about me
game theory
... more
Subscribe Weblog

Say something about IQ, as Tyler Cowen did over at Marginal Revolutions. Brad DeLong did something similar before, as well as Kevin Drum, and Andrew Sullivan. All you have to do is take the issue of IQ seriously, and bam!

The debate centers around three key issues that are somewhat independent: IQ is 1) important (has socio-economic consequences) 2) heritable (say 50%) and 3) varies systematically between different genetic families (eg, races). These factual hypotheses--and though racists believe in them, they are still statements about reality--and people on both sides have strong opinions, but the IQ crowd is generally more knowledgeable about what they are speaking because if you believe in IQ, you read up on it, while if you think it is meaningless, you don't. So when the anti-IQ crowd makes silly comments that they think definitively refute the above points, they face an avalanche of commenters who know much more about the issue than they do (usually these double to quadruple and average econ blog's comment).

The standard anti-IQ cliches include about how poor natives are so smart (eg, Jared Diamond's assertion that New Guinea tribesmen are smarter than Europeans), or that because race explains only 15% of genetic variability that races don't exist (it's the correlation in the variation that matters), that IQ can be altered significantly through early education (Heckman now thinks it's the non-IQ components that can be changed, and they are economically important), that the Flynn Effect will save us all (it stopped), that it's not nature but the combination that is essential (who said 100% nature?), etc., are smacked down by a rabid pack who have been reading GNXP and Steve Sailer.

To Cowen's point--that natives are smart because they can fish and gossip very well--I think the key point, is that while tribesmen are much better than Cowen at living off the land, so is my dog, and I'd be hesitant to say he's got a higher IQ (though certainly more 'fit' if we were in the forest).
Paul N (guest) meinte am 19. Jul, 03:25:
What are the policy implications of believing in IQ? I can't think of any. The issue seems to exist only so people can argue. I agree with TC, I think his point is natives don't have practice at IQ questions so they won't do well on IQ tests, not that being able to fish makes you smart. I'd suggest that a random native New Guinean baby raised in the US would have an average 100 IQ but I'm not a IQ stats fan so I'd better not.

IQ may correlate with success but I know a lot of high IQ math nerds who still live in their parents' basements and a lot of hard working sub-120 IQ people who are doctors. 
TGGP (guest) antwortete am 22. Jul, 00:24:
Why it matters if IQ matters
Bryan Caplan explains why it is important here
Jeff D (guest) meinte am 19. Jul, 04:31:
I have always wondered how one could except that different racial groups are better at certain physical activities (for example, west Africans tend to be better sprinters than other groups and Kenyans who grow up in the Rift Valley area tend to be good long distance runners) can't believe that we also vary racially in terms of intelligence. 
ruth (guest) meinte am 20. Jul, 15:24:
i am not one of the pc crowd, yet i find your unqualified dog-native comparison deeply offensive. yes, i do know you did not want to say that papua-neuguineans have the same iq as dogs. if you don't understand where my problem is, imagine how a similar comparison would sound to you if you found it on a arabic muslim website directed at white americans. eg "infidels are chaste like a dog is chaste". remember also that verbally dehumanizing people has a rather long tradition in the context of kinds of behaviour, like mass murder, that presumably you do not want to be associated with. i am ill in bed at the moment and have been surfing around a bit, and let me tell you that on some german nazi blogs, discourse about race is less aggressive in tone and less full of contempt for races and cultures not one's own than i have recently found on a number of presumably non-nazi english-language websites.
i thought cowen wanted to say that mexican villagers seem and probably are intelligent, yet would fail on a standard written iq test because of lack of familiarity with written material and written tests in general, that is, for cultural reasons. if that was not his point, then indeed his comment was not particularly pertinent. (nor by the way was your dog comment, because dogs are instinctive hunters physically well equipped for hunting and humans aren't. humans use the their intelligence and general learning and problem solving abilities "to live off the land", so diamond's obeservation was quite legitimate and relevant.)
having read up the subject race/iq quite extensively in the scholarly journals a few years ago, my conclusion was that genetic iq differences between races are not as well attested as people like Lynn (who is extremely unknowleadgabe about a number of subjects he would have needed to research for his argument, like, for example, cultural, economic and demographic history) claim.
Reseach so far has, for obvious reasons, been largely epidemiological,
and epidemiological studies (eg showing population-level associations between nutrition and diabetes, between american nationality and aggression, african race and iq) are not particularly good at sorting out cause-effect relationships. in medicine, you use them for hypthesis-building rather than for hypothesis-testing. what you need is controlled experiments (especially with a trait like iq that is clearly influenced by more than one variable). twin/adoption studies come close to doing that for iq, but these studies are small and at least the ones i have read are contradictory.
also, there are those epidemiological data that don't quite fit the trend.
why, for example, do pakistani immigrant children have dismal educational achievement in britain, while children from an indian background excel, far surpassing the white british group? that doesn't fit the thesis that race is the main factor in population-level differences, or even that it is a significant factor, the difference in educational achievement between indians and pakistais in britain being rather more noticable than the black-white divide in the states. (Perhaps you will tell me there are large genetics difference between Pakistanis/Bangladeshis and Indians. Are there? Intuitively, it's abit unlikely. Also, lynn lumps them with the middle easterners as having low iq, so obviously, they have low average iq in india. or they have in the studies lynn cites, his selection is not always impeccably impartial.)
just as people on the contra side of the debate on racial differences in iq should drop their attitide that something they do not want to be true cannot be true, so people on the other side of the divide should be a bit more cautious and less gleeful. 
ben tillman (guest) antwortete am 20. Jul, 16:30:
"i am not one of the pc crowd, yet i find your unqualified dog-native comparison deeply offensive."

It wasn't a comparison. It was a reductio ad absurdum. 
TangoMan (guest) antwortete am 20. Jul, 19:54:
why, for example, do pakistani immigrant children have dismal educational achievement in britain, while children from an indian background excel, far surpassing the white british group? that doesn't fit the thesis that race is the main factor in population-level differences, or even that it is a significant factor, the difference in educational achievement between indians and pakistais in britain being rather more noticable than the black-white divide in the states.

You raise a very interesting case study, but I think that you're jumping to conclusions too quickly. Consider the genetic consequences that follow from cultural practices.

First the cultural practice that most dramatically distinguishes Pakistanis from Indians:
A Labour MP has called for an end to the practice of cousin marriage after a Newsnight investigation reports that British Pakistanis are 13 times more likely to have children with recessive disorders than the general population.

The same research shows that British Pakistanis account for 3.4% of all births but have 30% of all British children with recessive disorders.

It is estimated that at least 55% of British Pakistanis are married to first cousins.

Mrs Cryer's constituency is the Bradford area, where local doctors estimate that three quarters of marriages in Bradford's Pakistani community are between first cousins.
Secondly, consider the following study:
The 50 inbred children were products of marriages between first cousins; their mean age was 7.7 years (range, 6-11 years). A significant (p 0.001) negative association was found between inbreeding and score on the Weschler Intelligence Scale for Children (WISC-C). In addition, the weighted mean IQ of inbred children )88.4 + or - 1.37) differed significant (p 0.001) from that recorded among 50 noninbred controls of similar age and socioeconomic status (99.6 + or - 2.0).
Eric Falkenstein antwortete am 20. Jul, 20:00:
I didn't mean to imply American Indians are dogs, morally or intellectually. I did want to highlight the simple logical fallacy of saying that because someone has certain high skills, they must have a high IQ. IQ measures abstract thinking about language, geometry, logic, etc. It is not a skills test. I just wanted to highlight the absurdity of his logic, that certain skills imply a high "IQ" if we could only measure it on what someone is best at. What Tyler is really saying is that they are perhaps well suited for the environment, but if he's saying that they have high IQs, I think there isn't any data to back him up 
ruth (guest) antwortete am 21. Jul, 09:20:
well, Eric, thanks for not overreacting, i was a bit heated. 
ruth (guest) antwortete am 21. Jul, 10:01:

well, i was not really coming to any conclusions there, rather i wanted to say that it's a bit early for conclusions and that things are at the very least a bit more complicated that "everyone but whites and east asians that live in cold climates have low iqs", as lynn, rushton and co keep saying in the media.
Yes, i also think that cousin marriage may well be a factor in depressing pakistani's intelligence (and that of turks in germany). in fact, i recently wanted to write a cautionary pamphlet for distrubution among turks here (i am german), but in the end didn't because further research showed me that, again, the picture wasn't as clear as i had assumed. while afzal and co-workers consistently found an inbreeding depression for iq in offspring of first-cousin-marriages, neel et al using a much larger cohort and much better controlled study design didn't on Hirado island, Japan, and also came to the conclusion that an earlier study in hiroshima was flawed and only marginally statistically significant anyway. (it is notoriously difficult nowadays to separate inbreeding effects from socioeconomic status correlates.) note also that that earlier hisroshima study, if i remember rightly, had concluded that the effect is no longer vsisible in the first generation of non-consanguinous mating, ie there is certainly no lasting effect on the gene pool of a consanguinous group. note also that the widespread cultural practice of cousin marriage hasn't kept the Japanese from the kind of development their country had in the 20th century. again, these are multifactorial, highly complicated questions, and simplification is not warranted. 
ruth (guest) meinte am 20. Jul, 16:13:
when i said that lynn was not knowleadgable about "dempgraphic history",
i meant the r/K selection part of modern racial theories. but that is of course not lynn's brainchild, but rushton's. 
SerieALivescore meinte am 23. Jan, 07:41:
Experiment cricket has grown to become dull together with prosaic. Some defined proposals that will revamp any test cricket have already been suggested by just former Of india cricket footballer Ravi Sasthri. www.scorespro.com