I think you missed Evil Economist’s point. Namely, what these studies suggest is that children of black soldiers who are considered “culturally black” in their home and community environments underperform children of white soldiers, while children of black soldiers who are considered “culturally white” in the same context don’t underperform children of white soldiers. In other words, it’s not the genetic makeup but the cultural aspects of race that lead to underperformance.
Now, you might speculate that the black children in military schools may be on average more “genetically black” than the half-white children of black soldiers in the German schools, and that those additional “black genes” are why the former group underperform all-white kids more. But that still doesn’t explain why the considerable amount of “black genes” in the half-white group didn’t cause them to underperform at all with respect to the all-white kids.
In the first cite you say: "The main criticism of the study . . . was the fact that the source populations were not necessarily a representative cross section of either population. In particular, blacks were screened against military IQ standards, of which about 30% of blacks failed "
So, according to you, the illegitimate children of black soldiers raised in Germany scored the same on IQ tests as the illegitimate children of white soldiers because black soldiers are selected for high intelligence.
However, you then provide a cite that the children of black soldiers score lower than the children of white soldiers when they attend US Army schools. The parents of these children are the same black soldiers who you earlier said were selected for high IQ.
Right there is your contradiction.
Children of black soldiers who are raised in German households score the same on IQ tests as the children of white soldiers. Meanwhile, children of black soldiers who are raised in American schools get lower scores.
Bam, you just disproved the intelligence gene theory. Thanks.
But the issue is not how you personally feel about the Flynn effect (and remember, I’m talking about “Flynn effect” in the sense of rapid changes in IQ test scores over time even in racially homogeneous populations).
The issue is whether you understand that the existence of the Flynn effect casts serious doubt on any claims that IQ testing is adequately controlled for environmental factors.
Then you need to explain why environmental variables are still so poorly controlled that we can’t even reliably compare IQ test scores over time in the same population, even when there are no racial differences involved.
That kind of looks like trying to change the subject. Remember, my point about the Flynn effect is that it’s observed even when no race-based differences are involved. And it indicates pretty clearly that we don’t sufficiently understand the effects of non-genetic factors in IQ testing to control for them successfully.
My pleasure, although I don’t quite follow your reasoning.
The modern study shows normalization of opportunity does not normalize outcome in black-white SIRE groups of military kids taking an academic test. The studie’s “environmental equality” in this case was chosen by the Journal of Blacks in Higher expectation because they thought it was so nearly perfectly equal. The JBHE changed their mind only when the gap persisted. Both populations in this study were chosen using the same screening criteria (whatever that is for the military these days). No one is saying the military population is a representative cross section of their source SIRE groups, as far as I know. There is a presumption that both groups are selected using roughly the same criteria.
The WWII study shows that admixed children did not regress to a mean that was much lower than that for whites; the study was small and the mean for white girls atypical. The black fathers were not a representative cross section of their source population.
But if you think these two studies contradict one another and the “intelligence gene” theory is therefor disproved based on them, enjoy your conclusion.
Again, I’ll ask you to read a thread you jumped into after a few hundred posts to see the comments and how they have evolved. I do not believe I am changing the subject at all by asking you the two questions that you dodged.
If we say that IQ tests change rapidly over time, that IQ tests as currently administered and reported reflect average source population cognition ability, and that the putative explanation for this is environmental change, we’re stuck with research that gives us a black adult average IQ of 85 in 2002, 79 in 1972, and even lower prior to that. We’re stuck explaining why academic test scores didn’t budge in the same direction. I don’t have academic test scores for non-US populations. I don’t know how IQ tests are developed, normalized or administered as a mechanism for getting average population IQs.
For a given individual, I think an IQ tests tells us something about cognition. I do not pretend an expertise in how accurately they are supposed to measure the IQ of an entire population, as they are administered today. I do note that, as a far as I am aware, the general results don’t violate the typical pattern of asians on the top tiers and blacks on the lower tiers.
I have no idea, really, what to do with the Flynn effect. I do not find it particularly persuasive as an argument that real intelligence is increasing in whole populations. I find the evidence that the skillset for mastery of taught content as measured by academic test scores is genetically driven to be very strong. This gap has not closed beyond a stubbornly persistent performance gap that is not related to SES.
Here is a possibly useful blog about a recent Flynn lecture. In the middle section there is a summary of Flynn’s black-white data, along with some summary comments:
"Flynn doesn’t believe that blacks and whites are born with differences in intelligence. As he rightly points out, it wasn’t that long ago that some psychologists were arguing that Irish immigrants in the United States were genetically inferior. But when Irish Americans began to invest in education, they completely closed the gap. Instead, Flynn argues that these trends become cumulative, and problems are already evident in preschool. He proposes some environmental explanations, including differences in attitudes toward academic achievement.
He says he is fully aware of the controversial nature of his research and his ideas, but thinks these are serious issues that require rigorous investigation. He believes that IQ trends show us interesting social trends. He told the audience that he has suggested half a dozen studies that could shed light on this issue,
(quoting Flynn) ‘But you cannot say these things. They are forbidden. Which means of course we go on in ignorance of what actually causes group differences. Which means we can’t provide any solutions. When you turn your back on reality you lose the ability to manipulate reality. One would think that is self-evident…I didn’t go into this to not try to find the truth.’ "*
I am obviously less enamored than Flynn of these envrionmental differences as the putative causes for observed test score outcomes since they don’t hold up to scrutiny (see the links upthread for various studies). You are going to see black-white score differences whether the kids are rich or poor, attend a fancy or putzy school, or have all black or all white teachers. It’s a remarkably stubborn pattern, and remarkable consistent across SES and environments.
My two questions for you remain.
As a reminder, here are those two questions:
Do you accept Flynn’s data that the current adult black IQ in the US as of 2002 is 85, and was 79 in 1972, with even lower scores prior to that?
What nurturing influences would you like to advance as the ones most likely to explain why rich black kids with educated parents underscore poor whites and asians with uneducated parents?
This is a math question. (A + B + C +…)/X = 79 or 85 or whatever. Flynn’s data holds that black American’s IQ test scores in 1995 match that of white Americans of 1945. So the black-white gap matches the white-white (1945-1995) gap. Why are you so obsessed with this math question? If the answer to the equation is 85 or 79 or whatever for white Americans in the 30s or 40s, why do you keep harping on it again and again?
There are many possibilities, some of which have been advanced in this thread. I think it’s reasonable to propose that there may be something intrinsic to the “black experience” in America- perhaps some combination of media influences, cultural stereotypes and pressure, educational expectations, and other things, that transcend economics and parental education, and serve as obstacles in the path of academic and educational achievement for black Americans. But I’m not confident in any particular explanation- I’d like to see good evidence for one. There is no positive evidence for your genetic explanation, so I don’t treat it as any more likely than other explanations with no positive evidence, and less likely as explanations that we at least know exist and can have some effect (like teacher expectations, which we know exist, and we know have an effect on a child’s performance, though it may not explain all of the gap).
Hmmmm…I must not have explained myself well. Let me try again:
I can think of only three ways to explain the dual facts that 1) illegitimate children of black soldiers raised by German parents score the same on IQ tests as illegitimate children of white soldiers raised by German parents, and 2) legitimate children of black soldiers raised by black parents score lower than legitimate children of white soldiers raised by white parents. Those three potential explanations are:
[ol]
[li]Environmental differences, not genetic differences, cause the IQ gap, or[/li][li]Army admission standards during WWII were much higher than army admission standards today, or[/li][li]German women have magical vaginas.[/li][/ol]
You claim that the reason that illegitimate black children raised in Germany score the same on IQ tests as illegitimate white children raised in Germany is because army admission standards were so high that only super-smart blacks were accepted into the army. But now you must explain why children of black soldiers score lower then children of white soldiers today. We’ve established that black soldiers are super-smart, so the reason must be environmental, right? We can’t claim a genetic explanation without contradicting ourselves.
I disagree. You seem to be still either misunderstanding or purposefully downplaying the implications of the Flynn effect for IQ testing, which is the point I was originally making and which is crucial to any claims about drawing inferences from IQ test scores in general.
Sure, but the problem is that we can’t be sure what it tells us about cognition, because phenomena like the Flynn effect make it clear that IQ testing does not adequately control for non-genetic factors, even in racially non-diverse populations.
The honest thing to do with it would be to acknowledge that it casts doubt on claims that non-genetic effects are successfully controlled for in IQ testing (because the Flynn effect shows test score increases that can’t plausibly be explained by genetic factors, but have not been successfully analyzed or explained in terms of non-genetic factors).
The dishonest thing to do with it is to shrug it off and keep on making the same discredited claims that environmental factors are adequately controlled for in IQ testing.
Exactly. Because it shows that IQ testing is not yet sufficiently scientific to exclude the effects of transient social and cultural factors when it comes to measuring what we call “real intelligence”.
As I pointed out, there is no such discrepancy on any sound statistical basis. Nowitzki represents a far smaller % of Germans than American whites being able to reach the NBA, and Nash is outnumbered by black Canadian NBA players by about the same ratio as white US NBA players are.
But you speak of examining further, so let’s in this case. In their own ways Notwitzki and Nash, and US born players like for example Larry Bird and John Stockton are or were highly effective NBA players. But almost no Caucasian NBA player ever has had the style of game like the all time great NBA players, mainly 2-guard/small forward types whose games exploit tremendous leaping ability, foot speed and quickness in one- on-one or in the open court, players like Dr. J or Jordan in the past or Le Bron or Kevin Durant now. And there’s a long list of players of generally similar strengths descending very far from that until you find any white players with that general type of game (Bobby Jones or Dan Marlje in past, perhaps David Lee now? much less illustrious players). That’s certainly not the game of Nowitzki (the core of which is a great shooting touch at fully 7’) or Nash (canny passing).
Somebody from Mars without a PC agenda would apply Occam’s Razor and conclude that the type of exceptional physique from which a hard working athlete can develop a game like that of Dr. J, MJ, LeBron etc is more common in people of African heritage, since people with that type of game on anywhere near that level, who are not of African heritage, are essentially non-existent. Such a style of play is not the only way to be an NBA player, as shown by plenty of non-black NBA players, a few of whom at any given time are excellent ones (Lee was the only NBA all star this year lacking visible African heritage), nor the only way to win a basketball game v NBA Olympic teams, but the foreign Olympic teams which have occasionally beaten NBA teams with better teamwork and defense (overconfidence and lack of motivation have clearly played a role as well) simply don’t have individual players like that.
I don’t recall saying anything was too hard or took too long. I stated the obvious fact that such differences are more obscure when you look at averages than when looking at extremes, where they scream ‘self evident’, especially if we really look closely and objectively as in point 1, and not just look for any excuse to come up with the politically ‘right’ answer.
And the basic problem with ‘bring your evidence!’ is that our society already has highly intrusive, resource consuming, wealth destroying policies which assume that differences in achievement can’t be related to race, evidence of which you don’t have. So as often, on the thread and in general, the burden of proof is the wrong way around. I’m not proposing any interference in the natural interaction of people based on race. If by some process I can’t fathom (nor can you explain to me, I’m sure) we eventually see lots of top black mathematicians and physicists, or something more than .1% or 0% of black applicants scoring at the white/asian 99%-tile on grad school tests, or we see lots of Dr J/Le Bron style white NBA players, well then there we’ll be! I’m not proposing government interference to ‘fix’ something when there’s no real evidence it’s fixable. You, if defending status quo policies, are.
I claim the Eyferth study is small, suspect (because of the low IQs of the white females, which are out of line with their group), and shows (flaws aside) that biracial children raised by German mothers had a pretty normal IQ when their black fathers were selected from a subpopulation that left out the lowest-scoring 30% of all blacks.
I claim that the JBHE study does not look at biracial populations, but at SIRE groupings. There, both groups had nearly identical nurturing environments but the black SIRE group still underperformed the white SIRE group. I do not know how and whether the military selection process for inductees has changed in 70 years.
It does not surprise me that, if you weed out the lowest-scoring three deciles of the black population, the remaining population has IQs that are higher than the average IQ of all comers for the comparison population. Why wouldn’t children of such an admixture have IQs closer to the white average, if IQ is genetic?
The second study looks at academic scores. If this skillset is genetic, the expectation would be that two subpopulations (the ones screened out by the military) have similar scores for similar nurturing. They don’t. They have dissimilar scores. I would expect biracial children of these black servicemen to do somewhat better academicall, although nearly all biracial studies/adoption studies and so on are marred by selection biases.
So…aside from accusing me of being dishonest, and accusing me of going off on a tangent when it is you who jumped into the thread quite late, are you able to answer my two questions?
As a reminder, here are those two questions:
Do you accept Flynn’s data that the current adult black IQ in the US as of 2002 is 85, and was 79 in 1972, with even lower scores prior to that?
What nurturing influences would you like to advance as the ones most likely to explain why rich black kids with educated parents underscore poor whites and asians with uneducated parents?
When you answer the first one, I think we’ll get a better idea of which of us is more enamored than the other with Flynn’s research.
And? Do current military entrance requirements leave out the lowest 30% also? Because if they do, then your genetic explanation has a problem, if children of black soldiers underperform average white children.
So, you don’t know how the military selection process has changed? But you are implicitly claiming that it’s gotten less selective, because if not, and if there was a race IQ gene, then the black students in the JBHE study should have gotten the same scores as the white children, right?
So, you are claiming that during WWII only high-IQ black soldiers were accepted? You claim that the average black soldier in WWII would have scored higher on IQ tests than the average white person? Not just that, but if the black and white illegitimate children both scored the same on IQ tests, then it must be the case that the average black soldier must have had the same average IQ as the average white soldier, right?
Are you aware that the military applied IQ tests to its recruits during WWII? Do the results of those tests support your thesis?
But if they don’t, then how do you explain the German data? I mean you just explained it by saying that we had screened out the bottom 30% of blacks and were only left with the creme de la creme, who were equivalent to white soldiers. Now you’re citing a study to show that children of black soldiers underperform the children of white soldiers. Which must mean that black soldiers must not be the top 70%. But then how do you explain the German data?
Could you be helpful and let me know if black soldiers are in the top 70% of the population or not, please? And if you say no then please explain the German study. And if you say yes then please explain the JBHE study.
Without knowing the IQ of the parents I’m not sure this gets us very far either way. The military apparently has a cut-off below 92 IQ. Obviously that leaves a lot of scope for variation above that level.
There does seem to be evidence thatBlack and white children regress towards their respective population means. Also, there are a number of larger and more recent samples showing that mixed race Black-White kids perform intermediate to their parental populations (see item 19 in the linked summary).
I remain confused, and perhaps permanently so, about your point.
Sorry. How does the fact that current black military children underscore white military children mean their fathers are not in the top 70% (and I have no idea if they are, or are not).
If current military entrance requirements still leave out 30% of the lowest scoring blacks and 3 % of the lowest scoring whites (the numbers for the Eyferth study) then the expectation would be that including all applicants instead of excluding the lowest 3 deciles of blacks would make the black-white gap even larger in the JBHE study.
One of the criticisms of the Eyferth study is that it excluded the lowest 30% of blacks, leaving a black population not representative of a random cross section of the source population. If the skillset for IQ is genetically driven, one would expect offspring of this subpopulation of blacks not to have the same regression degree toward a lower SIRE average than would a cross section of all black fathers. The Eyferth study does not say the black IQs were equivalent to the white mothers; it only studied the IQs of the offspring. Nor does it compare the IQs of the black and white soldiers. It’s really just looking at a biracial outcome for IQ, and saying that biracial children of black fathers screened to be in the top 70% of the black population for IQ can father children of normal IQ with white German moms. I don’t find that outcome amazing, although I agree with criticism of the study.
The JBHE study does not look at a presumed regression point for offspring; it looks at a direct difference. And it is significantly less flawed than the Eyferth study. It’s just looking at a skillset for academic test taking, under conditions which closely control for nurturing. Neither of those things are true for the Eyferth study. The conclusion of the JBHE study is that there must be some environmental variable they didn’t look at when they made the assumption the environment was nearly identical. They make this conclusion because the gap persisted, not because they were able to identify what it was.
It’s not a math question. It’s a question of what it means functionally to have an IQ in the 70’s. I find the idea that the same adult white male with an IQ of 75 today would have an IQ of 100 during the 1940s to be ridiculous. IQ scores are created to correlate with functional cognitive ability. Someone with an IQ in the 70’s is fairly challenged, mentally (to put it politely).
There isn’t a shred of evidence elsewhere that a dumb guy today would have had average intelligence 60 or 70 years ago. Academic test scores have dropped over that time.
You can argue that IQ test scores themselves mean nothing wrt intelligence (and good luck with that, but as I’ve said, I’m no expert on IQ tests). But if you argue that, then rising IQ scores also mean nothing substantive. Or you can argue that something is fishy with the whole Flynn research, the data, the way tests are normalized, the way they reflect whole population IQs…whatever.
In any case, I don’t buy that we are getting more intelligent this fast. I don’t buy that blacks are 17 points below whites now. And I don’t buy that they were 25 or 30 points below whites 70 years ago. That’s a pretty frightening gap for an intelligence proxy like IQ–a much different phenomenon than a difference in taught material on an academic test (at least part of which can be readily explained by a difference in opportunity 70 years ago).
I’m still curious if Kimstu (who is a lot smarter than I am) considers the current or past Flynn numbers to be accurate for average adult black IQ scores in the US, and if she does, has some putative explanations for why the gap in academic scores doesn’t close even if we normalize SES. If she does consider the Flynn numbers accurate, but IQ scores not to mean much as far as intelligence, then maybe we can stop obsessing over some guy whose research I am unable to make heads or tails of. I’d be curious too, if she has any remarks on the disparity between IQ scores rising and academic test scores remaining flat or decreasing.
I might add one more study to consider if you are interested in pursuing the concept of biracial children and where they score…
The Minnesota Transracial Adoption Study, summarized here in Wikipedia, looked at natural children of white Minnesotans, along with adopted children from black, biracial, and asian parents.
The raw data, particularly when corrected for the Flynn effect are pretty much what a genetically-oriented theorist would predict:
At age 17, white adoptees had an IQ of 101.5, biracial adoptees 93, and black adoptees 84.
The authors of the study have vigorously defended against its use by those who want to jump to a conclusion that IQ is heavily determined by genes. You can read their 1994 defense against Levin and Lynn (for example) here. It should give you some encouragement that the authors of this study think environmental factors are at play despite the plain data that they themselves derived in their study. (I should say I find their example of the very low “asian” IQs as supportive for an environmental cause to be weak, but they do make a number of points that I think you would make, so you can just refer me to the second link instead of re-writing everything yourself.)
Still, it’s an example of a general pattern that is derived over and over again, and always with the demand that every environmental influence be chased down before we leap to a genetic conclusion. Never the other way around.
Another interesting tidbit in this study is that, if one asserts that some sort of environmental factor was at play because the adopted kids looked “black,” one has to account for the fact that a dozen or so of the biracial kids looked so “black” their adoptive parents didn’t realize they were biracial. Yet those biracial kids were in the group that scored an intermediate IQ at age 17…
Anyway, if you haven’t looked over this rather classic study, it’s probably worth a peek. You may also be interested in Scarr’s comments on Arthur Jensen’s integrity, quoted near the end of the Wikipedia article. “The test performance of the Black/Black adoptees [in the study] was not different from that of ordinary Black children reared by their own families in the same area of the country. My colleagues and I reported the data accurately and as fully as possible, and then tried to make the results palatable to environmentally committed colleagues. In retrospect, this was a mistake. The results of the transracial adoption study can be used to support either a genetic difference hypothesis or an environmental difference one (because the children have visible African ancestry). We should have been agnostic on the conclusions […].”
Ok, so you don’t “believe” in the uncontroversial and widely accepted, by pretty much everyone, Flynn effect. Got it. I’ll keep in mind that you have trouble accepting data that conflicts with your hypothesis, regardless of the evidence. I’ll repeat that, among experts, no one denies the Flynn effect, that people 50 years ago or so scored significantly lower on IQ tests.
I think you are misrepresenting my opinion about Flynn’s research, because it helps you make an implication that I simply discard any data which conflicts with mine. To the contrary, I frequently post alternate views, including a paper just above defending the Minnesota Transracial Adoption study against its use to promote a genetic hypothesis.
The trouble I have with Flynn’s research is not that I think he found at least some evidence that IQ test scores are rising. I believe others using similar techniques, have observed this phenomenon elsewhere. The more fundamental question is whether these rising scores are an accurate reflection of rising average intelligence. Here are the areas that confuse me in terms of what to make of his analysis:
SAT scores, broadly administered over very large segments of the population, are NOT rising. As a matter of fact, over the time period when these IQ test scores were rising, they were static, or even dropping. What accounts for this disparity? Why would smarter kids be performing worse on a standardized test?
IQ tests are designed to measure cognitive ability, not grasp of content. A given score is standardized such that scores in the 80s are considered “dull normal cognition” or some such similar terminology, and scores in the 70s are considered mild to moderate mental disability. This is quite a different thing from a crummy score on a test reflecting mastery of content. There is a very large difference between ignorance (I don’t know that fact) and lack of intelligence (I am cognitively unable to grasp that). So if rising IQ scores are supposed to be reflecting any real rise in intelligence, we can go back 60 years and ask this question: was the average person 60 years ago literally stupider than today, and was half the entire black population in the 1950s moderately disabled for cognitive function against the standard for mental ability at that time? Or the other way 'round: Would someone with an average intelligence today be way ahead of the curve if transported back 60 years to the dullards who lived then?
If the rising IQ test scores reflect a real rise in average intelligence across whole populations, when did it start and when will it end? If it’s related to environmental factors such as general SES, what accounts for the marked disparity in academic test scores, where rich black kids with educated parents get such crappy SAT scores? Are we now arguing that poor teacher expectations make those kids cognitively duller versus simply less inclined to learn something?
Do those who promote Flynn’s analysis as an argument against a genetic explanation for intelligence admit with equal alacrity that accepting his data means accepting the average black adult IQ in 1972 was 79, and that only 6 of 21 points has been gained in 40 years? I don’t see people trumpeting the Flynn effect concomitantly trumpet quite as enthusiastically that they think the average black adult IQ in the US is 85 today, and that since it’s a bell curve, half the adult black US population has an IQ below that. Instead, what I see are cautions about interpreting IQ studies, and cautions about extending what studies there are to infer an average for a whole population. But if those cautions are legitimate, then any significance of the Flynn effect is also diminished. You cannot have it both ways. Either IQ tests mean something substantive at a population level, and adult blacks are currently a full standard deviation behind whites and asians, or these rising IQ test scores don’t reflect population averages (or don’t measure cognitive ability) and therefore there is isn’t good evidence that adult blacks are substantially “dumber” than whites. Most genetic egalitarians have argued the latter, preferring to dismiss academic scores as a reflection of nurturing opportunity and not a reflection of fundamental cognitive ability, even if that adult cognition deficit was itself a product of the environment. In other words, egalitarians don’t seem to want to trumpet that IQ test scores currently reflect a real, structural, fixed difference of a full standard deviation in actual cognition between adult blacks and whites. But if they want to use the Flynn effect to argue their point, they have to do that.
Finally, to the extent that the Flynn effect does show a relatively more rapid rise in black versus white intelligence, it only argues for a degree of malleability. It does not create a case that intelligence is so completely malleable by environment that we can create geniuses at will with the right nurturing. It’s not a correct leap to say that, if some of an intelligence gap can be closed with better nurturing, all of it can. What our genes determine is the maximum potential an individual can attain with perfectly optimal nurturing. If relatively better average nurturing for blacks closes some of the gap, but an enormous gap remains, that simply fits a hypothesis that intelligence itself is a combination of hereditary and environmental influences. On that point, I think everyone agrees. At issue is not whether or not intelligence is malleable at all to environmental influences, but whether or not groups with disparate gene pools have disparate maximum potentials when exposed to similar environmental variables.
As always, the question is what causes a residual gap when environment is normalized? The Flynn effect does not resolve that question; it merely suggests (if correct) that some of what we measure in IQ tests is malleable by environment. But even there, it would be nice to see some sort of non-IQ test proxy that we are all getting smarter. Improved academic scores would be a nice starter proxy. I don’t see that in most of the academic test score studies I look at.Or here, for SAT historic scores.
are pretty much what a genetically-oriented theorist would predict: