User:Klortho/snapshots/Race and intelligence (test data)

The gaps found between the average intelligences of races or ethnicities varies depending on methods used for racial grouping, the method and setting used to test intelligence, the health and economic situation of the test takers, the interplay between the culture of the person taking the test and the culture of those who made the test, and the period in history when the test was performed.

Depending on the way intelligence is measured a variety of gaps may be found between different racial and ethnic groups. Some groups that perform well on one task may do poorly on others. For example, Moroccan and North American individuals were asked in a study by Richard K. Wagner to remember patterns of Oriental rugs and pictures of everyday objects, such as a rooster and a fish. Moroccans, who have long experience in the rug trade, seemed to remember rug patterns better than the North American individuals. Likewise, in 1979 Robert Serpell had Zambian and English children perform a number of tasks. He found that English children did better on a drawing task, but that Zambian children did better on a wire-shaping task.

Over the years, there has been variation in both the observed average IQ of groups, as well as the relative relationships between the average IQ of groups. Early 20th century measures typically found Blacks on the low end, and Whites on the high end. Based on studies from the 1960s and 1970s, Flynn found a slightly lower average IQ of Japanese- and Chinese American children compared to White counterparts, albeit with a significantly higher standard deviation. Recent contemporary measures place Blacks on the low end, and Asians on the high end. In the United States, based on a national assessment of adult literacy African Americans improved their scores more than any other racial/ethnic group in the years between 1992 and 2003. The survey measured three elements of literacy: prose, document, and quantitative literacy--which are reading, synthesizing information from documents and graphs, and basic math.

In the past 20 years the use of the IQ test as the sole measure of intelligence and the ability of intelligence tests to predict intelligence between people of different cultural backgrounds has fallen under increasing criticism.

“Perhaps the best way to achieve coherence in the field of intelligence is to recognize that no single correct “model” or “approach” is evident and that different ones elucidate different aspects of a very complex phenomenon (Sternberg, 2003).”

Direct and indirect measures of intelligence




The modern controversy surrounding intelligence and race focuses on the results of IQ studies conducted during the second half of the 20th century, mainly in the United States and some other industrialized nations. On average, a difference of approximately one standard deviation was observed in the US between the mean IQ score of Blacks and Whites as adults. Whether these gaps have narrowed or not, especially regarding children which may or may not later also cause a narrowing for adults, is, as noted below, debated.

Attempted compilations of average IQ scores by race published by Richard Lynn in Mankind Quarterly and Murray, C. and Herrnstein, R. J. in their book The Bell Curve place Ashkenazi Jews and East Asians at the top, followed by Whites, Arabs and Native Americans, sub-Saharan Africans and Australian Aboriginals.

In modern western science controversy surrounding race and intelligence focuses on the results of IQ studies conducted during the second half of the 20th century, mainly in the United States and some other industrialized nations. IQ studies outside these nations are few and small. It is uncertain what the average IQ or subgroup IQ tests scores would be with more complete studies in the developing world. IQ test scores in the developing world are affected by factors less important in the developed world such as nutritional deficiencies (see Health and intelligence) Most of the remainder of this article refers to studies attempting to explain race differences in IQ test scores in the US, and does not refer to the world as a whole.

Over the years, there has been variation in both the observed average IQ of groups, as well as the relative relationships between the average IQ of groups. Early 20th century measures typically found Blacks on the low end, and Whites on the high end. Based on studies from the 1960s and 1970s, Flynn found a slightly lower average IQ of Japanese- and Chinese American children compared to White counterparts. Recent contemporary measures place Blacks on the low end, and Asians on the high end.

IQ test score gap in the US
In the United States, the mean IQ score among Blacks has at times been measured as approximately 85 and the mean IQ score among Whites has at times has been measured as approximately 100; the mean IQ score of Latinos has been reported to be measured as approximately 89 for unspecified dates.

The mean score for people of East Asian and Jewish descent is usually higher than the mean score of Whites. However, several studies place the median IQ of Ashkenazi Jews (who make up the overwhelming majority of American Jews) at approximately one standard deviation above the mean for other Whites, with the primary Jewish advantage in verbal reasoning and the East Asian advantage primarily in spatial reasoning. In The Bell Curve, Herrnstein and Murray report mean IQ scores for East Asians and Jewish Americans of 103 and 113, respectively (on a scale where Whites = 100).

Similar gaps are seen in other tests of cognitive ability, including university admission exams such as the SAT and GRE as well as employment tests for corporate settings and the military

IQ test data from various sources
NOTE: The information in the following tables is referenced to credible sources; however, it should be noted that some of these studies liberally synthesized their results from different sources and test methods. In general comparisons should only be made between similar primary sources, but this is not always possible given the wide variety of IQ tests, the fluid and debatable nature of racial categorization, and the lack of large scale representative data sets. Among the factors that invalidate comparisons across the studies are environment of the test subjects as well as inherent biases in the test procedures.

Employment tests and school achievement
Gaps are seen in other tests of cognitive ability or aptitude, including university admission exams such as the SAT and GRE, as well as employment tests for corporate settings and the military. Measures of school achievement correlate fairly well with IQ, especially in younger children. In the United States, achievement tests find that by 12th grade black students are performing on average only as well as white students in 10th grade, and Asian students in 8th grade; Hispanic students do only slightly better than blacks.

There is wide agreement that the U.S. Black-White gap among children and adolescents on achievement tests narrowed in the 1970s and 1980s, but stalled during the 1990s. On the basis of these data, argues that the gap in g has also narrowed to "0.6-0.7 standard deviation or approximately 10 IQ points". Both and  argue that "gains in scholastic achievement do not equal gains in g, and the Black-White differences in g are as large as ever, even for measures of reaction time". Charles Murray argues that the U.S. Black-White gap on the SAT has increased in size to 1 SD from the 1990s to 2005.

Measures of school achievement correlate fairly well with IQ (uncorrected correlations are highest for younger children because of restricted ranges of IQ at higher educational levels). The National Assessment of Educational Progress in the United States, find that by 12th grade Black students are performing on average only as well as White and Asian students in 8th grade; Hispanic students do only slightly better than Blacks. Closing this achievement gap is one of the aims of the No Child Left Behind act in the United States. The gap has been attributed to academic disengagement of Black students and parents and to environmental causes. Poverty in early childhood has long-lasting negative consequences for cognitive development and academic outcomes, as shown by numerous studies.

School achievement gaps between ethnic groups do not remain constant over the years a child spends in school. In a recently publicized 2005 study Deborah Wilson, Simon Burgess and Adam Briggs found that in the UK, Black children enter school at the same level as their White peers, if one controls for a number of covariates. Black children then fall behind, even controlling for covariates, in the first two years of schooling, and stay behind until secondary school where they make greater progress then their White peers substantially, but not completely closing the gap. "This suggests more systemic factors. One such often proposed is the importance of aspirations and values inculcated by families and reinforced by communities. This involves the importance of education in general, and the role of education in getting on." In the later years of schooling the study found that White pupils make slower progress than other ethnic groups.

Attempts to redress the achievement gap in the U.S. include Head Start and related early intervention programs. Neither Head Start nor most other (more intensive) programs have been able to produce lasting gains in IQ or school achievement. Gains are lost after the programs end. Supporters note that they continue as long as the programs continue. The Abecedarian Early Intervention Project is an exception to this pattern, producing an IQ boost of approximately 5 points into adulthood.

The achievement gap






Richard Nisbett and others have argued that the Black-White gap on various ability tests has narrowed from the 1970s to the 1990s, including the Equality of Educational Opportunity (EEO) survey, the National Longitudinal Study, the High School and Beyond survey, the National Education Longitudinal Study, and the National Assessment of Educational Progress program (NAEP). The Journal of Blacks in Higher Education found that although the Black-White gap on the SAT declined from 1976-1988, it has been increasing since 1988. Other studies find that the gap has been decreasing for most of the 20th century and that this trend continued during the nineties.

Jensen has argued that the Black-White differences in g have not narrowed. In support of this claim, he presents evidence that, while there have been gains in measures of acquired competency such as scholastic achievement, these improvements do not indicate gains in g. Jensen also argues that Black-White differences in g seen in measures of reaction time have not narrowed. A large meta-analysis using 6,246,729 samples found a mean Black-White score difference of 1.1 standard deviations (ranging from 0.38 to 1.46 depending on the g loading of the test). As to whether the IQ gap is narrowing, the authors speculated that any reduction was "either small, potentially a function of sampling error ... or nonexistent for highly g loaded" tests. Gottfredson agreed that the Black-White gap observed in the National Assessment of Educational Progress test has narrowed from 1.07 to 0.89 standard deviations. However, she then argues that reduction stopped by the mid-1980s and is compatible with stable group differences in g.

A large (21,260 children) and probably the most recent (1998) study found that the Black-White gap for young children in reading and math scores was much smaller than in earlier studies, and that all of the remaining difference could be explained by a few environmental factors. One possible explanation is that the Flynn effect started earlier for Whites but has now stopped, while continuing for Blacks.

There is disagreement about whether the results of IQ tests show a narrowing of the IQ gap, or if they do who has benefited. and argue that "the Black-White differences in g are as large as ever". found that the recent U.S. Black-White gap in g among adults is 1.1 sd, similar to characterization of the historical U.S. Black-White gap. In an analysis of standardization samples for the WAIS, WISC, SB, and AFQT, find evidence that the U.S. Black-White gap shrunk between 3 and 6 points from 1972 to 2002. These conclusions were challenged by, and a rejoinder was made by. Subsequently, analyzed testing data from the children of the 1979 NLSY cohort, concluding that "the B-W difference did not diminish on either academic achievement or cognitive tests for children born from the mid 1970s through the mid 1990s." To reconcile the contradictory findings, Murray suggests that "the effect that Dickens and Flynn found was concentrated among subjects born before the late 1970s". In 2006, Flynn and Murray debated the shrinking Black-White IQ gap. Both agree that the gap shrunk for children born before the late 1970s, however Murray believes the narrowing stopped while Flynn believes it has continued.

Several sources have argued that the U.S. Black-White gap varies with age. estimate that the 2006 U.S. Black-White IQ gap is 0.31 SDs at age 4, 0.63 SDs at age 12, 0.87 SDs at age 18, and 1.1 SDs at age 24. Using data from, estimates that the U.S. Black-White gap is 0.70 SDs in early childhood, 1.00 SDs in middle childhood, and 1.20 SDs in early adulthood. A study of children aged eight to twelve months found a U.S. Black-White gap of 0.06 SD.

"'The whole distribution of black cognitive ability is moving up relative to whites,' says Dickens. 'There's no reason to believe [the gap] isn't going to get more narrow as we move forward and as measures of social equality continue to improve.'"

World-wide IQ estimates
One review of the global cognitive ability data is Richard Lynn's 2006 Race Differences in Intelligence, which organizes the data by nine global regions, surveying 620 published studies from around the world, with a total of 813,778 tested individuals. Lynn's meta-analysis lists

Average IQ in Races (Note: Population samples don't have the same quality of education and living standards)
 * East Asians (105)
 * Europeans (99)
 * Inuit (91)
 * Southeast Asians and Amerindians each (88)
 * Pacific Islanders (85)
 * Middle Easterners (including South Asians and North Africans) (84)
 * sub-Saharan Africans (67) and
 * Australian Aborigines (62).

Searchlight magazine criticizes Lynn's publisher, which is not an academic press and for publishing “classic” Aryan and eugenic tracts. Lynn has previously argued at length that nutrition is the best supported environmental explanation for variation in the lower range, and a number of other environmental explanations have been advanced (see below). Ashkenazi Jews score significantly higher than other groups (107-115) in the U.S. and Britain (See Ashkenazi intelligence), but estimates of the average IQ of Ashkenazim in Israel may be somewhat closer to the European mean. In other data, Hispanics average 91 and African Americans average 87, though the latter is debated. Lynn's survey is an expansion by nearly four times of the data collected in his 2002 IQ and the Wealth of Nations with Tatu Vanhanen. 'IQ and the Wealth of Nations, which dealt with the relationship between IQ and economic development, received strong criticism from some for both error and alleged bias. Sociologist Thomas Volken argues the IQ and the Wealth of Nations data for national IQs is "highly deficient," citing limited sampling and varying tests and years (Volken). In a 1995 review of The Bell Curve, critic Leon Kamin writes that "Lynn's distortions and misrepresentations of the data constitute a truly venomous racism, combined with scandalous disregard for scientific objectivity.". In contrast to Kamin's strongly worded attack on Lynn, W. D. Hamilton described Lynn in a review of another of Lynn's books as doing "an excellent job with the facts" and being "brave [and] thick-skinned ... to swim against ... popular antirealistic currents." Examples of problematic national IQ figures in IQatWoN include that the stated average IQ score of 59 for Equatorial Guinea is based on one test of 48 children aged 10-14 in 1984; the Ethiopian average is derived from a study of Ethiopians who immigrated to Israel a year prior, and whose low scores were thought by the original authors to be a reflection of temporary adjustment to a different culture and language (note that this data is not used in the averages presented above). Kamin also argued Lynn selectively excluded data showing a similar score in Whites and sub-Saharan Africans: "Lynn chose to ignore the substance of Crawford-Nutt's paper, which reported that 228 black high school students in Soweto scored an average of 45 correct responses on the Matrices--HIGHER than the mean of 44 achieved by the same-age white sample on whom the test's norms had been established and well above the mean of Owen's coloured pupils". 's checking of RDiI's data finds discrepancies that are "mostly minor. . . typically within a couple of IQ points" but concludes: "The citations and references were, on the whole, accurate. In short: Yes, the general trends in the tables are probably dependable, if the assumptions regarding Flynn effects, etc., are correct, but it is prudent (as always) to check with original sources before quoting particular results. . . Is this book the final word on race differences in intelligence? Of course not. But Richard Lynn is a major player, and it is good to have his extensive work on this topic together in one place. Future workers who address these matters under this or any other label will find that Lynn has done a lot of spadework for them..." Lynn argues the surveyed studies have high reliability in the sense that different studies give similar results, and high validity in the sense that they correlate highly with performance in international studies of achievement in mathematics and science and with national economic development.

Related groups sometimes vary greatly in IQ in different nations. Black Africans score much lower than blacks in the US, although Black Americans average about 7-20% European admixture. Recent studies indicate that the black–white gap is smaller in the UK than in the U.S. Differences between groups of whites can also be seen, ranging to the low 90s in SE Europe (with a decrease also seen in brain size). In Israel, large gaps in test scores and achievement separate Ashkenazi Jews from other groups such as the Sephardi.

The IQ scores vary greatly among different nations for the same group. Blacks in Africa score much lower than Blacks in the US. Studies on African Blacks have yielded results of 65 and 69. Lynn (1991a) suggests this value should be rounded up to 70. The book IQ and the Wealth of Nations gives values for African nations ranging from 59 for Equatorial Guinea to 77 for Zambia.

Studies on Native Americans yield results ranging from 70 to 90. Lynn (1991a) suggests that the mean of several studies, 89, may be a reasonable approximation. Native Americans score higher on tests of spatial than verbal reasoning, a trait also shared with East Asians.

Some reports indicate that the Black–White gap is smaller in the UK than in the U.S. Other examples are Jews who score much lower in developing nations and Koreans who score much lower in Japan. There are also other examples of IQ score differences between close neighbors in the same nation, for example between French vs. Flemish speakers in Belgium, Slovaks vs. Gypsies in Slovakia, Irish and Scottish vs. English in Great Britain, and white speakers of Afrikaans vs. white speakers of English in South Africa. In Mexico, Whites score higher than Mestizos, who score higher than Native Mexican Indians.

There has been at least one comparative study on IQ scores in different European nations. The difference between the highest and lowest average national IQ scores is 13.3 points. There is a not statistically significant correlation between the country's latitude and its average IQ score. Also, Southern Europeans have a larger variation than Northern Europeans. Norwegians have less than half of the variability of Whites in the US while Italians and Bulgarians have about 150%.

In Israel, there is a large gap in test scores and achievement between the Ashkenazi Jews and other groups such as the Sephardi. Gaps are seen between Israeli Jews and Ethiopian Jews. A similar gap in IQ scores and achievement is seen between the Chinese and Malays in Singapore.

Both Lynn and Rushton have suggested that high IQ is associated with colder climates. To test this hypothesis, Templer and Arikawa (in press [doi:10.1016/j.intell.2005.04.002]; see discussion ) compare the national IQ data from Lynn and Vanhanen with data sets that describe national average skin color and average winter and summer temperatures. They find that the strongest correlations to national IQ were −0.92 for skin color and −0.76 for average high winter temperature. They interpret this finding as strong support for IQ-climate association. Other studies using different data sets find no correlation.

World-wide scores
Cognitive ability scores for the ten global genetic clusters identified in previous genetic cluster analysis have been surveyed by Richard Lynn. Lynn regards these ten population groups as races, and extracts racial averages from 620 published studies surveying a total of 813,778 tested individuals. When taken as individual national averages, the data available, particularly regarding the developing world, is speculative due to limited sampling, year of testing, and varying type of cognitive ability test used. Lynn's previous work, including the trade book IQ and the Wealth of Nations with Tatu Vanhanen, have received widespread and strong criticism for both bias and error.

In general, Lynn lists East Asians as demonstrating the highest average IQ, indigenous Americans and other Eurasians with intermediate average IQ, and Africans and Australian Aborigines with the lowest average IQ. According to Lynn, when the studies are grouped and taken as averages for the ten racial groups, the argument for their reliability is that, though additional evidence may be required to confirm some of the more limited estimates, many have very high reliability in the sense that different studies give similar results, as well as that they correlate highly with performance in international studies of achievement in mathematics and science and with national economic development. Lynn argues established environmental hypotheses can explain a substantial amount of these differences. The data set for sub-Saharan Africans around the world, the one most often criticized, is drawn from 155 different studies with a combined sample of 387,286 people. 57 of the studies are from countries in Africa, 54 from western countries, and 16 from non-western countries outside of Africa.

Some Ashkenazi Jews score significantly higher than any other group. An IQ of 70 is often associated with mental retardation, but deficits in adaptive behavior, such as telling the time, interacting socially, and looking after oneself, have only limited correlation with IQ, and are more important than IQ in determining whether a person is able to live an independent life.

IQ scores vary greatly among different nations for related groups. Blacks in Africa score much lower than Blacks in the US. Some reports indicate that the Black–White gap is smaller in the UK than in the U.S. American Blacks average about 7-20% European admixture; UK admixtures are not as well-studied. Many studies also show differences in IQ between different groups of Whites. In Israel, large gaps in test scores and achievement separate Ashkenazi Jews from other groups such as the Sephardi.

IQ and Retardation
The low average IQ of Blacks in Sub-Saharan Africa is often met with incredulity. Among whites, most people with an IQ below 70 are considered retarded. However among groups with lower average IQs this association is not valid, because mental retardation describes both an inability to solve problems (what IQ measures) and an inability to perform everyday tasks (e.g. daily living skills, communication skills, and social skills). Many whites with very low IQ have a genetic disorder (e.g., Down syndrome), whereas that level of reasoning ability is nearer to the average for sub-Saharan Africans. Jensen (1998) describes this phenomenon at length, including his great surprise at finding that people with normal social skills can perform so poorly on tests of fluid reasoning.

An IQ of 70 is often associated with mental retardation, but deficits in adaptive behavior, such as telling the time, interacting socially, and looking after oneself, have only limited correlation with IQ, and are more important than IQ in determining whether a person is able to live an independent life (, p. 177). reported that on the basis of IQ alone 10 times as many Blacks as Whites would be classified as retarded, but when adaptive behavior measures are added to the criterion, this difference completely disappears. Some ethnic differences in cognitive ability appear in some aspects of cognitive ability and not others (see below;, p. 178). The Black-White disparity seen in IQ does not appear in some basic cognitive functions that don't involve more than minimal elaboration, transformation, or other mental manipulation (, p. 178; ).

High-achieving minorities
The book World on Fire notes the existence in many nations of minorities that have created and control a disproportionate share of the economy, a market-dominant minority. Examples include Chinese in Southeast Asia; Indians in the United States and Britain; Whites, Indians, African immigrants to the US and UK, Lebanese and Igbo people of Western Africa; Whites in Latin America; and Jews in pre-World War II Europe, modern America, and modern Russia. These minorities are often resented and sometimes persecuted by the less successful majority.

In the United States, Jews, Asian Indians, Japanese, and Chinese earn incomes 1.72, 1.42, 1.32, and 1.12 times the American average, respectively. Jews and East Asians have higher rates of college attendance, greater educational attainment, and are many times overrepresented in the Ivy League and many of the United States' most prestigious schools, even though affirmative action discriminates against Asians in the admissions process (relative to Whites as well as to other minorities) At Harvard, for example, Asian American and Jewish students together make up 51% of the student body, though only constituting roughly 6% of the US population. In various Southeast Asian nations, Chinese control a majority of the wealth despite being a minority of the population and are resented by the majority, in some cases being the target of violence.

African Immigrants to the US and UK
African immigrants to the US have the highest educational attainment rates of any immigrant group in the United States with higher levels of completion than the stereotyped Asian American model minority, raising further questions about the benefits of affirmative action programs based on race as well as stereotypes about the intellectual capacity of races. The data used in the study, however, does not differentiate between African immigrants to America who are native to Africa or are foreign settlers who were displaced due to conflicts in their adopted countries. However, Asian-American includes all South, Southeast, and East Asians which encompasses an extremely large and diverse population. Despite the ongoing controversy about IQ difference in the US. Gargi Bhattacharyya, Liz Ison and Maud Blair have found that IQ differences between black and white populations in the UK and elsewhere are virtually non-existent(cite?). However, other data contradicts their findings. Black Caribbean and Black African children and children for whom English is an additional language make relatively greater progress during preschool than White children or those for whom English is a first language. . In a study by done by Dennis Saccuzzo on information-Processing for African-American, Latino, Filipino, and White elementary school children enrolled in gifted versus non-gifted school programs, gifted African-Americans showed the fastest reaction times of the ethnic groups tested while nongifted African-Americans showed the slowest.

SOURCE: 2000 US CENSUS

Jews
Achievement in science, a high-complexity occupation in which practitioners tend to have high verbal IQ, also appears consistent with some group IQ disparity.

Only 0.25% of the world population is Jewish, but Jews make up an estimated 28% of Nobel prize winners in physics, chemistry, medicine, and economics. In the U.S., these numbers are 2% of the population and 40% of winners.

Their high verbal-sequential intelligence might explain their overrepresentation at chess. Over half of the world chess champions from 1886 to 2000 had at least one Ashkenazi Jewish parent.

Some studies have shown significant variation in IQ subtest profiles between groups. In one analysis of IQ studies on Ashkenazi Jews, for example, extraordinarily high verbal and mathematical scores, but only average visuospatial scores were found.

East Asians
East Asians demonstrated high visuospatial scores

Test bias
It has been suggested by Adrian Dove, Nathaniel D. McNiel, Karen Phalet, and others that IQ tests may be biased against minorities, and that this accounts for part or all of the IQ gap. Some claim that there is no evidence for test bias. IQ tests are equally good predictors of IQ-related factors (such as school performance) for U.S. Blacks and Whites. The performance differences persist in tests and testing situations in which care has been taken to eliminate bias. It has also been suggested that IQ tests are formulated in such a way as to disadvantage minorities. Controlled studies have shown that test construction does not substantially contribute to the IQ gap. Studies have shown that the race of the test administrator does not have an effect on the U.S. Black-White gap. Still, a 2007 study at Case Western Reserve University found that cultural differences in the provision of information account for racial differences in IQ. The study also found that test problems, similar to some problems found on conventional IQ tests, were only solvable on the basis of specific previous knowledge. Such specific knowledge based questions showed evidence of test bias since the performance on non-specific knowledge based questions did not always correlate with the performance on the knowledge based question.

How test bias works
The Black Intelligence Test of Cultural Homogeneity, or BITCH-100, was created by Robert Williams in 1972, and is oriented toward the language, attitudes, and life-styles of African-Americans. Unsurprisingly, white students perform more poorly on this test than blacks, indicating that there are important dissimilarities in the cultural backgrounds of blacks and whites. Some argue that these findings indicate that test bias plays a role in producing the gaps in IQ test scores. Similarly to the Williams test, the Chitling Intelligence Test is another example of a culturally biased test that tends to favor African Americans. Both of these tests demonstrate how cultural content on intelligence tests may lead to culturally biased score results. Still these criticisms of cultural content may not apply to "culture free" tests of intelligence. The BITCH-100 and the Chitling test both have explicit cultural assumptions, while normal standardized tests are only theorized to have implicit bias. The fact that a test can have bias does not necessarily prove that a specific test does have bias. However, even on cultural free tests, test bias may play a role since, due to their cultural backgrounds, some test takers do not have the familiarity with the language and culture of the psychological and educational tests that is implicitly assumed in the assessment procedure. Beverly Daniel Tatum writes that dominant cultures often set the parameters by which minority cultures will be judged. Minority groups are labeled as substandard in significant ways, for example blacks have historically been characterized as less intelligent than whites. Tatum suggests that the ability to set these parameters is a form of white privilege.

Cultural bias
While the existence of average IQ test score differences has been a matter of accepted fact for decades, during the 1960s and 1970s a great deal of controversy existed among scholars over the question of whether these score differences reflected real differences in cognitive ability. The critics of testing argued that IQ tests are predominantly measures of cultural variables, and that these variables have nothing to do with intelligence, but they do differ between racial-ethnic and socio-economic groups. If the culture of one group prevents them from developing the skills and abilities that IQ tests measure, then that group is culturally disadvantaged. In contrast, if people from different cultures score differently despite possessing the relevant skills and abilities to the same extent, then the IQ test is culturally biased.

Cultural bias can be distinguished from cultural disadvantage. To do this, a precise definition of bias is required. Bias as mean differences merely begs the question. IQ tests must be standardized against a representative sample of people, and so one suggestion was bias as improper standardization. This has been ruled out in part because re-standardization does not affect the IQ gap. Bias as content could exist if test items presuppose knowledge that is more common in one group. This has been ruled out in part because test item difficulty levels are nearly identical for Blacks and Whites. Arguably the most important condition is bias as differential predictive validity. If test scores predict performance on some other criteria (e.g., school grades) less well for one group than another, then the test shows differential validity. IQ tests have equivalent predictive validity for Blacks and Whites across a range of criteria. Other kinds of bias have been proposed, but none have been found. Of these, bias as motivation is the most difficult to resolve because empirical tests are difficult. It is argued that Blacks are less motivated than Whites to perform well on IQ tests, but if equally motivated the IQ score gap would disappear. At least two views exist on the question of motivation. One view is that intelligence is an abstract concept related to potential, and the possibility of motivation affecting IQ scores is a problem. The second and more common view is that intelligence is a behavior, and performance is more relevant than potential. Under this view, the equal predictive validity of IQ for Blacks and Whites implies that IQ tests are right to be sensitive to motivation if motivation affects performance in important life outcomes.

Labeling bias
Related to the idea of cultural bias, is the concept of "labeling bias", described by Jencks and Phillips (1998). They insist that there exists a labeling bias in the tests, stating that most psychologists agree that IQ tests measure developed ability rather than innate ability—although the tests supposedly measure innate ability. Their assertion is that non-cultural environmental factors cause gaps measured by the tests, rather than any possible innate difference based on genetics, and to use these tests as a measure of innate difference is misleading and improper. In a PBS interview, Jencks states, "If we change the names of the tests, they still measure the same thing but it wouldn't convey this idea that somehow you've gotten the potential of somebody when you measured their IQ. And I think that creates a big bias, because the people who do badly on the tests are labeled as people with low potential in many people's minds and they sometimes even believe that about themselves."

Predictive value of psychometric tests
Since the U.S. Supreme Court outlawed employee selection, including testing, which is "fair in form, but discriminatory in operation" (Griggs v. Duke Power Co., 1971; see this page on disparate impact), American companies have had a strong incentive to construct valid tests which do not produce an IQ gap between ethnic groups, called "selection bias" in employment. Despite this incentive, these efforts have generally failed. For example, in one case regarding a police selection test in Nassau County, New York, a scandal ensued when tests which showed no "selection bias" (Black-White score gap) were found to have been denuded of their ability to measure intelligence (, pp. 24-26 PDF).

The lack of test bias due to test construction or methodology is widely accepted in the research community. From the American Psychological Association's summary of their 1996 task force report, "Intelligence: Knowns and Unknowns": "The differential between the mean intelligence test scores of Blacks and Whites does not result from any obvious biases in test construction and administration, nor does it simply reflect differences in socio-economic status". From The Wall Street Journal: Mainstream Science on Intelligence (PDF): "Intelligence tests are not culturally biased against American Blacks or other native-born, English-speaking people in the U.S. Rather, IQ scores predict equally accurately for all such Americans, regardless of race or social class."

According to Denny Borsboom, mainstream contemporary test analysis does not reflect substantial recent developments in the field and "bears an uncanny resemblance to the psychometric state of the art as it existed in the 1950s." This applies to many areas of mainstream experimental and quasi-experimental research, such as research on personality, attitudes, cognitive development, and intelligence. For example, it reports that some of the most influential recent studies on race differences in intelligence, in order to show that the tests are unbiased, use outdated methodology, the results if anything indicative of that test bias exists. That is prediction invariance the ability of tests to predict economic success on average for members of a certain racial group is generally indicative of violations of measurement invariance the ability of a test to predict the intelligence of members of racial groups without test bias. If two groups differ in their latent means, and a test has prediction invariance across the levels of the grouping variable, it must have measurement bias with regard to group membership.

Reaction time
In 1991, Richard Lynn tested 1,468 9-year old children consisting of Blacks from South Africa, East Asians from Hong Kong and Japan, and Whites from Britain and Ireland. The content of the tests involved flipping a switch after one or more lights came on. Lynn found that the decision times (the time taken to make a decision about what to do) had a correlation of .65 with IQ data on Raven's Progressive Matrices tests also administered during the same study, and that movement times (the time taken to execute the decision) did not show any correlation. He found that the Asians had the fastest decision times, followed by the Whites, and then by the Blacks. He also determined that the Black children had movement times that were substantially faster than those of Whites and Asians on certain tests. Studies by Jensen have shown similar patterns in response time on tests of arithmetic and international studies by Lynn have also asserted the same trend.

has stated that this allows for an extension of Spearman's hypothesis to include reaction time. Based on these results, they have concluded that the observed East Asian-White-Black test differences have a neurological basis. Differences have been found between the reaction times of ethnic groups. In a study by done by Dennis Saccuzzo on information-Processing for African-American, Latino, Filipino, and White elementary school children enrolled in gifted versus non-gifted school programs, gifted African-Americans showed the fastest reaction times of the ethnic groups tested while nongifted African-Americans showed the slowest. The study, however, found only a "modest" correlation between reaction time and IQ.

Average racial differences have been asserted to exist on tests of response time, although it is also argued that some of these differences can be explained by cultural phenomena. Jensen has used reaction time as a proxy for IQ, despite the low correlation with IQ. The use of reaction time as a proxy for race has been disputed by others in the field. Leon J. Kamin and Sharon Grant-Henry used Jenson's work as an example of "claims about racial differences in intelligence that are not supported by scientific evidence" in their study Reaction Time, Race, and Racism.

Facial recognition
Facial recognition ability has shown differences by race. Richard Ferraro writes that facial recognition is an example of a neuropsychological measure that can be used to assess cognitive abilities that are salient within African-American culture. In the US Blacks' performance is significantly better than that of whites', and blacks are better at recognizing faces of whites than whites are at recognizing blacks. A 1991 study found that white subjects performed significantly more poorly on trials involving African American faces than on trials involving White faces, whereas no such difference was obtained among African American subjects. One possibility is that expertise in perceiving faces of particular races is associated with increased ability to extract information about the spatial relationships between different features. Further research using perceptual tasks could shed light on the specific cognitive processes involved in the other-race effect.

Brain size & structure

 * See also: Craniometry

Brain size & structure and IQ

 * See also: Neuroscience and intelligence

Many studies report that IQ has a moderate correlation with various measures of brain size. For example, a 2005 meta-analysis found that brain size correlates with IQ by a factor of approximately .40 among adults. The correlation was also found in some studies to hold true within families (where environmental factors can be considered to be similar), but in one study of 36 sibling pairs essentially zero correlation between brain size and IQ was found when comparing within families.

A correlation of 0.4 explains only 16% of the variance in IQ. In principle, assuming brain size and IQ are related, then all environmental factors that can affect IQ can also affect brain size. Better nutrition can affect brain growth, as can cognitive stimulation. A larger brain is not by itself evidence for genetics any more than IQ scores are.

Brain size & structure and race
Some historical and modern studies, using skull and head measurements, weighing of brains at autopsy, or more recently, magnetic resonance imaging, report racial differences in average brain size similar to those for IQ. Other studies have not found these results. These studies are usually accompanied by a great deal of controversy.

Professor Ulric Neisser (1997) states that the studies comparing brain size by race "exhibit many internal inconsistencies (and the within-groups variabilities are always much larger than the between-groups differences), there is indeed a small overall trend in the direction they describe. Even taken at face value, however, such a trend hardly constitutes evidence for a genetic interpretation".

Old studies
Several studies from the 19th century and early 20th century found racial differences. "Caucasoids" were consistently found to have the largest, followed by "Mongoloids" and then "Negroids", which was the expected view at the time. This is different from the more recent claims where whites only have an intermediate position.(Rushton, 1991, 2000). As noted later, brain size and shape has changed greatly in the last century.

Recent studies
There are several hotly debated studies which have concluded that there are race differences in brain size. These studies are criticized as having not found racially-based difference, with critics arguing instead the the variation better explained in terms of geography. A biogeographical variation in brain size is widely described as an evolutionary adaptation to climate. Lieberman (2001) states "...the relationship between latitude and cranial size is an example of Bergmann’s principle that crania are more spherical in cold climates because mass increases relative to surface area to conserve core temperatures...80 percent of body heat may be lost through our heads on cold days." Lieberman and other anthropologists have also made several other criticisms of claims of racial differences.

In 1970 Philip V. Tobias listed a number of difficulties involved in measuring and making meaningful comparisons of brain weight. These included equating subjects on age, sex, body size, temperature etc. In addition, brain development is plastic, and brain size may be affected by early environmental factors. Because of all these difficulties, Tobias concluded that no adequate racial comparative studies had actually been conducted. The brain size of American Blacks reported in Tobias’s summary were larger than any White group, (which include American, English and French Whites) except those from the Swedish sub sample (who had the largest brains of any of the 77 national groups measured), and American Blacks were estimated to have some 200 million more neurons than American Whites (See Tobias 1970; Weizmann et. 1990).

One other study showing similar patterns in average brain weight is which analyzed autopsy records from Case Western Reserve University (N=1,261) to determine the effects of age, gender, and race on brain weight. The authors found that while there was a large amount of overlap among the groups, there were also statistically significant differences between the groups, and found that White males had the largest brain weight, followed by Black males, White females, and Black females. For a discussion of gender differences in brain size see sex and intelligence. also discovered that mean brain mass decreased over a five year period fastest for White males, followed by Black males, White females, and Black females.

Perhaps the largest single study, analyzed more than 20,000 skulls from around the world, finding that brain size varied with latitude. "Beals, Smith, and Dodd emphasize that this relationship is independent of 'race'."







In his controversial 1995 work Race, Evolution, and Behavior, J. Philippe Rushton reported an average endocranial volume of 1,415 cm³ for Orientals, 1,362 for Whites, and 1,268 for Blacks. When adjusted for average body size, the differences become more pronounced; i.e., the encephalization quotients (EQ) display greater differences than do absolute brain sizes. found an EQ of 7.26 for East Asians as compared to 6.76 for Caucasians. Differences in brain size between Asians and Europeans sometimes do not appear until adjusted for body size. However, height differences between races, and perhaps brain size and weight, are influenced substantially from environmental factors. This study did not mention environmental factors, such as the differences in diet between the races. In some cases Europeans averaged higher absolute brain sizes than East Asians but lower relative brain sizes when adjusted for body size. It needs to be added that these brain sizes are "mean" averages, not actual brain size. East Asians have a much higher endocranial measurement than any other race. This means they have much more dura matter (the 3 layer membrane that protects the brain). (??? citation, please) That might explain the more spherical brain shape. This membrane does not contain neurons. Many contest that brain/body ratio's should only include actual brain size.

Other studies have shown the same pattern in average head size, including, , and the National Collaborative Perinatal Project (described by ) which collected anthropometric data, including head measurements and IQ, on approximately 35,000 children from 1959 to 1974 (although the study began with over 50,000 subjects, some attrition occurred as with many longitudinal studies). Analyses of the data found the East Asian ? White ? Black pattern in head size and IQ at 4 months, 1 year, and 7 years of age.

According to Lieberman, Rushton and others combining the correlations between race, IQ and brain size to establishes a correlation between race and IQ. This conclusion is strongly objected to by Lieberman and other scientists, both on the basis of the initial correlation asserted between race and brain size, and the progressively weak correlation as the studies are combined. Jensen's use of "brain size" as a proxy for intelligence was criticized by Ty Partridge. "Several of the indices reported by Jensen are of questionable validity ... For example, correlations between head circumference or global brain volume and intelligence are unreliably reported in the literature. Head circumference is such a gross measure of cranial capacity that it is hardly useful. Intracranial endocasts are often used to determine the cognitive capabilities of early hominids; however, this is due primarily to the lack of more accurate measures rather than the validity of the method...Indeed there is ample evidence that simple measures of brain size are of little utility in understanding individual differences in 'intelligence'."

East Asian brains have greater width and breadth (i.e., are more brachycephalic) and are more spherically shaped than those of Europeans (it should be noted that theirs has more angles or a diamond shape, as well as a more developed frontal region of the skull). Africans tend to have longer and narrower (more dolichocephalic) brains. Beals et al. proposed that the longer and narrower African brain evolved for better heat dissipation in a warmer climate, while East Asians and Europeans evolved comparatively shorter and wider brains for thermoregulatory purposes in a cooler climate. (pp. 612-613) question the thermoregulatory hypothesis, instead positing that brachycephalization and sphericalization allow for greater brain size. At the same time, Rushton and Ankney believe it is possible that the need to thermoregulate in Africa may have selected against increasing brain size. found a pattern of descending prognathism, glabella size, postorbital constriction, and temporal fossae in African, European, and East Asian skulls and propose that these structures shrank over the course of evolution to allow greater brain size.

Rushton has been accused by other researchers of misrepresenting the data. When they have reanalyzed the data, Zack Cernovsky et al. argue that many of Rushton's claims are incorrect. Extensive statistical analyses by Beals et al. showed that cranial size varies primarily with climatic zones (e.g.,distance from the equator), not race. According to Beals et al., the correlations of brain size to race are spurious: smaller crania are found in warmer climates, irrespective of race. Several other studies found that North American Blacks were superior to American Whites in brain weight, that they had cranial capacities that compared favorably to the average for various samples of Caucasians, and that they had excess neurons larger than many groups of Caucasoids, for example, the English and the French. In general, skulls from people in countries with poverty and infant malnutrition are smaller regardless of race. Cernovsky argues that it is only by pooling their data with data for Negroids from countries in hot climatic zones (notorious for famine and infant malnutrition) that Rushton obtained an illusory support for his postulates.

In a 2005 review of the literature, Rushton and Jensen claim that MRI studies show racial differences in brain size. The cited study compared brain size between a group of 26 persons with bipolar disorder and a group of 48 persons with schizophrenia and also included 34 healthy controls. The distribution of age and race was approximately equal across the three groups and consisted of Caucasians, Blacks, and West Indians (p. 692, table 1). Harvey et al. write, "Over the entire sample intracranial volume was related to height (P=0.0001), gender (P=0.001), ethnicity (P=0.007), and IQ (P=0.03), but not age; larger volumes generally occurred in those subjects who were taller, male, and Caucasian' (p. 693)." (Note that East Asians were not included in the study, that there was a sex ratio imbalance among the schizophrenic group (p. 692), that intracranial volume is not the same thing as brain size, and that the sample where most have serious mental disease may not be representative for population as a whole.)

Possibly related to the Flynn effect is, as also previously observed by Boas, increase in cranial vault size and change in shape during the last 150 years in the US. Cranial vault size and shape have changed greatly during the last 150 years in the US. These changes must occur by early childhood because of the early development of the vault. The explanation for these changes may be related to the Flynn effect. Brain size has increased for both American Blacks and Whites probably due to improved health and nutrition.

Appendix - IQ Data from various sources
NOTE: The information in the following tables is referenced to published sources; however, it should be noted that some of these studies liberally synthesized their results from different sources and test methods. In general comparisons should only be made between similar primary sources, but this is not always possible given the wide variety of IQ tests, the fluid and debatable nature of racial categorization, and the lack of large scale representative data sets. Among the factors that invalidate comparisons across the studies are environment of the test subjects as well as inherent biases in the test procedures.

