Saturday, November 26, 2011

Culturally Biased IQ Tests

Someone commented on my Hebrew so I took an old grammar book (see scan) to the club to refresh it. Shockingly, there were words I could not understand! ולהלם - how do you read that? Then I realized that this old book had been written in the nineteen forties for immigrants from Germany and was not culturally neutral but strongly biased to implant in the minds of the "Yekkes" the doctrines of the socialist MAPAM party.

When I arrived in this country, there many Yekkes speaking grammatically refined Hebrew with comically German accent. Already there were rumors about the Yemenite Jewish children that had been secretly given in adoption to childless European families while their biological parents were told that they had died of disease. Talking to a Yekke acquantaince, she not only knew the facts but strongly defended what had happened. In those times, Israel wanted to be ruled by science and its first President was a scientist, Chaim Weizmann (Einstein had refused). All children were vacunated and scientifically treated (some irradiated - it was believed to be beneficial) and their IQ tested. The test had shown that Yemenite children were all retarded, with IQs well under the level considered (by European science of the time) "normal". The results amazed the political leadership of the country and it had consequences that is better to forget. My acquaintance was convinced that those poor children were "retarded" because of lack of modern education and a healthy socialist environment, so she justified their bringing up in good European families. By the way, exactly the same thing happened in Australia with aborigenes, and I believe it was a common practice all over the civilized world. Later, in the army, I met more than one Shapira and Goldstein who looked (and behaved) everything but Ashkenazim.

Why I am bringing this up? Because of this grammar book. Imagine a Yemenite child who only yesterday was playing in the dirt of some savage, isolated mountain village in Yemen, and today is subjected to a IQ questionnaire with Wilhelms and atomic submarines in the North Pole. Of course he is classified as shockingly retarded and sent to a special school.

However, I cant explain why many of them continue to fail also in modern culturally neutral IQ tests.

27 comments:

IHTG said...

Most likely the Shapiras and Goldsteins you met had Yemeni moms and Ashkenazi dads.

J said...

That's more likely today than in the past. Moreover the specific Shapira I am talking about told me that he was adopted. He had an auto repair shop in Sgulah.

Anonymous said...

The explanation is genetic. It has nothing to do with culture bias. If you use tests that are purely neutral the differences are just the same. Ashkenazi Jews are an unusually smart bunch of people - not many other races measure up. As you saw with your Army buddies, changing the environment, even at birth, does not affect inherited IQ. The IQ of adopted children is more correlated with their birth families than it is with their adoptive families.

K

Anonymous said...

Yemenites have substantial African racial mixture. Like Egyptians - 75% negro. There is the low IQ. That is why ancient Egyptians were smarter than modern Egyptians.

Anonymous said...

Perhaps the bible is right; blacks are cursed race by Yehova?

Anonymous said...

That's more likely today than in the past.

Don't Askenazim (and Russian Israelis) still prefer Ashkenazim, albeit less than past?

Anonymous said...

"However, I cant explain why many of them continue to fail also in modern culturally neutral IQ tests."

I think J was was employing irony.

Anon.

J said...

No irony: it is a real question.

But the question is not why Yemenite Jews, after two generations in Israel and after much improvement of the tests, they continue failing.

The question is why those old culturally-biased tests generated more or less the same results as current PC purified tests.

That I cant explain.

Maybe bias in the tests is unimportant? Maybe those fifties doctors were less prejudiced than we presume they were?

IHTG said...

Not scoring as high as an Eastern European Jew is a pretty harsh definition of "failure".

Anonymous said...

J, surely you cannot be serious.

It is well known amongst the intelligence specialists that IQ is substantially genetic and apart from avoiding gross physical insults like head injuries, birth trauma, central nervous system infections, lead poisoning, and severe malnutrition, that it cannot be changed by known social interventions, either up or down.

All you get from intensive interventional programs like "Headstart", etc, is a few extra IQ points in childhood that disappear by late adolescence.

There is a lot of research and interest in brain training programs, and some, like neurofeedback, and (in particular) working memory training, might actually work in a modest but measurable way. Drugs like Ritalin also enhance performance, and of course, who knows what other miracles might emanate from the pharmaceutical industry in the future.

But the blank-slaters and egalitarians are simply (and demonstrably) wrong. They are only able to push their message because of the draconian PC thought-control apparatus. Paradoxically, the more you get rid of cultural bias in the tests, the more you expose the native, genetically-driven physical differences. And it is not a pretty picture.

Anon.

Anonymous said...

Anon is correct - not only do the "culturally biased" tests produce the SAME results, but if you test purely on culture free items such as block arrangement, the differences are even greater. Scientists talk about "g" which is pure intelligence and cannot be measured directly, like blood sugar (at least not yet). All the different IQ tests try to measure "g" and some are better than others - the more "g loaded" the test, the greater the racial differences are. Some parts of common IQ tests ask "general knowledge" type questions - "How far is it from London to New York?" Since the answer to this can be taught, it is actually LESS "g loaded" than "culture free" questions.

Likewise, a few years ago in the US, there was an analogies section on the college admission SAT test. "Boy is to man as girl is to ____." The test writers made the mistake of writing an analogy involving the words yacht and regatta and this gave the blank slaters an opportunity to pounce - the reason blacks do 100 points (out of a possible range of 600 - a very significant difference) worse than whites is because folks in the ghetto have no personal familiarity with yachts (never mind that I don't either). As a result they ended up completely doing away with the analogies section. The only problem is that on the sections that are not "culturally biased" (unless you consider algebra to be culturally biased) such as math, blacks do even worse than they did on the analogies task.

K

Anonymous said...

Short of lying about the data, or not testing on anything "g-loaded", there is no solution for the blank slate egalitarians. This is why they will move to suppressing this data.

On forward memory (eg remembering a sequence of numbers in the same direction in which it was given, for instance)there isn't much difference between anybody; but if you have to reverse the sequence, the test becomes much more g-loaded and the differences come in out in proportion to IQ.

IQ can be roughly gauged in ways that are completely non-cultural: reaction times. The well-known disparities still occur. Jensen has written extensively on this, and in his own way, so has Volkmar Weiss ("clocking the brain").

There is nothing about the human body that isn't subject to optimization for different environments, and one can document all sorts of differences in different racial groups, because different races evolved in vastly different places and did not come into contact for tens of thousands of years, even hundreds of thousands of years, thus limiting gene flow. Consequently, blood groups vary, fast-twitch muscle groups vary, height varies, nasal passages vary, so does skin, sweat glands, even ear wax, to say nothing of adaptations to milk-based diets, etc.

To suppose the brain is immune from evolutionary pressure is sheer magical thinking.

Anon.

Anonymous said...

But...but ... but, Jefferson said that all men are created equal. Never mind that he did not mean equal in intelligence (Jefferson himself wrote about the differences between blacks and whites but these writings are considered out of character and held against him). Difference which favor blacks such as in athletic abilities are not only tolerated but encouraged - only in areas where non-blacks excel is it necessary to pretend that there are no differences.

K

Anonymous said...

"Yemenites have substantial African racial mixture. Like Egyptians - 75% negro. There is the low IQ. That is why ancient Egyptians were smarter than modern Egyptians."

Yemenites and Egyptians both have significant sub-Saharan admixture, but it is not even close to 75% for either population. It is still a minority component. For non-Jewish Yemenis, around 25% is average, and Yemeni Jews have less than that although substantially more than other subgroups of Jews.

J said...

The paradox is solved: Culturally biased tests contain concepts like "yacht" and "regatta" and "atomic submarines in the North Pole" that can be taught. Doing away with the sections that are "culturally biased" all that remains is math, and low IQ groups do even worse on those tasks.

Another paradox is that IQ of 70 in an European is a visibly retarded person who can barely manage in life. Yet an African of the same IQ appears vital and brilliant, and one would never guess that his mental capabilities are so limited.

J said...

To fix in our minds the concept that solves what I felt was a paradox: "Culturally biased IQ tests are friendlier to low IQ groups than high IQ groups; "pure" IQ tests increases the gap between them.

Those who reject IQ tests because they are culturally biased and prejudiced are either honestly mistaken or deliberately insincere.

The cultural bias argument looks reasonable, so I let say that its champions are honest but wrong. (Thanks K.)

Anonymous said...

Yes, you've got it now, but the non-biased tests are not all "math" - they can be things like arranging blocks in patterns and rotating diagrams in space. Then again these can be considered forms of geometry which a form of math. Or they can be memory speed/processing tasks like repeating number sequences backward.

K

Anonymous said...

Vide infra my comment on Zoomer, the athletic snail, for some thoughts on culture-free IQ testing methods.

Anon.

J said...

Anon,

Meteor, the holder of Israeli 20 cm long distance championship, is achieving very good times in a culture-free competition. But I cannot enter him in the World Competition because it is open only to British snails. Talk about biased tests and racism...

Anonymous said...

The funny thing is that in folk culture, IQ was always associated with speed - someone who is of deficient intelligence is said to be "slow".

If you imagine the brain to be some form of computer, then we know that computers with a faster "clock" are more capable of solving problems within a given time. Certain problems require solutions in "real time" (e.g. GPS navigation - it does no good if the computer tell you to turn 10 seconds after you have already passed the intersection) and so a slow clock is not just slower but completely useless for such tasks. If it were just "slowness" then a "slow" person could learn calculus - it would just take them a little longer, but it doesn't seem to operate this way - if you are listening to a lecture and can't process fast enough, the lecturer will have moved onto another point before you would have understood his 1st point and you won't understand his next point either - again you would have missed the proverbial intersection.

In the brain there must also be questions of short term memory retention in terms of time and quantity. Our long term memories are prodigious - we remember millions of facts and figures - entire languages, plots of books and movies, places where we have been, etc. etc. But our short term memory buffers are quite limited - a handful of numbers or names that are quickly replaced with the next items we need to process. Probably if you are too "slow" some of the necessary data set will fall out of the buffer before you have had a chance to fully process it and the problem at hand will then be completely unsolvable no matter how much time you are given.

This also explains one of J's other non-paradox paradoxes. African villagers with 70 IQs do not appear retarded because they have sufficient processing speed and power to solve the tasks needed for their daily living (this is one of the reasons they have 70 IQs to begin with - they were under no evolutionary pressure to do better). With an IQ of 70, you can plant maize, build a grass hut, etc. Only living in a complex urban technological society do the higher level skills become necessary. The Ashkenazim were lucky because they were, by accident of history , divorced from agrarian life sooner than most other groups and the traits that they developed "living by their wits" in the Middle Ages happened to line up with the needs of a post-Industrial Revolution technological society - the great great grandson of the Jew Moses the money lender/ talmudic scholar has the intellectual chops to be Morris the Nobel prize winning physicist. In modern times, one would have expected the American black population to have been placed under similar evolutionary pressure once their agricultural skills were no longer sufficient for survival, but the modern state with its welfare programs completely undermines any positive evolutionary pressure in that direction.

K

Anonymous said...

K is quite correct in that both processing speed (independent of working memory) as well as the capacity of working memory, together predict IQ very closely.

It seems that neurological speed cannot be altered by psychosocial interventions, but the capacity of working memory can be trained up, and this might genuinely allow an improvement in a variety of tasks in which "fluid intelligence", ie "g", is important. In other words, measured IQ might go up and in a manner that is apparently sustainable; if one believes the hype, that is.

See, for example: Lehrl, S. and B. Fischer (1990). A basic information psychological parameter (BIP) for the reconstruction of concepts of intelligence. European Journal of Personality 4, 259-286.

Anon.

Anonymous said...

Another point re: 70-IQ Africans. Among races with 100+ IQ medians, only severe medical problems can cause people to have IQ's in the 50-70 range - Down syndrome, Williams syndrome, and so forth. (Bottom 2%, obviously this will be disproportionately occupied by people with various problems.)

But among Africans with IQ medians of 70, low IQ occurs among medically normal people, simply because low medians relative to non-Africans are normal in that population.

Likewise, Jews with IQ's of 85 are likely to have other medical problems (bottom 2%,) while non-Jewish whites with IQ's of 85 are simply on the slow side of normal (bottom 1/6.)

So, a non-Jewish white with an IQ of 70 probably has various problems in addition to low IQ. These produce social and physical impairments (bottom 2%.) But an African black with an IQ of 70 is average and thus unlikely to suffer from a syndrome.

To make this less emotionally triggering, use a height analogy. A Dutchman who's 5'0" likely has medical problems, be they malnourishment, endocrine problems, dwarfism, etc., because the Dutch height median is around 6'2" for men under modern nutritional conditions. On the other hand, an African pygmy who's 5'0" is likely medically normal. He might grow a bit taller than his father if his diet is rigorously supplemented, but it's not normal for pygmies to grow tall, because their population has a low median height.

-s.n.

Anonymous said...

As regards discrimination against Meteor in the snail competition, J, all I can say is that the whole thing looks to me like a shell game.

Anon.

B said...

I love the idea of beating people on the head with a stick as an IQ test.

>"the great great grandson of the Jew Moses the money lender/ talmudic scholar has the intellectual chops to be Morris the Nobel prize winning physicist."

It remains to be seen whether the Ashkenazim are smart enough to keep their intelligence from making them too stupid to reproduce.

I think there's a lot more interesting stuff going on on the right hand side of the Bell curve than clock speed. Richard Feynman wrote about having synesthesia-the symbols in an equation would have their own colors, tastes, etc. to him. I have it too (though I'm no Feynman,) as do many smart people I've talked to. Weird things happen in that territory that make people more perceptive and engaged in abstractions, often at the expense of the real world. Judaism seems to be a good way to square that circle-I've never seen so many functional nerds as in Ashkenazi shul.

J said...

I think in machines operating in space.

Anonymous said...

Anon. sn at 1115 PM, re. the height of Dutchman. The younger generation of Dutch is indeed quite tall, but not 6'2". The average height of young Dutch men is between 6'-6'1", I believe.

Anonymous said...

Anon 1:21 AM,

I must have misremembered the article I read on it. Shame on me for not looking it up. Thanks for the correction.

-s.n.