Arthur R. Jensen (2000) A Fuzzy Boundary of Racial Classification Attenuates iq Difference. Psycoloquy: 11(022) Intelligence g Factor (34)

Volume: 11 (next, prev) Issue: 022 (next, prev) Article: 34 (next prev first) Alternate versions: ASCII Summary
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 11(022): A Fuzzy Boundary of Racial Classification Attenuates iq Difference

Reply to Jorion on Jensen on Intelligence-g-Factor

Arthur R. Jensen
Educational Psychology
School of Education
University of California
Berkeley, CA 94720-1670


Because about 25 percent, on average, of the gene pool of the American Black population is derived from European ancestry, the definitional boundary between the White and Black populations in the United States is said to be "fuzzy" in terms of the proportions of African and European admixture of any given person who is self-identified and socially perceived as Black. This fact does not in the least invalidate the mean differences in mental test scores observed between groups that are thus conventionally referred to as "Blacks" and "Whites." There is no merit to Jorion's (1999) claim that these facts somehow invalidate or render artifactual the reported average phenotypic White-Black differences on IQ and other mental tests.


behavior genetics, cognitive modelling, evoked potentials, evolutionary psychology, factor analysis, g factor, heritability, individual differences, intelligence, IQ, neurometrics, psychometrics, psychophyiology, skills, Spearman, statistics
1. America's Black population, with approximately 25 percent European admixture, received most of its European genes during the period of slavery, and ever since then the Black population has become increasingly homogeneous genetically; that is, across many generations the variance in the degree of admixture has decreased in the Black population and genetic linkage has approached equilibrium. Hence designations of racial admixture such as mulatto, quadroon, and octoroon are virtually inapplicable today.

2. In every large-scale mental testing of representative samples of non-Hispanic Whites and Blacks in the United States from World War I to the present, there has been found a White-Black (W-B) mean difference of at least one standard deviation. This is equivalent to an IQ difference of about 15 IQ points. In normalized distributions, this means that approximately 16 percent of Blacks score above the White mean and 16 percent of Whites score below the Black mean. There is 60 percent total overlap of IQ scores between the White and Black distributions. Equating the racial groups for socioeconomic status reduces the mean difference to about 12 IQ points. A difference at least that large is found in virtually every comparison ever published of reasonable sized W and B samples, at every age above 3 years, within every level of education, socioeconomic status, and occupation, in every region of the country, and at every time in history since mental ability tests were invented. I am referring here to phenotypic differences, that is, differences in test scores per se, without any inference about their cause. The evidence for these phenotypic differences and their universality is so overwhelming that there is no longer any controversy about these facts among those who are familiar with the evidence, such as the committee of experts appointed by the American Psychological Association to evaluate the evidence on this and other issues related to intelligence (Neisser, et al., 1996).

3. It is also a fact that the size of the W-B mean standardized difference is not at all uniform across different mental tests, but differs quite markedly across various tests. I (Jensen, 1998, Chapter 11; 1999; Nyborg & Jensen, 2000) have shown conclusively what Spearman (1927, p. 379) originally hypothesized: the size of the standardized W-B difference on various tests is a direct function of the tests' g loadings. W-B differences are small or nonexistent on other psychometric factors. This finding has been acknowledged as one of the "knowns" about the W-B difference (Neisser et al., 1996). From the regression of the W-B differences for a great many diverse tests on the g loadings of those tests, we can statistically estimate the size of the mean W-B difference on a hypothetically pure measure of g. This difference is about 1.3 standard deviations, which is equivalent to about 20 IQ points. Cultural bias has been ruled out as a cause of this difference (Jensen, 1980; 1998, pp. 360-369).

4. I take up the kind of criticism raised by Jorion (1999) in my earlier book "Educability and Group Differences" (Jensen,1973). It is pertinent to the whole of Jorion's argument. There I wrote:

    "Those social scientists who insist that there are no racial
    genetic differences in ability are often the most critical of
    studies which have used a social criterion of race rather than more
    precise genetic criteria. . . .  [These critics} seem not to have
    considered the idea that if the observed IQ differences between
    racial groups are due only to social-environmental factors, then
    the social definition of race should be quite adequate, and, in
    fact, should be the only appropriate definition. If it is argued
    that two socially defined racial groups which differ in mean IQ are
    not racially "pure" and that one or both groups have some genetic
    admixture of the other, it can mean only that the biological racial
    aspect of the IQ difference, if such exists, has been
    underestimated by comparing socially, rather than genetically,
    defined racial groups" (p. 219).


Jensen, A.R. (1973). Educability and group differences. London: Methuen.

Jensen, A.R. (1980). Bias in mental testing. New York: Free Press.

Jensen, A.R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger.

Jensen, A.R. (1999). Precis of: "The g Factor: The Science of Mental Ability" PSYCOLOQUY 10(23). psyc.99.10.023.intelligence-g-factor.1.jensen

Jorion, P.J.M. (1999). Intelligence and race: The house of cards. PSYCOLOQUY 10(064). psyc.99.10.064.intelligence-g-factor.12.jorion

Neisser, U., Boodoo, G., Bouchard, T.J., Jr., Boykind, A.W., Brody, N., Ceci, S.J., Halpern, D.F., Loehlin, J.C., Perloff, R., Sternberg, R.J., & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 66-101.

Nyborg, H., & Jensen, A.R. (2000). Black-white differences on various psychometric tests: Spearman's hypothesis tested on American armed services veterans. Personality and Individual Differences, 28, 593-599.

Spearman, C.E. The abilities of man. London: Macmillan.

Volume: 11 (next, prev) Issue: 022 (next, prev) Article: 34 (next prev first) Alternate versions: ASCII Summary