Critics Highlight Flaws in "National IQ" Studies, Citing Researcher-Based Selection Effects

Luca Dellanna, a prominent management advisor and author known for his work on risk management and behavioral psychology, recently drew attention to a critical issue in scientific research: "researcher-based selection effects." In a social media post, Dellanna stated, "A reminder that plenty of studies are actually researcher-based selection effects. An example: 'National IQ' studies are often based on very flawed samples." This assertion underscores long-standing criticisms regarding the methodology and validity of certain large-scale studies, particularly those attempting to quantify national intelligence.

Selection bias, or selection effect, occurs when the selection of individuals or data for analysis is not random, leading to a sample that is unrepresentative of the target population. This systematic error can significantly compromise the validity and generalizability of research findings. Experts identify various forms of selection bias, including sampling bias, self-selection bias (or volunteer bias), and survivorship bias, all of which can distort results.

The criticisms leveled against "National IQ" studies, notably those pioneered by the late British psychologist Richard Lynn, frequently cite these methodological flaws. Lynn's "national IQ" datasets, which purport to show average IQ scores for various countries, have been widely scrutinized for their reliance on inadequate and unrepresentative primary data sources. For instance, critics like Professor Rebecca Sear of Brunel University point out that some national IQ estimates were based on extremely small samples, such as Angola's IQ being derived from just 19 individuals from a malaria study, or Eritrea's from children in orphanages.

Professor Jelte Wicherts of Tilburg University, a psychometrician, has extensively critiqued Lynn's work, suggesting a systematic exclusion of higher IQ scores in calculations for African countries. Wicherts noted that Lynn's "main inclusion criteria he had been using appeared to be the IQ itself, not objective measures like whether it was a normal, healthy sample. That’s quite a lethal indicator of bias." These studies often used diverse cognitive tests and populations, making cross-national comparisons problematic due to cultural specificity and varying educational exposures.

The concerns extend beyond academic rigor, as these "national IQ" datasets have been used to support controversial and often discredited theories, including those with eugenic implications. Academic publisher Elsevier recently announced a review of Lynn's research papers, citing "external feedback" and a need to address the reliability of his work. This move reflects a broader recognition within the scientific community of the need for transparent and unbiased research practices, especially when findings have significant societal implications.