I saw this yesterday, and while I found it interesting and the study seems, at least, to confirm my experiences as the eldest of four, I have to wonder: the difference in IQ that they're seeing is three points.
The article goes on to claim that that's the difference between a B+ and an A, the difference between Penn State Main and Penn... but surely that's a load of crock. Three points is well within the margin of error for these tests, isn't it? And even if the tests were that accurate, since when do grades and college admissions line up with intelligence?
The article, though, I thought was exactly right when discussing family dynamics. This is of course on the level of anecdote, but one of the frustrating things for my youngest sister, very precocious, very bright, is that however much she succeeds at school, she's already the third rerun. As a result, she decided she doesn't care about school, and that she's not good at it.
Yeah, I was doubting that 3 IQ points was going to get you a perceptible difference in life outcomes, as well.
This is an awesome photo to have been put with the story.
Nothing wrong with your basic point, Cala, but whether or not 3 points is within the margin of error isn't really relevent in this context. If, as LB points out you are to use this result to rebut claims of genetically based differences, it would be interesting to know the magnitude of population differences there, though.
especially if the trend doesn't materialize until adulthood, I don't see how this would affect college admissions.
4: If the differences are always within the margin of error (which they might be, I dunno) then wouldn't that be the best argument against all supposed genetic IQ disparities?
Bell Curve-esque claims about black/white measured IQ differences usually refer to a measured difference in average IQ between the population of about 15 points, although I believe there's research in the last year or so showing that the difference has shrunk.
I'm hoping cerebrocrat shows up to comment on this. My anecdotal experience was somewhat the reverse of Cala's; my oldest sister was a lot smarter in most ways and much more precocious but I was a much better student (which is saying something - I was a terrible student).
Nothing wrong with your basic point, Cala, but whether or not 3 points is within the margin of error isn't really relevent in this context.
Why not? The study says that eldest children have an average IQ of 103, compared to 100 for the second-born. If such a small difference would be enough (say) to rebut the claim that a population was genetically less intelligent than another, why isn't enough to make us skeptical that the environment has had that much of an effect?
4: If the intra-population disparities match the inter-population disparities, that pretty much sinks the claim.
My point was that the margin of error tells you something about the precision of a particular measurement but, assuming not systemic mistakes were made, the average over a large population isn't affected by this because the errors should average out. So it's the statistics of that sample (and the underlying distribution) that matter, not so much the underlying margin of error.
Gotcha. Let me put it another way, then, that doesn't misuse statistical terms (I think I didn't mean 'margin of error', but whatever the term is for 'not statistically significant.'):
3 points is a fucking sneeze on one of the timed tests for IQ. That's the effect of birth order? That's supposed to explain these birth order traits? Break me a fucking give.
Right. 'Margin of error' only means something in relation to the size of your sample -- if the sample is big enough, the margin of error can be as small as you like. 3 IQ points looks like a terribly unimportant difference in average IQ, but that doesn't make it within the margin of error if the sample's big enough.
It's pretty much been orthodoxy for ages that IQ isn't 100% heritable anyway. In fact, my recollection is that the usual number bandied about it is more like 50% heritability (give or take 10 or 20%).
However, I don't see how any finding that suggests that IQ isn't 100% heritable makes any difference to the claim that there are heritable differences in IQ between populations.
IQ doesn't have to be 100% heritable in order for there to be population differences. Height isn't 100% heritable either, but there are correlations between one's ethnic origin and one's height.
I say this not to advance some claim for Bell Curve style intelligence differences between populations. Just pointing out that this research doesn't support what you say it supports.
Near the end of the story there's a remark that makes it sound like the youngest siblings in larger families don't turn out to have lower IQs. Which seems to make sense intuitively -- the middle child gets the short end of the stick in lots of families.
But what about this:
Though the study was done in men, the scientists said the results would almost certainly apply to women as well.
...
Because sex has little effect on I.Q. scores, the results almost certainly apply to females as well, said Dr. Petter Kristensen, an epidemiologist at the University of Oslo and the lead author of the Science study.
Maybe from some point of view 'sex has little effect on IQ scores', but surely if we're looking at family dynamics as a root cause, sex has to be relevant.
This can't be right. I know I'm smarter than my older brother. He's a Republican, for god's sake.
Seriously, though, he's always hated me for being better in school, better behaved, larger vocabulary, ect. I probably made it worse without trying to when I was young, but now that he's all Christian Right and everything, I'm kinda glad I did.
Heritability isn't simply equivalent to 'how much of a quality is explained by genetics versus how much is explained by environment' other than in a fixed spectrum of environments. In a population where pretty much everyone is well nourished, height is going to be much more highly heritable than in a population where some people are well nourished and others are badly nourished. So looking at a number for heritability of IQ outside a particular population in which it was calculated doesn't work well.
More broadly, my point is only that you see Bell Curve type arguments being made of the form: (1) IQ is highly heritable (2) there is a significant measuable difference between average measured IQs of black and white Americans, therefore (3) at least a portion of that difference must be explained by genetic differences between the populations of black and white Americans. This study is a refutation of that argument, because (1) remains the same, and (2) there is a significant measurable difference between the average measured IQs of siblings raised as firstborns and as laterborns but (3) that difference can't be explained at all by genetic differences, because we're talking about siblings here.
9/10 cont. I crossposted with you, and should have elaborated (I was replying to 4, not your 9). None of this `disproves' the claim that genetics underlies IQ differences, but if the differences found there are of the same magnitude as in-family differences, they really can't claim that nothing else is likely to be the issue. This doesn't affect the claim that there are population differences being measured, but does make it really hard to claim that they are due to genetics.
This is also quite different from the question of whether or not a shift of 3 points has any significant effect on, say, college admissions or whatever. That would depend a lot on the correlation of IQ to admissions, or correlation with IQ to course grades, or whatever. So I'm not sure, but I may have misread the implication you were questioning.
What I was sayiing is that the relative magnitude of the shift of 3 point to the margin of error doesn't make any difference to the solidity of that claim of those 3 points though. If the statistics are solid, then the expected IQ of older siblings is a bit higher than the expected IQ of younger siblings. What outcome affect this has, if any, is much more complicated
crap. 18 preempted by 11 too. sorry!
sex has little effect on IQ scores
Where can I sign up for the study to test that hypothesis?
And they better not put me in the control group!
The scientific fact is that each parent is afforded only so much smart juice in his or her loins.
18: Yes. I misused the term. Sorry for the confusion. I meant to question the latter claim (see 11.), and fumbled the terms.
I mean, I wish I could just look up my students' IQs and have that be perfect predictors of their grades. It would save a lot of time in marking. So grant that the researchers have solid statistics. It's still just three points.
I have nothing to say about the genetic stuff right now, except that 17 seems to be correct.
I pre-empt everyone! Bow before the calabat, bitches!
The scientific fact is that each parent is afforded only so much smart juice in his or her loins.
You can buy supplements at the drug store.
That study still isn't a refutation of that Bell Curve argument. The Bell Curve argument is a dumb argument because, as you say, 3 doesn't follow from 1 and 2.
We already KNOW that IQ isn't 100% heritable. The people doing the Bell Curve work already knew that. This study makes no difference at all if we are interested in finding a way to refute the Bell Curve argument. It's just an illustration of the non-100% heritability of IQ.
This seems like a fine place to note that "tests of statistical significance should never be used."
Wrenae's older brother sounds like mine. A living refutation of the study.
On the other handIn addition, I'm one of four, with the following progression:
Eldest: Stanford
Second: Cal Berkeley
Third: Reed
Youngest: Colorado
Whatever it is that the oldest has, judgment and sense peaks with the second born.
Wrenae's older brother sounds like mine. A living refutation of the study.
On the other handIn addition, I'm one of four, with the following progression:
Eldest: Stanford
Second: Cal Berkeley
Third: Reed
Youngest: Colorado
Whatever it is that the oldest has, judgment and sense peaks with the second born.
That study still isn't a refutation of that Bell Curve argument. The Bell Curve argument is a dumb argument because, as you say, 3 doesn't follow from 1 and 2.
Sure it is. Just because it's a dumb argument doesn't mean it can't be refuted by a counterexample. If you mean that no one should need a counterexample to understand its dumbness, then that's fine, but it's a nice, clear easy way of doing it.
(3) that difference can't be explained at all by genetic differences, because we're talking about siblings here.
Interestingly, the difference could be inborn (though not genetic), because there could be some property of second and third and fourth births that puts those offspring at a disadvantage - something about the womb environment, say. The story I read about this study indicated that they dealt with that issue by studying kids whose older siblings had died. Turns out that those kids inherit the firstborn advantage.
I've already erased a couple entries here due to LB pwnage, so I'll say in advance: Everything LB has said or will say is correct and complete and perfect.
Also, I'm the oldest in my family. I'm also the only one who went to university.*
* this is not particularly significant, however, since I'm the only person in my family -- not including very distant cousins -- to go to university, ever.
re: 32
But my point is, that no-one involved in making these arguments seriously believes that IQ is 100% heritable. So pointing to an experiment that shows that IQ isn't 100% heritable doesn't do any work.
It may act as a refutation of some 'dumbWorld' version of the 'racial differences in IQ' claim but it isn't going to do any work against anything else.
But that's all there is to the 'racial differences in IQ have got to be genetic claim' -- there's nothing but the DumbWorld version to refute. (Oh, people admit that IQ isn't 100% heritable, but gloss over it with 'but it's highly heritable, so how could a significant population-level difference not have some genetic component.') Are you aware of a better argument that needs higher-level refutation?
I mean, to take other examples, we know that drastically impoverished nutrition alters IQ, that foetal alcohol syndrome alters IQ, etc, etc ad infinitum.
These would all constitute counterexamples to any 'racial' argument that rested on the 100% heritability of IQ.
re: 37
Well, one could certainly construct a better argument for it, that pointed to differences in particular genes in different populations, and then suggested that those genes are correlated with IQ scores. There's fairly recent research that could be turned to the 'dark side' if you were seriously interested in making some racially driven argument about intelligence.
it's highly heritable, so how could a significant population-level difference not have some genetic component.
You're right that that is a sneaky glossing over a huge array of necessary biology that would need to be done in order for their argument to go through. I'm not claiming the 'bell curve' type argument you describe is a good one. I'm just saying that examples of non-100% heritability just aren't going to do the contrary work either.
These would all constitute counterexamples to any 'racial' argument that rested on the 100% heritability of IQ.
ttaM, I think you're missing the import of Murray's argument. Granted, he's not arguing directly that intelligence is 100% heritable, but he is arguing that the environmental influences on intelligence are well-enough understood to make broad generalizations about racial categories - categories whose environmental circumstances are profoundly different.
Anything that undermines that claim, undermines Murray's argument - and, of course, there was always plenty of evidence undermining that claim. Still, it's interesting to see an environmental influence like this come from completely out of left-field. The idea that the environment of siblings is so different that it produces measurable IQ differences is a pretty dramatic rebuttal of Murray's assumption that we can know the cross-race impact of environment.
ttaM, I think you're missing the import of Murray's argument.
And even more of the popular understanding of Murray's argument. I am, sadly, rehashing stupid arguments I've had with Bell Curve fans in the past with people who don't comment here, but trust me when I say the DumbWorld argument is out there and could use a good refutation.
Sometimes a younger sibling, with the aid of a kitten, is able to steal the smart juice from an older sibling's mouth while the older sibling slumbers. This explains the occasional exceptions.
Older siblings are advised to sleep on their stomachs.
re: 41
I suppose. I'm just not overly concerned with factual evidence that confirms what we already know, and which anyone with a half-way passing understanding of even the rudiments of the science already knows.
It's of the 'satellite photographs prove Earth round' variety. Yes, it's a nice example to use, but hardly a cutting edge piece of new research for some hitherto un-established fact.
29 is completely wrong. Slolenr should be banned, and possibly sent to Guantanamo.
17: 3) doesn't follow from the first two, as there are other variables not properly accounted for (racism, fetal environment eg). Wait for studies of particular alleles for the genetics to become clearer; This paper suggests a way to look for interesting alleles (high frequency + homozygous deleterious implies heterozygous advantage, like CFTR). Also interesting is PMID: 17220170 which suggests that Microcephalin is not the place to look.
ttaM, you clearly live in some happy country where dumb arguments don't rule the roost. Here in America we need those satellite photos.
With regards to the earlier margin of error discussion, there are actually two margins of error that are relevent here.
The first is the margin of error on a single IQ test. That is, suppose I take a test and get a 100, and you take an IQ test and get a 103. If the margin of error is +/- 5, then there's really no clear basis to say that you're smarter than I.
The other margin of error is the margin of error of this study, which is hopefully lower than the margin of error for a single IQ test. (Based on the same concept that if you pull two marbles from a bag, one's red and one's blue, then conclude that the bag is 50% red and 50% blue, you aren't being very exact, while if you pull 30 red and 30 blue, then you have much more confidence in your claim).
If the study's margin of error were 3 points or more, then any claim it made that first siblings were smarter than second siblings would be suspect. But, as soubzriquet mentions earlier, it doesn't matter what the margin of error of an individual test is.
None of this is intended to refute Cala's comment about "who the hell cares about a 3 point difference." Though, as I recall, the Bell Curve found something like a 3 or 4 point difference between whites and asians (asians higher, of course).
Anyhow, my IQ is clearly like 15 points higher than my little sister's. And if she argues otherwise, I'll give her a noogie.
If there is someone out there claiming that IQ is 100% genetic, then indeed, that person will be perplexed by the findings of this study. Like ttaM, I don't think there is anyone worth arguing with who is making that claim. Murray and Herrenstein certainly weren't.
29 is completely wrong. Slolenr should be banned, and possibly sent to Guantanamo
Do you want to explain why the linked article is incorrect?
Ergo, in families with kittens, one is likely to find imbecile first-born children.
In my next chapter, I will discuss doggies.
The trouble with DumbWorld arguments is that satellite photos don't always refute them as far as their adherents are concerned. It's hard to know what if anything will be effective.
It's very often personal, like the broad hatred of liberals itself. Certain people in my extended family can be convinced by arguments, particularly if facts they may not be aware of may be introduced and accepted. Other people are never going to be convinced by me, and if their coming to terms with things–as–they–are is going to happen, it will be by steps that either seem blindingly obvious to me, or else the reasoning will be just as faulty as the previous reasoning was. Whatever.
48: Right, it's just really really hard to figure out what was meant to be interesting or useful about the Bell Curve's discussion of racial differences in average IQ without that claim.
re: 52
Yeah, it is a dumb argument absent some plausible explanation of why the environments of blacks and whites are sufficiently and relevantly similar to render the genetic explanation the most likely. The fact that such an explanation hasn't been provided, though, doesn't mean that one couldn't in principle be provided.*
Which is why I'd want to rest my argument against Bell Curve type reasoning on something more substantial than counterexamples to 100% heritability.
* of course, it won't be. I don't believe that there are genuine large-scale genetically determined inter-population differences in IQ.
49: The first thing that struck me is that they advise researchers to include "effect sizes, confidence intervals, replications/extensions, and meta-analyses" while avoiding p-values. But confidence intervals also assume a given distribution, so they'd be no more accurate than approximate p-values! Even if you're mostly concerned about poorly defined tails to the distribution (which is a very legitimate concern), that doesn't come up in most scientific use of p-values that I've seen. Typically, researchers will merely note whether the result is significant at the 5% or 1% or any higher level, which is functionally equivalent to seeing if the null hypothesis is outside the 95% or 99% confidence intervals, so confidence intervals would have the exact same problems.
Mind you, I have not read the article, only the abstract. I couldn't find a link to the article there, but I'll take a poke around. If they are implying that poor experimental design renders even the best of statistics to be useless or possibly misleading, I have no qualms with that. But I would still think that is a larger problem with proper teaching of statistics use and experimental design. Throwing out all statistical significance measures would be terrible, since such measures give gauges of results' significance that are far far more accurate than human estimation from a few descriptive statistics.
Speaking of bad arguments, the Polish negotiating team at the current EU summit are employing creative use of counterfactuals.
52 lost me. I haven't read The Bell Curve, and so can't state what, if anything, is interesting or useful about it. I would venture the following:
1. One needn't believe that a trait is 100% genetically determined to believe that genetic differences can have a role in explaining differences in that trait between two populations.
2. It would be weird if someone who explicitly rejects the hypothesis that IQ is 100% genetic (as I believe Murray does) relied on that hypothesis to draw conclusions.
47
"The first is the margin of error on a single IQ test. That is, suppose I take a test and get a 100, and you take an IQ test and get a 103. If the margin of error is +/- 5, then there's really no clear basis to say that you're smarter than I."
If by margin of error you mean the error in each result is normally distributed with a SD of 5 then assuming the errors are independent (and that I have calculated correctly) the odds would be about 2-1 that the person with the 103 score is actually smarter.
I never figured out what the whole huge deal about the Bell Curve was. Even if it is correct that there are slight deviations in IQ between ethnic groups, which I don't know that there are. What does that actually lead to. As far as I can tell you can't make any actual decisions based on that information since it doesn't apply to individuals. If the science or math in the book is wrong it should be refuted, but I think the backlash probably gave the book more credence then it would have otherwise had. The only real negative is the DumbWorld arguments, but as someone pointed out evidence isn't really important to those people anyway.
56: 1. Right, but you'd want some evidence for that belief (that they do, rather than might have such a role.)
2. Sure would be.
53
"* of course, it won't be. I don't believe that there are genuine large-scale genetically determined inter-population differences in IQ. "
Why not? Unless you believe IQ is equally valuable in all the primitive environments in which these populations arose (which seems unlikely) such differences appear inevitable. Or by "large-scale" do you mean such differences are small?
I think the backlash probably gave the book more credence then it would have otherwise had.
No, I think it was more the cover story in The New Republic and that sort of thing.
JAC, email me your email (my moniker, gmail), and I'll send you the article. I'm not invested in the conclusion (not really my field), but I find it interesting and would like to know if someone more knowledgeable than I finds it persuasive.
Credence was probably the wrong word. I probably should have said publicity. The New Republic probably gave it some initial buzz, but I think all of the fury gave it a lot more staying power. I am sure the New Republic has talked about a lot of books I have never even heard of, but I keep hearing about the Bell Curve. For the most part I keep hearing about it in ways that don't address any flaws in the book, not talking about this post, but in ways that claim that it is obviously racist tripe without ever saying why it is wrong.
re: 58
Even if it is correct that there are slight deviations in IQ between ethnic groups, which I don't know that there are. What does that actually lead to.
Yes, I think that's true. Even if the IQ difference claim were true, the real world policy claims that advocates of those claims want to make don't follow.
re: 60
Unless you believe IQ is equally valuable in all the primitive environments in which these populations arose (which seems unlikely)
Why does that seem unlikely? it seems highly likely to me, actually, that most human environments are in fact highly similar in this respect. Especially if, as lots of people argue, the primary driver for human intelligence is us, i.e. our social environment. In that respect, all of our environments are, over 'evolutionary' time spans, pretty much identical.
There is some recent research that points to some more recent changes in the genome -- raised frequencies in particular genes -- that may undermine that slightly. But generally, it doesn't seem at all implausible to me that the nature of our evolutionary past would in fact make it highly unlikely that significant intelligence differences would arise.
it seems highly likely to me, actually, that most human environments are in fact highly similar in this respect. Especially if, as lots of people argue, the primary driver for human intelligence is us, i.e. our social environment.
Well, there was one interesting claim in the past that the social environment caused one particular group to develop higher intelligence. But I agree that the notion of a minimal social interaction level for early humans (hunting/gathering with family, etc.) could produce a sort of floor on the intelligence that selection pressure would push for.
The only real negative is the DumbWorld arguments, but as someone pointed out evidence isn't really important to those people anyway.
You know, I think there's something valuable to refuting the DW arguments that hang off a facially respectable piece of work like TBC. It got a huge amount of publicity back in the 90's as if it were going to have important policy implications. The thing is that the the arguments the book made explicitly don't have any useful policy implications, but the superficially similar DW arguments that were temporarily made respectable by their association with TBC do have policy implications.
Sensible people went apeshit refuting the DW arguments (and pointing out the flat-out errors in TBC that made it appear to more strongly support the DW arguments). And the response to the sensible people was "God, I don't know why you people get so excited; it's not like the book was actually saying anything as stupid as you say. All it was saying was some boring minor shit that doesn't have any policy implications. Geez, you people are tense." And now, no one remembers TBC as having been influential in the formation of policy.
The sensible people look as though we were overexcited and unfair to TBC, but TBC didn't actually shape anyone's policy judgments. We won that one.
We won that one.
If that's true, it's only partially true, and if it's true, it's only true because people like Gould stepped all over TBC ASAP.
66 also addresses the question posed by CJB. The attention this thing got - and continues to get - was not a result of it being denounced.
69: Right -- Gould would be one of the sensible people who went apeshit. As a result of the things he, among others, wrote, the defense of TBC has had to be "TBC was a mild, unassuming, harmless book making no interesting policy claims" rather than "TBC was right about its policy prescriptions resulting from the ineradicable differences in average intelligence between racially different populations."
Although if you follow the link in 66 and re-read Gould, you will remember how full of malarkey is the phrase, "TBC was a mild, unassuming, harmless book making no interesting policy claims."
Oh yeah, the advocates of TBC totally wanted policy prescriptions to follow from the book. Those policy prescriptions DON'T follow, but I'm fully aware that an awful lot of people really wanted them to.
73: Isn't this usually true when you start of with a policy idea and then go shopping around looking for a justification?
Eh, I got pretty heavily invested in arguing about TBC when it came out and afterwards, and as I recall the book, it was (IMO) disingenously but fairly completely hedged with weasel words. If you were willing to read it with an absolute unwillingness to draw conclusions from anything is said, you could say it made no interesting policy claims.
The Bell Curve arrived in my life as a long synoptic article in The Wilson Quarterly, sometime in the mid-eighties, if my sense of time and place are correct. It was obvious that it had full-bore neocon sponsorship and support, and was intended as a blow to the legitimacy and efficacy of the Great Society. That was the context in which I read it then and still think of it.
LB's take on this, about the need and effectiveness of the refutation, is quite right.
Here are two articles that give the current state of "bell curve" arguments:
http://taxa.epi.umn.edu/~mbmiller/journals/pppl/200504/2/235-2.html
http://taxa.epi.umn.edu/~mbmiller/journals/pppl/200504/2/302-2.html
That Solloway guy is a douchebag:
http://www.marginalrevolution.com/marginalrevolution/2004/03/born_to_sue.html
64
"... But generally, it doesn't seem at all implausible to me that the nature of our evolutionary past would in fact make it highly unlikely that significant intelligence differences would arise."
This is dependent on what you mean by significant. Would you consider a 1 SD (15 IQ point) difference significant?
As for our evolutionary past remember IQ doesn't come for free. IQ is being traded off against other desirable traits like disease resistence, starvation resistence, cold resistence, heat resistence, high altitude adaptation etc. I doubt its relative adaptive value was exactly the same while the value of all these other traits was varying among primitive environments.
As for social environments these varied widely as well (for example between agricultural and hunter/gatherer cultures) and I would not expect IQ to have been of equal value in all of them.
As for our evolutionary past remember IQ doesn't come for free. IQ is being traded off against other desirable traits like disease resistence, starvation resistence, cold resistence, heat resistence, high altitude adaptation etc.
Um, other than 'there must be tradeoffs for everything', have you got any evidentiary basis for any of that? Something showing that Pre-Columbian Native Americans had higher IQs because they evolved in an environment with less disease, or anything?
I swear to god, I wish the entire concept of IQ had never even been invented.
75
My memory of TBC is that it primarily argued that a substantial number of people were inherently unsuited (because of genetically determined relatively low IQs) for lots of intellectually demanding high status jobs. Are you saying it didn't say this or that this is not interesting or what?
81: Then we'd just be back to talking about "intelligence", though, right? Maybe there'd be some advantage to not even being able to pretend we could precisely measure it, I guess. But the economists would probably then argue that the best proxy we have must be income (since at a rough estimation that's what everyone's obviously using their intelligence to try and obtain), and then our income distribution charts would be effectively be our intelligence distribution charts. I can't see that being better than IQ charts.
>Um, other than 'there must be tradeoffs for everything', have you got any evidentiary basis for any of that? Something showing that Pre-Columbian Native Americans had higher IQs because they evolved in an environment with less disease, or anything?
The article that JAC linked to hints at something like this:
http://www.nytimes.com/2005/06/03/science/03gene.html?ei=5088&en=b9f98c383be17b51&ex=1275451200&partner=rssnyt&emc=rss&pagewanted=all
81: Testify. These discussions drive me crazy.
When I was a little kid, some friend of my parents gave me an IQ test. The only thing I remember is being frustrated at not being able to answer an analogy question, "letter is to envelope as gun is to?" So I had no idea what a holster was; why the fuck should I? This bullshit is supposed to determine people's intelligence in some meaningful way? Fuck that.
IIRC, M&H didn't quite make the claim 'genetically determined'. They claimed 'immutable' (based on evidence purporting to show the failure of interventions like Head Start in raising IQs) and 'highly heritable', and left those of us willing to draw obvious conclusions to assume they meant 'genetically determined'.
If you drop the 'genetically determined', what you're left with is a claim that stupid people will not be successful doing intellectually demanding jobs. I'd call that uninteresting.
67: if few alleles have an effect, the frequency of these few alleles matters. Consider convergent evolution of lactase persistence as a model for recent adaptation; What is not known is how many distict mutations confer advantage, and when in human history they appeared. If many mutations matter or the mutation was >200k years ago, population variation is rounding error. If not, not.
86: Hmm... I've never read the book, I've only read around it, but I thought the "interesting" claim was that those stupid people who can't do intellectually demanding jobs are mostly black. And there's nothing you can do to change that.
Not so much that stupid people are mostly black, but that black people are mostly stupid. And yes, that there's nothing you can do to change that.
80
Practically any trait you can measure is observed to vary between populations. It is hard to see why IQ should be an exception. However the specific tradeoffs are usually not very well understood. In the case of IQ there is a recent paper claiming that higher IQ in European Jews is associated with certain diseases and is the result of high selection pressure for increased IQ.
If you read the article, the paper doesn't claim to have shown that higher IQ is associated with those diseases. It proposes that association as a hypothesis worthy of testing.
84: The Hardy article does more than hint, it provides a falsifiable hypothesis. You'd expect heterozygote advantage in intelligence for at least some mendelian diseases common to ashkenazim. A study analogous to the microcephalin/aspf study for some of these is the thing to watch for.
That is, the NY Times article describing the paper you refer to linked in comment 84.
86
"If you drop the 'genetically determined', what you're left with is a claim that stupid people will not be successful doing intellectually demanding jobs. I'd call that uninteresting."
Suppose we don't drop genetically determined? Or suppose we add the claim that we don't know how to turn stupid people into smart people?
Slolenr I haven't read the paper, but I've read similar ones. _I_ am not confused by tests of statistical significance, so it annoys me when I'm told I can't be trusted with them. The difference between tests of statistical significance and confidence intervals is basically cosmetics, I fail to see how one can be a proper method, and the other is not. Replications and reporting effect sizes are clearly important, but are not substitutes for some expression of statistical significance. I don't know how meta-analyses fit into his argument.
Well, on 'genetically determined' we're back to the siblings article listed in the post -- clearly, the existence of a difference in average measured IQ between two populations does not establish that the difference is 'genetically determined', because there is such a difference between children raised as firstborns, and children raised as laterborns, and it can't possibly be genetic. So I'm either going to drop it or dismiss anyone claiming that such a genetic cause has been established as a nitwit, your choice.
On 'we don't know how to turn stupid people into smart people', the question is what follows from that. I won't take it as a valid argument for refusing to support social programs, or presuming that it's just and inevitable that children should be limited by the socioeconomic status of their parents, because their parents are poor due to their own stupidity and the kids are doomed to the same. But you may have some other interesting conclusion to draw from that claim.
Suppose we don't drop genetically determined? Or suppose we add the claim that we don't know how to turn stupid people into smart people?
Then you are left with the foolish argument that TBC was trying to promote. The simplest way to rebut this is to note the IQ gains in populations that experience new prosperity.
I swear to god, I wish the entire concept of IQ had never even been invented.
If colleges are going to opt out of the US News & World Report rankings (yay!), people can opt out of intelligence testing. It's already happening on an individual level. Next stop: organizations.
Went to lunch so addressing arguments upthread
68: I absolutely think that DW arguments should be refuted. It is just the level of reaction that I don't understand. It seems like the better approach would be to do the refutation on the merits like some of the stuff linked here and then basically move on. If people bring it up later point to the factual refutations. You probably won't convince them anyway, but it paints you in the more reasonable light.
From what I have seen of people bringing it up to support policy it seems it is only brought up to support policy that the book itself doesn't even support. So it might be worth pointing that out as well. I just don't like that a lot of the criticisms to TBC that I see are very reactionary. I also think that if you have to rely on responses like that you don't have a better argument. Which in this case isn't true.
I'm not following what you mean by 'reactionary'.
100: It seems like a lot of the arguments were based on the fact that the book would dare to say that intelligence was ethnically linked with out checking whether it was in fact true or not. Maybe this wasn't the case, but that was the feeling I got from a lot of the people who came out against the book.
I think I would have used reactive there, to avoid the explicit political history and meaning of reactionary.
Well, yeah. A book that says 'intelligence is immutably ethnically linked', when it doesn't have good evidence for that (evidence that 'we don't have reliable methods for raising the intelligence of an individual' is not evidence that 'changes in societal factors cannot possibly affect differences in average measured IQ between ethnic groups') is going to piss people all the way off, and rightly so. I'm not quite seeing why it shouldn't.
103: As long as people check whether the evidence is good I am fine with pissed off. It seemed to me a lot of people went all righteous fury without doing any checking of the evidence at all.
96
"Well, on 'genetically determined' we're back to the siblings article listed in the post -- clearly, the existence of a difference in average measured IQ between two populations does not establish that the difference is 'genetically determined', because there is such a difference between children raised as firstborns, and children raised as laterborns, and it can't possibly be genetic. So I'm either going to drop it or dismiss anyone claiming that such a genetic cause has been established as a nitwit, your choice."
You are conflating two issues, whether there are people with relatively low IQ genes and the extent to which we can predict who such people are based on who their parents are. The claim that there are genetic differences in IQ between individuals (like siblings) does not depend on there being genetic differences between populations at all.
No, the evidence got all the checking it deserved. The evidence for 'immutability' of IQ was based on the fact that we don't have social programs that have shown broad success at raising IQ significantly. That is self-evident crap as evidence that differences between ethnic groups can't possibly be affected by social factors. And trying to rely on that sort of evidence to support a claim that differences between ethnic groups are immutable deserves an angry and contempuous response.
re: 104
Well, it's not always about the evidence, if the arguments themselves are piss-poor. We can go after the factual premises in the argument or the structure pf the argument itself. Or both, if both are piss-poor.
105: Huh? There's a difference in the average measured IQ between two populations: first borns and later borns. This difference is not explained by a genetic difference between those populations. Therefore it is possible for a difference in average measured IQ between two populations to be explained by environment rather than genetics.
What am I conflating?
97
"... The simplest way to rebut this is to note the IQ gains in populations that experience new prosperity."
Stupid and smart are relative terms. Raising everybody's IQ doesn't help low IQ people compete with high IQ people for most intellectually demanding jobs.
The problem I have with angry and contemptuous response is that it can let the other side take the moral high ground and say things like "your just angry because the facts don't match your world view and you aren't giving the argument a fair chance" if you are angry and refute the arguments based on the facts of their argument then it takes that away.
109: No, but if one population increases in prosperity relative to another population then its members may also increase in IQ relative to the other population. Really, very simple.
110: The point is to do both. A reasonable response to someone making a badly supported argument that concludes that one ethnic group is genetically intellectually superior to another is to say: (A) Your argument fails to stand up because of X, Y, and Z; (B) someone trying to justify a racist conclusion by means of a weakass argument sure looks like a racist shithead.
The problem I have with angry and contemptuous response is that it can let the other side take the moral high ground and say things like "your just angry because the facts don't match your world view and you aren't giving the argument a fair chance" if you are angry and refute the arguments based on the facts of their argument then it takes that away.
In the real world, this is exactly wrong. It is the noisy, angry folks who rule public discourse these days. Murray is an overt racist who promotes falsehoods as science. He has been treated much too politely, and that's why he is still such a prominent public figure - not because people are unduly mean to him.
108
My original statement in 82 was:
"My memory of TBC is that it primarily argued that a substantial number of people were inherently unsuited (because of genetically determined relatively low IQs) for lots of intellectually demanding high status jobs."
You are objecting that genetic differences between populations have not been demonstrated. But this is not a claim about populations. If IQ is determined by a moderately large number of genes with roughly independent additive effects and high IQ and low IQ alleles are both common then even if the population is homogeneous and mating is completely random some individuals will get more high IQ alleles than other individuals by pure chance. The claim is the genetically disfavored individuals (the bottom 10% say) are inherently unsuited for many intellectually demanding high status jobs.
112: Right, but you have to do A in order to do B. You can't just say an argument is weak, well you can but I think you are shooting yourself in the foot if you do, you have to show why.
Anyway I have to go do some landscaping and I think we pretty much agree anyway.
If we're not talking about populations, what's interesting about that? Sure, stupid people are going to do poorly in intellectually demanding jobs, and sure, for any given stupid person their low IQ is likely to be explained at least partially by their genes. This seems to me right up there with "Pope seen attending Mass; in other news, bear emerges from woods looking relieved. Film at 11."
I think I've lost grasp on what is being debated here. I haven't read TBC, and am not terribly interested in defending the man Charles Murray. That said, I think the following three points are not terribly controversial:
1. differences between individuals in IQ derive both from environment and from genetics
2. current estimates suggest that the influence of genetics on IQ is substantial.
3. we know much less about IQ differences between groups than we do about IQ differences between individuals
With this as background, how ought we to investigate observed differences in IQ between groups? I would have thought we are not justified in starting out with a strong bias that genetics > environment, environment > genetics, or environment = genetics. We don't know, and should subject each to equal scrutiny. Right?
Further, it seems like we'd want to separate the question of the source of difference in IQ between people or groups with the question of whether and how IQ can be changed. It seems to me like I could grant that my IQ is lower than Einstein's simply and only because he has better genes than me, but still think it would be useful for me to do IQ-raising exercises.
Comity?
You're a second child, aren't you baa?
117: I would add to your list of factors:
4. There is a long history of using flawed science to establish that one socially disfavored group is innately intellectually inferior to another group.
(And actually, I'm not so sure about 3. We know that intelligence is explained partially by genetic and partially by environmental causes. But I'm not sure that we do know much more than that about what makes one person more intelligent than another -- I'm not aware that 'smartness' genes have been identified, or much in the way of measurable biological facts that correllate with intelligence. But I don't think that 3 is important to any argument you're making.)
So we go ahead and do research in any direction that looks interesting, but when you get someone claiming to have evidence that one ethnic group is innately superior or inferior intellectually to another, you remember 4 and go through their evidence with a fine-tooth comb. It's an area where people have a long history of getting it wrong.
I am hugely in favor of running the statistics to determine which ethnic groups are intellectually inferior, and which are superior. To prevent repeating the measuring errors of the past, however, we'll first need to create a society in which everyone achieves his or her full intellectual potential. For the time being, let's focus our energies on that for now. Once we get there, the data should be relatively easy to collect, and its interpretation should be straightforward.
116
So you are saying it is not interesting because it is obviously true. In a sense I agree but in fact it is not obvious to everybody. Rival "blank slate" type claims still have wide support and influence. For example that the educational system can and should prepare everybody for intellectually demanding high status jobs.
Because of 4, I'm not sure you do go ahead and do research in any direction that looks interesting. Or maybe not because of 4, but because that research--into the genetic roots of intelligence--isn't really doing anybody any good.
That would be silly. The idea that the educational system can and should provide every child with the opportunity to develop their intellect fully in order to discover whether they are well fitted for intellectually demanding high status jobs, on the other hand, that makes sense to me.
As long as people check whether the evidence is good I am fine with pissed off. It seemed to me a lot of people went all righteous fury without doing any checking of the evidence at all.
I'm always amazed by this argument when I see it. "Well, yes, everything the detractors say is true, and they are right to be furious about it, but expressions of that fury are inherently suspect."
This was the exact argument used to malign Iraq war opponents who, history has shown, were in fact much too timid - or were shouted down by nitwits who know how to communicate in this modern world.
It's a Bush/Limbaugh/O'Reilly world, and will be until something is done about it. I see some hope in the fact that decent people are finally learning to recognize that fact and are starting to express outrage at the outrageous, without worry, per CJB, if they are following the right protocol in doing so.
122 assumes that "stupid" equals "stupid from birth." Whereas "stupid from birth" is very much in dispute.
It's funny; my memory of the BC controversy is that quite a lot of people (everyone from professionals like Gould to laypeople like LB) checked the evidence and then got righteously angry. And then a whole lot of people (such as me) trusted their interpretations and felt justified getting angry without doing our own analysis. I didn't even read the darned book.
Although my anger was more of the "How dare they disingenuously pretend to be neutrally exploring an interesting topic when it has such an explosive history?" I mean, there is room for legitimate research on explosive topics. Prefacing it with an acknowledgment of the explosivity is a good way to start.
The idea that the educational system can and should provide every child with the opportunity to develop their intellect fully in order to discover whether they are well fitted for intellectually demanding high status jobs, on the other hand, that makes sense to me.
Hmm. I'm not sure I agree with the "can," which makes me very uncomfortable.
Rival "blank slate" type claims still have wide support and influence.
Did you read Pinker's Blank Slate? Never have so many straw men met such a gruesome fate. This is very Pinker-like:
For example that the educational system can and should prepare everybody for intellectually demanding high status jobs.
Pure straw. The real-life version of this claim is that such an education should be made available to everyone. And equal opportunity in education isn't even on the political agenda in this country.
129: Oh, fair enough. Throw in an 'equal' there, and that fixes it -- I'm not saying that we can create a perfect system, just one that gives every child a fair shot at reaching their potential.
127
"122 assumes that "stupid" equals "stupid from birth." Whereas "stupid from birth" is very much in dispute."
Really? As I understand LB that many people are stupid from birth is too obvious to be interesting.
There's a difference between saying that a given person may have low intellectual potential for largely genetic reasons ('stupid from birth') and saying that we can reliably pick out children like that for the purpose of giving them lesser educational opportunities. The first is uncontroversial, the latter is wildly controversial.
"As I understand LB [INSERT COMMA] that many people are stupid from birth is too obvious to be interesting."
As you will note, LB only says that many people are stupid.
124
And what do you do with the substantial number of children that are not suited for intellectually demanding jobs? At what point do you shunt them onto more suitable tracks?
125: I am not sure you are understanding the idea I am trying to get a crossed. That is probably my fault reading through my previous posts. I am OK with people getting angry with something that is in fact wrong. Most of the time when I see arguments about this book happen I see the people arguing that it is wrong without ever saying why it is wrong. It boils down to it is wrong because I say it is which isn't a convincing argument. This thread was very good. People pointed to actual rebuttals. This is a much more convincing argument. So I am not saying don't get angry. I am saying if your going to get angry back it up with something other then just your personal anger.
As to the war I also didn't buy that Iraq was a terrorist threat because Bush said so. Same kind of thing.
You know, we've got enough problems with the unequal provision of educational opportunity to children long before it is apparent what their academic potential is, that I'm going to worry about equalizing that before addressing when we need to write off the ones who aren't up to the demands of grad school.
When the grade schools are being run equitably, come back and ask me how to track older kids and adolescents. I should have time to get a couple of graduate degrees in the subject by then.
You want an angry rant? Here's one. Can I just say that the thought that a 15-point difference in IQ (assuming for the sake of argument that it tracks intelligence and not whether your parents were able to buy you the right sort of preschool) is going to determine the very course of your life's success is about the most enwhitled notion ever, suitably only to people who think that you'll be a failure if you have to attend NYU instead of Harvard? Let me count the ways that assumption is really goddamn dumb.
1) You'd have to believe that IQ tracks success, and that success is defined mo more broadly than "must get into elite universities." And is this the sort of success we mean when we talk about getting kids out of poverty? Is it broadly true that people who do well on IQ tests are likely to make more money than those who do less well? Are you going to commit yourself to the idea that all successful businessmen are the smartest in their class? If you think these things, I suggest, respectfully, that you get out of your overeducated bubble and talk to a small businessman who made a good life without attending an elite university.
2) But you'd need to believe more than that. You'd need to believe that not only does IQ track success, but that it tracks success so perfectly that a 15-point average IQ difference is enough to Make You Fail, because that's what's explaining why those ethnic minorities* aren't doing well. And surely this only a proposition that is even remotely plausible to you if you don't know anyone who attended a state university. An anecdote, to be sure, but a 30-point IQ difference, at least, separates me from my youngest sister. Yet she is off to a middlingly decent school in the fall.
Why is she not doomed to failure? Because she's lucky to have parents who value education and are able to scrape together the funds to send her to a school, and she went to a good school district. If we're talking 'how to make policy ameliorate the lives of the urban poor', we don't need to get them all into Harvard. Pitt-Johnstown's a good start.
*None of the people arguing that IQ is so important that it should determine policy towards black people would accept for a minute a system that took away funds from their kids to give them to little Asian kids. No, instead, we decide, we must learn to study like the Japanese and value knowledge as those parents do.
I'm not sure you do go ahead and do research in any direction that looks interesting
I'm Straussian enough to be sympathetic with that, actually. A problem occurs when the Straussian "topics we should not discuss or examine too closely" becomes an intellectual "presuppositions we should have." I (mistakenly, I think) took LB to be arguing above that you want more evidence for a genetic-predominates thesis than you want for an environment-predominates thesis. That's not justified.
As to the war I also didn't buy that Iraq was a terrorist threat because Bush said so.
Well, okay. You don't provide any reasons here for this belief, but as long as you don't get too angry about it, I guess you are permitted to express this opinion without providing detailed supporting material.
140: Woohoo, excellent rantage!
None of the people arguing that IQ is so important that it should determine policy towards black people would accept for a minute a system that took away funds from their kids to give them to little Asian kids.
Perfect. So true.
30-point IQ difference, at least,
The "at least" here is funny, as if you're not quite sure about the actual numbers, but damn, it's gotta be at least 30, right? Just listen to her...
I was thinking. Is anyone else familiar with Dexter's Laboratory? I'm now picturing Calasis as the balletic big sister.
Hehe. No, my parents never told us what we scored when the school tested us, but you were tracked into different programs based on how you scored (if they thought you were bright enough to bother administering the test to), so I have an idea of where the cutoff lines were, but not where we fell within the distribution. Youngest calasister was tested only at my parents' insistance (she has to be smarter than that, look at her sisters...).
I would be willing to bet, however, that she will be more successful, by most material measures, than I will be. She's nicer, she's better with people, she's very beautiful and can charm people into doing her bidding. And she has a very practical look at the world. I'm good at tests. This means you end up in grad school.
Oh, but that IQ! The problem is that IQ theories are designed by smart people who have an interest in believing they end up on top and who haven't talked to anyone who isn't an academic in 20 years.
139, is, alas, somewhat unresponsive. It's true that we have many problems of unequal education provision at the grade school level. This fact does not make it reasonable to ignore that some educational efforts may be wasted. It's a very, very delicate issue, I grant you, because the danger is that you write people off. And it seems like many unscrupulous people are just looking for an excuse to write people off.
Given this, I understand the position "because I do not trust the motives of my opponents, and fear the truth will be misused in the pursuit of bad policies, I will not seek the truth on this matter, or will assign a very low priority to finding the truth on it." Nonetheless, it would be very helpful to know, if our goal is improving education, what type of effects we can reliably hope to achieve, and whether we can identify circumstances in which education $ are more or less likely to be helpful.
142: OK PF if you were arguing against someone who thought that the Iraq war was a great idea you are telling me you would argue that it was bad just because it is. You aren't going to list off about 50 reasons why it in fact was a bad idea? Because if I didn't know anything about the Iraq war (which I am assuming most people haven't heard of TBC or what it is about) and the pro side listed a bunch of reasons why it was a grand idea and you came back with you are a warmongering fuckwit. I would probably side with the pro war side. I do know about the Iraq war, so I think the pro side is in fact wrong but your argument wouldn't have helped me make a decision.
There's this myth that this sort of research isn't being done, and it's very comforting to some unscrupulous types to believe that it's being supressed, but you know, if I go to scholar.google.com and query for race and IQ, I get 18,400 academic citations. Now, I haven't gone through and checked for duplicates or ensure they weren't talking about racing on a track, but still, I call shenanigans. It takes a specific sort of whiny-ass wonk to whine that the research isn't being done just because he hasn't found one study that magically gives him the answers.
Go and do the research. Just don't base your policies on one poorly researched book and then whine about the liberal academy when your policy looks dumb.
some educational efforts may be wasted.
I'd say it's almost a guarantee. The question is: Do we have the ability to sort out waste from well-spent money? Maybe in terms of good program/bad program. Definitely not in terms of worthy human being/unworthy human being.
148: I'm still not getting why 'figuring out which students to write off' is a bigger emergency than 'giving them all the same opportunities'. If they haven't had the same opportunties, I'm not going to have any equitable way of deciding who to write off, and I don't see the need for focusing on it immediately.
Where's the fire?
151: Yes, yes, yes. One of the things that made me spitting mad about the Summers imbroglio was the response that "How come he was getting in trouble for just saying that research should be done--why should sex differences be off limits?" when of course he was speaking to an audience including many people doing exactly that sort of research.
And, to echo LB, why are we writing people off in elementary school? Of the people I've known and worked who had disabilities, not one would have benefitted from being written off in grade school as incapable of higher achievement.
And of the so-called highly intelligent people I've worked with, a small but significant subset of them had been diagnosed with disabilities when they were grade-school aged. Oddly enough, those youthful "definitive" diagnoses had not actually turned out to be a good predictor of their abilities.
I didn't say anything about which was a bigger emergency. Which is a bigger emergency is a) irrelevant to the what the truth of the matter is, and b) a false choice. We don't have to choose between redressing inequity and finding out what kinds of educational efforts are useful and which are wasted. Indeed, if both were being done in good faith, they would help each other.
156: Regardless everything else going on in the thread, I really don't like the odds of either of them being done in good faith, honestly.
That and he obviously has his head up his ass from the start. Gee, maybe there are no Catholic investment bankers because Jews are naturally better with money.
My more considered opinion on this IQ-policy brouhaha is that above a certain threshold, intelligence doesn't correlate well with success more than, say, your environment, your upbringing, and your opportunities. And I think that threshold is somewhere around "able to tie one's own shoes." Not that there aren't tradeoffs, or questions of how best to spend the money, or good questions to ask about whether we should push people to a lower-tier college if they might be more suited for a trade. It's just that IQ doesn't have much to do with it.
156: I was under the assumption "which is the bigger emergency" is a pretty typical policy question. We don't have time or money or energy to explore everything. Triage is unavoidable.
Gee, maybe there are no Catholic investment bankers because Jews are naturally better with money.
Arrrgh! Cala, you're a wonderful person, must you make me defend Larry Summers who -- just to be clear -- said nothing whatsoever like this?
Don't bother. His argument was pretty dumb even without that.
OK PF if you were arguing against someone who thought that the Iraq war was a great idea you are telling me you would argue that it was bad just because it is.
Pretty much, yeah. I mean, obviously, there are different levels of detail that you go into under different circumstances, but - to take a current example - look at an eminent, reasonable, well-spoken intellectual such as Norman Podhoretz.
It is entirely appropriate to look at his elaborate, politely stated, deeply intellectual argument for why we should bomb Iran and respond, in your elegant phrase, "What a warmongering fuckwit !"
and 163 was me, dammit. sorry
finding out what kinds of educational efforts are useful and which are wasted
I'm interested in your thoughts on this. I don't have any good ideas for figuring out which students shouldn't be given the opportunity for further education, but there must be some interesting propositions out there.
We'll have to agree to disagree. I think his argument may have been wrong, but it wasn't dumb. Looking back at his comments, I remain pretty amazed at the fall-out.
baa - I don't get it. You suggest that LB failed to answer James's question in 137, but of course she did. Still, where is your answer? Where is James's answer?
To remind you, the question was not what research is acceptable (although LB made clear that her position is the exact opposite of the one you attribute to her). James's question was:
And what do you do with the substantial number of children that are not suited for intellectually demanding jobs? At what point do you shunt them onto more suitable tracks?
By asking this question, James implies that the information is already available for an informed answer. LB quite legitimately answered that it was not. But okay - let's take your implication and say the info is out there, and LB merely wants to suppress it.
What is your answer to James's question?
163: I might say something like that to people I knew because I already know what their views are and am not trying to convince them of anything. But if I am actually trying to convince someone that my side is right it seems counterproductive to respond like that. Mostly because I assume that the argument that someone is making is the best argument that they have. And since that isn't really a rebuttal to any argument made I would have to assume that you don't have a rebuttal.
any good ideas for figuring out which students shouldn't be given the opportunity for further education
Well, obviously, merit scholarships already exist. Exam schools like Stuyvesant already exist. Both of these are ways of trying to target resources to students who will benefit more. Those who don't get them aren't "given that opportunity." We also have special ed classes, in which kids are certainly taught differently. I recall that one pretty well-done study found that the benefit of smaller class size was predominantly an effect on kids of low socio-economic status, so maybe small class size is a resource to target away from certain kids. Whenever someone gets more/different, we can describe this as denying opportunity to others.
But in general, I tend to agree that we have really bad data on what works and what doesn't. This is, I think, a good argument for more experimentalism, and more attempts to try things that are different. Maybe there's a great literature of randomized special ed vs. mainstreaming. Or a literature of exam schools vs. normal schools. These are both "randomized trials" that I would try. I would also like to take an existing exam school and try adding 10% kids who wouldn't have made it based on the exam criteria, and see if they are, at the end, distinguishable from the ones who hit the exam criteria.
And (perhaps) needless to say, what I personally favor isn't "we" deciding what "they" get. Rather, it's a situation where money follows the student (perhaps on a sliding scale like the negative income tax where the poor get more), and is controlled by their family. Whether this occurs in a private or public context matters a lot less to me.
Then all this research ends up being information to parents about how best to customize education for their kids. And information that kid can use, at the age of 18, to decide whether to take on the real and opportunity cost of college.
Oh, I figured Shearer was asking for something more than "what we do now" and "someone should figure that out."
But, yeah, I agree that someone should figure out how to best assure that each child gets a fair opportunity to get access to an education that suits their intellectual potential and needs.
He also misuses "beg the question."
That's infuriating, I grant you.
And since that isn't really a rebuttal to any argument made I would have to assume that you don't have a rebuttal.
That is correct. You made a characterization of my view and I elaborated a bit on that characterization, but I basically agreed with it.
But your response was (or at least, I took it to be, and maybe this was just a mistake) "I won't consider this question until ... utopia" That's what seemed unfortunate. If your point is: I want people to have a chance to decide for themselves what they want to do -- lawyer track or car mechanic track -- then we're all singing Kumbaya together, I think...
I think you took my response out of context. Shearer's question was, rephrased, "What should schools do with students who are incapable of succeeding at challenging academic work?" and my response was intended to indicate that identifying such students was contingent on giving them a fair shot at academic success first -- any sorting mechanism that starts from a wildly unequal provision of opportunity before the students are sorted out is going to be unjust.
So, various types of tracking, magnet schools, vocational schools, and so forth may all be good or bad based on their individual merits, but globally it's not a good system unless all the kids have equal opportunities going into it.
And of the so-called highly intelligent people I've worked with, a small but significant subset of them had been diagnosed with disabilities when they were grade-school aged. Oddly enough, those youthful "definitive" diagnoses had not actually turned out to be a good predictor of their abilities.
My dad used to point out that I was diagnosed as "retarded" in some impossibly early grade. I don't know if some special ed—they wouldn't have called it that in the fifties—recommendation went with that, but he felt free to disregard it, and laugh about it later.
But his point always was that the teacher who told us that did so in complete good faith, convinced she was doing the hard but necessary work of telling people the unpalatable truth, lest they get their hopes up, and helping assure that educational resources were directed towards those best able to make use of them.
Ok, sounds like a big deal over nothing then. I agree that sorting presupposes some adequate baseline. My apologies (blog commenting: enormously rewarding!)
Of course, as per usual, I think a truly equal baseline is basically a phantom. There's just feasible way to truly "equalize" the difference between family background, for example. Given this, the question becomes what level of opportunity counts as good enough or adequate.
175
So how many times do you make them fail at challenging academic work before you allow them to try something else? The system should be fair to dumb students as well as smart students and pushing dumb students down an academic track on the slim chance they are actually smart is not fair to them.
Rules like the NCAA Proposition 16 which put artificial obstacles in the path of athletically gifted but dumb kids are very unfair to them. Although contrary to the link they are unfair because there is no good reason to prevent a dumb kid from pursuing an athletic career not because some groups are disproportionately affected.
As for equal opportunity I would deny that the schools alone are currently wildly unequal. If you are including early home environment and the like it makes little difference to the schools whether a kid is dumb because of bad genes or bad parenting there is little they can do about either.
158
"My more considered opinion on this IQ-policy brouhaha is that above a certain threshold, intelligence doesn't correlate well with success more than, say, your environment, your upbringing, and your opportunities. And I think that threshold is somewhere around "able to tie one's own shoes." ... "
So anybody smart enough to tie his own shoes won't be at a disadvantage in an intellectually demanding career? Do you actually believe that? Does anybody else?
147
And if your parents had refused to believe the test and continued to push an academic career on your sister (perhaps with tutors etc.), would they have been doing her a favor?
177: We don't need to go as far as equalizing opportunity in order to some good. Ensuring that classrooms are minimally functional, that there are enough textbooks for the students and some science equipment is a decent start. It might not be enough, and we might not be able to do enough to help the kid who has a horribly deprived childhood get out and up in the world, but what bothers me about these discussions is that it seems as though there's a presumption that if we can't guarantee equality, we should throw up our hands from the start.
178: Shearer, you seem to be assuming that we can tell a priori who the dumb students are, when the actual situation seems to be more like that a poor school will not have, say, a functioning chemistry lab and we therefore conclude that all of the students are just not naturally gifted at chemistry and so we shouldn't waste our time.
So how many times do you make them fail at challenging academic work before you allow them to try something else? The system should be fair to dumb students as well as smart students and pushing dumb students down an academic track on the slim chance they are actually smart is not fair to them.
Have you got policy recommendations here, James? Standards by which to identify the incapable, ages at which they should be tracked into vocational work?
I'm fascinated if you do, but the rhetorical questions about why I don't have mercy on the poor dumb kids who are being given every opportunity in the world to succeed but just won't ever be able to handle calculus aren't going to lead to any interesting conversation.
180: They did refuse to believe the test, and that's why she's going to university rather than dropping out. Again, if you frame the choices as between 'a high-powered career in the academy' and 'a doctor or a lawyer' I think you're missing the point. Higher education educates more than just the elites. Most people who are college graduates aren't elites. If we are talking about how to design an educational system, we really aren't designing it only for the elites.
Moreover, yes, I think it has been good for my sister's mental health that my parents have not treated her as the dumb calasister, but as the sister who's bright and engaging and not very good at standardized tests. She was a very precocious child, especially verbally, but she just doesn't have the knack for decoding exams. This lead to a lack of confidence, and that explains the exam scores. (IQ is just a test score. Tests can be cracked.) I think she is better off going to university than she would be being told that because she wasn't Harvard-bound, she should go to trade school.
The truth is that the great majority of status careers aren't as much intellectually demanding as assiduousness demanding. So, let's assume that IQ really measures something, that something is, for the most part, wasting everyone else's time. What they need from you just isn't that complicated.
As for equal opportunity I would deny that the schools alone are currently wildly unequal.
Some assertions are debate-stoppers, and this is one. Maybe you just haven't seen a lot of schools up close, but gosh, this seems very naive.
Maybe there's a way to convey this to you by putting it into conservo-speak: I disagree with "school choice" as a matter of public policy, but the premise of that slogan - that the difference between schools matters - seems self-evident to me. You don't think so?
My 4-year-old was diagnosed with all kinds of development delays when he was 2. I think those diagnoses were justified. After a couple of years of state-sponsored assistance, he's getting ready to attend a regular kindergarten as a regular kid.
It's always interesting to me that the Shearers of the world are so eager to identify innate inequality in order to single out inferiors so that they can get less assistance. I think that tells you what motivates this point of view.
There are two potential lessons one could take away from my son's experience: That normal human variability makes it tough to pigeonhole kids, or that kids benefit from assistance. Both of these are obviously true, and both argue strongly against encouraging policies that, in James's words, "shunt them onto more suitable tracks."
The truth is that the great majority of status careers aren't as much intellectually demanding as assiduousness demanding.
Isn't this that, 5% inspiration, 95% perspiration thing?
Bérubé's Life as We Know It gave me (yes, yes, unscientific and anecdotal) new faith in what can be done with community support, without so much IQ.
re: 188
I don't think I've ever worked in a (non-academic) environment where being really really smart made much difference over just being not dumb. I'd be willing to bet that even quite large IQ differences translate into not much difference at all above a certain threshold; with the 'perspiration' effect swamping anything else.
Now that James Shearer has demonstrated his ability yet again to get you to argue on any topic of his choosing, can we move on to more important matters? Like what to do with slolernr for deviating from the statistical significance orthodoxy. At first, I thought Gitmo was the only solution, but I learn from Jim Henley the more moderate possibility of trying slolernr at a military tribunal in Leavenworth. Apparently, this is a UN-approved way to handle crimes against statistics, first used at Nuremberg to handle Nazi Bayesians.
190: I think academia gives an unusually heavy weight to "clever" but even there, persistence seems to me the thing that pays off. As one of my friends -- someone who works in your neighborhood, in fact -- used to say, you wait in the taxi rank long enough, eventually you get a cab.
Oh, Walt, I'll extend to you the same offer I extended JAC -- email me your email, and I'll send you the articles. I'm not actually invested in the conclusion -- as I say, it's not my field -- but I think it's pretty funny that it gets certain people riled up.
In my field there are rarely enough data sets for meta analysis.
re: 192
Yes, I think that's probably right.
I'm pretty sure, and I say this without intending to be bitchy, that I know people who have success in the academic world who really aren't, relatively speaking, that bright, but who work hard, and/or have very good interpersonal skills.
I should read some Berube. I agree with what ttaM said--above a certain threshold, it's a rarity in any career where those extra IQ points (assuming they are anything) matter.
In the rare case where those IQ points have mattered, those who weilded them had to be brave or crazy enough to position themselves such that they weren't just in the way.
And I agree with slo. I agree with all. And I will try to read some Berube. Is "Life as We Know It" where I should start?
Huh. I'd say more brains matter in lawyering, but they're far from being the only thing that matter. (That is, being 'really smart' is a particular quality -- there are good lawyers who aren't, some very good, and 'really smart' lawyers who aren't very good at the other 80% of the job. Ideally, they team up, so you've got one person handling the flashes of brilliance, and someone else managing the much more important 'not fucking up.')
I agree with all, too, but disagree, strongly, with text in #196. This is known as quantum comity.
Only now do I see that slo was being backhanded. No matter, those extra IQ points wouldn't have done me any good.
text, Bérubé's Life as We Know It is about bringing up his Downs-syndrome son. It's not high-powered textual analysis Bérubé, which is quite good; I think his What's Liberal About the Liberal Arts is very good.
you've got one person handling the flashes of brilliance, and someone else managing the much more important 'not fucking up.'
According to trial lawyers of my intimate acquaintance, the latter is vastly more important in winning cases.
Rules like the NCAA Proposition 16 which put artificial obstacles in the path of athletically gifted but dumb kids are very unfair to them. Although contrary to the link they are unfair because there is no good reason to prevent a dumb kid from pursuing an athletic career not because some groups are disproportionately affected.
What's wrong with this scene is that colleges got conned into the business of training professional athletes. The NBA, NFL, etc. should just suck it up and pay for a minor league like baseball does.
Then perhaps I'll read it. I'm branching into non-fiction. It occurred to me that 189 could be read in a clever, backhanded way. I clearly don't know anything about Berube. I remember that he debated Horowitz or something.
Too much working and not enough reading unfogged today, so I may have missed my opportunity to argue with LB about education for the proles. But if you're looking to compare the US method of "everyone should have a chance at going to college" vs. say the German Gymnasium/Hauptschule/Realschule system, the existence or absence of differences in intelligence surely has an impact.
Says someone who with only a B.S. is probably in the 25th percentile of educational attainment in the Unfoggedtariat.
So strong is the force of ttaM's personality, that the mere presence of him on a thread causes other people to transpose "i" and "e" in emulation.
It occurred to me that 189 could be read in a clever, backhanded way.
Naw, I was just trying to distinguish between the levels of Bérubé's goodness. It's all good, though.
204: Like I said, 'much more important'.
The NBA, NFL, etc. should just suck it up and pay for a minor league like baseball does.
Ah, but why should they when they don't have to? Shearer and the rest of his big-government allies have won the day on this issue - and Shearer's endorsement of further expanded subsidies is much more likely to prevail than your sensible libertairian view.
The NBA, NFL, etc. should just suck it up and pay for a minor league like baseball does.
The NBA has a minor league; it's called the NBDL.
Good times. Now let's chop those programs. Seven figure salaries for fucking college coaches makes me want to hurt people.
185
"Maybe there's a way to convey this to you by putting it into conservo-speak: I disagree with "school choice" as a matter of public policy, but the premise of that slogan - that the difference between schools matters - seems self-evident to me. You don't think so?"
No, I don't the difference matters a lot in most cases. And to the extent it does matter it is mostly because of the composition of the student body not the quality of the teachers or the physical facilities.
Conseratives like to criticize the performance of public schools when educating stupid kids because it is a handy (if cynical) way to bash the teacher's union. It is more effective than it should be because liberals find it difficult to argue that there are stupid kids and that a school full of stupid kids should not be expected to perform as well (in terms of absolute student achievement) as a school full of smart kids and that for the most part such "poor" schools are actually doing as about as well as "good" schools at educating kids relative to their potential.
211
Actually I would like to reduce the subsidy by eliminating the ability of big time college sports to get away with not paying the talent.