Independent assessment · Sourced
The WorldPopulationReview IQ Data Controversy, Explained
WorldPopulationReview's “Average IQ by Country” ranking is built primarily on Richard Lynn and David Becker's 2019 national IQ dataset, supplemented by self-selected web surveys. The Lynn data has been the subject of peer-reviewed methodological criticism (Sear, Lawson, Kaplan & Shenk, 2022), retraction calls reported by STAT News, and ethical concerns documented by the Southern Poverty Law Center. Defensible alternatives — PISA scores, average years of schooling, R&D intensity — produce a different ranking and come from official OECD, UNESCO, and World Bank sources.
The dataset itself
Richard Lynn (1930–2023) and David Becker compiled the most-cited “national IQ” dataset over several decades, with the most recent revision published in 2019. The dataset assigns a single IQ figure to each country by averaging results from a heterogeneous collection of underlying studies — often using different IQ tests, different age groups, different sampling methods, and, for many countries, samples of fewer than a few hundred people. For countries with no underlying study, Lynn imputed values from geographic or demographic neighbors.
WorldPopulationReview's “Average IQ by Country” page presents this single per-country number as authoritative, in the same template and typography as its Census-sourced population figures. The methodology disclosure on the page itself is minimal; the underlying source studies are not linked.
The peer-reviewed critique
The most substantive academic challenge to the Lynn dataset is Sear, Lawson, Kaplan & Shenk (2022), which argues that national IQ datasets of this kind cannot accurately measure population intelligence because the underlying studies are too heterogeneous, the samples are often too small or unrepresentative, and the imputation strategies introduce large biases. The authors recommend that researchers stop treating these aggregated “national IQ” figures as defensible measures.
Other peer-reviewed critiques have noted similar concerns about the underlying study quality and Lynn's imputation methods. The methodological objections are independent of the political or ethical concerns raised about Lynn's broader work — even setting those aside, the data quality is contested.
The retraction calls
STAT News reported in June 2024 (“Researchers want academic journals to retract racist research from a leading psychologist”) that researchers had called for the retraction of multiple papers based on Lynn-derived national IQ data. The Society for Genetic Education and the Evolution and Human Behavior Association (EHBEA) have publicly distanced themselves from Lynn-derived research. As of 2026, several journals have issued expressions of concern; full retractions vary by journal and paper.
For a reader trying to evaluate the reliability of a current third-party page that re-publishes Lynn-derived figures, the practical signal is: the data has been formally challenged in the peer-reviewed literature, and the journalistic record documents an ongoing retraction effort.
Defensible alternatives
For cross-country comparison of cognitive or educational outcomes, four primary-source datasets are well-established and widely cited:
- PISA scores (OECD Programme for International Student Assessment) — tests 15-year-olds in reading, math, and science across 80+ countries every three years. Standardized methodology, published by the OECD with full documentation. oecd.org/pisa
- Mean years of schooling (UNESCO + World Bank) — average years of formal education for adults 25+. A direct measure of educational attainment. data.worldbank.org
- R&D intensity (R&D expenditure as % of GDP) — national investment in research and development, published by the World Bank.
- Tertiary education attainment (OECD Education at a Glance) — share of adults with college or university qualifications.
None of these is a proxy for “intelligence” in the way Lynn used the term. They are measures of educational attainment and research investment — which is, in most cases, what policy-relevant cross-country comparisons actually need.
What this means for a reader of WorldPopulationReview
A page that re-publishes the Lynn dataset without surfacing any of the documented methodological concerns presents the figures as more authoritative than the underlying evidence supports. That isn't the same as the figures being “wrong” — they reflect real (though disputed) underlying studies — but it does mean those figures should not be treated as comparable to Census Bureau population data or World Bank GDP data, which are based on systematic sampling with documented methodology.
For any decision that touches policy, journalism, or academic citation, use PISA, education-years, or other primary-source measures and link to the source. For demographic and economic data, the rest of WorldPopulationReview is generally a fine second-source for re-published Census/UN/World Bank figures.
Frequently Asked Questions
WorldPopulationReview's "Average IQ by Country" page draws primarily on Richard Lynn's national IQ datasets, supplemented by the "International IQ Test" — a self-selected web survey. Lynn's methodology has been challenged in peer-reviewed work (notably Sear, Lawson, Kaplan & Shenk, 2022) showing that national IQ datasets of this kind are not statistically reliable measures of population intelligence. The Southern Poverty Law Center has documented Lynn's broader work, and STAT News reported in 2024 on retraction calls aimed at journals that published Lynn-derived studies. Media Bias/Fact Check explicitly downgrades WorldPopulationReview to "Mostly Factual" because of this data.
Richard Lynn (1930–2023) was a British psychologist who published a long series of papers and books claiming to measure cognitive differences across countries and ethnic groups. His national IQ datasets were assembled from a heterogeneous mix of small-scale studies, often with minimal sampling controls. Wikipedia's biographical article on Lynn documents his academic affiliations and the criticism his work attracted; the Southern Poverty Law Center maintains a profile cataloguing his ties to extremist publications.
Sear, Lawson, Kaplan & Shenk (2022) — a peer-reviewed paper in evolutionary social science — argues that national IQ datasets of the type Lynn published cannot accurately measure population intelligence because the underlying studies use different IQ tests, different age ranges, different sampling methods, and often very small samples (sometimes under 100 people for an entire country). The authors conclude that the resulting "national IQ" figures are not statistically meaningful comparisons. Other critiques cite the way Lynn imputed missing-country values from neighboring countries, which introduces large biases.
STAT News reported in June 2024 that researchers and academics had called for the retraction of multiple papers based on Lynn's national IQ datasets. The Society for Genetic Education and Genetics Education and Outreach Network (EHBEA) has made public statements distancing itself from Lynn-derived research. As of 2026, several journals have issued expressions of concern; full retractions vary by journal and paper. The relevant point for readers: any current page that re-publishes Lynn-derived national IQ figures is republishing data that has been formally challenged in the peer-reviewed literature.
For cross-country cognitive comparison, the defensible primary sources are: (1) PISA scores from the OECD, which test 15-year-olds in reading, math, and science with a standardized methodology across 80+ countries; (2) average years of schooling, published by UNESCO and the World Bank; (3) R&D intensity (R&D as % of GDP) from the World Bank; (4) tertiary-education attainment rates from OECD. None of these are proxies for "intelligence" in the way Lynn used the term, but they are well-sourced measures of educational attainment and research investment, which is usually what policy-relevant comparisons actually need.
For the bulk of pages — population, demographics, income, education, housing, country-level economic data — WorldPopulationReview re-publishes US Census ACS, UN World Population Prospects, and World Bank data, and those figures should match the original federal/UN sources. The IQ rankings, crime-perception rankings, cost-of-living composites, and "happiness" rankings carry more methodological baggage than a pure-Census-data page and should be cross-checked against primary sources before citation. Media Bias/Fact Check rates the site "Mostly Factual" with explicit concern about the IQ data.
Related
This page summarizes publicly documented academic and journalistic concerns about the Lynn national-IQ dataset. We link directly to the peer-reviewed critique, the retraction-call coverage, and the defensible primary-source alternatives. We do not allege bad faith on the part of WorldPopulationReview; we point readers to where the methodological issues are documented and to better data.