The Inequality Taboo by Charles Murray

commentary ^ | September 2005 | Charles Murray

[garbanzo]

Yes, he does measure this, as relating IQ scores to, say, being an indicator of academic success; job success; and so on.

I'll take some issue with this in that are we to say that Shaq and Tom Cruise are more intelligent than most people given the level of success in their respective careers? If you think about it, Shaq and a number of NBA players probably would do fairly well on a number of spatial organization and perceptual tasks.

In other words, I think job success is a poor proxy for intelligence and Murray seems to admit that academic success can be coached or acheived with hard work and focus. So it's not clear to me what exactly the definition of intelligence is that an IQ test is measuring other than proxies that themselves are subject to some interpretation.

[Tribune7]

bump for later

[GSlob]

Thank you for the posting. Full article has a very interesting bibliography.

[Map Kernow]

Nothing in the article mentions any issue with the precision of a given test (e.g. is a difference of a single point between two individuals a significant difference or not) nor any real accepted standard for intellegence.

Nonsense, you're simply refusing to read the entire report, or cherry-picking for language that makes the science concerning intelligence and its testing more imprecise than it is. I doubt you understand the language you quoted. I suspect that's why you used the weasel wording "real accepted standard for intellegence [sic]"

This passage after the one you quoted explains Spearman's concept of "g" or general intelligence:

...all such correlations [between various tests of intelligence] are typically positive and form what is called a "positive manifold." Spearman (1927) showed that in any such manifold, some portion of the variance of scores on each test can be mathematically attributed to a "general factor." or g. Given this analysis, the overall pattern of correlations can be roughly described as produced by individual differences in g plus differences in the specific abilities sampled by particular tests. In addition, however, there are usually patterns of intercorrelation among groups of tests.

What follows in the linked Report concerns the predictive value of intelligence tests, such as how well IQ testing predict performance in school, etc., but I have always found Dr. Thomas Sowell's statement in his contemporaneous book "Race and Culture" on the predictive value of IQ tests to be best:

"Existing knowledge of the structure and nature of intelligence is far too preliminary to answer many of the questions raised about race and mental potential. The more immediate, practical use of mental tests as aids to prediction raises fewer and more limited questions. The relevant issue is not how accurate the predictions are on some absolute scale, or even relative to predictions in other fields, such as economics or weather forecasting. The question is whether alternative means of accomplishing the same tasks as mental tests have as good a record. The evasion of this crucial empirical question has been a consistent pattern among critics of mental testing."
---"Race and Culture," p. 184

[Publius6961]

Strangely enough, everyone that I know who has read the book thinks it's about race.

Well, you can't say that any more.

I read it after the first wave of criticism, and do not agree that it's about race.
I don't have it handy, but as I recall, it had roughly 12 pages on race out of 600 odd pages.

As I recall, it firmly and unequivocally argued that economic status was more important and indicative than race.

[Map Kernow]

Can I tell anybody who takes an IQ test how many what a score of 100 versus a score of 101 means? Is such a difference significant or not? How many points difference between two test scores is sigificant and how does that relate to Murray's basic argument?

Good Lord, you really don't understand IQ or statistics, do you? That, or the meaning you attach to the concept of "significance" is as imprecise as you allege IQ testing to be.

Go here for a description of the concept of "statistical significance." For IQ and IQ testing, you really need to read the literature in depth, such as Jensen's "Bias in Mental Testing," and "The g factor," although there are many others you could profit from, including the excellent bibliography in the Murray article you are attempting to obliquely disparage. You are simply ignorant about the field, and either you go inform yourself more, or get out of the discussion: people who are interested in the subject don't have the time, charity, or inclination to educate someone as obviously prejudiced on the issue as yourself.

[Lancey Howard]

My wife is brilliant. She was president of her school's National Honor Society and got a science degree, top of her class, etc. But she can't straighten a picture hanging on the wall.

[Publius6961]

It may help to look up the scientific uses of words like "accuracy" and "precision" first. Accuracy is the how close a measure comes towards an accepted standard. Precision involves the repeatability of a measurement.

Wrong.

Accuracy has nothing to do with "standards", other than agreement as to the units of measurement. Accuracy can only be "proven" by repeatability of measurement.
Precision is an absolute value, not subject to agreement, and can be either correct or wildly wrong.

Intelligence can't be measured with a micrometer, and I would never expect Einstein to test at 165 IQ one time and 47 another time. Its consistency within the limitations of the process are well established, and repeatable.

Precision, on the other hand can be exquisitely correct or a joke. I will leave the truth of that last statement up to the technically savvy readers.

[Bon mots]

Academic success does not necessarily lead to real-world success.

Remember the old adage:

"A" students work for "C" students.
"B" students work for the government.

[GOPJ]

The European Union’s immigration policy has, willy-nilly, decided that now you can move to Denmark and become Danish or move to France and become French. Is this true? Everyday experience suggests that Denmark’s culture works because it fits the characteristics of Danes, that France’s culture works because it fits the characteristics of the French, and that these ethnic characteristics are importantly different and deeply rooted, whether in genes or in habits of the heart. Replace a large proportion of French with Danes—let alone peoples more distant—and French culture will be profoundly changed. But it is taboo among the elites to talk about such things (although ordinary people sense what is at stake), and so a momentous social experiment is under way without any reason to think that its assumptions are correct, many historical reasons for thinking they are wrong, and recurring stories on the evening news suggesting that the social fabrics of Europe will be shredded before the elites can make themselves come to grips with what they have been doing.

One more reason the Department of Defense should expand to include "cultural democracy teams" to interpret the ideas of freedom within Middle Eastern cultural norms.

[garbanzo]

What follows in the linked Report concerns the predictive value of intelligence tests, such as how well IQ testing predict performance in school, etc.

That wasn't the question asked...the question asked was if an IQ test purports to be a measure of intelligence, how precisely does an IQ test measure intelligence and how accurate is any given test. I take measurements and analyze them for a living - part of my job is understanding whether or not a potential signal is an artifact or significant. These are the questions hard scientists ask from data sets. What is the reliability of the signal? Nothing in the link you posted actually gave any data on the question like what is the standard deviation for a given individual on a given test if taken multiple times? Do different tests give identical results for the same taker? If not, how big are the differences? While many different tests cross-correlate, what is the gold standard for an intelligence test? We have a gold standard so to speak for the definition of a meter - where's the equivalent standard for intelligence?

[garbanzo]

Intelligence can't be measured with a micrometer,

Good finally someone admits it that it's not a real measurement then.

and I would never expect Einstein to test at 165 IQ one time and 47 another time.

And your evidence for this is something other than your supposition? Like for example published records of Einstein's IQ test results?

Its consistency within the limitations of the process are well established, and repeatable.

So are the results for the Scientologist's e-meter readings.

[LS]

Again, you're trying to debate without reading it. This can only go so far. Its FAR more than "job success," but that is one component. It's academic success, it's related to crime, it's related to unemployment, it's related to illegitimate births. About EVERY indicator you can identify (and he identifies many, many) correlates with intelligence.

Please, let's not keep going unless you have specifics from his book to discuss. I read it years ago, and it's not my job to go dig up all his arguments.

[garbanzo]

About EVERY indicator you can identify (and he identifies many, many) correlates with intelligence.

Or do they correlate with IQ tests? To be honest, I'm really not that deeply interested enough to go buy and read the book. All I'm asking is for a couple of cites - and if you're defending his arguments you should at least be able to cite them properly. I'm not asking for a book report - just short relevant cites supporting your position. My point is that I'm not suggesting that IQ tests results have no predictive value - I'm just dubious that they measure intelligence.

[LS]

Well, if you don't want to check out a copy from the library, I can't help you. I don't have the book here at home--it's at school, and it's not my job to provide you with cites when they wrote a whole damn book full of evidence.

[neutrino]

That is what sets America apart, we grant rewards based on ability and not a scientific notion.

Umm...two words. Affirmative action.

It has been awhile since rewards were actually based on ability.

[A. Pole]

I'm a capable, self assured female. However, I have observed a lot of the differences in male and female capability in spatial types of math - and in figuring out which way is North .

My opinion that the main biological difference between men and women is related to the child rearing focus of the later.

Men an women have different personality structure - while women tend to be more detail oriented, cautious and perceptive, men are more audacious, aggressive and innovative.Women can be great scientists when the particular subject of research requires more feminine traits like it was in the case of Maria Sklodowska (married Pierre Curie) and Rosalind Franklin (the true discoverer of DNA

In other words - it is all in testosterone :) ).

I also suspect that the normal curve for the IQ might be more dispersed for men (more extremely bright and more retarded). It would give advantage for the top science achievements.

[Stentor]

Since then, I have deliberately not published anything about group differences in IQ, mostly to give the real topic of The Bell Curve—the role of intelligence in reshaping America’s class structure—a chance to surface.

The point is still being missed on this thread. It helps to have read the book.

[garbanzo]

My reading list is long enough as it is and it's not something I have strong axe about enough to bump it up to the top. Again if you're going to argue half-remembered stuff it's not my fault either. In any event, my argument isn't contra Murray, but rather on the basic metrology of intelligence. All I've asked here is whether or not intelligence tests measure intelligence in the same way that metersticks measure distance, clocks measure time, and balances measure weight. All I've been presented with is obfuscation and insults.

The closest evidence I've seen is pair-correlations between different people. I've read articles stating that there exists no common accepted definition of what neurological basis of intelligence is, only fairly vaguely defined concepts like "g" and debate over whether or not "g" is the most meaningful definition of intelligence.

It's pretty simple - can I measure intelligence like I measure weight or distance? If so, how accurate is a given test and how precise is it in terms being able to resolve differences in intelligence. It's not a hard question and data should be readily available like it is with any high performance scientific equipment.

[mmercier]

>> That is what sets America apart, we grant rewards based on ability and not a scientific notion.

>>> Umm...two words. Affirmative action.

Good point, we also grant some rewards based on race. The larger point (now that you popped my fondly held notion) is that America has no structure that prevents those of talent and ability from obtaining (on their own) the rewards of their hard work and talent. The only structure I can think of that holds talented people back in this country is the rat party; those who cripple themselves mentally by believing in them.