April 26, 2006
Fryer and Levitt on Race and Intelligence
Brad Delong points to this paper by Roland Fryer and Steven Levitt. (The whole working paper is available at Fryer's site.)
In tests of intelligence, Blacks systematically score worse than Whites, whereas Asians frequently outperform Whites. Some have argued that genetic differences across races account for the gap. Using a newly available nationally representative data set that includes a test of mental function for children aged eight to twelve months, we find only minor racial differences in test outcomes (0.06 standard deviation units in the raw data) between Blacks and Whites that disappear with the inclusion of a limited set of controls. The only statistically significant racial difference is that Asian children score slightly worse than those of other races. To the extent that there are any genetically-driven racial differences in intelligence, these gaps must either emerge after the age of one, or operate along dimensions not captured by this early test of mental cognition. A calibration exercise demonstrates that the observed patterns in the data can be generated by a model in which there are extremely small mean differences in intelligence across races, but where there are large racial differences in environmental factors that grow in importance as children age.
Posted by Robin Varghese at 10:23 AM | Permalink
TrackBack
TrackBack URL for this entry: http://www.typepad.com/services/trackback/6a00d8341c562c53ef00d834bb335169e2
Listed below are links to weblogs that reference Fryer and Levitt on Race and Intelligence:
Comments
A single study? Come on, the literature is vast. Which is why is why the overwhelming majority of psychologists who specialize in this field would disagree with the implications of this article. Use Google and find out for yourself.
Posted by: Luke Lea | Apr 26, 2006 10:44:32 AM
Luke
I think you overstate the agreement in the profession regarding the relationship between genetics and IQ. The main problem I see with the study is the difficulty involved in measuring IQ at such an early age. The authors, quite rightly, mention this in their discussion.
A good question to ask is why the Bayley is not more highly correlated with later IQ scores. Does it have to do with cultural insensitivity (in the good sense)? Or is it just that there is not way to measure the important cognitive functions that haven't come on line at such an early age.
Worthy of discussion.
John
Posted by: John | Apr 26, 2006 3:24:57 PM
John,
I am hardly an expert, but I do follow the literature.
Granted, the cause of the differences are not as well understood as their existence, and their persistence both in time (since iq tests started being widely used) and space (ie, the same groups in different countries and parts of the world).
The youngest age at which measurable differences start showing up, using proxies (reaction times), is I believe around 2 years, but I could be wrong about this.
Steve Sailer is an honest, responsible journalist -- and a fearless one -- who reports on these issues regularly and without bias; certainly he is mindful of the best interests of all groups, and has many constructive suggestions. (E.g., he has been recommending nutrition supplements for the peoples of Africa for years now, which he thinks is a significant factor for the abysmally low scores on that continent.) I recommend his web site highly: isteve.com
BTW, the only reason I think these issues are important in the first place is that realism is a requirement -- maybe the first requirement -- of moral responsibility. Without a committment to the critical and unbiased search for truth, we will never find our way to a better, fairer, or more just social order, in my opinion.
Posted by: Luke Lea | Apr 27, 2006 12:22:30 AM
Luke,
I agree wholeheartedly with your statement, "Without a committment to the critical and unbiased search for truth, we will never find our way to a better, fairer, or more just social order...".
I was deep into issues of test bias and test fairness earlier in my career. Part of the research issues we were dealing with was developing good statistical techniques to determine whether test items or whole tests were biased or fair for one group over another. Initially, my experience was in educational research and later in the corporate world in human resources research. I attended many lectures and conferences dealing with the views of Nobel Laureate William Shockley, and a major lecture by Psychologist Arthur Jensen at the APA's annual convention.
After many years of dealing with issues of race differences on intelligence, aptitude and ability tests, I have formed the following personal and professional views:
1. There is a significant genetic component that accounts for INDIVIDUAL DIFFERENCES on typical tests of general mental abilities (IQ). These typical IQ tests usually have a verbal and a performance component. This is true for everyone and for all race groups.
2. There is NO basis for a significant genetic component that accounts for GROUP DIFFERENCES among races on typical test of general mental abilities (IQ).
3. Very often, research supporting a significant genetic component for INDIVIDUAL DIFFERENCES, is used as an explanation for observed GROUP DIFFERENCES. INDIVIDUAL DIFFERENCES on IQ does not generalize to GROUP DIFFERENCES among races.
4. If there is any truth to a significant genetic component for GROUP DIFFERENCES on IQ, it must rest on a biological/genetic construct of distinct races within our human species. If there is no viable biological/genetic construct of races (as we typically tend to view them), that can accomodate differences in mental functioning, then observed race group differences cannot be attributed to genetics.
5. Recent research on the biological/genetic construct of race does not support genetic differences on mental functioning. See the Lewontin link, below, that supports the notion that race group differences, "...are only skin deep."
6. Realism suggests we have to deal with and accept manifest group differences. It also obligates us to come up with a better understanding of the causes, remediation, recovery and prevention of such disparities. It should also enlighted social policy.
Here is a collection of links on the biological-genetic side of the discussion:
http://evolvethought.blogspot.com/2006/04/races-geography-and-genetic-clusters.html
I highly recommend a streaming video lecture by Richard Lewontin at:
http://webcast.ucsd.edu:8080/ramgen/UCSD_TV/8456.rm
There is also a discussion on the topic at:
http://scienceblogs.com/pharyngula/2006/04/that_question_of_race.php
Posted by: Norman Costa | Apr 27, 2006 4:01:52 PM
Dear Norman Costa,
I wish I could agree with you but I can't.
Posted by: Luke Lea | Apr 28, 2006 12:59:04 AM
Dear John and Norman Costa and other readers,
Rather than spouting off my personal opinions, allow me to offer a couple of links to sites that deal with this controversy in an unbiased way. The first is Gene Expression, moderated by two South Asian immigrants to the U.S.. One a Bengladeshi of Muslim descent (though now an atheist) and the other of Hindu descent (also an atheist). Both are able to walk through the mindfield of this controversial topic in a sure footed way, a rare accomplishment indeed: here is the link to all their posts on the subject of IQ and race from a Google search: http://www.google.com/search?num=100&hl=en&lr=lang_en&safe=off&c2coff=1&domains=gnxp.com&q=IQ+RACE&btnG=Search&sitesearch=gnxp.com
The other link is to Steve Sailer's website, mentioned above. First a search under the heading IQ: http://www.isteve.com/Articles_IQ.htm
The second is a search under the heading of race: http://www.isteve.com/Articles_Race.htm
These are three fearless writers who do indeed engage in the critical and unbiased search for truth unblinkered by the dogma's of political correctness. They are among the smartest guys on the planet, certainly in their verbal talents and analytical abilities.
Another interesting site is by a character who goes by the moniker of La Griffe du Lion. His work combines mathematical sophistication with literary art, and takes the form of a Socratic dialogue: http://www.lagriffedulion.f2s.com//
I defy anyone who is unfamiliar with the academic liturature on this subject to drink deeply of these three sites -- or even drink lightly -- and not come away with a sense that the mainstream media's conventional wisdom on this subject is itself highly biased, with tragic results for public policy and the general welfare of our society for all groups in it.
Posted by: Luke Lea | Apr 28, 2006 9:16:01 AM
Luke,
Thanks for your post. I think we may have the makings of a very good discussion here. I'd like to hear more of your personal opinions on the subject. There's nothing like taking the time to carefully articulate our views and exposing them to analysis and comment. That's what I find so enjoyable in a great post-counterpost with informed people.
I took a look at some of your links and there is a lot of stuff there (non-fluff, serious posts as far as I can tell). Perhaps you could focus on a few for a discussion.
Here's what I distilled from the links I checked out. (It's been a few years since I was knee-deep in the subject, but the fundamental issues haven't changed.)
1. Some researchers like to focus on an underlying, and more fundamental, construct for IQ called 'g'. Let's call this hypothetical construct pure IQ, for convenience.
2. Specific IQ tests are attemping to capture some manifestation of 'g'.
3. There is a long history of research supporting a signicant genetic component for 'g'. In science, nothing is ever absolutely certain, but this is about as certain as you can get.
4. The support for a genetic component for 'g' rests solely--and this is very important--on INDIVIDUAL DIFFERENCES in IQ. The best evidence comes from twin studies, both fraternal and identical. Put another way: Q. Why are some people smarter than others? A. In part, individual differences in genetic makeup. If you asked the question this way: Q. Why are some African-American people smarter than other African-Americans? The answer would still be the same.
5. Manifest group differences among races have been observed over expanses of time and geography. These observed differences are not disputed, but the causes and interpretations are.
6. The question that is debated is whether or not GROUP DIFFERENCES on IQ, or 'g', can be attributed, at least in part, to genetic differences among the races.
7. The argument for a genetic component to race differences in 'g' is based on the following: (A) The persistence of observed differences over time and geography. (B) Taking evidence for a genetic component for INDIVIDUAL DIFFERENCES in 'g' as defacto evidence for a genetic component for GROUP DIFFERENCES in 'g'. (C) The following logical analysis, " What are the chances that historically separated populations will have all the exact same aptitudes?...[I]t is unlikely, that by chance and genetic drift various human populations will develop their own frequencies of various alleles that code for phenotypes. They will have their own response to environments and adapt both culturally and biologically. 'tis nature as work. Man is a social animal, but biology is still the bedrock of our nature." [taken from a post by razib on gene expressions]. (D) Races exist as a biological/genetic construct. (E) Race as a biological/genetic construct can contain 'g' as race-linked DNA, in exactly the same way we speak of sex-linked genetic compositions.
The case for a genetic component for GROUP DIFFERENCES on any characteristic can never rest on evidence of genetic INDIVIDUAL DIFFERENCES.
The persistence of group disparity over time and distance is compelling observation, but by itself is not evidence for a genetic component.
The logical analysis that separated human groups will develop separate aptitudes is a reasonable hypothesis, but it must be supported by genetic evidence that race groups have, in fact, developed sufficiently in isolation to manifest profound variations in 'g'. The work of Richard Lewontin and others is finding such tiny differences in race group genetics as to completely dislodge the viability of the biological/genetic construct of race within the human species. They state their conclusion this way, "Race is only skin deep."
If race is not a sufficiently viable construct then it follows that it is not sufficiently viable to describe other characteristics as 'race-linked.'
Posted by: Norman Costa | Apr 28, 2006 1:18:57 PM
Luke,
I did take a closer look at the le griffe du lion website. I hesitate to use sweeping generalizations, but it contains a lot of racist claptrap that mascerades as imitation academic discussion. The isteve website seems earnest but poorly equipped to do the proper analysis and reporting on issues of race and IQ.
Posted by: Norman Costa | Apr 29, 2006 1:58:56 AM
However you juggle the statistics, it always seems to turn out that black populations, whether indigenous or immigrant, do less well economically than Asian or European populations. Look at the success of the Asians brought to East Africa and the Carribean by the British, or that of the traumatized Vietnamese boat people who started to thrive economically as soon as they arrived in the US, leading to resentment from Afro Americans. Its acceptable to say that Black people have a greater sense of musical rythm or greater physical strength than whites, but its taboo to suggest that other differences may be to their disadvantage.
Posted by: aguy109 | Apr 29, 2006 4:38:15 PM
Aguy109,
It was not always the case that black populations did less well than Asian or European populations. The great Nile civilizations of 5,000 years ago, the Islamic empire and civilizations of 600s to 1200s, and the great southern India civilizations, were all examples of high achievement, empire and civilizations that eclipsed the Europeans and rivaled the Mongol and Chinese empires. There are others.
Jared Diamond deals with the passage into recent times of the ascendency of European civilization in his book and PBS program, "Guns, Germs, and Steel: The Fates of Human Societies". It's not taboo to suggest that other differences may be to the detriment of one group versus another. Diamond does this very well in ground breaking theory and research. For example, he shows how geography and latitude accounts for why some groups have more stuff than others.
Remember, observed group differences that are not explained by causes A, B, and C, do not automatically default to a presumed cause of genes. A default answer is not an answer. Also, genetic differences between race groups are tiny compared to within race group differences. This is debunking the biological/genetic construct of race as a viable reference for anything we might label "race-linked" as we do for the label "sex-linked" in denoting significant genetic differences.
Posted by: Norman Costa | Apr 29, 2006 6:35:16 PM
Norman Costa:
I am not an expert nor do I pretend to be. I simply follow the academic literature. There are a number of studies beginning to appear that actually identify somne of the specific genes that are linked to intelligence -- the Cochran-Harpending paper on Ashkenazi Jews, in particular, received attention in the NYT. Once a number of such genes have been identified, it becomes an empirical question whether these genes are spread evenly among the various population groups that make up the human family. There is obvious human biodiversity in physical abilities, for instance, long and short-distance running. Will the same turn out to be the case with mental abilities? We will just have to wait and see.
In the meantime, best to develop a philosophy of the general welfare that can take it either way: all human beings are morally equal, in the sense that they enjoy the same rights, and that anybody's happiness and welfare is just as important as anybody else's, and a democratic government ought to govern accordingly.
Posted by: Luke Lea | Apr 30, 2006 11:38:30 PM
"5. Recent research on the biological/genetic construct of race does not support genetic differences on mental functioning. See the Lewontin link, below, that supports the notion that race group differences, "...are only skin deep."
See comment on Lewontin Fallacy.
"Further technical comment: you may have read the misleading statistic, spread by the intellectually dishonest Lewontin, that 85% percent of all human genetic variation occurs within groups and only 15% between groups. The statistic is true, but what is often falsely claimed is that this breakup of variances (larger within group than between group) prevents any meaningful genetic classification of populations. This false conclusion neglects the correlations in the genetic data that are revealed in a cluster analysis. See here for a simple example which shows that there can be dramatic group differences in phenotypes even if every version of every gene is found in two groups -- as long as the frequency or probability distributions are distinct. Sadly, understanding this point requires just enough mathematical ability that it has eluded all but a small number of experts.) Update: see here for an explanation in pictures of Lewontin's fallacy. I also edited the paragraph above for clarity.
On the other hand, for most phenotypes (examples: height or IQ, which are both fairly heritable, except in cases of extreme environmental deprivation), there is significant overlap between different population distributions. That is, Swedes might be taller than Vietnamese on average, but the range of heights within each group is larger than the difference in the averages. Nevertheless, at the tails of the distribution one would find very large discrepancies: for example the percentage of the Swedish population that is over 2 meters tall (6"7) might be 5 or 10 times as large as the percentage of the Vietnamese population. If two groups differed by, say, 10 points in average IQ (2/3 of a standard deviation), the respective distributions would overlap quite a bit (more in-group than between-group variation), but the fraction of people with IQ above some threshold (e.g., >140) would be radically different. It has been claimed that 20% of all Americans with IQ > 140 are Jewish, even though Jews comprise only 3% of the total population.
...The imbalance continues to increase for still higher IQ’s. New York City’s public-school system used to administer a pencil-and-paper IQ test to its entire school population. In 1954, a psychologist used those test results to identify all 28 children in the New York public-school system with measured IQ’s of 170 or higher. Of those 28, 24 were Jews.
There is no strong evidence yet for specific gene variants (alleles) that lead to group differences (differences between clusters) in behavior or intelligence, but progress on the genomic side of this question will be rapid in coming years, as the price to sequence a genome is dropping at an exponential rate.
What seems to be true (from preliminary studies) is that the gene variants that were under strong selection (reached fixation) over the last 10k years are different in different clusters. That is, the way that modern people in each cluster differ, due to natural selection, from their own ancestors 10k years ago is not the same in each cluster -- we have been, at least at the genetic level, experiencing divergent evolution.
In fact, recent research suggests that 7% or more of all our genes are mutant versions that replaced earlier variants through natural selection over the last tens of thousands of years. There was little gene flow between continental clusters ("races") during that period, so there is circumstantial evidence for group differences beyond the already established ones (superficial appearance, disease resistance)."
http://infoproc.blogspot.com/2008/01/no-scientific-basis-for-race.html
Posted by: B Pearle | Jan 13, 2010 12:43:02 AM
It is obviously very important to know when these supposed racial differences emerge, and especially what we're measuring and how good/fair our methods of measurement are, but knowing that at they occur after age one says nothing obvious about causes.
You wouldn't find differences in breast size between boy and girl toddlers either. Man-boobs notwithstanding, I hope no-one thinks that difference among adults is explained by nurture or sexism. Rinse and repeat for family history and susceptibility to Alzheimer's or heart disease. Meanwhile, Gujarati and Korean children at age one already distinguish different sets of phonemes, which has nothing to do with genes.
Posted by: prasad | Jan 13, 2010 5:25:29 AM
Post a comment