[Paleopsych] Commentary: Charles Murray: The Inequality Taboo
checker at panix.com
Sun Aug 28 20:06:03 UTC 2005
Charles Murray: The Inequality Taboo
When the late Richard Herrnstein and I published The Bell Curve eleven years
ago, the furor over its discussion of ethnic differences in IQ was so intense
that most people who have not read the book still think it was about race.
Since then, I have deliberately not published anything about group differences
in IQ, mostly to give the real topic of The Bell Curve-the role of intelligence
in reshaping America's class structure-a chance to surface.
The Lawrence Summers affair last January made me rethink my silence. The
president of Harvard University offered a few mild, speculative, off-the-record
remarks about innate differences between men and women in their aptitude for
high-level science and mathematics, and was treated by Harvard's faculty as if
he were a crank. The typical news story portrayed the idea of innate sex
differences as a renegade position that reputable scholars rejected.
It was depressingly familiar. In the autumn of 1994, I had watched with dismay
as The Bell Curve's scientifically unremarkable statements about black IQ were
successfully labeled as racist pseudoscience. At the opening of 2005, I watched
as some scientifically unremarkable statements about male-female differences
were successfully labeled as sexist pseudoscience.
The Orwellian disinformation about innate group differences is not wholly the
media's fault. Many academics who are familiar with the state of knowledge are
afraid to go on the record. Talking publicly can dry up research funding for
senior professors and can cost assistant professors their jobs. But while the
public's misconception is understandable, it is also getting in the way of
clear thinking about American social policy.
Good social policy can be based on premises that have nothing to do
with scientific truth. The premise that is supposed to undergird all
of our social policy, the founders' assertion of an unalienable right
to liberty, is not a falsifiable hypothesis. But specific policies
based on premises that conflict with scientific truths about human
beings tend not to work. Often they do harm.
One such premise is that the distribution of innate abilities and
propensities is the same across different groups. The statistical
tests for uncovering job discrimination assume that men are not
innately different from women, blacks from whites, older people from
younger people, homosexuals from heterosexuals, Latinos from Anglos,
in ways that can legitimately affect employment decisions. Title IX
of the Educational Amendments of 1972 assumes that women are no
different from men in their attraction to sports. Affirmative action
in all its forms assumes there are no innate differences between any
of the groups it seeks to help and everyone else. The assumption of
no innate differences among groups suffuses American social policy.
That assumption is wrong.
When the outcomes that these policies are supposed to produce fail to
occur, with one group falling short, the fault for the discrepancy
has been assigned to society. It continues to be assumed that better
programs, better regulations, or the right court decisions can make
the differences go away. That assumption is also wrong.
Hence this essay. Most of the following discussion describes reasons
for believing that some group differences are intractable. I shift
from "innate" to "intractable" to acknowledge how complex is the
interaction of genes, their expression in behavior, and the
environment. "Intractable" means that, whatever the precise
partitioning of causation may be (we seldom know), policy
interventions can only tweak the difference at the margins.
I will focus on two sorts of differences: between men and women and
between blacks and whites. Here are three crucial points to keep in
mind as we go along:
1. The differences I discuss involve means and distributions. In all
cases, the variation within groups is greater than the variation
between groups. On psychological and cognitive dimensions, some
members of both sexes and all races fall everywhere along the range.
One implication of this is that genius does not come in one color or
sex, and neither does any other human ability. Another is that a few
minutes of conversation with individuals you meet will tell you much
more about them than their group membership does.
2. Covering both sex differences and race differences in a single,
non-technical article, I had to leave out much in the print edition
of this article. This online version is fully annotated and includes
extensive supplementary material.
3. The concepts of "inferiority" and "superiority" are inappropriate
to group comparisons. On most specific human attributes, it is
possible to specify a continuum running from "low" to "high," but the
results cannot be combined into a score running from "bad" to "good."
What is the best score on a continuum measuring aggressiveness? What
is the relative importance of verbal skills versus, say, compassion?
Of spatial skills versus industriousness? The aggregate excellences
and shortcomings of human groups do not lend themselves to simple
comparisons. That is why the members of just about every group can so
easily conclude that they are God's chosen people. All of us use the
weighting system that favors our group's strengths.1
The technical literature documenting sex differences and their
biological basis grew surreptitiously during feminism's heyday in the
1970's and 1980's. By the 1990's, it had become so extensive that the
bibliography in David Geary's pioneering Male, Female (1998) ran to
53 pages.2 Currently, the best short account of the state of
knowledge is Steven Pinker's chapter on gender in The Blank Slate
Rather than present a telegraphic list of all the differences that I
think have been established, I will focus on the narrower question at
the heart of the Summers controversy: as groups, do men and women
differ innately in characteristics that produce achievement at the
highest levels of accomplishment? I will limit my comments to the
arts and sciences.
Since we live in an age when students are likely to hear more about
Marie Curie than about Albert Einstein, it is worth beginning with a
statement of historical fact: women have played a proportionally tiny
part in the history of the arts and sciences.4 Even in the 20th
century, women got only 2 percent of the Nobel Prizes in the
sciences-a proportion constant for both halves of the century-and 10
percent of the prizes in literature. The Fields Medal, the most
prestigious award in mathematics, has been given to 44 people since
it originated in 1936. All have been men.
The historical reality of male dominance of the greatest achievements
in science and the arts is not open to argument. The question is
whether the social and legal exclusion of women is a sufficient
explanation for this situation, or whether sex-specific
characteristics are also at work.
Mathematics offers an entry point for thinking about the answer.
Through high school, girls earn better grades in math than boys, but
the boys usually do better on standardized tests.5 The difference in
means is modest, but the male advantage increases as the focus shifts
from means to extremes. In a large sample of mathematically gifted
youths, for example, seven times as many males as females scored in
the top percentile of the SAT mathematics test.6 We do not have good
test data on the male-female ratio at the top one-hundredth or top
one-thousandth of a percentile, where first-rate mathematicians are
most likely to be found, but collateral evidence suggests that the
male advantage there continues to increase, perhaps exponentially.7
Evolutionary biologists have some theories that feed into an
explanation for the disparity. In primitive societies, men did the
hunting, which often took them far from home. Males with the ability
to recognize landscapes from different orientations and thereby find
their way back had a survival advantage. Men who could process
trajectories in three dimensions-the trajectory, say, of a spear
thrown at an edible mammal-also had a survival advantage.8 Women did
the gathering. Those who could distinguish among complex arrays of
vegetation, remembering which were the poisonous plants and which the
nourishing ones, also had a survival advantage. Thus the logic for
explaining why men should have developed elevated three-dimensional
visuospatial skills and women an elevated ability to remember objects
and their relative locations-differences that show up in specialized
Perhaps this is a just-so story.10 Why not instead attribute the
results of these tests to socialization? Enter the neuroscientists.
It has been known for years that, even after adjusting for body size,
men have larger brains than women. Yet most psychometricians conclude
that men and women have the same mean IQ (although debate on this
issue is growing).11 One hypothesis for explaining this paradox is
that three-dimensional processing absorbs the extra male capacity. In
the last few years, magnetic-resonance imaging has refined the
evidence for this hypothesis, revealing that parts of the brain's
parietal cortex associated with space perception are proportionally
bigger in men than in women.12
What does space perception have to do with scores on math tests?13
Enter the psychometricians, who demonstrate that when visuospatial
ability is taken into account, the sex difference in SAT math scores
Why should the difference be so much greater at the extremes than at
the mean? Part of the answer is that men consistently exhibit higher
variance than women on all sorts of characteristics, including
visuospatial abilities, meaning that there are proportionally more
men than women at both ends of the bell curve.15 Another part of the
answer is that someone with a high verbal IQ can easily master the
basic algebra, geometry, and calculus that make up most of the items
in an ordinary math test. Elevated visuospatial skills are most
useful for the most difficult items.16 If males have an advantage in
answering those comparatively few really hard items, the increasing
disparity at the extremes becomes explicable.
Seen from one perspective, this pattern demonstrates what should be
obvious: there is nothing inherent in being a woman that precludes
high math ability. But there remains a distributional difference in
male and female characteristics that leads to a larger number of men
with high visuospatial skills. The difference has an evolutionary
rationale, a physiological basis, and a direct correlation with math
Now put all this alongside the historical data on accomplishment in
the arts and sciences. In test scores, the male advantage is most
pronounced in the most abstract items. Historically, too, it is most
pronounced in the most abstract domains of accomplishment.17
In the humanities, the most abstract field is philosophy-and no woman
has been a significant original thinker in any of the world's great
philosophical traditions. In the sciences, the most abstract field is
mathematics, where the number of great women mathematicians is
approximately two (Emmy Noether definitely, Sonya Kovalevskaya
maybe). In the other hard sciences, the contributions of great women
scientists have usually been empirical rather than theoretical, with
leading cases in point being Henrietta Leavitt, Dorothy Hodgkin, Lise
Meitner, Irène Joliot-Curie, and Marie Curie herself.
In the arts, literature is the least abstract and by far the most
rooted in human interaction; visual art incorporates a greater
admixture of the abstract; musical composition is the most abstract
of all the arts, using neither words nor images. The role of women
has varied accordingly. Women have been represented among great
writers virtually from the beginning of literature, in East Asia and
South Asia as well as in the West. Women have produced a smaller
number of important visual artists, and none that is clearly in the
first rank. No female composer is even close to the first rank.
Social restrictions undoubtedly damped down women's contributions in
all of the arts, but the pattern of accomplishment that did break
through is strikingly consistent with what we know about the
respective strengths of male and female cognitive repertoires.
Women have their own cognitive advantages over men, many of them
involving verbal fluency and interpersonal skills. If this were a
comprehensive survey, detailing those advantages would take up as
much space as I have devoted to a particular male advantage. But,
sticking with my restricted topic, I will move to another aspect of
male-female differences that bears on accomplishment at the highest
levels of the arts and sciences: motherhood.
Regarding women, men, and babies, the technical literature is as
unambiguous as everyday experience would lead one to suppose. As a
rule, the experience of parenthood is more profoundly life-altering
for women than for men. Nor is there anything unique about humans in
this regard. Mammalian reproduction generally involves much higher
levels of maternal than paternal investment in the raising of
children.18 Among humans, extensive empirical study has demonstrated
that women are more attracted to children than are men, respond to
them more intensely on an emotional level, and get more and different
kinds of satisfactions from nurturing them. Many of these behavioral
differences have been linked with biochemical differences between men
Thus, for reasons embedded in the biochemistry and neurophysiology of
being female, many women with the cognitive skills for achievement at
the highest level also have something else they want to do in life:
have a baby. In the arts and sciences, forty is the mean age at which
peak accomplishment occurs, preceded by years of intense effort
mastering the discipline in question.20 These are precisely the years
during which most women must bear children if they are to bear them
Among women who have become mothers, the possibilities for high-level
accomplishment in the arts and sciences shrink because, for innate
reasons, the distractions of parenthood are greater. To put it in a
way that most readers with children will recognize, a father can go
to work and forget about his children for the whole day. Hardly any
mother can do this, no matter how good her day-care arrangement or
full-time nanny may be. My point is not that women must choose
between a career and children, but that accomplishment at the
extremes commonly comes from a single-minded focus that leaves no
room for anything but the task at hand.21 We should not be surprised
or dismayed to find that motherhood reduces the proportion of highly
talented young women who are willing to make that tradeoff.
Some numbers can be put to this observation through a study of
nearly 2,000 men and women who were identified as extraordinarily
talented in math at age thirteen and were followed up 20 years
later.22 The women in the sample came of age in the 1970's and early
1980's, when women were actively socialized to resist gender
stereotypes. In many ways, these talented women did resist. By their
early thirties, both the men and women had become exceptional
achievers, receiving advanced degrees in roughly equal proportions.
Only about 15 percent of the women were full-time housewives. Among
the women, those who did and those who did not have children were
equally satisfied with their careers.
And yet. The women with careers were four-and-a-half times more
likely than men to say they preferred to work fewer than 40 hours per
week. The men placed greater importance on "being successful in my
line of work" and "inventing or creating something that will have an
impact," while the women found greater value in "having strong
friendships," "living close to parents and relatives," and "having a
meaningful spiritual life." As the authors concluded, "these men and
women appear to have constructed satisfying and meaningful lives that
took somewhat different forms."23 The different forms, which directly
influence the likelihood that men will dominate at the extreme levels
of achievement, are consistent with a constellation of differences
between men and women that have biological roots.
I have omitted perhaps the most obvious reason why men and women
differ at the highest levels of accomplishment: men take more risks,
are more competitive, and are more aggressive than women.24 The word
"testosterone" may come to mind, and appropriately. Much technical
literature documents the hormonal basis of personality differences
that bear on sex differences in extreme and venturesome effort, and
hence in extremes of accomplishment-and that bear as well on the male
propensity to produce an overwhelming proportion of the world's crime
and approximately 100 percent of its wars. But this is just one more
of the ways in which science is demonstrating that men and women are
really and truly different, a fact so obvious that only intellectuals
could ever have thought otherwise.
Turning to race, we must begin with the fraught question of whether
it even exists, or whether it is instead a social construct. The
Harvard geneticist Richard Lewontin originated the idea of race as a
social construct in 1972, arguing that the genetic differences across
races were so trivial that no scientist working exclusively with
genetic data would sort people into blacks, whites, or Asians. In his
words, "racial classification is now seen to be of virtually no
genetic or taxonomic significance."25
Lewontin's position, which quickly became a tenet of political
correctness, carried with it a potential means of being falsified. If
he was correct, then a statistical analysis of genetic markers would
not produce clusters corresponding to common racial labels.
In the last few years, that test has become feasible, and now we know
that Lewontin was wrong.26 Several analyses have confirmed the
genetic reality of group identities going under the label of race or
ethnicity.27 In the most recent, published this year, all but five of
the 3,636 subjects fell into the cluster of genetic markers
corresponding to their self-identified ethnic group.28 When a
statistical procedure, blind to physical characteristics and working
exclusively with genetic information, classifies 99.9 percent of the
individuals in a large sample in the same way they classify
themselves, it is hard to argue that race is imaginary.
Homo sapiens actually falls into many more interesting groups than
the bulky ones known as "races."29 As new findings appear almost
weekly, it seems increasingly likely that we are just at the
beginning of a process that will identify all sorts of genetic
differences among groups, whether the groups being compared are
Nigerian blacks and Kenyan blacks, lawyers and engineers, or
Episcopalians and Baptists. At the moment, the differences that are
obviously genetic involve diseases (Ashkenazi Jews and Tay-Sachs
disease, black Africans and sickle-cell anemia, Swedes and
hemochromatosis). As time goes on, we may yet come to understand
better why, say, Italians are more vivacious than Scots.
Out of all the interesting and intractable differences that may
eventually be identified, one in particular remains a hot button like
no other: the IQ difference between blacks and whites. What is the
present state of our knowledge about it?
There is no technical dispute on some of the core issues. In the
aftermath of The Bell Curve, the American Psychological Association
established a task force on intelligence whose report was published
in early 1996.30 The task force reached the same conclusions as The
Bell Curve on the size and meaningfulness of the black-white
difference. Historically, it has been about one standard deviation31
in magnitude among subjects who have reached adolescence;32 cultural
bias in IQ tests does not explain the difference; and the tests are
about equally predictive of educational, social, and economic
outcomes for blacks and whites. However controversial such assertions
may still be in the eyes of the mainstream media, they are not
controversial within the scientific community.
The most important change in the state of knowledge since the
mid-1990's lies in our increased understanding of what has happened
to the size of the black-white difference over time. Both the task
force and The Bell Curve concluded that some narrowing had occurred
since the early 1970's. With the advantage of an additional decade of
data, we are now able to be more precise: (1) The black-white
difference in scores on educational achievement tests has narrowed
significantly. (2) The black-white convergence in scores on the most
highly "g?-loaded" tests-the tests that are the best measures of
cognitive ability-has been smaller, and may be unchanged, since the
first tests were administered 90 years ago.
With regard to the difference in educational achievement, the
narrowing of scores on major tests occurred in the 1970's and 80's.
In the case of the SAT, the gaps in the verbal and math tests as of
1972 were 1.24 and 1.26 standard deviations respectively.33 By 1991,
when the gaps were smallest (they have risen slightly since then),
those numbers had dropped by .37 and .35 standard deviations.
The National Assessment of Educational Progress (NAEP), which is not
limited to college-bound students, is preferable to the SAT for
estimating nationally representative trends, but the story it tells
is similar.34 Among students ages nine, thirteen, and seventeen, the
black-white differences in math as of the first NAEP test in 1973
were 1.03, 1.29, and 1.24 standard deviations respectively. For
nine-year-olds, the difference hit its all-time low of .73 standard
deviations in 2004, a drop of .30 standard deviations. But almost all
of that convergence had been reached by 1986, when the gap was .78
standard deviations. For thirteen-year-olds, the gap dropped by .45
standard deviations, reaching its low in 1986. For
seventeen-year-olds, the gap dropped by .52 standard deviations,
reaching its low in 1990.
In the reading test, the comparable gaps for ages nine, thirteen, and
seventeen as of the first NAEP test in 1971 were 1.12, 1.17, and 1.25
standard deviations. Those gaps had shrunk by .38, .62, and .68
standard deviations respectively at their lowest points in 1988.35
They have since remained effectively unchanged.
An analysis by Larry Hedges and Amy Nowell uses a third set of data,
examining the trends for high-school seniors by comparing six large
data bases from different time periods from 1965 to 1992. The
black-white difference on a combined measure of math, vocabulary, and
reading fell from 1.18 to .82 standard deviations in that time, a
reduction of .36 standard deviations.36
So black and white academic achievement converged significantly in
the 1970's and 1980's, typically by more than a third of a standard
deviation, and since then has stayed about the same.37 What about
convergence in tests explicitly designed to measure IQ rather than
academic achievement?38 The ambiguities in the data leave two
defensible positions. The first is that the IQ difference is about
one standard deviation, effectively unchanged since the first
black-white comparisons 90 years ago. The second is that harbingers
of a narrowing difference are starting to emerge. I cannot settle the
argument here, but I can convey some sense of the uncertainty.
The case for an unchanged black-white IQ difference is
straightforward. If you take all the black-white differences on IQ
tests from the first ones in World War I up to the present, there is
no statistically significant downward trend. Of course the results
vary, because tests vary in the precision with which they measure the
general mental factor (g) and samples vary in their size and
representativeness. But results continue to center around a
black-white difference of about 1.0 to 1.1 standard deviations
through the most recent data.39
The case for a reduction has two important recent results to work
with. The first is from the 1997 re-norming of the Armed Forces
Qualification Test (AFQT), which showed a black-white difference of
.97 standard deviations.40 Since the typical difference on
paper-and-pencil IQ tests like the AFQT has been about 1.10 standard
deviations, the 1997 results represent noticeable improvement.41 The
second positive result comes from the 2003 standardization sample for
the Wechsler Intelligence Scale for Children (WISC-IV), which showed
a difference of .78 standard deviations, as against the 1.0
difference that has been typical for individually administered IQ
One cannot draw strong conclusions from two data points. Those who
interpret them as part of an unchanging overall pattern can cite
another recent result, from the 2001 standardization of the
Woodcock-Johnson intelligence test. In line with the conventional
gap, it showed an overall black-white difference of 1.05 standard
deviations and, for youths aged six to eighteen, a difference of .99
There is more to be said on both sides of this issue, but nothing
conclusive.44 Until new data become available, you may take your
choice. If you are a pessimist, the gap has been unchanged at about
one standard deviation. If you are an optimist, the IQ gap has
decreased by a few points, but it is still close to one standard
deviation. The clear and substantial convergence that occurred in
academic tests has at best been but dimly reflected in IQ scores, and
at worst not reflected at all.
Whether we are talking about academic achievement or about IQ, are
the causes of the black-white difference environmental or genetic?
Everyone agrees that environment plays a part. The controversy is
about whether biology is also involved.
It has been known for many years that the obvious environmental
factors such as income, parental occupation, and schools explain only
part of the absolute black-white difference and none of the relative
difference. Black and white students from affluent neighborhoods are
separated by as large a proportional gap as are blacks and whites
from poor neighborhoods.45 Thus the most interesting recent studies
of environmental causes have worked with cultural explanations
instead of socioeconomic status.46
One example is Black American Students in an Affluent Suburb: A Study
of Academic Disengagement (2003) by the Berkeley anthropologist John
Ogbu, who went to Shaker Heights, Ohio, to explore why black students
in an affluent suburb should lag behind their white peers.47 Another
is Black Rednecks and White Liberals (2005) by Thomas Sowell, who
makes the case that what we think of as the dysfunctional aspects of
urban black culture are a legacy not of slavery but of Southern and
rural white "cracker" culture.48 Both Ogbu and Sowell describe
ingrained parental behaviors and student attitudes that must impede
black academic performance. These cultural influences often cut
across social classes.
From a theoretical standpoint, the cultural explanations offer fresh
ways of looking at the black-white difference at a time when the
standard socioeconomic explanations have reached a dead end. From a
practical standpoint, however, the cultural explanations point to a
cause of the black-white difference that is as impervious to
manipulation by social policy as causes rooted in biology. If there
is to be a rapid improvement, some form of mass movement with
powerful behavioral consequences would have to occur within the black
community. Absent that, the best we can hope for is gradual cultural
change that is likely to be measured in decades.
This brings us to the state of knowledge about genetic explanations.
"There is not much direct evidence on this point," said the American
Psychological Association's task force dismissively, "but what little
there is fails to support the genetic hypothesis."49 Actually, there
is no direct evidence at all, just a wide variety of indirect
evidence, almost all of which the task force chose to ignore.50
As it happens, a comprehensive survey of that evidence, and of the
objections to it, appeared this past June in the journal Psychology,
Public Policy, and Law. There, J. Philippe Rushton and Arthur Jensen
co-authored a 60-page article entitled "Thirty Years of Research on
Race Differences in Cognitive Ability."51 It incorporates studies of
East Asians as well as blacks and whites and concludes that the
source of the black-white-Asian difference is 50- to 80-percent
genetic. The same issue of the journal includes four commentaries,
three of them written by prominent scholars who oppose the idea that
any part of the black-white difference is genetic.52 Thus, in one
place, you can examine the strongest arguments that each side in the
debate can bring to bear.
Rushton and Jensen base their conclusion on ten categories of
evidence that are consistent with a model in which both environment
and genes cause the black-white difference and inconsistent with a
model that requires no genetic contribution.53 I will not try to
review their argument here, or the critiques of it. All of the
contributions can be found on the Internet, and can be understood by
readers with a grasp of basic statistical concepts.54
For those who consider it important to know what percentage of the IQ
difference is genetic, a methodology that would do the job is now
available. In the United States, few people classified as black are
actually of 100-percent African descent (the average American black
is thought to be about 20-percent white).55 To the extent that genes
play a role, IQ will vary by racial admixture. In the past, studies
that have attempted to test this hypothesis have had no accurate way
to measure the degree of admixture, and the results have been
accordingly muddy.56 The recent advances in using genetic markers
solves that problem. Take a large sample of racially diverse people,
give them a good IQ test, and then use genetic markers to create a
variable that no longer classifies people as "white" or "black," but
along a continuum. Analyze the variation in IQ scores according to
that continuum. The results would be close to dispositive.57
None of this is important for social policy, however, where the issue
is not the source of the difference but its intractability. Much of
the evidence reviewed by Rushton and Jensen bears on what we can
expect about future changes in the black-white IQ difference. My own
thinking on this issue is shaped by the relationship of the
difference to a factor I have already mentioned-"g"-and to the
developing evidence for g's biological basis.
When you compare black and white mean scores on a battery of
subtests, you do not find a uniform set of differences; nor do you
find a random assortment. The size of the difference varies
systematically by type of subtest. Asked to predict which subtests
show the largest difference, most people will think first of ones
that have the most cultural content and are the most sensitive to
good schooling. But this natural expectation is wrong. Some of the
largest differences are found on subtests that have little or no
cultural content, such as ones based on abstract designs.
As long ago as 1927, Charles Spearman, the pioneer psychometrician
who discovered g, proposed a hypothesis to explain the pattern: the
size of the black-white difference would be "most marked in just
those [subtests] which are known to be saturated with g."58 In other
words, Spearman conjectured that the black-white difference would be
greatest on tests that were the purest measures of intelligence, as
opposed to tests of knowledge or memory.
A concrete example illustrates how Spearman's hypothesis works. Two
items in the Wechsler and Stanford-Binet IQ tests are known as
"forward digit span" and "backward digit span." In the forward
version, the subject repeats a random sequence of one-digit numbers
given by the examiner, starting with two digits and adding another
with each iteration. The subject's score is the number of digits that
he can repeat without error on two consecutive trials.
Digits-backward works exactly the same way except that the digits
must be repeated in the opposite order.
Digits-backward is much more g-loaded than digits-forward. Try it
yourself and you will see why. Digits-forward is a straightforward
matter of short-term memory. Digits-backward makes your brain work
The black-white difference in digits-backward is about twice as large
as the difference in digits-forward.60 It is a clean example of an
effect that resists cultural explanation. It cannot be explained by
differential educational attainment, income, or any other
socioeconomic factor. Parenting style is irrelevant. Reluctance to
"act white" is irrelevant. Motivation is irrelevant. There is no way
that any of these variables could systematically encourage black
performance in digits-forward while depressing it in digits-backward
in the same test at the same time with the same examiner in the same
In 1980, Arthur Jensen began a research program for testing
Spearman's hypothesis. In his book The g Factor (1998), he summarized
the results from seventeen independent sets of data, derived from 149
psychometric tests. They consistently supported Spearman's
hypothesis.62 Subsequent work has added still more evidence.63 Debate
continues about what the correlation between g-loadings and the size
of the black-white difference means, but the core of Spearman's
original conjecture, that a sizable correlation would be found to
exist, has been confirmed.64
During the same years that Jensen was investigating Spearman's
hypothesis, progress was also being made in understanding g. For
decades, psychometricians had tried to make g go away. Confident that
intelligence must be more complicated than a single factor, they
strove to replace g with measures of uncorrelated mental skills. They
thereby made valuable contributions to our understanding of
intelligence, which really does manifest itself in different ways and
with different profiles, but getting rid of g proved impossible. No
matter how the data were analyzed, a single factor kept dominating
By the 1980's, the robustness and value of g as an explanatory
construct were broadly accepted among pyschometricians, but little
was known about its physiological basis.66 As of 2005, we know much
more. It is now established that g is by far the most heritable
component of IQ.67 A variety of studies have found correlations
between g and physiological phenomena such as brain-evoked
potentials, brain pH levels, brain glucose metabolism,
nerve-conduction velocity, and reaction time.68 Most recently, it has
been determined that a highly significant relationship exists between
g and the volume of gray matter in specific areas of the frontal
cortex, and that the magnitude of the volume is under tight genetic
control.69 In short, we now know that g captures something in the
biology of the brain.
So Spearman's basic conjecture was correct-the size of the
black-white difference and g-loadings are correlated-and g represents
a biologically grounded and highly heritable cognitive resource. When
those two observations are put together, a number of characteristics
of the black-white difference become predictable, correspond with
phenomena we have observed in data, and give us reason to think that
not much will change in the years to come.70
One implication is that black-white convergence on test scores will
be greatest on tests that are least g-loaded. Literacy is the obvious
example: people with a wide range of IQ's can be taught to read
competently, and it is the reading test of the NAEP in which
convergence has reached its closest point (.55 standard deviations in
the 1988 test). More broadly, the confirmation of Spearman's
hypothesis explains why the convergence that has occurred on academic
achievement tests has not been matched on IQ tests.
A related implication is that the source of the black-white
difference lies in skills that are hardest to change. Being able to
repeat many digits backward has no value in itself. It points to a
valuable underlying mental ability, in the same way that percentage
of fast-twitch muscle fibers points to an underlying athletic
ability. If you were to practice reciting digits backward for a few
days, you could increase your score somewhat, just as training can
improve your running speed somewhat. But in neither case will you
have improved the underlying ability.71 As far as anyone knows, g
itself cannot be coached.
The third implication is that the "Flynn effect" will not close the
black-white difference. I am referring here to the secular increase
in IQ scores over time, brought to public attention by James Flynn.72
The Flynn effect has been taken as a reason for thinking that the
black-white difference is temporary: if IQ scores are so malleable
that they can rise steadily for several decades, why should not the
black-white difference be malleable as well?73
But as the Flynn effect has been studied over the last decade, the
evidence has grown, and now seems persuasive, that the increases in
IQ scores do not represent significant increases in g.74 What the
increases do represent-whether increases in specific mental skills or
merely increased test sophistication-is still being debated. But if
the black-white difference is concentrated in g and if the Flynn
effect does not consist of increases in g, the Flynn effect will not
do much to close the gap. A 2004 study by Dutch scholars tested this
question directly. Examining five large databases, the authors
concluded that "the nature of the Flynn effect is qualitatively
different from the nature of black-white differences in the United
States," and that "the implications of the Flynn effect for
black-white differences appear small."75
These observations represent my reading of a body of evidence that is
incomplete, and they will surely have to be modified as we learn
more. But taking the story of the black-white IQ difference as a
whole, I submit that we know two facts beyond much doubt. First, the
conventional environmental explanation of the black-white difference
is inadequate. Poverty, bad schools, and racism, which seem such
obvious culprits, do not explain it. Insofar as the environment is
the cause, it is not the sort of environment we know how to change,
and we have tried every practical remedy that anyone has been able to
think of. Second, regardless of one's reading of the competing
arguments, we are left with an IQ difference that has, at best,
narrowed by only a few points over the last century. I can find
nothing in the history of this difference, or in what we have learned
about its causes over the last ten years, to suggest that any faster
change is in our future.
Elites throughout the West are living a lie, basing the futures of
their societies on the assumption that all groups of people are equal
in all respects. Lie is a strong word, but justified. It is a lie
because so many elite politicians who profess to believe it in public
do not believe it in private. It is a lie because so many elite
scholars choose to ignore what is already known and choose not to
inquire into what they suspect. We enable ourselves to continue to
live the lie by establishing a taboo against discussion of group
The taboo is not perfect-otherwise, I would not have been able to
document this essay-but it is powerful. Witness how few of Harvard's
faculty who understood the state of knowledge about sex differences
were willing to speak out during the Summers affair. In the
public-policy debate, witness the contorted ways in which even the
opponents of policies like affirmative action frame their arguments
so that no one can accuse them of saying that women are different
from men or blacks from whites. Witness the unwillingness of the
mainstream media to discuss group differences without assuring
readers that the differences will disappear when the world becomes a
The taboo arises from an admirable idealism about human equality. If
it did no harm, or if the harm it did were minor, there would be no
need to write about it. But taboos have consequences.
The nature of many of the consequences must be a matter of conjecture
because people are so fearful of exploring them.76 Consider an
observation furtively voiced by many who interact with civil
servants: that government is riddled with people who have been
promoted to their level of incompetence because of pressure to have a
staff with the correct sex and ethnicity in the correct proportions
and positions. Are these just anecdotes? Or should we be worrying
about the effects of affirmative action on the quality of government
services?77 It would be helpful to know the answers, but we will not
so long as the taboo against talking about group difference prevails.
How much damage has the taboo done to the education of children?
Christina Hoff Sommers has argued that willed blindness to the
different developmental patterns of boys and girls has led many
educators to see boys as aberrational and girls as the norm, with
pervasive damage to the way our elementary and secondary schools are
run.78 Is she right? Few have been willing to pursue the issue lest
they be required to talk about innate group differences. Similar
questions can be asked about the damage done to medical care, whose
practitioners have only recently begun to acknowledge the ways in
which ethnic groups respond differently to certain drugs.79
How much damage has the taboo done to our understanding of America's
social problems? The part played by sexism in creating the ratio of
males to females on mathematics faculties is not the ratio we observe
but what remains after adjustment for male-female differences in
high-end mathematical ability. The part played by racism in creating
different outcomes in black and white poverty, crime, and
illegitimacy is not the raw disparity we observe but what remains
after controlling for group characteristics. For some outcomes, sex
or race differences nearly disappear after a proper analysis is done.
For others, a large residual difference remains.80 In either case,
open discussion of group differences would give us a better grasp on
where to look for causes and solutions.
What good can come of raising this divisive topic? The honest answer
is that no one knows for sure. What we do know is that the taboo has
crippled our ability to explore almost any topic that involves the
different ways in which groups of people respond to the world around
them-which means almost every political, social, or economic topic of
Thus my modest recommendation, requiring no change in laws or
regulations, just a little more gumption. Let us start talking about
group differences openly-all sorts of group differences, from the
visuospatial skills of men and women to the vivaciousness of Italians
and Scots. Let us talk about the nature of the manly versus the
womanly virtues. About differences between Russians and Chinese that
might affect their adoption of capitalism. About differences between
Arabs and Europeans that might affect the assimilation of Arab
immigrants into European democracies. About differences between the
poor and non-poor that could inform policy for reducing poverty.
Even to begin listing the topics that could be enriched by an inquiry
into the nature of group differences is to reveal how stifled today's
conversation is. Besides liberating that conversation, an open and
undefensive discussion would puncture the irrational fear of the
male-female and black-white differences I have surveyed here. We
would be free to talk about other sexual and racial differences as
well, many of which favor women and blacks, and none of which is
large enough to frighten anyone who looks at them dispassionately.
Talking about group differences does not require any of us to change
our politics. For every implication that the Right might seize upon
(affirmative-action quotas are ill-conceived), another gives fodder
to the Left (innate group differences help rationalize compensatory
redistribution by the state).81 But if we do not need to change our
politics, talking about group differences obligates all of us to
renew our commitment to the ideal of equality that Thomas Jefferson
had in mind when he wrote as a self-evident truth that all men are
created equal. Steven Pinker put that ideal in today's language in
The Blank Slate, writing that "Equality is not the empirical claim
that all groups of humans are interchangeable; it is the moral
principle that individuals should not be judged or constrained by the
average properties of their group."82
Nothing in this essay implies that this moral principle has already
been realized or that we are powerless to make progress. In
elementary and secondary education, many outcomes are tractable even
if group differences in ability remain unchanged. Dropout rates,
literacy, and numeracy are all tractable. School discipline, teacher
performance, and the quality of the curriculum are tractable.
Academic performance within a given IQ range is tractable. The
existence of group differences need not and should not discourage
attempts to improve schooling for millions of American children who
are now getting bad educations.
In university education and in the world of work, overall openness of
opportunity has been transformed for the better over the last
half-century. But the policies we now have in place are impeding, not
facilitating, further progress. Creating double standards for
physically demanding jobs so that women can qualify ensures that men
in those jobs will never see women as their equals. In universities,
affirmative action ensures that the black-white difference in IQ in
the population at large is brought onto the campus and made visible
to every student. The intentions of their designers notwithstanding,
today's policies are perfectly fashioned to create separation,
condescension, and resentment-and so they have done.
The world need not be that way. Any university or employer that
genuinely applied a single set of standards for hiring, firing,
admitting, and promoting would find that performance across different
groups really is distributed indistinguishably. But getting to that
point nationwide will require us to jettison an apparatus of laws,
regulations, and bureaucracies that has been 40 years in the making.
That will not happen until the conversation has opened up. So let us
take one step at a time. Let us stop being afraid of data that tell
us a story we do not want to hear, stop the name-calling, stop the
denial, and start facing reality.
CHARLES MURRAY is the W.H. Brady Scholar in Freedom and Culture at
the American Enterprise Institute. His previous contributions to
COMMENTARY, available online, include "The Bell Curve and Its
Critics" (May 1995, with a subsequent exchange in the August 1995
My thanks go to Michael Ashton, Thomas Bouchard, Gregory Carey,
Christopher DeMuth, David Geary, Linda Gottfredson, Arthur Jensen,
John Loehlin, David Lubinski, Kevin McGrew, Richard McNally, Derek
Neal, Steven Pinker, Philip Roth, Philippe Rushton, Sally Satel,
Christina Hoff Sommers, Hua Tang, Marley Watkins, Lawrence Weiss, and
James Q. Wilson for responding to questions or commenting on drafts.
Their appearance on this list does not imply their endorsement of
anything in the essay.
* This is a fully annotated version of the article that appears in
the September 2005 issue of COMMENTARY.
1 If you think this is mushy nonjudgmentalism, try a thought
experiment: Suppose that a pill exists that, if all women took it,
would give them exactly the same mean and variance on every dimension
of human functioning as men-including all the ways in which women now
surpass men. How many women would want all women to take it? Or
suppose that the pill, taken by all blacks, would give them exactly
the same mean and variance on every dimension of human functioning as
whites-including all the ways in which blacks now surpass whites. How
many blacks would want all blacks to take it? To ask such questions
is to answer them: hardly anybody. Few want to trade off the unique
virtues of their own group for the advantages that another group may
Sometimes these preferences for one's own group are rational,
sometimes not. I am proud of being Scots-Irish, for example, even
though the Scots-Irish group means for violence, drunkenness, and
general disagreeableness seem to have been far above those of other
immigrant groups. But the Scots-Irish made great pioneers-that's the
part of my heritage that I choose to value. A Thai friend gave me an
insight into this human characteristic many years ago when I remarked
that Thais were completely undefensive about Westerners despite the
economic backwardness of Thailand in those days. My friend explained
why. America has wealth and technology that Thailand does not have,
he acknowledged, just as the elephant is stronger than a human.
"But," he said with a shrug, "who wants to be an elephant?" None of
us wants to be an elephant and, from the perspective of our own
group, every other group has something of the elephant about it. All
of us are right, too.
2 Geary (1998).
3 Pinker (2002). A non-technical book-length treatment is Rhoads
(2004). Halpern (2000) and Kimura (1999) are good one-volume
discussions of cognitive differences between the sexes. An up-to-date
summary of neuro-physiological findings about sex differences in the
brain appeared in last May's Scientific American, Cahill (2005).
Baron-Cohen (2003) is an ambitious attempt to tie together known sex
differences into an overall theory. Those who want to compare these
accounts with defenses of the no-innate-differences position can look
at Valian (1999) and a set of essays weighted toward social
explanations of math differences in Gallagher and Kaufman (2005).
4 My discussion of women and accomplishment in the arts and sciences
is in Murray (2003): 265-293. For a complementary discussion, see
Simonton (1999): chapter 6.
5 For the story on grades, see Kimball (1989). For a review of the
literature on male-female differences in means and methods of
mathematical processing, see Geary, Saults, Liu et al. (2000). For
discussions of sex stereotyping, see Brown and Josephs (1999), Stipek
and Gralinski (1991), and several of the essays in Gallagher and
6 This ratio is based on the percentages of boys and girls from
Talent Search who later, as high-school students, got the top
possible score in the SAT-Math (12.7 percent of males and 1.9 percent
of females, given in Lubinski, Benbow, Shea et al., 2001). Julian
Stanley, who has been associated with Talent Search for many years,
is said to have asserted in an interview that the male:female ratio
among such students has dropped to about 3 to 1. I have not been able
to locate the interview or any data substantiating that ratio. In any
case, here is a reminder: currently, the 800 top score in the
SAT-Math is only about 2.6 standard deviations above the mean-that
is, it includes about one in 200 test-takers. This is nowhere close
to the extreme right end of the bell curve from which top
mathematicians are drawn.
7 Nyborg (in press) finds a sex difference in the general mental
ability g, not just in spatial skills, and evidence that the male
advantage increases exponentially as distance from the mean increases.
8 For a review of studies about sex differences in throwing ability,
see Geary (1998): 213-14, 284-85. For a presentation of the
evolutionary explanation, see Jones, Braithwaite, and Healy (2003)
and Kimura (1999): 11-30. It has also been argued that spatial skills
were an advantage in tool-making. See Wynn, Tierson, and Palmer
9 Geary (1998): 286-90; Kimura (1999): 43-66.
10 A continuing problem for evolutionary biology is the accusation
that its scholars observe human characteristics today and work
backward into a rationale that fits. But a sex difference in
visuospatial abilities is found in many other animals besides humans,
always favoring males-which gives good reason for thinking that in
this case we are observing something more than a just-so story. See
Jones, Braithwaite, and Healy (2003).
11 For a review of the evidence that male and female IQ is the same,
see Jensen (1998): 536-42. The underlying problem is that the
subtests in IQ tests have been developed and normed in ways that tend
to push male and female IQs toward the same mean IQ (for example,
items that show a large sex difference are usually discarded). For
the evidence that men have a higher mean IQ than women, see Ankney
(1992), Lynn (1999), Lynn and Irwing (2004), and Nyborg (in press).
12 See Goldstein, Seidman, Horton et al. (2001) and the
interpretation of those findings in Cahill (2005). This is far from a
settled issue. Research into the neurophysiology of sex differences
is exploring a variety of trails. For example, Gron, Spitzer, Tomczak
et al. (2000) discovered that men and women activate different parts
of the brain when they are working out navigation tasks, and do so in
patterns consistent with the proposition that navigation is
cognitively more difficult for women. Consistent evidence also links
the size of brain regions with level of capability (Cahill 2005).
This relationship between specific parts of the brain and capability
holds at an aggregate level as well: IQ is correlated with brain size
(adjusted for body size). The relationship of brain size to IQ has
often been derided (e.g., Gould 1981), and indeed brain size was a
problematic measure when it had to be based on skull size or
post-mortem data. But magnetic resonance imaging (MRI) studies of
brain size have ended the uncertainty about the existence of its
relationship with IQ. For meta-analyses of MRI and other in vivo
studies, see Jensen (1998): 147, which puts the correlation between
brain size and IQ at about .40, and McDaniel (2005), which puts it at
13 E.g., Johnson (1984), Casey, Nuttall, Pezaris et al. (1995), and
Geary, Saults, Liu et al. (2000). There has been dispute on this
point. Friedman (1995) argues that performance in math tests is more
strongly related to verbal ability than to visuospatial abilities.
Royer, Tronsky, Chan et al. (1999) present evidence that the real
source of the male advantage is faster retrieval of arithmetic facts
from long-term memory. A third line of argument has been that the
apparent male advantage is actually mediated by IQ (e.g., Linn and
Peterson 1985). Geary, Saults, Liu et al. (2000) controlled for IQ
and found that both visuospatial abilities and the computational
advantage found by Royer, Tronsky, Chan et al. (1999) were at work.
14 Casey, Nuttall, Pezaris et al. (1995).
15 Pinker (2002): 344-45.
16 Visuospatial skills are helpful across the entire range of items
(see Geary, Saults, Liu et al. 2000), but good verbal skills can
substitute in solving the less difficult items.
17 For a more detailed presentation of the evidence about the pattern
of female accomplishment in the arts and sciences, see Murray (2003):
18 Geary (1998): 20-28, 97-120.
19 For an analysis of sex differences in nurturing, written by a
committed feminist who is also a scientist (an anthropologist), see
Hrdy (1999). For a short review of studies on the importance of
children and of the biological sources of nurturing differences, see
Rhoads (2004): 190-222.
20 Simonton (1984): chapter 6.
21 Ochse (1990), Simonton (1994): chapter 5.
22 Benbow, Lubinski, Shea et al. (2000).
23 Ibid., 479. The figures in the text combine the data reported for
two separate cohorts.
24 For a meta-analysis of sex differences in risk-taking, see Byrnes,
Miller, and Schafer (1999). For a discussion of the role of
testosterone, see J.M. Dabbs and M.G. Dabbs (2000).
25 Lewontin (1972).
26 For a technical description of what has been labeled "Lewontin's
fallacy," see Edwards (2003). For a nontechnical statement of how the
understanding of this issue has been changing, see Leroi (2005).
27 Studies incorporating some variant of this type of analysis
include Bamshad, Wooding, Watkins et al. (2003), Bowcock,
Ruiz-Linares, Romfohrde et al. (1994), Calafell, Shuster, Speed et
al. (1998), Mountain and Cavalli-Sforza (1997), Rosenberg, Pritchard,
Weber et al. (2002), and Stephens, Schneider, Tanguay et al. (2001).
28 Tang, Quertermous, Rodriguez et al. (2005). The self-identified
ethnic groups consisted of non-Hispanic black, non-Hispanic white,
East Asian, and Hispanic. The statistical procedure was cluster
analysis. The algorithms in cluster analysis are not trying to find
groupings that correspond to any pre-identified characteristic of the
people in the sample-that is, the researchers did not use any
information about the physical characteristics that humans use to
identify ethnicity. Cluster analysis simply looks for
interrelationships among the genetic markers that identify
statistically distinct entities.
29 In Tang, Quertermous, Rodriguez et al. (2005), "Hispanic"
corresponded to a cluster, even though no one thinks of "Hispanic" as
a race. People do not need to belong to different races,
conventionally defined, to be genetically distinct.
30 Neisser, Boodoo, Bouchard et al. (1996).
31 The standard deviation is a statistic that (simplified) expresses
the average difference of all the scores from the mean. More
precisely, the standard deviation is calculated by squaring the
deviation from the mean for each score, summing all those squared
deviations, finding the mean of that sum, then taking the square root
of the result. Given a normal distribution-a bell curve-someone who
is one standard deviation above the mean is at the 84th percentile.
Two standard deviations above the mean put that person at the 98th
percentile. IQ tests are normed to have a mean of 100 and a standard
deviation of 15.
32 The black-white difference emerges as early as IQ can be tested,
but the gap is usually smaller in pre-adolescence. Among
pre-schoolers, the gap can be just a few IQ points. Why does it
increase with age? One obvious hypothesis is inferior schooling-e.g.,
Fryer and Levitt (2004). But black children attending excellent
schools also fall behind their white counterparts, as discussed
subsequently in the text and in note 14. The alternative explanation
is that the heritability of IQ increases with age for people of all
races, and this is reflected in black IQ scores in adolescence and
adulthood. See Jensen (1998): 178.
33 My analysis of its annual College-Bound Seniors report,
distributed as printed material prior to 1996 and available online
from 1996 onward.
A word about the method of calculating the difference. When comparing
scores from two groups, the preferred method is to divide the
difference in the two scores by the pooled standard deviations of the
two groups. The equation is
where N is the sample size, X is the sample mean, ? is the standard
deviation, and the subscripts a and b denote each group. When the
black-white difference for a specific test is reported subsequently
in the text, this equation has been used to compute it.
34 The Long Term Trend Study with consistent data for the NAEP from
the early 1970's through 2004 is now available in mathematics and
reading for students tested at ages nine, thirteen, and seventeen.
35 For nine-year-olds, the gap in reading scores expressed as points
was smaller in 2004 (26 points) than in 1988 (29 points), but the
difference in standard deviations was fractionally larger (.76
standard deviations in 2004 as compared with .74 in 1988).
36 Hedges and Nowell (1998): 154.
37 I will venture a prediction that a variety of academic achievement
measures in elementary and secondary school will soon show renewed
convergence because of the No Child Left Behind Act, which puts
schools under intense pressure to teach to the test in basic skills.
If students are drilled on limited ranges of subject matter, scores
will tend to rise. The more basic the tests are (that is, the easier
they are), the more that improvements among the least skilled will
affect the mean. Also, the higher the stakes facing a school-and the
No Child Left Behind Act makes those stakes very high indeed-the
greater will be the incentives for administrators to use some of the
many resources at their disposal to make the results come out right,
through the judicious manipulation of suspensions and absences, and
through outright cheating (yes, it has been known to happen). Some
convergence in black and white test scores will probably occur, but
partitioning that effect among the competing explanations is a task
that will take a few years. Insofar as the convergence has been the
result of teaching to the test and of artifacts, it will be temporary.
38 In a given year, IQ tests and academic tests administered to the
same sample will produce similar results. Thus, it is possible to
make a reasonably good guess about a person's IQ based on his SAT
score compared to the distribution of SAT scores in a given year, and
after taking the composition of the SAT population into account. But
the results of academic tests are sensitive to changes in academic
achievement, whereas IQ tests are explicitly designed to measure a
general mental factor, g, that is independent of academic
achievement. A notorious illustration of the way that academic test
scores can drop is the period during the 1960's and 1970's when SAT
scores declined substantially, even after accounting for changes in
the pool of test-takers (Murray and Herrnstein 1992). The
intelligence of American youth was not declining, just their academic
39 The significance of g-loadings is discussed later in the text. In
terms of interpreting trends over time, the problem is that tests are
not equally good measures of g. They go from poor (e.g., a basic
reading test) to excellent (the most highly g-loaded, individually
administered IQ test). It is as if you were trying to measure changes
in average height with measuring tapes of varying accuracy. For a
statement of the no-change position, see Gottfredson (2005a), or a
summary of her argument in Gottfredson (2005b).
40 The .97 figure comes from my analysis of the proxy AFQT score in
the most recent release of the 1997 cohort of the National
Longitudinal Study of Youth (NLSY). I call it a proxy score because,
eight years after the test battery was administered, the Armed Forces
still has not gotten around to creating an official AFQT score. The
version created by the NLSY staff is a composite of the same subtests
used for previous versions of the AFQT, and takes the subject's age
into account. The NLSY has released the percentile scores, which I
converted to standard scores. The analysis used the NLSY's sample
weights to make the results representative of the national
population. The NLSY data can be downloaded online.
41 I take the 1.10 figure from Roth, Bevier, Bobko et al. (2001), a
meta-analysis of the black-white difference in both achievement tests
and IQ tests. The Roth et al. results are necessarily reflective of
pencil-and-paper tests, because that is where the overwhelming
majority of published test data come from. With rare exceptions, the
data on individually administered IQ tests such as the Wechsler,
Stanford-Binet, and Woodcock-Johnson are limited to their periodic
standardization samples. The number of such studies is small. These
results are overwhelmed in a meta-analysis by the many more studies
based on pencil-and-paper tests.
The previous re-norming of the AFQT occurred in 1979, when the AFQT
was administered to the 1979 cohort of the NLSY. Herrnstein and
Murray (1994) put the black-white difference for that cohort at 1.21
standard deviations. Compared with that figure, the improvement in
the 1997 cohort (a .97 black-white difference) is .24 standard
deviations. But Neal (in press) has uncovered patterns in the answers
of black members of the 1979 cohort that indicate the 1979 cohort
produced an artificially low black mean.
First, some background: Any test that tries to measure cognitive
ability has to make assumptions about baseline skills. If a person
can read, even if not very well, then an IQ test can make use of
written items; if the subject is illiterate, it cannot. Similarly, if
a person knows numbers and the principles of basic arithmetic, even
if not very well, then an IQ test can make use of numeric problems;
but if the subject is innumerate, it cannot.
Neal argues that the pattern of answers for the 1979 cohort indicates
that "a substantial fraction of the NLSY79 sample of black males who
took the ASVAB test lacked the basic math and reading skills covered
by the exam, lacked any motivation to put forth effort during the
exam, or both," with a similar situation, not quite as bad, for black
females (Neal, in press: 13) Given the convergence in academic test
scores during the 1980's, it is likely that the proportion of the
1997 NLSY cohort so completely lacking in the basic skills was
smaller than in the 1979 cohort. If so, this change alone, not an
increase in cognitive ability, would produce convergence in the
black-white difference in the AFQT. In addition, the administration
of the ASVAB in 1997 was computer-adaptive. Instead of being
confronted with pages of questions (105 of them) as in the
traditional paper-and-pencil ASVAB (the kind used in 1979), subjects
saw one question at time, and the difficulty of each subsequent
question was adapted to the subject's previous answer-a method less
likely to provoke the kind of give-up response that Neal found in the
1979 data. Neal did not try to estimate the magnitude of the artifact
in the 1979 data, but if a "substantial fraction" of the NLSY males
had unrealistically low scores, some figure lower than 1.21 standard
deviations would be appropriate as a baseline for comparing the 1997
AFQT results. The overall black-white difference of 1.10 standard
deviations as found in the meta-analysis is the natural choice.
42 The black and white means on the WISC-IV's measure of full-scale
IQ were 91.7 and 103.2 respectively (Prifitera, Weiss, Saklofske et
al. 2005: 24). Standard deviations for computing the black-white
difference were supplied by the Psychological Corporation, which
produces the Wechsler tests.
43 The 1.05 and .99 figures come from my analysis of data for the
2001 standardization sample for the Woodcock-Johnson III (WJ-III)
test of cognitive ability, provided courtesy of the Woodcock-Munoz
Foundation. The results from the WJ-III are noteworthy because the
WJ-III provides the best known statistical estimate of g. Uniquely
among the major standardized tests, the scoring system for the WJ-III
uses principal-components analysis to find the best weighted
combination of subtests instead of treating all subtests equally
(Schrank, McGrew, and Woodcock 2001).
44 Two resourceful defenders of the environmental hypothesis about
the black-white difference, James Flynn and William Dickens, are
working on their own analysis of the black-white difference over time
that should materially add to the state of knowledge when it is
released. Here are a few examples of the ambiguities that complicate
the assessment of whether the IQ difference has changed, and that
have prevented me from stating a confident conclusion:
Example 1. One of the few sources that has several data points over
time with a consistent measure is the General Social Survey (GSS)
available online, conducted annually by the National Opinion Research
Center; which in most years through the 2000 survey, it included a
ten-item vocabulary test.
Example 2. The Kaufman Assessment Battery for Children (K-ABC) is a
test that has consistently shown smaller black-white differences than
other IQ tests. There are a number of reasons for this, one being
that subtests showing large black-white differences were excluded
(the K-ABC includes forward-digit span but not backward-digit span,
for example). See Jensen (1984) for a full discussion. But though the
black-white difference is smaller, it has not changed. In the manual
for the original standardization published in 1983, the means on the
"Mental Processing Composite" ( K-ABC's version of an IQ score) for
the white and black samples were 102.0 and 95.0 respectively (A. S.
Kaufman and Kaufman 1983: 152). Twenty-one years later, those means
were both within a point of their 1983 values-102.7 and 94.8
respectively (A.S. Kaufman and N.L. Kaufman 2004: 96). Which is more
meaningful? The smaller black-white difference shown by the K-ABC? Or
the absence of any convergence over time?
Example 3. In trying to discriminate between increases in IQ and
improvements in academic achievement, one strategy is to explore
which parts of the distribution of scores show the most change.
Convergence that occurs because of improvements at the bottom of the
distribution is likely to reflect remediation of fundamental
educational deficits, which could leave the IQ distribution more or
In their analysis of six major cross-sectional databases spanning the
period from 1965 to 1992, Hedges and Nowell (1998) found that "Racial
disparities have diminished over time in the lower tail, but not in
the upper tail" (159). In the NAEP, they found that "From 1980 to
1988 there was a substantial increase at all points on the black
distribution, with much greater change in the lower percentiles" for
the reading scores, and a similar pattern for math scores (161).
Another analysis, however, finds that almost all of the improvement
in scores has occurred among black students in the upper half of the
black distribution. For example, the AFQT math score of a black male
age 15-17 at the 70th percentile of the black distribution in 1980
was equivalent to the score of a white male at about the 28th
percentile of the white distribution (Neal, in press, Figure 2a). In
1997, a black male at the 70th percentile of the black distribution
had risen to about the 40th percentile of the white distribution.
Neal finds a similar result for math scores in the NAEP in the period
1978-1992/96 (Figures 2c and 2d). In contrast, Neal has found almost
no increases among students in the bottom half of the black
How can the results from two analyses be so different? The apparent
contradiction-it is not a real contradiction-arises from the fact
that almost all of the improvement of blacks in the upper half of the
black distribution represents improvement in scores in the lower half
of the national distribution of scores. But return to the example of
the AFQT: even in 1997, a black subject with a score that put him at
the 50th percentile of the white distribution-in other words, a
little above the overall national mean-was at about the 80th
percentile of the black distribution. In 1980, a black student had to
be at about the 90th percentile of the black distribution to have a
score above the national mean.
Which analysis should one use? That depends on the topic for which
one wants information. If the question is, "Who improved their
scores relative to whites, the students at the bottom of the black
distribution or the students at the top of it?," Neal's analysis
provides the correct answer. If the question is, "Did most of the
improvement in black scores occur at the bottom or the top of the
national distribution of scores?," then Hedges and Nowell's approach
provides the correct answer.
In deciding whether IQ has risen, how does one balance these results?
I am an optimist about the recent past. To me, the various ambiguous
indicators add up to the likelihood that a reduction in the IQ gap
has occurred alongside the reduction in the academic-achievement gap.
Forced to make a bet, I would guess that the black-white difference
in IQ has dropped by somewhere in the range of .10-.20 standard
deviations over the last few decades. I must admit, however, that I
am influenced by a gut-level conviction that the radical improvement
in the political, legal, and economic environment for blacks in the
last half of the 20th century must have had an effect on IQ. To
conclude that no narrowing whatsoever has occurred raises the
question, "How can that be?" One would have to argue that all of the
gains in some aspects of the environment have been counterbalanced by
new deficits in other aspects, and that those new deficits affect
different socioeconomic classes similarly. If the argument is
restricted to environmental changes, I cannot imagine how that case
might be made.
Another possibility is that improvement in the environmental causes
of IQ has been counterbalanced by what is known as "dysgenic"
fertility. For several decades at least, women with the highest IQs
have been having the fewest babies, and black women have been no
different from anyone else (Herrnstein and Murray 1994: chapter 15).
But the problem is especially acute among blacks because it is not
just black women above the national average IQ who are having the
fewest babies but women above the black average. Consider the results
for the women of the 1979 NLSY cohort, whose childbearing years are
effectively over (they ranged in age from thirty-eight to forty-five
when these numbers were collected). Using a nationally representative
subsample for the analysis, one finds that the mean AFQT score of the
black women was 85.7. Sixty percent of the children born to this
cohort were born to women with AFQT scores below that average.
Another 33 percent were born to women with scores from 85.7 to 100.
Only 7 percent were born to women with IQs of 100 and over.
Did the children do better? A total of 716 of them were tested with a
highly g-loaded verbal test, the Peabody Picture Vocabulary Test
(revised). The mean of the subset of mothers whose children were
tested was 83.7. The mean of their children was 80.2. The mothers and
children were tested with different instruments, so it should not be
concluded that the black mean actually went down in the new
generation. But these data certainly give no reason to think it went
It is thus technically possible that black IQ could have remained
about the same during the last half-century despite the revolutionary
changes for the better in the status of black Americans. Deciding
whether that in fact happened requires more evidence than I have
When I try to forecast the future, I become a pessimist. Here is how
I read the overall patterns of change in the academic achievement
tests versus the IQ tests:
In a world where Rushton and Jensen are right and the black-white
difference is 50- to 80-percent genetic, academic performance and IQ
will both improve as the environment improves, and for the same
reason: environment plays a role in both measures. Academic test
scores will begin to rise before IQ does, because academic
performance can improve immediately upon getting a better education
whereas the environmental factors affecting IQ are more diffuse. For
a related reason-changes in the quality of education can cause
substantial increases or drops in academic achievement, whereas IQ
cannot be changed much by any known discrete, time-limited
environmental change-convergence will be greater in academic
achievement than in IQ. Since the environmental role is only 20 to 50
percent of the total, the improvements in both academic and IQ test
scores will eventually level off as the limits of environmental
change are reached.
To me, the pattern we have observed since good longitudinal data
became available in the early 1970's is consistent with these
expectations. The only surprise is that evidence for convergence in
IQ scores has been so slow to emerge and so spotty. I interpret the
pattern as indicating that convergence is nearing an asymptote and
that not much will change in the future.
45 Blacks and whites have different distributions of socioeconomic
status (SES), and SES is correlated with IQ among both blacks and
whites. When the difference in black and white SES distributions is
statistically controlled, studies have typically found that the
black-white difference is reduced by about a third of a standard
deviation. But when blacks and whites of similar socioeconomic status
are compared with each other, the difference as measured in standard
deviations remains the same or increases as SES goes up. For a review
of the evidence on this point, see Herrnstein and Murray (1994):
46 I put aside here the explanation that has received the most
publicity in recent years, the phenomenon labeled "stereotype
threat." Its discoverers, Claude Steele and Joshua Aronson,
demonstrated experimentally that test performance by academically
talented blacks was worse when a test was called an IQ test than when
it was innocuously described as a research tool (Steele and Aronson
1995). Press reports erroneously interpreted this as meaning that
stereotype threat explained away the black-white difference. In
reality, Steele and Aronson showed only that it increases the usual
black-white difference; if one eliminates stereotype threat, the
usual difference remains.
The misrepresentation of these results in the mainstream media was
grotesque. For example, the narrator of the PBS television program
Frontline told his viewers that "blacks who believed the test was
merely a research tool did the same as whites." The Boston Globe
reported that "Black students who think a test is unimportant match
their white counterparts' scores." Newsweek reported that "blacks who
were told that the test was a laboratory problem-solving task that
was not diagnostic of ability scored about the same as whites." Such
claims have now infiltrated major psychology texts. The third edition
of Psychology by Davis and Palladino (2002) reports that "The results
revealed that African-American students who thought they were simply
solving problems performed as well as white students." Similar
statements have appeared in scientific journals. All of the above
examples are taken from Sackett, Hardison, and Cullen (2004). Sackett
et al. also have a nice description of how the research results
should have been described: "In the sample studied, there are no
differences between groups in prior SAT scores, as a result of the
statistical adjustment. Creating stereotype threat produces a
difference in scores; eliminating threat returns to the baseline
condition of no difference" (9).
Readers may follow the latest in the debate by reading a set of
responses to Sackett, Hardison, and Cullen (2004) in the April 2005
issue of American Psychologist, but nothing in the critiques
overturns the above description. The existence of stereotype threat
has indeed been demonstrated. It is an interesting phenomenon, and
some claims have been made that reducing stereotype threat can
improve scores on certain tests (Good, Aronson, and Inzlicht 2003),
but the widespread assertion that stereotype threat explains a
significant part of the observed black-white difference is wrong. The
dissemination of that false assertion is perhaps understandable in
the case of journalists who are not supposed to be sophisticated
about such topics. It is less easily explained away when done by
authors of technical articles and textbooks.
47 Ogbu (2003).
48 Sowell (2005).
49 Neisser, Boodoo, Bouchard et al. (1996): 95.
50 Neisser, Boodoo, Bouchard et al. (1996): 95. In truth, the closest
thing to direct evidence involves brain size, which is known to have
a correlation with IQ (see note 12) and to be different for blacks,
whites, and East Asians. See J.P. Rushton and E.W. Rushton (2003) for
a recent literature review of the evidence. But the task force did
not mention brain size. There is also no mention of IQ in sub-Saharan
Africa, the results of transracial adoption studies, the correlation
of the black-white difference with the g-loadedness of tests,
regression to racial means across the range of IQ, or other relevant
data. What the task force chose to define as "direct evidence" was a
study of children of American black soldiers born to German women
after World War II, and studies that use blood-group methods to
estimate the degree of African ancestry in American blacks. Both are
discussed at length in Rushton and Jensen (2005a) and Nisbett (2005).
51 Rushton and Jensen (2005a).
52 The other articles are Sternberg (2005), Nisbett (2005), Suzuki
and Aronson (2005), Gottfredson (2005b), and Rushton and Jensen
53 The ten categories, following Rushton and Jensen's wording, are as
follow: (1) the world-wide evidence of a consistent black-white-Asian
difference, (2) the greater black-white difference on g-loaded
subtests than on culture-bound subtests, (3) the greater black-white
difference on highly heritable subtests than on culturally malleable
subtests, (4) the association of the black-white-Asian difference
with differences in brain size, (5) the persistence of the
black-white-Asian difference among trans-racial adoptees, (6) the
consistency of the black-white difference with studies of racial
admixture, (7) regression of black and white relatives (offspring or
siblings) to their respective racial means, (8) consistency of the
black-white-Asian IQ differences with differences in 60 other
behavioral traits, (9) consistency of the black-white-Asian
differences with evolutionary explanations, and (10) the inability to
explain black-white-Asian differences with a zero-genetic model or
even with a 50-percent environmental model.
54 Rushton has posted all of the articles at his website.
55 Chakraborty, Kamboh, Nwankwo et al. (1992), Parra, Marcini, Akey
et al. (1998).
56 A variety of studies, summarized in Rushton and Jensen (2005a):
260-61, generally show that the IQs for mixed-race children are about
midway between those of children with two white and two black
parents. On the other hand, studies that characterized racial
composition based on blood group do not predict IQ (Nisbett 2005:
57 The results of such a study would be especially powerful if the
study also characterized variables like skin color, making it
possible to compare the results for subjects for whom genetic
heritage and appearance are discrepant. For example, suppose it were
found that light-skinned blacks do better in IQ tests than
dark-skinned blacks even when their degree of African genetic
heritage is the same. This would constitute convincing evidence that
social constructions about race, not the genetics of race, influence
the development of IQ. Given a well-designed study, many such
hypotheses about the conflation of social and biological effects
could be examined.
58 Spearman (1927): 379.
59 The average adult gets a digits-backward score of 5 (Jensen 1998:
263). You may compare your own score with the highest I have
observed, 13 and 12, achieved respectively by José Zalaquett, former
chairman of Amnesty International, and the political analyst Charles
Krauthammer. Zalaquett's score might have been higher if he had not
been in a car weaving through traffic at 70 miles per hour on the New
Jersey Turnpike. Krauthammer's score might have been higher if he
hadn't been driving.
60 Jensen (1998): 370.
61 A similarly clean example of a black-white difference is produced
by reaction-time tests, in which two different measures are taken:
the time it takes for the subject to respond to the lighted buttons
that constitute the stimulus (a g-loaded measure) and the time it
takes to move one's finger from the home button to the appropriate
lighted button (no g-loading). Black subjects have faster movement
times and slower response times-once again a contrast, consistent
with Spearman's hypothesis, produced at the same time with the same
examiner in the same setting. None of the usual ways to explain away
the black-white difference through cultural causes applies. See
Jensen (1998): 389-93.
62 Jensen (1998): 369-402.
63 Nyborg and Jensen (2000). It should also be noted that one test of
Spearman's hypothesis has been conducted comparing East Asians and
whites. The better the measure of g, the greater the advantage of
East Asians over whites. See Nagoshi, Johnson, DeFries et al. (1984).
64 Jensen's evidence has been accompanied by a debate over his method
of correlated vectors for testing Spearman's hypothesis. P.H.
Schönemann has argued, most extensively in Schönemann (1997), that
Jensen's evidence was no more than a statistical artifact, a claim
refuted by Dolan and Lubke (2001). But other ways in which the method
of correlated vectors might yield spurious results are still being
debated; e.g., Dolan (2000), Lubke, Dolan, and Kelderman (2001),
Dolan, Roorda, and Wicherts (2004), Ashton and Lee (in press). These
arguments are being carried on at an arcane methodological level. I
am making a limited claim about what Jensen has established beyond
dispute: when you take a battery of mental tests, subject them to a
factor analysis, and correlate the loadings on the first factor with
the size of the black-white difference, the correlation will average
about .6. The actual method of correlated vectors is more complicated
than this, and is described in Jensen (1998): 372-74.
65 Factor analysis can be conducted in many different ways, which has
led to widespread popular acceptance of one of Stephen Jay Gould's
allegations in his best-selling book, The Mismeasure of Man (1981),
namely, that g is a statistical artifact that appears only when
certain analytic choices are made. Actually, the opposite is true. A
single factor, typically explaining about three times as much
variance as all the other factors combined, emerges under all of the
normal methods of conducting a factor analysis. The only exception
occurs if the factor-analysis program is explicitly instructed to
apportion the variance in such a way that a single factor does not
emerge. But if you do that and then try to publish your results, the
reviewers will point out that if you hadn't issued that instruction,
you would have gotten a dominant single factor. As Richard Herrnstein
liked to say, "You can make g hide, but you can't make it go away."
For a review of this issue with sources, see the Afterword to the
softcover edition of The Bell Curve (559-62). For a technical
demonstration of the convergent results from alternative ways of
conducting a factor analysis, see Ree and Earles (1991). For a
wide-ranging set of articles about the current role of g in
understanding intelligence, see the articles in the special section
of the January 2004 issue of Journal of Personality & Social
Psychology commemorating the 100th anniversary of Spearman's
discovery of g. An overview is given in Lubinski (2004).
66 Gould (1981) still shapes the lay received wisdom about IQ tests,
but his denunciation of g was already technically outdated when it
was published. For an account of the differing ways in which The
Mismeasure of Man was assessed by the media and by scholars, see
Davis (1983). For a recent discussion of the nature of g and the
issues that Gould was wrong about, see Bartholomew (2004).
67 Jensen (1998): 182-89.
68 Jensen (1998): 137-68.
69 Haier, Jung, Yeo et al. (2004); Thompson, Cannon, Narr et al. (2001).
70 Let it be clear: I am not asserting that putting these two facts
together proves that the black-white difference is genetic. The logic
of the situation was memorably converted to an analogy in Lewontin
(1970) and adapted in Herrnstein & Murray (1994): 298. If you take
two handfuls of genetically identical seed corn and plant one in Iowa
and the other in the Mohave Desert, you will get a large group
difference in results despite the high heritability of the traits of
corn. William Dickens and James Flynn have operationalized the
analogy through a simulation model that produces a large black-white
difference from environmental factors even given high heritability
(Dickens and Flynn 2001). The validity of that model was subsequently
disputed by Loehlin (2002) and Rowe and Rodgers (2002), with a reply
by Dickens and Flynn (2002). But that debate does not pertain here.
The implications I describe follow simply from knowing that g is
highly heritable among blacks, as it is among all groups, and that
the black-white difference is largely a difference in g.
71 See te Nijenhuis, Voskuijl, and Schijve (2001), who also found
evidence, as did Neubauer and Freudenthaler (1994), that coaching
also reduced the g-loadedness of the test, and for the obvious
reason: noise has been introduced into the IQ score, changing the
score but not the thing that makes an IQ test predictive, g. An
athletic analogy may be usefully pursued for understanding these
results. Suppose you have a friend who is a much better athlete than
you, possessing better depth perception, hand-eye coordination,
strength, and agility. Both of you try high-jumping for the first
time, and your friend beats you. You practice for two weeks; your
friend doesn't. You have another contest and you beat your friend.
But if tomorrow you were both to go out together and try tennis for
the first time, your friend would beat you, just as your friend would
beat you in high-jumping if he practiced as much as you did.
72 Flynn (1984) is an early statement. Over the years since The Bell
Curve was published, it has been especially exasperating to be told,
or to see it written, that Herrnstein and I were wrong because we did
not know about the Flynn effect. We not only provided the first
discussion of the Flynn effect aimed at a general audience; we named
it (Herrnstein and Murray 1994: 307-09). Some scholars, notably J.
Philippe Rushton, have subsequently called it the "Lynn-Flynn
effect," thereby acknowledging Richard Lynn's role in identifying the
rise in IQ scores.
73 Flynn (1998).
74 An early statement of this evidence, based on analysis of the g
loadings of subtests, is Jensen (1998): 320-21. Rushton (1999)
elaborates, disputed in Flynn (1999) and Flynn (2000), with a
rejoinder in Rushton (2000). Since then the evidence that the Flynn
effect does not consist of increases in g has been augmented by an
independent method, multigroup confirmatory factor analysis (MGCFA),
which permits a test for factorial invariance between cohorts. In
less technical terms, the method tests for whether differences in IQ
scores between groups reflects true differences in g. See Lubke,
Dolan, Kelderman et al. (2003) for a description of the method and
its uses. Wicherts, Dolan, Hessen et al. (2004) used MGCFA on five
large databases: Dutch adults in 1967/68 and 1998/99; Danish draftees
in 1988 and 1998; Dutch high-school students in 1984 and 1994/95;
Dutch children in 1981/82 and 1992/93; and Estonian children 1934/36
and 1997/98. The authors found that the hypothesis of factor
invariance was untenable, and that the gains in intelligence-test
scores were not manifestations of increases in g. Previously, Dolan
(2000) and Dolan & Hamaker (2001) had used the MGCFA to test for
factor invariance between blacks and whites on IQ tests, and had
concluded that the results passed the MGCFA test. In other words, the
black-white differences were consistent with a difference in g. It
was this contrast in results that led Wicherts and his colleagues to
conclude that the Flynn effect would have little effect on the
75 Wicherts, Dolan, Hessen et al. (2004): 531.
76 In the text I ignore Europe, where both academic and political
elites have suppressed the discussion of group differences even more
effectively than in America. Contemporaneously, the European Union
has revolutionized free movement within Europe. That, combined with
immigration from outside Europe, legal and illegal, has produced
unprecedented population change in countries that historically have
been ethnically homogeneous.
Immigration poses problems for European countries that are
qualitatively different from those faced by the United States.
Becoming an American requires only that immigrants buy into a set of
American ideals. You can move to America from anywhere in the world,
be of any ethnicity, social class, or race, and become an American.
Assimilation is what America does-not as well as it used to, but
still pretty well. The European Union's immigration policy has,
willy-nilly, decided that now you can move to Denmark and become
Danish or move to France and become French. Is this true? Everyday
experience suggests that Denmark's culture works because it fits the
characteristics of Danes, that France's culture works because it fits
the characteristics of the French, and that these ethnic
characteristics are importantly different and deeply rooted, whether
in genes or in habits of the heart. Replace a large proportion of
French with Danes-let alone peoples more distant-and French culture
will be profoundly changed. But it is taboo among the elites to talk
about such things (although ordinary people sense what is at stake),
and so a momentous social experiment is under way without any reason
to think that its assumptions are correct, many historical reasons
for thinking they are wrong, and recurring stories on the evening
news suggesting that the social fabrics of Europe will be shredded
before the elites can make themselves come to grips with what they
have been doing.
77 A few systematic examinations of this issue have been published;
e.g., Lott (2000) on the effects of affirmative action on policing.
For a journalistic account of the effects of political correctness on
the Los Angeles Police Department, see Golab (2005).
78 Sommers (2001).
79 Satel (2002).
80 For examples of the effects of controlling for group differences
on a variety of outcomes and groups, see Herrnstein and Murray
(1994): chapter 14, Nyborg and Jensen (2001), and Kanazawa (2005).
81 See Pinker (2002): chapter 16 for a discussion of how politics
interacts with the acceptance of group differences.
82 Pinker (2002): 340.
Ankney, D. 1992. "Sex differences in relative brain size: The
mismeasure of woman, too?" Intelligence 16: 329-36.
Ashton, M.C. and K. Lee. in press. "Problems with the method of
correlated vectors." Intelligence.
Bamshad, M.J., S. Wooding, W.S. Watkins, et al. 2003. "Human
population genetic structure and inference of group membership."
American Journal of Human Genetics 72: 578-89.
Baron-Cohen, S. 2003. The Essential Difference: Male and Female
Brains and the Truth about Autism. New York: Basic Books.
Bartholomew, D.J. 2004. Measuring Intelligence: Facts and Fallacies.
Cambridge: Cambridge Univ. Press.
Benbow, C.P., D. Lubinski, D.L. Shea, et al. 2000. "Sex differences
in mathematical reasoning ability at age 13: Their status 20 years
later." Psychological Science 11: 474-80.
Bowcock, A.M., A. Ruiz-Linares, J. Romfohrde, et al. 1994. "High
resolution of human evolutionary trees with polymorphic
microsatellites." Nature 368: 455-57.
Brown, R.P. and R.A. Josephs. 1999. "A burden of proof: Stereotype
relevance and gender differences in math performance." Journal of
Personality & Social Psychology 76: 246-57.
Byrnes, J.P., D.C. Miller, and W.D. Schafer. 1999. "Gender
differences in risk taking: A meta-analysis." Psychological Bulletin
Cahill, L. 2005. "His brain, her brain." Scientific American, May.
Calafell, F., A. Shuster, W.C. Speed, et al. 1998. "Short tandem
repeat polymorphism evolution in humans." European Journal of Human
Genetics 6: 38-49.
Casey, M.B., R. Nuttall, E. Pezaris, et al. 1995. "The influence of
spatial ability on gender differences in mathematics college entrance
test scores across diverse samples." Developmental Psychology 31:
Chakraborty, R., M.I. Kamboh, M. Nwankwo, et al. 1992. "Caucasian
genes in American blacks." American Journal of Human Genetics 50:
Dabbs, J.M. and M.G. Dabbs. 2000. Heroes, Rogues, and Lovers:
Testosterone and Behavior. New York: McGraw-Hill.
Davis, B.D. 1983. "Neo-Lysenkoism, IQ, and the Press." The Public
Interest no. 73: 41-59.
Dickens, W.T. and J.R. Flynn. 2001. "Heritability estimates versus
large environmental effects: The IQ paradox resolved." Psychological
Review 108: 346-69.
-. 2002. "The IQ paradox is still resolved: Reply to Loehlin (2002)
and Rowe and Rodgers (2002)." Psychological Review 109: 764-71.
Dolan, C.V. 2000. "Investigating Spearman's hypothesis by means of
multi-group confirmatory factor analysis." Multivariate Behavioral
Research 35: 21-50.
Dolan, C.V. and G.H. Lubke. 2001. "Viewing Spearman's hypothesis from
the perspective of multigroup PCA: A comment on Schoenemann's
criticism." Intelligence 29: 231-45.
Dolan, C.V., W. Roorda, and J.M. Wicherts. 2004. "Two failures of
Spearman's hypothesis: The GATB in Holland and the JAT in South
Africa." Intelligence 32: 231-45.
Edwards, A.W.F. 2003. "Human genetic diversity: Lewontin's fallacy."
BioEssays 25: 798-801.
Flynn, J.R. 1984. "The mean IQ of Americans: Massive gains 1932 to
1978." Psychological Bulletin 95: 29-51.
-. 1998. "IQ gains over time: Toward finding the causes" in The
Rising Curve: Long-term Gains in IQ and Related Measures, edited by
U. Neisser. Washington: American Psychological Association.
-. 1999. "Evidence against Rushton: The genetic loading of WISC-R
subtests and the causes of between-group IQ differences." Personality
and Individual Differences 26: 373-79.
-. 2000. "IQ gains and fluid g." American Psychologist 55: 543.
Friedman, L. 1995. "The space factor in mathematics: Gender
differences." Review of Educational Research 65: 22-50.
Fryer, R.G. and S.D. Levitt. 2004. "Understanding the black white
test score gap in the first two years of school." Review of Economics
and Statistics 86: 447-64.
Gallagher, A.M. and J.C. Kaufman (eds). 2005. Gender Differences in
Mathematics: An Integrative Psychological Approach. Cambridge:
Cambridge Univ. Press.
Geary, D.C. 1998. Male, Female: The Evolution of Human Sex
Differences. Washington: American Psychological Association.
Geary, D.C., S.J. Saults, F. Liu, et al. 2000. "Sex differences in
spatial cognition, computational fluency, and arithmetical
reasoning." Journal of Experimental Child Psychology 77: 337-53.
Golab, J. 2005. "How racial PC corrupted the LAPD." American Enterprise.
Goldstein, J.M., L.J. Seidman, N.J. Horton, et al. 2001. "Normal
sexual dimorphism of the adult human brain assessed by in vivo
magnetic resonance imaging." Cerebral Cortex 11: 490-97.
Good, C., J. Aronson, and M. Inzlicht. 2003. "Improving adolescents'
standardized test performance: An intervention to reduce the effects
of stereotype threat." Journal of Applied Developmental Psychology
Gottfredson, L.S. 2005a. "Implications of cognitive differences for
schooling within diverse societies" in Comprehensive Handbook of
Multicultural School Psychology, edited by C. L. Frisby and C. R.
Reynolds. New York: Wiley.
-. 2005b. "What if the hereditarian hypothesis is true?" Psychology,
Public Policy, and Law 11: 311-19.
Gould, S.J. 1981. The Mismeasure of Man. New York: W. W. Norton.
Gron, G., M. Spitzer, R. Tomczak, et al. 2000. "Brain activation
during human navigation: Gender-different neural networks as a
substrate of performance." Nature Neuroscience 3: 404-08.
Haier, R.J., R.E. Jung, R.A. Yeo, et al. 2004. "Structural brain
variation and general intelligence." NeuroImage 23: 425-33.
Halpern, D.F. 2000. Sex Differences in Cognitive Ability. Mahwah, NJ:
L. Erlbaum Associates.
Hedges, L.V. and A. Nowell. 1998. "Black-White Test Score Convergence
since 1965." Pp. 149-81 in The Black-White Test Score Gap, edited by
C. Jencks and M. Phillips. Washington: Brookings Institution Press.
Herrnstein, R.J. and C. Murray. 1994. The Bell Curve: Intelligence
and Class Structure in American Life. New York: Free Press.
Hrdy, S.B. 1999. Mother Nature: A History of Mothers, Infants, and
Natural Selection. New York: Pantheon Books.
Jensen, A.R. 1984. "The Black-White difference on the K-ABC:
Implications for future tests." Journal of Special Education 18:
-. 1998. The g Factor: The Science of Mental Ability. Westport, CT:
Johnson, E.S. 1984. "Sex differences in problem solving." Journal of
Educational Psychology 76: 1359-71.
Jones, C.M., V.A. Braithwaite, and S.D. Healy. 2003. "The evolution
of sex differences in spatial ability." Behavioral Neuroscience 117:
Kanazawa, S. 2005. "Is 'discrimination' necessary to explain the sex
gap in earnings?" Journal of Economic Psychology 26: 269-87.
Kaufman, A.S. and N.L. Kaufman. 1983. Kaufman Assessment Battery for
Children: Interpretive Manual. Circle Pines, MN: AGS Publishing.
-. 2004. KABC-II Manual. Circle Pines, MN: AGS Publishing.
Kimball, M.M. 1989. "A new perspective on women's math achievement."
Psychological Bulletin 105: 198-214.
Kimura, D. 1999. Sex and Cognition. Cambridge, MA: MIT Press.
Leroi, A.M. 2005. "A family tree in every gene." Pp. 21 in New York
Times, March 14.
Lewontin, R.C. 1970. "Race and intelligence." Bulletin of the Atomic
Scientists 26: 2-8.
-. 1972. "The apportionment of human diversity." Evolutionary Biology
Linn, M.C. and A.C. Peterson. 1985. "Emergence and characterization
of sex differences in spatial ability: A meta-analysis." Child
Development 56: 1479-98.
Loehlin, J.C. 2002. "The IQ paradox: Resolved? Still an open
question." Psychological Review 109: 754-58.
Lott, J.R. 2000. "Does a helping hand put others at risk? Affirmative
action, police departments, and crime." Economic Inquiry 38: 239-77.
Lubinski, D. 2004. "Introduction to the special section on cognitive
abilities: 100 years after Spearman's (1904) '"General intelligence,"
objectively determined and measured'." Journal of Personality &
Social Psychology 86: 96-111.
Lubinski, D., C.P. Benbow, D.L. Shea, et al. 2001. "Men and women at
promise for scientific excellence: Similarity not dissimilarity."
Psychological Science 12: 309-17.
Lubke, G.H., C.V. Dolan, and H. Kelderman. 2001. "Investigating group
differences using Spearman's hypothesis: An evaluation of Jensen's
method." Multivariate Behavioral Research 36: 299-324.
Lubke, G.H., C.V. Dolan, H. Kelderman, et al. 2003. "On the
relationship between sources of within- and between-group differences
and measurement invariance in the common factor model." Intelligence
Lynn, R. 1998. "Has the black-white intelligence difference in the
United States been narrowing over time?" Personality and Individual
Differences 25: 999-1002.
-. 1999. "Sex differences in intelligence and brain size: A
developmental theory." Intelligence 27: 1-12.
Lynn, R. and P. Irwing. 2004. "Sex differences on the progressive
matrices: A meta-analysis." Intelligence 32: 481-98.
Matarazzo, J.D. 1972. Wechsler's Measurement and Appraisal of Adult
Intelligence. New York: Oxford Univ. Press.
McDaniel, M.A. 2005. "Big-brained people are smarter: A meta-analysis
of the relationship between in vivo brain volume and intelligence."
Intelligence 33: 337-46.
Mountain, J.L. and L.L. Cavalli-Sforza. 1997. "Multilocus genotypes,
a tree of individuals, and human evolutionary history." American
Journal of Human Genetics 61: 705-18.
Murray, C. 2003. Human Accomplishment: The Pursuit of Excellence in
the Arts and Sciences, 800 B.C. to 1950. New York: HarperCollins.
Murray, C. and R.J. Herrnstein. 1992. "What's really behind the
SAT-score decline." The Public Interest no. 106: 32-56.
Nagoshi, C.T., R.C. Johnson, J.C. DeFries, et al. 1984. "Group
differences and first principle-component loadings in the Hawaii
family study of cognition: A test of the generality of 'Spearman's
hypothesis.'" Personality and Individual Differences 5: 751-53.
Neal, D. in press. "Why has black-white skill convergence stopped?"
in Handbook of Economics of Education, edited by E. Hanushek and F.
Welch. New York: Elsevier.
Neisser, U., G. Boodoo, T.J. Bouchard, Jr., et al. 1996.
"Intelligence: Knowns and unknowns." American Psychologist 51: 77-101.
Neubauer, A.C. and H.H. Freudenthaler. 1994. "Reaction time in a
sentence-picture verification test and intelligence: Individual
strategies and the effects of extended practice." Intelligence 19:
Nisbett, R.E. 2005. "Heredity, environment, and race differences in
IQ: A commentary on Rushton and Jensen (2005)." Psychology, Public
Policy, and Law 11: 302-10.
Nyborg, H. In press. "Sex-related differences in general intelligence
g, brain size, and social status." Personality and Individual
Nyborg, H. and A.R. Jensen. 2000. "Black-white differences on various
psychometric tests: Spearman's hypothesis tested on American armed
services veterans." Personality and Individual Differences 28: 593-99.
-. 2001. "Occupation and income related to psychometric g."
Intelligence 29: 45-55.
Ochse, R. 1990. Before the Gates of Excellence: The Determinants of
Creative Genius. Cambridge: Cambridge Univ. Press.
Ogbu, J. 2003. Black American Students in an Affluent Suburb: A Study
of Academic Disengagement. Hillsdale, NJ: Lawrence Erlbaum Assoc.
Parra, E.J., A. Marcini, J. Akey, et al. 1998. "Estimating African
American admixture proportions by use of population specific
alleles." American Journal of Human Genetics 63: 1839-51.
Pinker, S. 2002. The Blank Slate: The Modern Denial of Human Nature.
New York: Viking Penguin.
Prifitera, A., L.G. Weiss, D.H. Saklofske, et al. 2005. "The WISC-IV
in the clinical assessment context." Pp. 3-32 in WISC-IV Clinical Use
and Interpretation: Scientist-Practitioner Perspectives, edited by A.
Prifitera, D. H. Saklofske, and L. G. Weiss.
Ree, M.J. and J.A. Earles. 1991. "The stability of convergent
estimates of g." Intelligence 15: 271-78.
Rhoads, S.E. 2004. Taking Sex Differences Seriously. San Francisco:
Rowe, D.C. and J.L. Rodgers. 2002. "Expanding variance and the case
of historical changes in IQ means: A critique of Dickens and Flynn
(2001)." Psychological Review 109: 759-63.
Rosenberg, N.A., J.K. Pritchard, J.L. Weber, et al. 2002. "Genetic
structure of human populations." Science 298: 2381-85.
Roth, P.L., C.A. Bevier, P. Bobko, et al. 2001. "Ethnic group
differences in cognitive ability in employment and educational
settings: A meta-analysis." Personnel Psychology 54: 297-330.
Royer, J.M., L.N. Tronsky, Y. Chan, et al. 1999. "Math-fact retrieval
as the cognitive mechanism underlying gender differences in math test
performance." Contemporary Educational Psychology 24: 181-266.
Rushton, J.P. 1999. "Secular gains in IQ not related to the g factor
and inbreeding depression-unlike Black-White differences: A reply to
Flynn." Personality and Individual Differences 26: 381-89.
-. 2000. "Flynn effects not genetic and unrelated to race
differences." American Psychologist 55: 542-43.
Rushton, J.P. and A.R. Jensen. 2005a. "Thirty years of research on
race differences in cognitive ability." Psychology, Public Policy,
and Law 11: 235-94.
-. 2005b. "Wanted: More race realism, less moralistic fallacy."
Psychology, Public Policy, and Law 11: 328-36.
Rushton, J.P. and E.W. Rushton. 2003. "Brain size, IQ, and
racial-group differences: Evidence from musculoskeletal traits."
Intelligence 31: 139-55.
Sackett, P.R., C.M. Hardison, and M.J. Cullen. 2004. "On interpreting
stereotype threat as accounting for African American-white
differences on cognitive tests." American Psychologist 59: 7-13.
Satel, S. 2002. PC, M.D.: How Political Correctness is Corrupting
Medicine. New York: Basic Books.
Schrank, F.A., K.S. McGrew, and R.W. Woodcock. 2001. Technical
Abstract (Woodcock-Johnson III Assessment Service Bulletin No. 2).
Itasca, IL: Riverside Publishing.
Schönemann, P.H. 1997. "Famous artifacts: Spearman's hypothesis."
Cahiers de Psychologie Cognitive 16: 665-94.
Simonton, D.K. 1984. Genius, Creativity, and Leadership. Cambridge,
MA: Harvard Univ. Press.
-. 1994. Greatness: Who Makes History and Why. New York: Guilford Press.
-. 1999. Origins of Genius: Darwinian Perspectives on Creativity.
Oxford: Oxford Univ. Press.
Sommers, C.H. 2001. The War Against Boys: How Misguided Feminism Is
Harming Our Young Men. New York: Simon & Schuster.
Sowell, T. 2005. Black Rednecks and White Liberals. San Francisco:
Spearman, C. 1927. The Abilities of Man. New York: Macmillan.
Steele, C.M. and J. Aronson. 1995. "Stereotype threat and the
intellectual test performance of African Americans." Journal of
Personality & Social Psychology 69: 797-811.
Stephens, J.C., J.A. Schneider, D.A. Tanguay, et al. 2001. "Haplotype
variation and linkage disequilibrium in 313 human genes." Science
Sternberg, R.J. 2005. "There are no public policy implications: A
reply to Rushton and Jensen (2005)." Psychology, Public Policy, and
Law 11: 295-301.
Stipek, D.J. and J.H. Gralinski. 1991. "Gender differences in
children's achievement-related beliefs and emotional responses to
success and failure in mathematics." Journal of Educational
Psychology 83: 361-71.
Suzuki, L. and J. Aronson. 2005. "The cultural malleability of
intelligence and its impact on the racial/ethnic hierarchy."
Psychology, Public Policy, and Law 11: 320-27.
Tang, H., T. Quertermous, B. Rodriguez, et al. 2005. "Genetic
structure, self-identified race/ethnicity, and confounding in
case-control association studies." American Journal of Human Genetics
te Nijenhuis, J., O.F. Voskuijl, and N.B. Schijve. 2001. "Practice
and coaching on IQ tests: Quite a lot of g." International Journal of
Selection and Assessment 9: 302-08.
Thompson, P., T.D. Cannon, K.L. Narr, et al. 2001. "Genetic
influences on brain structure." Nature Neuroscience 4: 1-6.
Valian, V. 1999. Why So Slow? The Advancement of Women. Cambridge,
MA: MIT Press.
Wicherts, J.M., C.V. Dolan, D.J. Hessen, et al. 2004. "Are
intelligence tests measurement invariant over time? Investigating the
nature of the Flynn effect." Intelligence 32: 509-37.
Wynn, T.G., F.D. Tierson, and C.T. Palmer. 1996. "Evolution of sex
differences in spatial cognition." Yearbook of Physical Anthropology
More information about the paleopsych