Some of my colleagues have a new paper out:
-
Te Nijenhuis, J., Pesta, B. J., & Fuerst, J. G. R. (n.d.). General mental ability testing and adverse impact in the United Kingdom: A meta-analysis with more than two million observations. European Journal of Work and Organizational Psychology, 0(0), 1–14. https://doi.org/10.1080/1359432X.2024.2377780
We review ethnic group differences on high-stakes General Mental Ability (GMA) tests based on 21st century UK data. Thereafter, we meta-analyse scores on 23 occupational, public sector, educational, military, or general-public selection tests, with a sample size exceeding two million. Relative to White GMA, the grand meta-analytic effect sizes (Cohen’s d) by major ethnic groups were: Mixed d = .14 (k = 24, N = 67,114), Blacks d = .65 (k = 32, N = 112,975), Asians d = .33 (k = 32, N = 311,695), and Other d = .49 (k = 24, N = 42,846). Further, although Chinese residents outscored White British residents (d = −.15, k = 20, N = 18,897), all other Asian ethnic groups scored slightly to substantially lower. For example, South Asians as a whole averaged d = .37; k = 13, N = 67,566. By subgroups, these averages were: Indians (d = .17, k = 10, N = 28,236), Pakistanis (d = .49, k = 9, N = 19,371), and Bangladeshi (d = .55, k = 7, N = 19,772). Implications for practice and theory are discussed.
As it says, these are based on selection or entrance tests. That is, people who apply for some job or want to be selected or approved for something else. Checking their supplementary materials, we see these are from:
- Public Sector Tests
- Military Tests
- Public Driving Tests
- College Entrance Tests & Qualifications
- Law School Tests
- Medical School Tests
- Business School Tests
It is fun to see the driving theory exam being used as a source. Here’s the results for that particular test:
Gaps are not that large, but they look to be roughly in the right directions. Of course, everybody can and most will practice on the theory test. That’s the point, learning the rules of traffic! Because of this, many subjects will hit the ceiling and the gaps will be diminished artificially.
Anyway, aggregating all their samples, here’s their main results:
It avoids saying the IQ equivalents, but you can just multiply the d value by 15 to get the gap to Whites (mostly Brits). Thus we see that Chinese have an advantage of 0.15 d or 2.25 IQ. Smaller than in the USA, maybe due to language bias (this group has low English proficiency) and weaker elite migrant selection (they prefer the USA). The overall Black-White gap is 0.65 d or 9.75 IQ. We also see the superiority of the Indian immigrants (0.17 d gap) over the very closely related Muslims from Pakistan and Bangladesh (d’s 0.49 and 0.55, or about 8 IQ). Indians seem to be over-represented among elites in the UK (e.g. prime minister), so presumably the distribution of Indian IQs is wider and has a strong smart fraction.
There is no correction for measurement error / reliability, so these are lower bounds values. Again, since these are also based on self-selected samples (selection tests) and explicitly calls for training on the test (driver’s theory test), these are smaller than the population gaps. However, they should have about the same relative differences, and they do:
For the 18 narrow ethnic groups, d values reported in our and the corresponding d values computed from the data for all adults reported by Pesta et al. (Citation2023; Table S4) correlated at r = .84.
They unfortunately don’t provide a plot, so here’s one I made:
The general population samples are the estimates from the same authors’ prior study of UK ethnic gaps in IQ, which I apparently didn’t cover before, so let’s do it here because it is also important:
- Pesta, B. J., te Nijenhuis, J., Fuerst, J. G., & Shibaev, V. (2023). Links between Ethnicity, Socioeconomic Status, and Measured Cognition in Diverse Samples of UK Adults. Comparative Sociology, 22(6), 785-823.
In the UK, immigrant groups frequently have lower mean socioeconomic status (SES) than do White British, which is a source of concern for the British government. Group-level SES tends to show positive relationships with cognitive ability scores. Thus, the authors estimate the mean cognitive and SES scores of various ethnic groups and test empirically if they correlate. They compute SES and cognitive ability scores using high-quality representative samples of adults. They then computed correlations between the two measures. General SES and group-cognitive ability correlated strongly at r = .59 to r = .79 (N = 18 groups). Finally, the authors computed cognitive scores predicted by the nation or region-of-origin of the ethnic groups and calculated correlations between these expected scores and the measured scores. The predicted and measured scores correlated strongly at r = .93 (N = 16 groups). The authors conclude that ethnic differences in SES are partly linked to differences in cognitive ability.
They did a few things:
- Meta-analyzed fairly representative samples of adults from various ethnicities on various IQ tests.
- Computed ethnic gaps on socioeconomic variables for the same groups.
- Matched the ethnicities to countries or regions of origin and their test scores.
Doing this leads to this main result:
So higher IQ ethnicities have higher general socioeconomic outcomes, r = 0.79. The meritocracy works well! They also plotted each individual social variable against IQ:
Unfortunately, they have mistakenly swapped the axes (IQ should be on X as it is the cause), but the correlations will be the same.
Finally, if we were to guess ethnicities’ average IQs in the UK based on their origins:
Again, we are not surprised. The correlation is absurdly good: sqrt(.857) = 0.93.
In the new study, the IQ means of applicants also correlate well with country of origin predictions:
However, ethnic d values were correlated strongly and statistically significantly with the predicted region of origin test scores reported by Pesta et al. (r = −.73; N = 16) and the predicted region of origin test scores computed by the present authors (r = −.71; N = 18). The correlations remained high when dropping ethnic groups for whom a substantial percentage of the group had unclear region of origin (i.e., Other Black, Gypsy, Roma and Irish Travellers, and Other Mixed) (r = −.79 to −.80; N = 15) or when additionally dropping all other mixed groups (i.e., Caribbean-White, African-White, and Asian-White) (r = −.82 to −.83; N = 12) or when additionally dropping fairly heterogeneous groups (i.e., Other White, Other Asian, any Other) (r = −.85 to .-86; N = 9). Excluding British Whites, on the grounds that they are not a migrant group, did not substantially change the results (r = −.67, N = 17, to r = .83, N = 8). These results tentatively suggest that ethnic groups from countries and regions with more developed educational infrastructures (e.g., Irish, Other White, and Chinese) tend to outperform those from regions with less developed educational systems (e.g., Pakistanis, Caribbeans, Africans, and Arabs).
The country of origin values used don’t seem to be in the article or its supplements, but they are maybe the same as those in the plot from the prior study.
Conclusions
- UK ethnic/race differences have been challenged based on data from A-levels and the GCSE. See here, here, here, here, here, here.
- However, actual adult data confirm the usual pattern of results.
- This new study shows that this is also true for people who apply for jobs, take entrance tests, or driver’s licenses.
- The hereditarian expectations borne out: social status and country of origin scores correlate extremely well with the results from inside the UK.