Gaps around the world

I’ve decided to take Richard Lynn up on his 2010 challenge:

“If a multiracial society is found where these race differences in intelligence are absent, the evolutionary and genetic theory of these differences would be falsified. Those who maintain that there are no genetic differences in intelligence between the races are urged to attempt this task.”

As such, I’m going to start collecting new data (i.e., from 2000 on), starting with secondary data, on the ethnic gaps in multiethnic societies. If you’re aware of any good sources feel free to let me know. I am generally only interested in data which includes a standardized measure of ability. I’m going to take my time on this, so this post will be continually updated until I find a society that substantially falsifies Lynn’s evolutionary hypothesis.


Sackett et al. Perspectives From Twenty-Two Countries on the Legal Environment for Selection (Chapter 30). Farr and Tippins (Eds.), 2010. Handbook of Employee Selection.

Baron et al., 2005. Ethnic group differences and measuring cognitive ability. International Review of Industrial and Organizational Psychology


(Summary: Gaps much smaller than reported in Lynn (2008); aborigine/non-aborigne gaps about 1 SD)

Bradely et al. The magnitude of educational disadvantage amongst indigeneous minority groups in Australia

De Bertoli and Thomson, 2009. The achievement of Australia’s Indigenous students in PISA 2000 – 2006

Dalton, 2011. US Educational Achievement on International Assessments: The Role of Race and Ethnicity. See: Table 3. TIMSS 2007 achievement differences between majority and minority students in Australia, New Zealand, and the United States

Leigh and Gong, 2008. Estimating Cognitive Gaps Between Indigenous and Non-Indigenous Australians
(See table 1. Summarizes scores from tens of thousands of individuals)

Thomson et al., 2010. Challenges for Australian Education: Results from PISA 2009 (see p. 63, 189, and 233 for reading, math, and science non-aborigine/aborigine gaps, respectively)

Zubrick SR, Silburn SR, De Maio JA, Shepherd C, Griffin JA, Dalby RB, Mitrou FG,
Lawrence DM, Hayward C, Pearson G, Milroy H, Milroy J, Cox A. 2006. The
Western Australian Aboriginal Child Health Survey: Improving the Educational
Experiences of Aboriginal Children and Young People.
Perth: Curtin University of
Technology and Telethon Institute for Child Health Research. Volume III, Chapter 5.
Available at

(Summary: Can’t find any good data; generally, “visible minorities” underperform Whites)

Chung-Yan and Cronshaw, 2005. Information Exchange Article A Criterion-Related Validation Study of Transit Operators

Hausdorf et al., 2005. Cognitive Ability Testing and Employment Selection- Does Test Content Relate to Adverse Impact?

Jain, et al. Recruitment, selection and promotion of visible-minority and aboriginal police officers in selected Canadian police services

Finnie and Meng, 2003. Minorities, Cognitive Skills and the Incomes of Canadians

Vanderpool and Catano, 2008. Comparing the Performance of Native North Americans and Predominantly White Military Recruits on Verbal and Nonverbal Measures of Cognitive Ability

(Summary: Turks don’t seem to do so bad.)

Dustmann et al., 2012. Education of second generation immigrants

(Summary: Gaps are narrowing; some differences are clearly due to psychometric bias; minorities over-perform on measures of reaction time.)

Gijsberts, et al., 2012. Bijlagen Jaarrapport integratie 2011

Helms-Lorenz et al., 2002. Cross-cultural differences in cognitive performance and Spearman’s hypothesis- g or c?

de Meijer et al., 2008. Criterion-Related Validity of Dutch Police-Selection Measures and Differences Between Ethnic Groups

Stevens et al., 2009. Researching race:ethnicity and educational inequality in the Netherlands- a critical review of the research literature between 1980 and 2008

Meijer and Marise, 2006. Applicant and Method Factors Related to Ethnic Score Differences in Personnel Selection: A Study at the Dutch Police

te Nijenhuis et al., 2004. Are Cognitive Differences Between Immigrant and Majority Groups Diminishing?

(Summary: Ethiopian immigrants underperform other Israelis by around 1.3 SD on PISA tests.)

Nesher, 2012. Report shows widening gap between Ethiopian pupils and other Israelis. Haaretz.

(Summary: Ambiguous results; generally UK blacks perform below UK whites; “Asian” scores are often not disaggregated)

See references in:

Occidentalist, 2011. The General Mental Ability (GMA) of Black British

Occidentalist, 2012. Is (global) “race realism” still tenable?

Occidentalist, 2012. Partially falsified

Frederickson and Petrides, 2005. Ethnic, gender, and socio-economic group differences in academic performance and secondary school selection: A longitudinal analysis

USA (immigrant data)
(Summary: 2nd generation Blacks and Hispanics underperform 2nd and 3rd generation Whites and Yellows.)

See references in:

Occidentalist, 2012. More immigrant IQ

South Africa
(Summary: Gaps, as reported in IO, are smaller than as reported in Lynn (2008).)

See also discussion in Baron et al. (2005) (above)

Kriek and Dowdeswell. Adverse Impact & Validity Evidence in South Africa- 12 Years of Data Trends

Quirion, 2011. The predictive validity of psychological test battery for the selection of cadet pilots in a commercial airline

(Summary: Gaps clearly due, in part, to psychometric bias.)

Kvist, 2011. Interpretation of Cognitive Test Scores in Relation to Swedish and Immigrant Groups

Kvist and Gustafsson, 2008. The relation between fluid intelligence and the general factor as a function of cultural background: A test of Cattell’s Investment theory

New Zealand
(Summary: Relative to Europeans, Pacific Islanders perform as they do in the US)

Telford and May, 2010. Pisa 2009: Our 21st century learners at age 15 (p. 12, 28, and 38 for reading, math, and science respectively)

This entry was posted in Uncategorized. Bookmark the permalink.

23 Responses to Gaps around the world

  1. catperson says:

    Any particular reason why the “Super-duper selection” thread was deleted?

    As for Lynn’s 2010 challenge, the white > South Asian IQ gap is not found in the United States, indeed it is reversed with South Asians scoring higher than whites.

    I have unpublished IQ data I collected myself comparing different races though sample sizes were small and the tests administered under very sub-optimum conditions.

    • Chuck says:

      “Any particular reason why….”

      It looked really bad — so I was kind of embarrassed about it, which is why I buried it here: If I get a chance, I’ll rewrite it.

      “I have unpublished IQ data…”

      Could you describe it some more? Or put it on an excel file and send it to me? (I don’t use the email associated with this blog, so, if yes, I’ll have to send you that email address.)

      • Catperson says:

        It might take a long time to put it in an excel file as it consists of several sub-scores and an overall score for hundreds of people. I also don’t know if you would want it as it was just an unpublished unprofessional test administered under unprofessional conditions to an adult sample of convenience of wildly different ages and occupations. It could never be published in a peer reviewed journal. Also my sample only consisted of about a dozen and a half unmixed blacks.

        Anyway, if you have any more questions about it, my email is:

    • Chuck says:

      Basically, I’m rather skeptical of super-duper selection explanations. It sounds like an epicycle. Given this logic, I don’t see why East Asians in the West, who are also super duper educationally selected (refer again to Docquier and Marfouk (2005)), don’t have super-duper IQs. I offered Steve a possible account here:
      But this presupposes that the National IQ score differences represent actual phenotypic g differences. This is something that has yet to be established.

      • Catperson says:

        I just don’t believe immigrants from china are anywhere near as selected as immigrants from third world places like India and sub-sahara.   Indians and Africans moving to the west represent a rags to riches story of upward mobility so of course they are going to be more impressive than the people in their home countries, but going from a rich country like china to another rich region like the west is not enough of an achievement to select china’s best and brightest. 

        Check out the following link.  It shows the education of immigrants from china who moved to Canada.  

        Table 3 in the total chinese column shows 13.3% of the Chinese immigrants to Canada have a Bachelors degree, 3.2% have a masters and 0.9% have a doctorate.  So a grand total of 17.4% of Chinese immigrants to Canada (during the 1980s and 1990s) have university degrees.

        Meanwhile roughly 8% of china is college educated,

        So a college degree puts you at the 92%ile (+1.38 sigma) of education for china but at the 84%ile (+1 sigma) for Chinese immigrants (to Canada).  That’s a difference of only 0.38 SD in education.  That’s vastly smaller than the 1.63 SD education gap you calculated (in the comment section of the super-duper selection thread) between sub-Saharans in Africa and sub-Saharan immigrants to the UK.   So immigrants from china are simply not analogous to immigrants from sub-sahara because those who get out of Africa are going somewhere better and thus are better than who can’t leave.  By contrast, china is almost the world’s greatest power, so those who leave are not going anywhere better so they are hardly more impressive than the average Chinese.

        Also, the Chinese in north America are not just the children of recent Chinese immigrants.  Many are descended from low status laborers who came to north America during the gold rush back in 1849, and if anything, these were NEGATIVELY selected for intelligence.


      • Steve Sailer says:

        East Asian test scores are pulling away over the last 10 years in California.

  2. Kiwiguy says:

    @ Catperson,

    Jason Richwine had an article in Forbes about Indians being the “new model minority”. Sailer also blogged about Richwine’s findings.

    • JL says:

      On the CAT test, Indians in Britain score similarly to whites. On the WISC, Indians in India score about 20 points below the British mean. However, it’s difficult to say what this suggests about the “genetic IQ” of Indians because the difference could be due to immigrant selection, lack of measurement invariance between tests administered in the UK and India, adverse environmental conditions in India, etc.

      Chinese (in China) have apparently scored similarly or better than whites on IQ tests for decades, even though China was until recently as poor as India, so people from the developing world aren’t necessarily at a disadvantage when taking IQ tests. The difference between India and China could be that communists in China have built a more equitable society. Communism has traditionally been strong in the state of Kerala in India, and it has a first-world-like Human Development Index score, even though it’s dirt poor. Kerala’s mean IQ could be very informative about human capital in India in general.

      • Catperson says:

        I believe India has a genotypic IQ in the mid 90s. There are 2 reasons for adopting this estimate:

        (1)according to Lynn, sub-Saharan Africa has a phenotypic IQ of 67 but a genotypic IQ of 80. This suggests that the sub-optimum nutrition of third world countries subtracts 13 points from IQ. If 13 points have been subtracted from india’s measured IQ of 81, the genotypic IQ is 94.

        (2) Indians in south Africa (probably descended from average Indians) have an IQ 86, 3 points higher than coloreds who score 83. Since both Indians and coloreds in south Africa live in similar environments (more advantaged than blacks but less advantaged than whites) the 3 point IQ gap favoring Indians is probably genetic. However the genotypic IQ of coloreds is probably 90 since Lynn argues this is the genotypic IQ of mulattos. This gives a genotypic IQ of 93 (90+3) for Indians.

      • Jason Malloy says:

        The Chinese had high IQ well before Communism (IQ studies span back to the early 20th century). However these samples are often urban; there are plenty of lower IQ rural pockets. The difference is that are few to no pockets of high IQ in India. Outside of India, OTOH, Indian immigrants score better than whites in a number of nations.

        Kerala norms on the progressive matrices in 1970 were about 10th percentile. Matrices studies from the 2000s don’t seem to reveal much improvement.

  3. JL says:

    Racially disaggregated PISA results for New Zealand are available here (p. 12, 28, and 38 for reading, math, and science respectively).

    • Chuck says:


      Can anyone get a free copy of this paper: “Changes in student populations and average test scores of Dutch primary schools”?

      I would like to see how West Indian Blacks are currently doing in Holland.

      • JL says:

        Chuck, I have access to that paper, but it does not contain test scores by ethnicity; all students with at least one non-Western parent are lumped together, and the comparisons are between schools with different percentages of minorities, not between ethnic groups.

    • Chuck says:

      The Pacific Islander-European gap looks to be about 1 SD (see the SDs here, for comparison). This is comparable with the Pacific Islander-Euorpean US NAEP gap. In the Australian data, the European-Aboriginal gap is much smaller than one might expect, were Lynn’s estimated IQs accurate. How admixed are aborigines these days?

      • JL says:

        About 70% of Y haplogroups in Aboriginals are non-indigenous. Autosomally they’re probably a lot more indigenous than that.

        • Chuck says:


          Could you help me look for more recent CITO scores? See here — post under construction. I think Holland might be the first multicultural society to substantially violate Lynn’s theory (given, Lynn’s own falsification criterion, estimates, etc.)

          I think I might be able to put a substantial genetic hypothesis to rest.

      • Gilbert Pinfold says:

        There are two questions I would ask before accepting the summary on the Australian data. One: how aboriginal are ‘aborigines’? The ‘one drop rule’ is back in vogue for self identification. And two: how white are non-aborigines? Migration has been heavily third world in the last decade or so.

        On the South African data, the white flight, brain-drain effect is likely significant.

  4. Chuck says:

    What’s the fair market price for a mole of ATP?

    Thanks for offering but, no, that’s not an issue.

  5. Catperson says:

    The figure of 64 (U.S. norms) or 62 (white norms) comes from Richard lynn’s review of the literature in the 2006 book “races differences in intelligence”. An average IQ of 62 may sound low, but rushton found that Roma have an average IQ around 70. Also against today’s norms, whites from WWI probably had IQ’s around 80, so when you put it in context it’s not low. Lynn claims Bushmen have an average IQ of 54.

  6. Ryan Faulk says:

    Richard Lynn is wrong, it wouldn’t be falsified, since an immigrant group can be an elite sample. And there’s never zero selection effects on immigrants.

  7. 猛虎 says:

    Apparently, according to Karin Taht and Olev Must (2013)

    PISA scores are not comparable, i.e., measurement invariant. Their model 3 which sets invariance in both factor loadings and intercepts shows poor fit :

    The third model, M3 for scalar invariance, has fit indices which do not show adequate fit on the data. Changes in goodness of fit indices showed that changes were too large (for goodness of fit) between M3 and M2 (Table 5). Thus, there is no scalar invariance across the 55 nations using the PISA scales of educational achievement and learning attitudes. Or in other words, the measurement is invariant up to factor loadings.

    • Chuck says:

      Thanks. I saw the paper but I didn’t have access to it. From what I have read it’s difficult to establish MI with international studies. So, it would have been surprising if MI was found. That said, this is kind of another blow to Lynn’s hypothesis. See also: Meisenberg, G., & Woodley, M. A. (2013). Are cognitive differences between countries diminishing? Evidence from TIMSS and PISA. Intelligence.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s