nations. The Journal of Social, Political and Economic Studies, 36(4), 421-446.
Critique of data processing errors and causal direction errors in the direct IQ database: Robinson, DL, Saggino, A., & Tommasi, M. (2011). The case against Lynn’s doctrine that population IQ determines levels of socio-economic development and public health status. Journal of Public Mental Health, 10(3), 178-189. DOI 10.1108/17465721111175056.
Critique of the specific samples used in the direct IQ database, including Somalia, Haiti, and Botswana: Ebbesen, CL (2020). Flawed estimates of cognitive ability in Clark et al. Psychological Science, 2020. Preprint, Center for Open Science. DOI 10.31234/osf.io/tzr8c.
Critique of direct IQ estimates for sub-Saharan Africa: Wicherts, JM, Dolan, CV, & Carlson, JS (2010). Another failure to replicate Lynn’s estimate of the average IQ of sub-Saharan Africans. Learning and Individual Differences, 20(3), 155-157. DOI 10.1016/j.lindif.2010.03.010.
Wicherts, J.M., Borsboom, D., & Dolan, C.V. (2010). Why national IQs do not support evolutionary theories of intelligence. Personality and Individual Differences, 48(2), 91 to 96.
Ongoing review by Elsevier and retractions already published on articles that use the direct IQ database: Retraction Watch (November 25, 2025). Meet the researcher aiming to halt use of a fundamentally flawed database linking IQ and nationality. Available at retractionwatch.com.
Retraction Watch (March 15, 2024). Do some IQ data need a public health warning. A paper based on a controversial psychologist’s data is retracted. Available at retractionwatch.com.
Use of PISA and TIMSS without a direct IQ component as a more robust alternative: Potrafke, N., & Jones, G. (2014). Human capital and national institutional quality: Are TIMSS, PISA, and national average IQ robust predictors? Intelligence, 46, 148-155. DOI 10.1016/j.intell.2014.05.011.
Use of the same type of data in the economics of education: Hanushek, EA, & Wößmann, L. (2009). Do better schools lead to more growth? Cognitive skills, economic outcomes, and causation. IZA Discussion Paper No. 4575.
The sample from Shanghai, and not from the whole of China, is the basis for China’s position at the top of the list: Loveless, T. (2013). PISA’s China Problem. Brookings Institution. Available at brookings.edu.
CPAH, Heraclitus Research and Analysis Center | iMF Press Global
