← Autodidact Archive · Original Dissent · FadeTheButcher
Thread ID: 10208 | Posts: 1 | Started: 2003-10-02
2003-10-02 22:37 | User Profile
Lets go through the cultural theories one by one.
Miele: There are hundreds if not thousands of articles and books that say exactly the opposite. We can't examine every nongenetic or culture-only explanation for the Black-White difference in average IQ, but I'd like to present ten of the best-known, one-by-one, and have you respond with the evidence you believe disproves them.
Jensen: Go ahead.
Miele: Culture-Only Theory #1: Blacks and Whites differ significantly in their average socioeconomic status (SES), and since SES is a determinant of IQ, it explains the average Black-White IQ difference.
Jensen: Racial differences in SES cannot explain the average IQ difference. When statistical procedures are used to remove the effect of the difference in SES or when Blacks and Whites are simply matched on measures of SES, the Black-White IQ difference is reduced, but only from 15 to 12 points. And not all of that three-point reduction is due to SES, because SES differences within each racial group also have some genetic component. Therefore matching Blacks and Whites on SES to some extent also matches them genetically in terms of the g factor.
Miele: Culture-Only Theory #2: School facilities of Blacks are well below those of Whites.
Jensen: Educational inequality can't explain away the average IQ difference either. In the last 30 years or so many school systems have been racially integrated and now provide the same facilities and instructional programs for Blacks and Whites alike, yet the average differences of about one standard deviation in IQ and scholastic achievement remain.
Miele: Culture-Only Theory #3: IQ only measures knowledge of the "core culture" and therefore the tests are inherently biased against minorities.
Jensen: The claim that the Black-White IQ difference is a result of culturally biased tests has been disproved. A detailed explanation, which requires a working knowledge of psychometrics, is presented in my 1980 book Bias in Mental Testing. But you need not take my word on it. Following publication of my book, a special committee of the National Academy of Sciences and the National Research Council examined the question and reached essentially the same conclusions.
The most widely used mental tests today have the same reliability for Blacks and Whites, which means that if you give the test to a group of people and then test them again at a later date, they get about the same scores. The important point is that there is no evidence that the test scores for Blacks are unstable or erratic. Whatever the tests measure, they measure it just as reliably for Blacks as they do for Whites.
Mental tests also have the same predictive validity for Blacks as they have for Whites. This means that they predict other important real-life criteria, such as school grades and job performance, with the same accuracy for both groups. If you are trying to predict how well someone will do in college and they have an IQ of 125, it makes no difference whether they are Black or White, or anything else.
Tests also have the same factor structure in both groups. So if you factor analyze the test scores of either Blacks or Whites on a battery of mental tests, you will find the g factor at the top, followed by the group factors, and then the special factors. The g factor is indeed real and just as important for Blacks as it is for Whites, and indeed, for any group.
The item-to-item correlations are the same for both groups, and so is the rank order of item difficulties. Simply stated, the items that are easiest for whites are also the ones that are easiest for Blacks. This is important because it would not be true if some types of items were specially biased against Blacks - the way vocabulary items, for example, are biased against recent immigrants who are unfamiliar with English. Blacks and Whites even make the same types of errors and get fooled into picking the same distractor items in multiple-choice tests.
The evidence on each of these points is so overwhelming that no one in the field any longer argues the point.
Miele: Culture-Only Theory #4: African-Americans are being tested in a language other than their own.
Jensen: Insufficient familiarity with standard English and the use of "Black English" was a popular claim in the 1960s and 70s. But the Black-White IQ differences are as large or larger on a variety of non verbal tests that make no use of alphanumeric symbols as on verbal tests. And children who are born deaf and hence have had virtually no exposure to spoken language do not show any deficiet on non verbal tests.
Miele: Culture-Only Theory #5: Nutrition plays an important role in mental development, as does exposure to toxic chemicals and Blacks and Whites differ on these measures also.
Jensen: I've never claimed that the Black-White difference in average IQ is 100 genetic, only that both genetic and environmental factors are involved, just as with individual differences within each race, and probably to around the same degree. Nutritional factors do account for some part of the average racial IQ difference. Even when there is no evidence of poor nutrition, however, there is still a Black-White IQ difference.
In the Black underclass, nutrition does have a measurable effect on IQ. On a per-capita basis, prematurity and low birth weight are much more prevalant amongst Blacks than Whites. The difference between mother's milk and baby formulas also makes a significant difference in the IQs of low-birth-weight infants of either race by the time they reach school age. Unfortunately, at this point in history, a smaller percentage of African-American mothers breast feed their babies. Fortunately, that's one thing that could probably be changed at relatively little cost by making the information better known in communites, Black or White, that are at highest risk for low birth weight babies.
Miele: Culture-Only Theory #6: Blacks and Whites differ markedly in their historical experience.
Jensen: To my knowledge, no one has ever demonstrated that a group’s past history, independent of its earlier genetic history, affects their present-day average IQ. Some racial and ethnic minorities that historically have been victimized by discrimination and persecution, such as Jews in Europe, East Indians in Africa and in Britain, and Chinese and Japanese in the United States, actually have higher average IQs than the White or other majority population they live among.
Miele: Culture-Only Theory #7: The totality of these cultural, environmental, and nutritional factors interact step by step, and nutritional factors interact step by step, from conception to adolescence, to construct our entire cognitive structure.
Jensen: This hypothesis could be tested by rearing black children in middle-class or upper middle-class White families. That is what was done in the Minnesota Transracial Adoption Study. The researchers compared infants with two Black parents (BB or Black in everyday parlance), infants with a White mother and a Black father (WB, or mixed race), as well as a control group of white children (WW), all adopted into two-parent White upper-middle-class homes. The adoptive parents were mostly college graduates with managerial and professional jobs.
All of the adoptees, Black (BB), mixed race (WB), and White (WW), were given IQ tests and scholastic achievement tests at age seven years and again at age 17. When tested at age 7, average IQs for the BB and WB children were several points higher than the average Black children reared in the same community, indicating a beneficial effect of the middle-class, White home environment on the IQ of these adoptees. By age 17, however, the average IQ for the Black adopted children was about 16 points below the White average. This is not significantly different from the national average IQ for Black youths. So even growing up in a White middle-class home did not produce a lasting reduction in the familiar one standard deviation Black-White difference in average IQ.
Miele: Culture-Only Theory #8: The lower average IQ of Americans of African ancestry is the result of racist America’s “self-fulfilling prophecy” of discrimination against Blacks and even supposedly “scientific theories” about “genetic inferiority” based on the color of their skin, not the level of the g factor.
Jensen: To some extent you can test hypothesis as well by a more detailed analysis of the results of the Minnesota Transracial Adoption Study. The average IQ of the mixed race (WB) adoptees, when they were tested at age 7 and again at age 17, was just about halfway between the average IQs of the WW adoptees and of the BB adoptees. Yet there was no mistaking the African ancestry of the mixed-race children from their appearance. In fact, a WB child would probably be considered Black in America today, as are well known individuals of mixed ancestry such as Halle Berry or Tiger Woods. Its hard to explain the intermediate position of the WB adoptees in purely environmental terms, including the consequences of societal racism, but it is what you could predict beforehand from the Default Hypothesis. The complete results of the Transracial Adoption Study showed no evidence that, by age 17, being raised from infancy in an upper middle class White family raised the Black adoptees average IQ or their overall level of scholastic performance above that of Black children reared by their biological parents.
Miele: Culture-Only Theory #9: Experimental programs of intensive early cognitive intervention such as the classic "Miracle in Milwaukee" successfully increased the IQ of Black children.
Jensen: The results of these experimental attempts to raise the IQs of Black infants at risk for Low IQ are quite consistent with the results of the Transracial Adoption Study. In the highly publicized Milwaukee Project, a trained staff gave Black children from poor homes intensive all-day environmental enrichment and training in mental skills from infancy to age six, at which time they entered regular public schools. The special training raised IQs quite markedly above those of a control group of similar children who did not receive the training. But the training did not significantly raise the scholastic achievement and other kinds of performance that are typically correlated with IQ. So the gains in the test scores were "hollow" with respect to the g factor. The most likely explanation is that the results came from "teaching the test," and not really raising the level of g.
Miele: Culture-Only Theory #10: The Milwaukee Project was an early attempt at experimental cognitive intervention and naturally had its limitations. But the more recent Abecedarian Project produced lasting gains in the IQ and school achievement of at-risk Black children. So both the claim in your 1969 "HER article that "compensatory education has been tried and apparently it has failed*" and your Default Hypothesis about intelligence, race, and genetics have now been disproved.
Jensen: You're correct that the criticisms of "teaching the test" and "hollow gains" cannot be made of the more recent Abecedarian Project, which involves intensive and prolonged educational training of children at risk for low IQ. The project raised the IQ of the children who received it about five points, on average, above a control group, who did not. You're also correct that his gain still held up when the children were retested at age 14 and that it was accompanied by a comparable gain in scholastic performance. And this is all well and good. But it should also be noted that even the most intensive cognitive intervention program yet devised provided during all the children's preschool years only rduced the national Black-White difference in average IQ by about one-third (that is, 5 out of 15 points). This is the best evidence we have of the extent in which improving the cognitive environment of at-risk groups can increase their IQ. The results of this Abecedarian Project in no way disprove the Default Hypothesis of genes and environment; they are fully consistent with it.
Frank Miele, Intelligence, Race, and Genetics: Conversations with Arthur R. Jensen (Oxford, 2002), pp.127-33