[It is commonly said that “the difference within the races or between siblings is as large/larger than the differences between races.” ]

Maybe this is not the space to put forward a slightly technical argument, but it seems to me this statement is statistical nonsense.

It seems to me thatIQ difference *within* a race could only be the range between the lowest and highest IQ’s found among individuals of that race. That is, two individuals would determine this number. I.e. for whites.

IQ (smartest white ) - IQ (dumbest white).

Now, two individuals do not a scientific sample make. Average IQ level for the different races are measured via samples of *thousands* of individuals. The *average* naturally tends to the mean of each group. When people say that IQ varies more within races than between them, they are comparing statistical apples and oranges.

That argument has never made sense to me either; nevertheless, it has been very effective in making people imagine that racial differences really don’t matter. That’s why it’s important to nail the issue down. And the best way to do that is to compare the two bell curves in their totality and point to the black-white differentials above IQ 115 or IQ 130.

Well, the statistics aren’t going to tell you about actual individuals at all, only about populations. Trying to find the one actual black or white person at this or that end of the distribution doesn’t make any sense.

Mr. Auster’s approach to it is exactly right. It is abstractly possible for there to be very narrow distributions for blacks versus whites (or asians, etc) such that there is virtually no overlap among the everyday Joes. It is also possible (and in fact the case) for there to be a large overlap (so the differences between random siblings are similar to the differences between any random pair of everyday Joes of different races). What goes unsaid is that none of that matters. What matters is the percentage of each race is found in the higher bands of intelligence where society’s leaders disproportionately come from, and in the lower bands where society’s troubled disproportionately come from. In the high band there are six times as many whites as blacks on a percentage basis. On the lower end of the spectrum the same is true in reverse: there are six times as many blacks as whites on a percentage basis. This is an objective fact about the world, not some racially supremecist moral assertion.

Of course the things that become political issues are boundary conditions like these. Whether there is or is not large overlap in the everyday Joe category — and there no doubt is — is irrelevant to the most critical political and social implications.

Exactly my point. To say that IQ varies more within races than between races is sheer nonsense, because total IQ variance *within* a race is a matter of individuals — or in the case of a sample of individual cases — while any comparison between races is a matter of comparing means of the two groups, and thus the law of “regression to the mean” begins to take effect.

I suspect you would start seeing the effect with rather small samples, say N=30. That is, if you took 30 blacks and 30 whites, you would already be seeing a noticeable difference in *average* intelligence much more frequently than seeing a roughly equal level of intelligence. This is important because people look at a small organization — a firm, a school, etc — and see blacks in lower level jobs and whites in higher positions. Discrimination is assumed, but it is really the effects of differential IQ among groups.

I will say that there is an element of ‘social construction’ in all this. After all, Latin countries like Mexico and Brazil confront the same situation. In those countries the ‘color line’ was not so fixed, so more mixing took place. Whiter people still come out on top, darker on the bottom, but the difference is much more subtle than the US case. (Steve Sailor has written much on this phenomenon).

There is yet another issue. Whites are not the smartest group. East Asians have higher average IQ’s. Does that mean that whites should welcome the current high rate of Asian immigration as a way of bettering the country? I say no. I think it unfair that the whites who paid for the building of the UC system, for example, to see their children and grandchildren being squeezed out from that system by immigrants from Asian and children of Asian immigrants, many of whom were quite well off in their home countries. I don’t propose any affirmative action for whites, but the question then becomes; Are limits to immigration a sort of affirmative action for present day Americans, shielding them from competition at both the high level of education and the low level of manual work? I think this is the neocon position — neocons see immigration restriction as a sort of affirmative action for Americans.

Indeed the larger issue is that whites are being squeezed out of the entire state of California if they are not willing to put up with third world conditions and pay. Sorry for meandering away from the topic.

What then are people actually referring to when they speak of the variance within races and say that it is greater than the mean difference between races?

On the question of differences betweeen siblings, that would seem less logically problematic than the difference within races, though still doubtful. What is apparently meant is that the average difference between two siblings is as great as the mean difference between whites and blacks—15 IQ points. I haven’t read up on that, but it seems very unlikely.

Mr Auster writes:
“What is apparently meant is that the average difference between two siblings is as great as the mean difference between whites and blacks—15 IQ points. I haven’t read up on that, but it seems very unlikely.”

I don’t have a comment on whether siblings typically differ by 15 points or so, since I don’t know the factual answer to that factual question. But I do know that we can stipulate it and it doesn’t really change anything. A large standard deviation doesn’t eliminate the impact of a large difference in the mean.

Also Mr. Young’s point about the statistical significance of a sample size of thirty or even less is well made: in a sample of 15 randomly chosen blacks and 15 randomly chosen whites we should expect the average IQ of the whites to be 15 points higher than the average IQ of the blacks. Companies and organizations aren’t random samples, but different roles reflecting different average intelligence should be expected pervasively across all organizations, even small ones.

Note that Jared Taylor will be appearing on MSNBC this Memorial Day at 6:30 pm Eastern to discuss diversity. The editor of the American Renaissance is a guest on the Buchanan & Press show. Maybe the readers of Lucianne.com will take notice of it?

I find it puzzling when I read of “asian superiority” or that “asians are the smartest because of the highest average IQs”. People get this argument from the “Bell Curve” which shows the asian IQ mean at about 103, where the caucasian mean is 100. Well, if you look at the charts it shows that the asian bell curve is very steep and clusters around the average whereas the caucasian curve is much more spread out. According to this chart the steep asian curve pretty much stops at 120 where the much flatter caucasian curve extends above 140. So according to this chart, the vast majority of people with IQs above 20 (highly intelligent and genius) are caucasian. So if this “Bell Curve” were true, how someone could say that asians are superior when they have such a low representation of IQs over 120 is beyond me! Some whites argue that this is the reason there is virtually no innovation in asia in terms of great technological advances. Personally, I think that the whole IQ thing is pretty much meaningless. I know a lot of very intelligent asians, very intelligent whites, and very intelligent blacks. I just find it insulting to be called “inferior” to another race and think that people who argue that whites are “inferior” based on the bell curve, should take another look at the chart.

>>>>I find it puzzling when I read of “asian superiority” or that “asians are the smartest because of the highest average IQs”. People get this argument from the “Bell Curve” which shows the asian IQ mean at about 103, where the caucasian mean is 100. Well, if you look at the charts it shows that the asian bell curve is very steep and clusters around the average whereas the caucasian curve is much more spread out. According to this chart the steep asian curve pretty much stops at 120 where the much flatter caucasian curve extends above 140. So according to this chart, the vast majority of people with IQs above 20 (highly intelligent and genius) are caucasian. So if this “Bell Curve” were true, how someone could say that asians are superior when they have such a low representation of IQs over 120 is beyond me! Some whites argue that this is the reason there is virtually no innovation in asia in terms of great technological advances. Personally, I think that the whole IQ thing is pretty much meaningless. I know a lot of very intelligent asians, very intelligent whites, and very intelligent blacks. I just find it insulting to be called “inferior” to another race and think that people who argue that whites are “inferior” based on the bell curve, should take another look at the chart.>>>>

First off The Bell Curve never showed an asian bell curve chart. Second, the chart you got was from this wasn’t it. http://www.lagriffedulion.f2s.com/sftfi1{image3}.gif
http://www.lagriffedulion.f2s.com/sft.htm

Notice that they are hypothesing with this chart. Read CAREFULLY what they say here:

“An Asian distribution with a standard deviation 39 % of the white brings the two smart fractions into coincidence. An even narrower distribution would be required to explain the Lynn-Vanhanen data. Either way, there is NO EVIDENCE to support such a difference in spread. I’m afraid we will have to look elsewhere for an explanation.”

As you can see, they have discredited this theory.

The fact is there have been absolutely no evidence to show that asians have a lower standard deviation and if anything have been shown to have a higher one. This can be shown by SAT scores where the average gap is asians/pacific islander scoring about 10 points above whites. But asians also had a higher standard deviation at 123 compared to whites at 100. So when you took at the 99th percentile of both groups, which is people scoring about 1500 or so, the gap increases to about 100 points asians over whites. So it seems that not only do asians have a higher mean average but also have more intelligent people on the far end of the spectrum.

Here is the source. Go to table 4-1.
http://www.collegeboard.com/prod_downloads/about/news_info/cbsenior/yr2003/pdf/2003_TOTALGRP_PRD.pdf

So anyways Mr. Anonymous person to sum it all up. Yes the Bell Curve is probably true. No the Bell Curve never stated that Asian IQ stopped at 120. That hypothesis was based on some random person after reading “IQ and Wealth of Nations” and was quickly discredited. And so no asians do not have a lower standard deviation but probably have a higher one.

After engaging in a detailed discussion of racial differences in IQ based on differentials in standard deviation of different populations, Knight tells us that “Personally, I think that the whole IQ thing is pretty much meaningless.” Huh?

Also, I skimmed the article on the Smart Fraction Theory of the Wealth of Nations. One thing doesn’t make sense. They say, assuming an Asian mean IQ five points higher than the white, that an East Asian SD 39 percent the size of the white SD would be required to bring down the smart fraction of Asians to equal that of whites. But this assumes a definition of the smart fraction as everyone above IQ 108. Yet this whole discusion (so I thought) assumed that the difference in racial/national performance depends on the percentage of a population above IQ 120. So it seems to me that that explains the puzzle raised in that discussion of why East Asian performance doesn’t match its higher mean IQ. A narrow Asian distribution DOES lead to an equal-to-the-white proportion of the Asian population with IQ over 108, but it doesn’t lead to an equal-to-the-white proportion of the Asian population with IQ over 120.

I may have misunderstood this by skimming it instead of reading it carefully (the inherent problem of reading complex materials on a computer screen), so correct me if I am wrong.

Sounds like IQ is important but is being oversold. There are cultural differences between Asians and Anglo-Americans that affect economic performance. Many Asian nations tried to remain closed societies for centuries, which cuts down on intellectual cross-pollination, so to speak. The pressures for conformity in a country such as Japan are not conducive to creativity.

Asian languages are cumbersome and inefficient means of communication, by whatever historical accident related to their development. I once read a travelogue by a famous computer scientist (the late Edsger Dijkstra of the Netherlands) who said, based on oral translations given at computer conferences overseas, that it seems that for every 5 words used in English to express something, it took 7 words in Russian and 10 words in Japanese. These languages are also poorly suited for the modern era of computers, keyboards, printers, the web, etc.

Various non-IQ factors have an effect on economic performance, leading to seemingly unsound conjectures about standard deviations being improbably narrow. This is the problem always encountered when someone overextends a good insight into a comprehensive “theory of everything”.

My mistake, it was “Anonymous,” not Knight who said that IQ doesn’t matter. Knight had quoted Anonymous’s entire post.

To Lawrence: In lack of a quote button I tend to use this >>>> >>>>. It might be confusing but it’s due to this site. They need more advanced messaging system. Quote features and bold and italic and color fonts. :) And smiley faces.

There are numerous factors that constitute the wealth and technological advancement of a nation. Intelligence is one of them yes but it is NOT everything nor close to it.

America is far richer than Germany, France, England but I doubt anyone will make the argument that it’s due to because Americans are smarter or that they have a higher percentage above 120. A political and economical structure has a vast impact of a country’s wealth. Not to mention resources of the land you’re on.

It should be pointed out that the use of the word “Asian” here is very imprecise. Apparently, East Asians or “people of Mongoloid race” is usually meant, excluding Indians, who are Asians but Caucasoids who differ from Europeans only in the genes for color. Yet Indians also do exceptionally well in certain intellectual fields. Are we seeing an effect that is common to all people with a cultural heritage of high civilizations rather than a racial one?

IQ SD could be very narrow in every Asian country and still be wide overall. The chart by Lynn and Vanhanen (whose name I hope everyone is pronouncing in the correct Finnish manner, with hard stress on the first syllable) shows two widely divergent groups of East Asian lands. One’s range is 103-7 and the other 86-92. (China, the 800-lb panda, sits squarely at 100.)

New Worlders are descended almost entirely from restless migrants, and their departure would have subtracted quite a bit of restlessness from their Old World homelands. This may explain some of the economic disparity.

Money has negligible explanatory significance for the cultural distinction of nations and larger (and smaller) groups. Italians in the 15th century were not richer or healthier than the people of Africa or China today. Composers have not required contemporary incomes, and the growth of average incomes seems to cut down the number of distinguished ones. It is the same way with painting and other subjects. Medicine would seem to be an area that every culture places high priority on; yet in all of these fields, the important achievements show European predominance in ratios on the order of one hundred-to-one (as against the non-whites). Immigrants’ self-selection effects have been mentioned, and these would explain how we can get high-scorers to clump in certain countries. Even so, the east-asian average on conceptual ability (verbal intelligence as distinct from mathematical and allied skills) can be assumed to be lower in order to account for the high-achievement differential. Neither standard deviation, nor money, nor cultural priorities, can explain this towering discrepancy.

“Even so, the east-asian average on conceptual ability (verbal intelligence as distinct from mathematical and allied skills) can be assumed to be lower in order to account for the high-achievement differential. Neither standard deviation, nor money, nor cultural priorities, can explain this towering discrepancy.”

Sounds like pure assumption and conjecture. We don’t know that they have lower conceptual ability until it is measured. We don’t know about their standard deviations until they are accurately measured. We don’t know all of the effects of culture, and they will be VERY hard to measure.

The whole conversation is rife with assumption followed by proof-by-assertion.

When faced with an extraordinarily high performance discrepancy at the topmost levels, and if explanations like per capita income, cultural priorities and health differentials are inadequate; the assumption of equality has a burden of proof on it by now. The conjecture that racial and ethnic differences can’t be genetic to any great degree, needs evidence. With the asian-white comparison; we have studies including japanese-americans, whose families have been here for around a century, showing the same pattern of high math scores, and verbal scores a little below the majority mean, taking into account some of the effects of better socio-economic backgrounds, not all of which would come from genetics. If there is any measurement error in Nobel prizes in science and medicine; then the non-white share is the same in the 1st half of the 20th century as in the second half: 4%+ or -2=4%+ or -2. Yet enormous changes in culture, life expectancy and incomes have occurred over this period, especially in Asia. [the Nobel figures are from Murray’03 p.284 Human Accomplishment]