Jump to content

Talk:Race and intelligence/Archive 5

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 3Archive 4Archive 5Archive 6Archive 7Archive 10

Introduction rewrite

In response to numerous complaints I've seen on this talk page, I've rewritten the introduction and "Overview" section. The old intro was non-NPOV, did not concisely summarise the contents of the article, and mostly comprised a debate on the validity of race that's already handled (in 56 kilobytes of agony) at the article Race and needn't be discussed here. -- Schaefer 15:39, 18 Dec 2004 (UTC)

I'm removing the "Debates" section as well, as it introduces content already in the introduction, and most of it is handled in detail later in the article (the role of genetics, whether tests are culturally biased, etc.). And the rest is just non-NPOV ("[...] most race-based claims about intelligence were not derived from the results of value-free scientific testing, but rather from testing biased by racial prejudice [...]"). -- Schaefer 16:05, 18 Dec 2004 (UTC)

Revisions to Intro and 'IQ Gap'

The first sentence has serious problems:

Many researchers have claimed significant correlations between race and intelligence to exist.

First, race is not numeric and thus cannot be correlated with anything. Second, the point is not what the researchers' claims are, but what their studies have shown; neutrally speaking, regardless of their claims, the test results show differences. Third, "to exist" is redundant. I am changing this to:

Many studies of cognitive ability have shown that racial groups differ in measured intelligence.

I'm still not thrilled by this (one could probably remove "of cognitive ability," though I think the precision is helpful), but it fixes the above-cited problems.

The second sentence,

Such claims have sparked public debates concerning not only the truth of the claims themselves, but the validity and fairness of intelligence tests in general and extent to which intelligence is determined by genetics

also needs work. The "truth of the claims" is a red herring, and the revision to the first sentence obviates this phrase. Also, the broad debate is not just over genetics, but over social and biological determinants of intelligence. Accordingly, I have changed this sentence to:

These results have sparked public debates concerning not only the reliability of the studies and the motives of their authors, but the validity and fairness of intelligence tests in general and extent to which measured intelligence is determined by biological and social factors.

I could live with removal of "and the motives of their authors," though in my experience this is a major part of the debate for some people.

Which brings me to the 'IQ gap between races' section, which opens with a statement that Jensen was the first to claim that the black-white IQ gap was genetically determined in part. By this point in the article, 'IQ gap' has not been introduced, and no statement about black/white differences has been made, making the Jensen paragraph look like a non sequitur. I rearranged the section to first introduce the gap, and moved Jensen toward the bottom where explanations are entertained.

I don't know J.R. Baker's work, but this:

John Randal Baker presented a lengthy argument in favor of innate racial differences

seems implausible. Did he really argue in favor of racial differences? Or did he just argue that the differences were innate? I suspect the latter, and hope a knowledgable soul can remove the ambiguity. --DAD 20:32, 18 Dec 2004 (UTC)

Table of IQs

I am not happy with the table of IQ scores recently posted. No references are provided in the References section. The table is incomplete. Given the inflammatory nature of the topic, I'd be much more comfortable if only peer-reviewed work or readily verifiable work (e.g. the APA report) were cited (e.g. not IQ and the Wealth of Nations, and if (consistent with the introductory paragraph in the section) all studies were from the latter half of the 20th century.

Some studies are performed on narrow parts of the population (e.g. South African schoolchildren, by Lynn and Holmshaw), and the table feels like over-summarization.

Other people have thoughts on this? Not sure why I feel so squirmy about this table. -- DAD 21:15, 19 Dec 2004 (UTC)

The table seems a good one, it cites sources. If you'd like to cite sources criticzing these sources, that would be fine too. [[User:Sam Spade|Sam Spade Wants you to vote!]] 22:15, 19 Dec 2004 (UTC)
Whenever possible, it would be best to give full citations in the Reference section. Are there any review papers that have a table like this that we could get references from? --Rikurzhen 22:29, Dec 19, 2004 (UTC)
Result of IQ testing
Group Avg. IQ Spatial Verbal
Ashkenazi Jews 110¹-118¹ 98²-105³ 117²-125³
East Asian 105-106 110-113 ?
Caucasian 100
World est. 90⁴
African American 85
African 60-80⁴
1. Bachman (1970), Vincent (1966), Brill (1936), et.al
2. Levinson (1960)
3. Levinson (1958), on yeshiva students
4. IQ and the Wealth of Nations
What needs cited better? [[User:Sam Spade|Sam]] Spade wishes you a merry Christmas! 22:46, 19 Dec 2004 (UTC)
Preferable but not absolutely necessary: article title, journal title and volume, and page numbers -- in the "Reference" section of the article. Also, because those may (but I have no real idea) be values that are out of line with other studies, it would be even better if we had a review paper that examined lots of studies and gave a composite number. What would be really cool would be to have "n" values for those studies. --Rikurzhen 23:14, Dec 19, 2004 (UTC)
For example, Richard Lynn has a 2004 paper in Personality and individual differences with this result:
The results are shown in Table 1. Reading from left to right, the columns show the numbers in the four groups, the mean vocabulary scores, standard deviations and conventional IQs based on a gentile white mean of 100 and standard deviation of 15. Thus, expressed in this way, the Jewish group obtains a mean IQ of 107.5, significantly higher than the gentile whites (t=5.82); the blacks obtain a mean IQ of 89.7, significantly lower than that of gentile whites (t=17.89); the "others" obtain a mean IQ of 98.6, not significantly different from that of gentile whites.
So any particular study might be a bad source, and we'd be best citing averages of studies. --Rikurzhen 23:26, Dec 19, 2004 (UTC)
That would be original research, and of highly dubious merit regardless. If you have something against a particular study, cite one which contridicts it, or an expert who shares your opinion. [[User:Sam Spade|Sam]] Spade wishes you a merry Christmas! 23:32, 19 Dec 2004 (UTC)
No, I mean we should find someone who has published an average of studies or a range from studies and cite that number instead. --Rikurzhen 23:56, Dec 19, 2004 (UTC)
Oh, well that would be fine, minus the instead. Citing it along w the other data is all good, but when different studies come up w different answers the honest thing to do is cite them all, and provide expert explanations and opinions of the differences. [[User:Sam Spade|Sam]] Spade wishes you a merry Christmas! 00:07, 20 Dec 2004 (UTC)

Sam, I think I'm not explaining myself well. The table does not seem to use a consistent method to present scores. Some are given as ranges, some as single points. Some are cited, some are not. --Rikurzhen 02:07, Dec 20, 2004 (UTC)

OK. Then the uncited info should go, eh? Unless we can cite it. But cited info should likely stay. Perhaps the chart could be cleaned up? I happen to like it, it presents the info pretty clearly (even if said info is inconsistant and/or uncited, as you say). In summary, cited info should stay, uncited info should go (or be cited) and the chart should be consistant (alth.o I won't be cleaning it up, I'm bad w these HTML charts). [[User:Sam Spade|Sam]] Spade wishes you a merry Christmas! 12:23, 20 Dec 2004 (UTC)

Sam, maybe I am mistaken but I think you are still missing Rikurzhen's point. You are responding to a perceived concern about citations, and you are certainly right -- citations are important and verifiable work in general can belong in articles. But I think Rikurzhen is making a separate point: if you are going to provide comparative data, you have to be comparing comparable things. At first glance the table above seems -- seems -- to do this: it is comparing IQs of Blacks with IQs of whites. But this is only a superficial similarity. How were the IQs (of individuals) measured? How were the IQ results sampled? How were these aggregate or average figures calculated? Unless the answers to these three questions are the same for the White and African and African-American figures, we cannot compare them. You do indeed provide a citation for the African IQ. But how was it calculated? Until we know that, we have no way of knowing whether this figure can be compared to a figure from another study. Rikurzhen, I don't mean to put words in your mouth, and if I am wrong I apologize, Slrubenstein 18:58, 20 Dec 2004 (UTC)

I hadn't thought it through all the way, but that's the logical extension of my point. Perhaps we should maybe make the table have columns for each prominent study and rows for each group. The Lynn (2004) data I pasted above for example covers three of the groups. That minimizes the amount of editorial manipulation. AndyCapp's ideas below also sound good. --Rikurzhen 20:16, Dec 20, 2004 (UTC)

I think one thing we could do is cite a range, from the lowest figure any study gives, to the highest figure any study gives, and then cite the studies. If somebody finds a study with a figure outside this range, or a new study is conducted that gives a figure outside the range, someone can update the range and cite the new study. Another thing we could do is cite some of the most prominent studies or the range between them. Or we could simply use another author's data and then cite that author as the reference. AndyCapp 20:02, 20 Dec 2004 (UTC)

Is it worth distinguishing studies from surveys of studies? For example, some of the work that Lynn has been criticized for are not original studies, but surveys of studies condicted by others. Critics say that he misread the underlying research. -Willmcw 02:12, 21 Dec 2004 (UTC)

I notice that some entries in the IQ Table have no citations. Where did they come from? If they are not cited they should be removed. -Willmcw 21:30, 30 Dec 2004 (UTC)

I cited them to the Lynn article in Mankind Quarterly the table links to. We should probably get full citations for these sources in the references section eventually. -- Schaefer 22:30, 30 Dec 2004 (UTC)
S., thanks for adding the citations to the uncited numbers. Perhaps the next time someone edits this table a large footnote can be created containing the references. It may need to be a subsection of the article, as the studies and surveys themselves should be annotatated. -Willmcw 00:39, 31 Dec 2004 (UTC)

Can anyone tell me what component of the table is drawn from this reference?

Brill, M. (1936). Studies of Jewish and non-Jewish intelligence. Journal of Educational Psychology, 22, 331-352

We should not use a citation from 1936 for an article about plate tectonics or for intelligence and race either. Science does move forward. A reference that old about a field like this is prima facie obsolete, in my opinion. It also makes me question the immediate source of this info. Did an editor here read Brill? Or is all of this this copied from a book? If so, what book? -Willmcw 05:55, 2 Jan 2005 (UTC)

I've commented-out the table until we get usable references for the material in it. Levinson 1960 is not a usable reference. -Willmcw 06:07, 5 Jan 2005 (UTC)

Just because I'm looking at this page...I have a vague intent to do something about this table. Working up to it, I suppose. When the table reappears, I suggest (if I'm not the one who ends up restoring it) that it report IQ only, not sub-abilities. There are tens of second-order factors; selecting just a few steals clarity and carries the scent of selection bias. --DAD 06:36, 19 Jan 2005 (UTC)

IQ determines income?

"Third, many scientists believe that IQ determines income, and not the other way around."

Is there a reference to this? It seems to completely fly in the face of, well, all known reality. I can't even count the number of time I've met people with IQ's over 135 (sometimes even as high as 170) living in poverty and squalor, while some very financially well-off folks have trouble following a simple multi-linear syllogism (and in some cases, even the most basic uni-linear syllogism).

I might believe there was an inverse relationship between income and IQ, but not a direct one. Put simply, a lot of successful folks are laughably dim-witted.

Secondly, in my Psych 101 class, I learned than an individual can get a score up to 20 points lower if he/she is suffering from malnutrition.

Also, I think it's a little biased to say there's no evidence of bias in IQ tests or that they remove the potential for bias. That's ludicrous. What if someone doesn't know what a particular word means in a particular question? He/she might be an extremely bright and clever individual, and still end up getting a low score because of factors like this.

Now, I really have no reason to complain here, as my scores are in the low-end genius range, but I've seen a lot of very bright people get low scores (and do very poorly in school, and very poorly in the workplace), and I've seen complete morons get surprisingly high scores (and get straight A's in school, and be very successful financially). Granted, this is the exception to the rule, but it still happens far too often to put any sort of genuine confidence in IQ scores, or academic performance, or financial success.

I think there's really only one good reason for the discrepency in IQ scores among races and income-levels: the folks who design the test are wealthy whites who did well in school. A poor minority who did poorly in school is at an unfair disadvantage when taking one of those things -- schools are designed to keep underpriviledged children (especially minorities) from performing well academically, IQ tests are designed to predict academic performance, the workplace favors those who did well in school. It's a 3-pronged system of keeping ol' Whitey Morebucks at the political/academic/financial helm.

And this page has an obvious white, middle-class, honor-roll student POV --Corvun 09:42, 7 Jan 2005 (UTC)

Your confusion stems from the difference between statistical and annecdotal evidence. The existence of counter examples is expected for a statistical phenomena where correlations are less than 1. In this case, the correlation between IQ and income is in the "moderate" to "high" range, but not 100%. One excellent reference for the income-IQ causation question is a 1998 paper by Charles Murray -- I'll find the full reference. Second, the question of "bias" is very certainly resolved. Bias has a technical meaning, which has been extensively tested. In essense, IQ tests predict outcomes equally well for whites and blacks -- whereas, if there were bias against blacks one would expect IQ would under-predict outcomes for blacks. --Rikurzhen 18:30, Jan 7, 2005 (UTC)

The Murray reference is Income Inequality and IQ (PDF link). The essence of Murray's analysis is that IQ differences between siblings predict their relative earnings, educational attainment, and occupational attainment. Note this analysis has nothing to do with race, and explicitly controls for differences in upbringing (e.g. poor vs. rich family).

These results directly contradict your (Corvun's) assertion that income discrepancies are due to privilege differences. Moreover, they support the (already well-studied) notion that IQ determines outcomes, not the other way around. Causality can only run one direction for a simpler reason: by the time income can be measured (early 20's, typically), IQ has been stable for years. Income can't reach into the past and change IQ.

Rikurzhen's statements about test bias are consistent with the literature and with, for example, the findings of the American Psychological Association's task force report, "Intelligence: Knowns and Unknowns". If that page is too verbose, consider the press-release summary: "The differential between the mean intelligence test scores of Blacks and Whites does not result from any obvious biases in test construction and administration, nor does it simply reflect differences in socio-economic status." --DAD 19:11, 8 Jan 2005 (UTC)

I have updated the "Unfair/biased test" section to directly quote sources. Certainly one could have complained that the previous version included lots of assertions with few citations. Should be fixed now. --DAD 22:02, 8 Jan 2005 (UTC)

Citation request in "Caste-like minorities" section

Can someone here please cite the source of this passage from the article: "Another example cited is the Ashkenazi Jews that came to America from Russia had an extremely low IQ as manifested in tests that were done in the US army by Carl Brigham, the creator of SAT, but later reached their present extremely high average IQ-level. Similarly, Irish, Italian and Polish immgrants in the USA are reported to have all scored about 80 pts in the beginning of the 19th century, but now tend to reach 100." You could imagine how important this information would be in challenging the idea of IQ as something purely determined by genetics. Perhaps you could email me. DeanoNightRider tdevine(nospam)@(nospam)gmail.com

These early IQ test analyses are pretty much worthless. The tests Brigham was analysing didn't even provide data on which persons were Jewish. He was making silly generalizations from the scores of a population of WWI draftees, simply assuming that half of his Russian-born sample was Jewish and that the Jewish subset scores were equal to the Whites. The other study on American immigrants is almost certainly that of Henry Goddard, who used the test scores of immigrants who were pre-selected as being obviously unintelligent and tried to predict from those scores the average intelligence of the nations from which they emigrated. These conclusions only show the anti-Semitism of the analysts, and do not evidence any miraculous ascension of the Ashkenazic IQ from idiocy to brilliance during the twentieth century. Steven Jay Gould gives more detail exposing their errors and those of other early IQ testers in The Mismeasure of Man (despite how utterly wrong Gould happens to be concerning modern IQ testing).
Wikipedia shouldn't be using such outdated and obviously flawed studies. I'll be removing them if there are no objections here. -- Schaefer 22:03, 12 Jan 2005 (UTC)
No objections here. Goddard's work is indeed a plague on clarity. You could remove them, but consider rewriting and supplementing them with material similar to what you've written above, so that others can learn. --DAD 06:26, 19 Jan 2005 (UTC)

I removed the references to Goddard and Brigham. I also moved the whole "Caste-like minorities" section out of its own heading and into a paragraph under "Other explanations". At best, the "some people" citing research in favor of this are just Thomas Sowell, and if it's to stay it should be reworded as his opinion. At worst, it's original research. Note that the paragraph provides links to the research that "some people" supposedly cite, not the papers doing the citing. Either way, the theory's not terribly notable. -- Schaefer 10:34, 19 Jan 2005 (UTC)


NPOV?

This article is seriously imbalanced, purely describing the points of view of supporters of racially-based IQ differences, and selective has chosen scientists and theoreticians that support this point of view. Deserving of serious entire rewrite. Hauser 21:50 23 Jan 2005 ( NZEST )

Hopefully not by you, from the sounds of it. Those who oppose these ideas rarely conduct studies on them. Criticism is fine, but educated, expert criticism of this subject generaly comes from the advocates themselves, not the talking heads which disapprove of them. Sam Spade (talk · contribs) 12:42, 23 Jan 2005 (UTC)

No opponent to those ideas will conduct study on it because there's no need to answer to a bad question. All theses studies are based on weak and fuzzy concepts, they're biased from the beginnings. To conduct such a study you should admit :

  • 1 - That human races exists.
  • 2 - That intelligence can be measured.

In fact :

  • 1 - Human races don't exist from a scientific POV.
  • 2 - Intelligence as a objective concept can not be defined and thus can not be measured.

All theses studies will only show that there are different human cultures, a fact that nobody will deny.

Ericd 13:34, 23 Jan 2005 (UTC)

Actually, race is understood to involve language group, and blood type by present day experts. Intelligence is currently being evaluated in the means by which it can be measured. The current concensus involves types of intellegence, of which the historic IQ was one. Your "facts" are just as opinion-driven as anything else, and it astounds me that you would attempt to pass them off as anything more. Persons unwilling to conduct research or accept for the sake of argument postulated concepts have no place here, because they arn't experts of any related concepts. Sam Spade (talk · contribs) 13:51, 23 Jan 2005 (UTC)

Intelligence is currently being evaluated in the means by which it can be measured. The measure defines the concept... What did you wrote thereafter about postulated concepts ? If I define the measure of (male) intelligence by the measure of penis the intelligence is the length of penis ? Ericd 14:00, 23 Jan 2005 (UTC)

If you conduct some verifiable research correlating penis length and intellect, I would be amused to hear it, but so far the facts seem to show a negative correlation between the two (read The bell curve). In any case, silly ideas you come up with are original research, not expert witness, and are irrelevant here. Sam Spade (talk · contribs) 14:11, 23 Jan 2005 (UTC)

This is the first time I (and hopefully the last) I would refer to my PhD in social science on Wikipedia. This is not silly idea nor original research. Social scientists have adopted a pragmatic view that the test measures what it actually measures leaving the work to make a relationship between what is measured and a generally accepted concept to the reader... If you find that length of penis isn't a valid measure of intelligence that's your problem.

If I find some relationship between length of penis and and something else you are free to understand it as you want !

However the title of the article makes some difference :

  • Race and intelligence.
  • Race and intelligence measured by the length of penis test.
  • Relationship between Ethnic groups and intelligence measured by the length of penis test.

You can add your own

Ericd 14:36, 23 Jan 2005 (UTC)

Anyone writing about race and intelligence should certainly begin by a) defining what they mean by "race" and b) stating how they think intelligence can be measured. This is the same for any other concepts scientists might want to study or measure. However, unless I'm misreading him, Ericd seems to be implying that concepts, therefore, do not exist from a scientific POV. - Nat Krause 14:47, 23 Jan 2005 (UTC)

From a scientific POV, some concepts could exists as defined from a scientific POV. Ericd 15:01, 23 Jan 2005 (UTC)

This statement: "Those who oppose these ideas rarely conduct studies on them" is simply not true. The vast body of work on test bias and racial differences has been carried out to a significant extent by people who came to the field convinced that the key findings (significant racial differences, no evidence of test bias) were false. Nathan Brody (primary researcher and author of the textbook "Intelligence") has said that he was motivated to "prove Jensen wrong." His textbook contains a chapter on black-white differences consistent with the present Wikipedia entry. Linda Gottfredson was a confirmed egalitarian, but the data changed her mind.
Few primary researchers are concerned with the "definition of intelligence." Similarly, few biologists concern themselves with the "definition of life." Definitions aren't science, in general, but laypeople love to endlessly debate them (see above). In The Bell Curve, Herrnstein and Murray largely avoid the term "intelligence" specifically because it's so loaded. Cognitive ability is favored by most researchers I've read. Definitions and labels aside, the important point is that "IQ is strongly related, probably more so than any other single measurable human trait, to many important educational, occupational, economic and social outcomes" (Mainstream Science on Intelligence, WSJ; see entry). The same claim, substituting "genital size" for "IQ," has no empirical support. In short, one can't just make up a new measure of intelligence, because it is predictive validity that confers value on a measure.
I'm so amused by the claim that "human races don't exist from a scientific POV." Let's be very clear:
  • Human ethnic groups, fuzzy as they may be, show heritable phenotypic differences
  • Cognitive ability, as measured by IQ tests and their ilk, is a heritable phenotypic difference
  • Human ethic groups, fuzzy as they may be, show differences in mean cognitive ability, as measured by IQ tests and their ilk
I can't find anyone who disagrees with those statements -- they are well-established facts. That's all the foundation one needs to have an interesting discussion about race and intelligence. In my view, the present article properly accounts for the nuanced definitions of race and the definition of intelligence by referring to Race and Intelligence (trait). It also gives voice to contrary points of view throughout. One may argue that the evidence is not NPOV, but the same accusation can be leveled against evolutionary biology by those who deny evolution. The POV must reflect the facts; it's our job as community editors to make that happen. --DAD 19:07, 23 Jan 2005 (UTC)
Definitions aren't science. I disagree they are essential in social sciences. Ericd 23:13, 23 Jan 2005 (UTC)
I'm going to infer your meaning, as you wrote something that has exactly the opposite sense to what I believe you intended. That definitions are essential does not grant them special status re: science. Of course they're essential -- and not just in the social sciences. However, definitions are rarely hypotheses, or logical arguments, or evidence. Stating that intelligence should be defined in some substantively different way than "a general mental capability that involves the ability to reason, plan, solve problems, think abstractly, comprehend ideas and language, and learn" (from Intelligence (trait)) does not change the research on cognitive ability, only its name. --DAD 23:48, 23 Jan 2005 (UTC)

"IQ is strongly related, probably more so than any other single measurable human trait, to many important educational, occupational, economic and social outcomes" no reference to race ? Ericd 19:34, 23 Jan 2005 (UTC)

Read the statement for yourself here. It does discuss race and is consistent with the Wikipedia page. --DAD 22:36, 23 Jan 2005 (UTC)
I don't see any definition of race in this page ?
Ericd 23:02, 23 Jan 2005 (UTC)
This article is a subset of the large set of articles on race. We do not need to consider the definition of race on this page; it is treated elsewhere. --Rikurzhen 23:15, Jan 23, 2005 (UTC)
And if you were referring to the WSJ article, no definition of race is given. Race is generally self-reported, and the article discusses degree of admixture. Also keep in mind that race, at least at the level of white/black/Asian/Hispanic, is not a confusing concept to most people. When you say, "Blacks experience pervasive discrimination in the U.S.," everyone knows what you mean. I'm always surprised at the speed with which the concept becomes "undefined" or controversial once IQ is added to the mix. --DAD 23:56, 23 Jan 2005 (UTC)
No for any social science study the definition of the concept is more important than it's name. For instance the concept of a "tall person" has some sense to me like the concept of a "small person". But if I begin to split a human population between "small peoples" and "tall people", I have to operationalize the concept by giving some unambigous definition.
Please also notice that nobody outside the USA will consider that there's an hispanic race.
Ericd 00:57, 24 Jan 2005 (UTC)

"Cognitive ability, as measured by IQ tests and their ilk, is a heritable phenotypic difference". The question of "nature versus nurture" has finally found some scientific response ? Ericd 20:44, 23 Jan 2005 (UTC)

No, Eric. Stating that something is heritable does not mean a choice has been made between nature and nurture. Height is heritable, but is influenced by childhood nutrition. Heritability is measured as a squared correlation between genotype and phenotype (0 to 1), and for IQ heritability ranges from 0.4 to 0.8, increasing with age. This is considered highly heritable. --DAD 22:36, 23 Jan 2005 (UTC)

If the question of race and intelligence can be given some scientific answer I think the next step in progress of social sciences will be to answer to the question "Are blondes really stupid ?" Ericd 20:50, 23 Jan 2005 (UTC)

Cranial vault size ?

This one is great. It's a proven fact that size increase in developped countries. Thus cranial vault size increase. What else ? Did the US brain have more neuronal cells inside ? My personal advice is that a small brain in such a wide space will just get lazy :-) Ericd 16:25, 23 Jan 2005 (UTC)

B/W test score gap

"There is some evidence that the US black-white gap in reading and math scores is starting to converge"

This is my edit of a sentence added recently that originally read, "...the US black-white gap is starting to converge," and cited a recent paper by Fryer and Levitt. The implication, because this sentence appears in the "IQ gap" section, is that the convergence is in IQ, but in fact the paper only looks at reading and math scores, which are likely a correlate of IQ but not a substitute. Hence the edit. With the edit, though, this sentence looks out of place, since it does not actually address the IQ gap. I'd like to remove it or move it. Thoughts? --DAD 19:47, 23 Jan 2005 (UTC)

The easiest thing to do is follow that up with the counter-point. I think there have been a few recent studies arguing that the US IQ gap is not converging. --Rikurzhen 20:25, Jan 23, 2005 (UTC)
Counter-point is a better idea. I am not aware of the convergence studies. Got a ref? --DAD 20:32, 23 Jan 2005 (UTC)
here [1] --Rikurzhen 21:33, Jan 23, 2005 (UTC)
The sentence above it could also use a counter point. We need to mention that demographics may affect Africa and the UK -- e.g. many bright Africans leave Africa for the UK and US. Admixture will also complicate the interpretation of this data. --Rikurzhen 20:25, Jan 23, 2005 (UTC)
Agreed, though the migration hypothesis sounds like rank speculation without some data.
Sorry, I meant those more as un-controlled recognized confounding factors. For example, do a Google News search for "brain drain". It's mostly a political discussion; I don't know of any good scholarly sources. --Rikurzhen 21:42, Jan 23, 2005 (UTC)
If we can include brain volume and its relatively weak correlation to IQ, then we can also include math and reading scores. I changed the statment.Ultramarine 20:47, 23 Jan 2005 (UTC)
Yes. I edited to show what the actual correlations are, roughly. Also added a caveat about using properly validated IQ tests to measure the gap. Correlates are just not good enough. I wish the cited paper had included any sort of validation, but they were completely unconcerned with IQ, and interested only in scores and grades. This opens the interpretation that IQs are converging to a host of fundamental criticisms. For example, the reading and math tests may well have been constructed to reduce ethnic disparity -- the B-W differences they report are significantly smaller (about half a SD) than those studies that measure IQ directly. Just such an effect is seen in male-female comparisons, because many tests are constructed specifically to eliminate sex differences (e.g., an important property for the SAT's wide acceptance, but a hidden and fatal trap for researchers using SAT scores to examine gender differences in ability). I would strongly prefer to replace (or supplement) the citing of the Fryer/Levitt study with one that actually measures the IQ gap.
I'm not a big fan of the brain volume stuff, except inasmuch as it shows that these are real biological differences, not purely social constructs. The predictive validity of brain volume for actual outcomes (grades, job performance) is essentially nil. --DAD 21:24, 23 Jan 2005 (UTC)
I think the text is accurate now. The study indicate a strong possibility that the IQ gap is starting to diminish. This can only be refuted by a more recent IQ study. And I do no think there is one.Ultramarine 23:09, 23 Jan 2005 (UTC)
I would phrase this differently. The study can be used to argue that the IQ gap has diminished (the study itself says nothing about IQ). This argument can be refuted in any number of ways, including a demonstration that the tests or grades have declined in g-loading, a contemporaneous study actually measuring the IQ gap rather than reading and math, a demonstration that the correlation between these scores/grades and IQ is weaker than we infer, a demonstration that such deviations in the score gap are consistent with the deviations across studies that occur despite a stable mean difference, and so on.
That this study finds that the score gap goes to zero after controlling for a small number of social and economic factors directly contradicts the preponderance of research on the IQ gap that show it persists (widens!) at high socioeconomic levels and can't be eliminated by controlling for social or economic factors. To me, the obvious inference is not that this new study is much better than the other ones, or that it's picking up a new trend, but that it's not a study on the IQ gap and its findings are consistent with effects driven by non-cognitive factors. These factors are all the study actually entertains as a possibility anyway (and why is that?).
Again, I'd prefer an older study that actually measured the IQ gap over this new one that doesn't. --DAD 01:14, 24 Jan 2005 (UTC)
But there is no such refutation right now. Until then, the study is the most recent evidence available.
Actually, I would argue that this development should be expected from the Flynn effect. It seems likely that the Flynn effect started first for the most affluent (white) part of the population, and that it will also finish first for that part. Then there will be a period when the black catch up. We may have reached that period now. And the Flynn effect is not some non-cognitive artifact, evidenced indirectly by the great increase in cranial volume and more directly by several recent studies that show large increases in g from the Flynn effect. Ultramarine 01:30, 24 Jan 2005 (UTC)

We should probably add a sub-heading or two to the "IQ gap among races" to split up the text. --Rikurzhen 22:12, Jan 23, 2005 (UTC)

Désolé c'est en Français

http://www.er.uqam.ca/nobel/r30574/PSY1282/C2P1.html

Rought translation of a short part :

The name given to identify "a built" is a summary of the definition. For instance : The term "leadership" to speak about "the capacity of a member of a group to influence the other members as for the installation and realization of the goals of the group". This name is a summary, the name does not cover the extent of the definition.

Ericd 23:32, 23 Jan 2005 (UTC)

Some references

http://danny.oz.au/communities/anthro-l/debates/race-iq/

http://skepdic.com/iqrace.html

http://www.fsmitha.com/com/p-bell.htm

http://maxpages.com/raindrops/Psychology+tests

http://www.plp.org/pamphlets/riwc.html

http://www.hartford-hwp.com/archives/45/022.html

http://www.ferris.edu/isar/Institut/pioneer/kaukas.htm

spearman's hypothesis study

re: "More recent studies have questioned the statistical method used [2]"

a quick glance at the discussion section of this paper seems to indicate that they limit their conclusions to these particular tests because they were biased, but say that US tests are not biased and spearman's hypothesis holds for them --Rikurzhen 19:47, Jan 25, 2005 (UTC)

Read introduction. Ultramarine 20:26, 25 Jan 2005 (UTC)

Most importantly, this study doesn't concern whether group differences in g are genetic, so at the very least it shouldn't be in this section. The sentence says that the study is questioning his conclusion that group differences in intelligence are largely genetic, which it does not.

All they're stating is that the method Jensen uses to test the Spearman hypothesis on Black/White group differences in the U.S. cannot be used on two particular studies from South Africa and Holland, since they contain elements of subtest bias that make it impossible to assume factorial invariance, which is necessary for Jensen's method of correlated vectors. This does nothing to invalidate to Jensen's conclusion that the Black/White IQ gap in the United States is primarily a result of differences in the primary factor g, and the authors of the paper specifically mention this. What they are objecting to are the analyses of others that have used Jensen's correlated vector method to prove the applicability of the Spearman hypothesis to other data sets while assuming factorial invariance where it may not exist. -- Schaefer 21:31, 25 Jan 2005 (UTC)

From the introduction. "However, because various competing models fitted the data approximately equally well, it remained difficult to establish the central role of g in B-W differences with any confidence."
Ah. Why not reference one of Dolan's earlier papers that are more on-topic? They are all on his web site. --Rikurzhen 01:31, Jan 26, 2005 (UTC)
Good idea. Actually, how about linking the site? I will also move the sentance to a more appropriate paragraph.Ultramarine 02:49, 26 Jan 2005 (UTC)

"flynn effect"

if we're going to talk about the flynn effect -- *cringe* -- it may need it's own section so it can be done properly. --Rikurzhen 19:57, Jan 25, 2005 (UTC)

More on this and whether the Flynn effect increases g. [3] [4] [5][6] [7] [8] Ultramarine 22:14, 25 Jan 2005 (UTC)

This would be great for the article on the Flynn effect, but in my experience you'll be hard pressed to find anything like a scientific consensus on what causes the effect and what it means for the difference in test scores between races. (For example, I've seen speculaton that it's caused by everything from nutrition to heterosis.) For the most part, everyone has their own theory. Thus to be impartial, you'd have a to a massive literature review and bring out all of the conjecture on what it means. I would recommend keeping it simple and impartial about the current lack of a consensus about what Flynn effect means. --Rikurzhen 01:00, Jan 26, 2005 (UTC)

(cultural difference --> no genetic differences) fails as an argument

SLR, this argument is clearly flawed. I don't mind if you want to attribute it to someone, but it is pretty easy to explain why it fails. Evidence for cultural differences says nothing about the existence or not of genetic causes because they operate at different levels -- it is possible for cultural differences to exist while genetic differences are the ultimate cause of those differences. Moreover, the hypothesis that part of the gap is caused by genes is not affected by the discovery that another part is affected by culture -- indeed, that is expected under Jensen's hypothesis. Finally, is it not necessary that the the cause of the Asian-White gap is identical to the cause of the Black-White gap or any other gap. The only way to know one way or another is to do the genetic research, which has not been done and likely will not be done any time soon. --Rikurzhen 03:57, Feb 21, 2005 (UTC)

There is no strong evidence that the variation in IQ between different populations is mostly determined by genetics. Slrubenstein | Talk 16:32, 21 Feb 2005 (UTC)
SLR, That's a great opinion, but again an unattributed one -- and not logically connected to my criticism. So I'm going to remove the logically offending bit until someone can provide either attribution to a source or something that I'm not aware of that makes the argument obviously valid. --Rikurzhen 20:45, Feb 21, 2005 (UTC)

The source is Gould's The Mismeasure of Man. Frankly, I do not understand your argument. No one questiuons the fact that human intellegence has a genetic component, even a high one. The question is how much inter-group variation is accounted for my genetics. The evidence is that it is mostly accounted for by culture. Jensen simply misuses the concept of heritability to skew his analysis. Heritability applies to intragroup variation, but not intergroup variation. Let me give you a simple analogy. We can all agree that genes have a huge impact on one's vision. But actual differences in vision between White Brazilians and Indian Brazilians is largely cultural: White Brazilians have access to eyeglasses that correct for most cases of poor eyesight, but many Indian societies have no access to eyeglasses. Slrubenstein | Talk 20:50, 21 Feb 2005 (UTC)

SLR, I've done considerably reading on this topic, including a semester-long course in college. You can attribute the claim to Gould in the text, but it cannot stand as an obviously true assertion. The cultural differences offered as an explanation for the Asian-White gap is education attainment, yet (even if that were true that such a gap exists, I don't know for certain that it does) it doesn't demonstrate whether education causes the IQ difference or the IQ differences causes the differnece in educational attainment. (More likely, there is a bi-directional interaction between genes and environment, and that's why I'm comfortable with Jensen and others conclusions that the gaps are partly genetic.) In the end the data is very thin because so few people do good research in this area, and only indirect arguments can be given to support any particular theory -- that includes the cultural theory. --Rikurzhen 21:58, Feb 21, 2005 (UTC)
p.s. I and many researchers [9] think that Gould's book is rubbish. There are competent criticisms of Jensen's thesis, but Gould's book is not one of them. --Rikurzhen 21:58, Feb 21, 2005 (UTC)

Yours is one view that certainly should be represented. But there are many who stand by Gould's arguments and that too is a view that must be represented in the article. Just stick to our NPOV policy, that's all. Slrubenstein | Talk 22:09, 21 Feb 2005 (UTC)


I've gotten off topic, so let me bring this back to the point. The section presents evidence of differential educational attainment between races, and then argues that that evidence favors a culture only hypothesis over the partly genetic hypothesis. This is invalid reasoning. As the text states, Asians and Jews have a legacy of educational attainment and today show higher group-average IQs than whites. Yet this pattern is exactly what would be expected under the partly genetic hypothesis too. So finding that pattern does not favor one hypothesis over the other. That's not to say that no evidence could favor culture only over partly genetic. For example, if Asians or Jews had historically shown poor educational performance, but suddenly demonstrated exceptional achievement; then that might favor the culture only hypothesis. Or, if group-average IQ of certain races were very different in different countries, then that too would favor culture only over partly genetic. But the argument--as written--is clearly invalid, because the evidence presented does not favor one hypothesis over the other. You can feel free to attribute it to a ceratin author, but I'll be forced to present the counter point, which will just turn the article into a mess. I think this article has been fairly good about only presenting above-board, well-founded, conservative arguments; but this section fails that criteria. --Rikurzhen 00:58, Feb 23, 2005 (UTC)

Let me summarize: the argument presented in the text doesn't disfavor genetics, it ignores it. --Rikurzhen 01:42, Feb 23, 2005 (UTC)

I understand your point, and agree. But as far as I understand it, no one is claiming that there is no genetic basis for intelligenct -- only that most variation is caused by (or explained by) cultural practices. Slrubenstein | Talk 18:38, 23 Feb 2005 (UTC)


I believe maybe we've fundamentally miscommunicated, and perhaps an improvement to the article can come out of this. On the question of genetics, the actual question is: what fraction of the observed IQ gap is attributable to genetic differences, rather than non-genetic (e.g. cultural, socieo-economic, etc.) differences. There seem to be three categories of answers to this question:

  1. culture only: the contribution of genes is minimal (e.g., 0-15%)
  2. partly genetic: the contribution of genes is substantial (e.g., ~50%, >=20% and <=80%)
  3. mostly genetic: the contribution of genes is nearly total (e.g. 90-100%)

The view endorsed by Jensen, Rushton, et al. is not #3, but #2. My complaint about the new text is that by saying the evidence favors culture rather than genetics, is that it seems to be claiming that the evidence supports #1, and disfavors #2. But that kind of evidence can't tell the difference between #1 and #2. It would take much stronger evidence to differeniate #1 from #2. --Rikurzhen 01:50, Feb 24, 2005 (UTC)

Yes, I do think we are talking past one another. You are making a point about the argument; I am making a point about what our job is as editors. My "meta" point is that it does not matter wheat you or I consider a strong or weak argument. What matters (for the construction of an NPOV article) is that we make sure all major views are represented. There are indeed many people who believe that these differences are caused by cultural factors. Whether we agree or disagree, like or do not like, their arguments is irrelevant. This view has to be represented in the article. Slrubenstein | Talk 15:25, 24 Feb 2005 (UTC)
SLR, it is not enough to represent all major views. For a subject like this, the reasons (and reasoning) for those views are crucial, and the arguments matter. The paragraph under discussion reads like uncased sausage, mostly for the reasons Rikurzhen is concerned with: the arguments do not actually favor any particular hypothesis relevant to the topic. Instead, the observations about educational traditions and so on (utterly without substantiation, I might add) are being used as rhetorical devices to smuggle in opinions ("others argue", "some speculate"). It is not NPOV, but non sequitur. I recommend you guys return to source material and let it speak for itself. --DAD 17:32, 24 Feb 2005 (UTC)
Right. SLR, the fact that even Jensen and Rushton don't endorse #3 is good evidence that the vast majority of people think that culture/environment contributes at least in part to the IQ gap. But the reasons presented in the paragraph are not the reasons they would endorse. Indeed, I don't know anyone that would endorse the justifications given in the paragraph -- which is why I was telling you about the arguments and how they were so obviously flawed they shouldn't be there. If you can find someone that does endorse them, then we could pin it on them. You suggested Gould, but I don't recall any text that supports that claim. Evidence for culture comes from more substantial sources. Things like: comparisons between the average IQs of Africans, African-Americans and British-Africans; the heritability of IQ among the poor; the short-term successes of educational interventions; and the differential performance of black girls and boys in childhood. That's just a few examples off the top of my head. --Rikurzhen 19:15, Feb 24, 2005 (UTC)

I understand that both of you feel there needs to be a better explanation of the argument presented in this section. I agree with both of you in this regard. But the way you write, I think you may be suggesting something more, and if I understand you correctly, it is this something more I do not agree with. The something more is this: either you are saying that we need to put in an argument Jensen and Rushton would agree with -- and I reject this categorically; or you are saying that given the lack of adequate explanation for this argument, we should either cut it altogether, or note that it lacks support -- and I reject these alternatives too. The section opens by making a claim. The section should go on to say who makes the claim, and why. If the section does not do so, then the solution is to try to find more information and include it. Maybe this is all you are saying. If you are saying that the section does not comply with our verifiability/cite sources policies, I do agree with you and we should either find the sources and provide more information, or get someone else to do it. Please understand, I did not write this section, and do not know what sources it is based on -- and definitely would agree that the sources should be provided and explained. My only point has been that the argument should be dismissed just because any of the editors dismiss them (that would violate our No original research policy), or because some psychologists, like Jensen or Rushton, dismiss them (that would violate our NPOV policy). This is the only point I am arguing for.

On a separate note I see that the last section is on The Bell Curve versus Gould. But there was no explanation of Gould's views. So I put them in. Perhaps someone else can elaborate on the Bell Curve's arguments. Slrubenstein | Talk 21:24, 24 Feb 2005 (UTC)

On the first note. What I'm saying is that not only are the claims unattributed, but I suspect that they are original research -- because I can't imagine any scientists seriously making those claims -- they have the intellecutal integrity of a newspaper editoral. That's why I suggest we roll back the section. However, we can make an in-kind substiution with my other point -- that even Jensen and Rushton believe that part of the BW gap is cultural. Basically, there's no real argument among researchers that at least part of the gap is cultural. --Rikurzhen 22:10, Feb 24, 2005 (UTC)

Well, if this is indeed original research (and I think you may be right) it should be deleted. But I am sure that there are scholars out there who argue that variations in intelligence between different populations are largely (if not entirely) cultural in origin exist, and this view should be represented. If I have time I will dig around to see who has argued this but I do think others who have in the past contributed to this article might have a better idea than I. Slrubenstein | Talk 22:20, 24 Feb 2005 (UTC)

Aside from Gould, the best source may be Alexander Alland's Race in Mind (2002) Slrubenstein | Talk 22:45, 24 Feb 2005 (UTC)

adoption studies

from the article: Asian minorities score well on IQ tests and on average enjoy greater economic success than other minorities. However, many members of East Asian communities tested in the United States and other western nations come from households that had and maintain strong educational traditions.

This argument is rendered moot by the part that has been left out -- while many East Asians are from educated backgrounds, East Asian adoptees from impoverished backgroudns but raised from a young age by white parents in the US have high IQs. --Rikurzhen 04:05, Feb 21, 2005 (UTC)

Usually it is difficult to adopt a child. There is usually a screening process in order to remove couples not likely to be good parents. The adoption is often costly which requires some degree of wealth. Often the future foster parents have some control over which child to adopt and they usually demand a healthy child. Ultramarine 09:31, 21 Feb 2005 (UTC)
Usually that true, but in the cases I'm referring to -- sorry I've forgotten the reference right now -- this was not the case. Yet this is really just an example of the problem discussed above. The education history of East Asians is an interesting piece of data, but not one that supports any single theories about the White-Asian gap. Put another way, I know of no researcher who claims that it does. --Rikurzhen 20:51, Feb 21, 2005 (UTC)

Gould all over again

The Gould criticism is fully covered on the page devoted to The Mismeasure of Man. It should not be reproduced or paraphrased on this page, any more than the contents of The Bell Curve should be. A link is adequate. --DAD 03:07, 1 Mar 2005 (UTC)

Older discussions are archived at Talk:Race and intelligence/Archive 1, Talk:Race and intelligence/Archive 2, Talk:Race and intelligence/Archive 3, and Talk:Race and intelligence/Archive 4.

NPOV

I put material from Gould back in. Maybe it can be rearranged, or incorporated into other sections -- but it has to stay in this article. Without it, the article is wildly unbalanced and violates our NPOV rule. There is nothing wrong with having articles that overlap -- complete articles on The Origin of Species (the book) and "evolution" (the theory) will overlap considerably. Gould was writing about race and intelligence, so what he wrote belongs here. If you cut it out, then you will have to find other ways to incorporate those arguments into the article if you intend on complying with our NPOV policfy. Slrubenstein | Talk 18:08, 1 Mar 2005 (UTC)

No, SLR; the poster of the Gould material (now, that is you), by the NPOV policy, has the responsibility to add an equivalent section on The Bell Curve. Is that really what we want on this page? IMHO, those books have ample coverage on their own pages. We need not rehash all their arguments on this one -- that's what links are for. That the article is wildly unbalanced against Gould's argument is your opinion -- one which I do not share. --DAD 18:26, 1 Mar 2005 (UTC)

I have never deleted material concerning an opposing view.

NPOV policy does not require that one editor provide all views.

You are right that there should be more material on the Bell Curve and I will not stop you or anyone else from adding it. But it is not my obligation to add it. My obligation is only this: not to delete it. You are under the same obligation. Never delete material just because you don't like it. Slrubenstein | Talk 19:00, 1 Mar 2005 (UTC)

SLR: I agree with you in principle. In my defense, the Gould content had been previously eliminated -- with some implicit consensus, given the months that have elapsed since then -- because of its redundancy. Hence my edit title, "Gould all over again." My deletion was simply restoring an edit which I believed had already been agreed upon, but I began a new topic on the Discussion page in case someone wanted to talk it over. I'm not deleting material just because I don't like it. Furthermore, I do not agree that deletions are somehow prohibited by NPOV. NPOV alone does not justify duplication of content just because someone is afraid readers won't click through a link. --DAD 05:03, 2 Mar 2005 (UTC)
Gould's material is old and inapplicable to most modern research. The same can be said about the Bell curve. As such should neither have a prominent place in this article. I suggest that you read more modern research and critique. For example, the introduction to this study [10]. Ultramarine 19:08, 1 Mar 2005 (UTC)

As far as Gould is concerned, you are mixing apples and oranges. And here to we come to an NPOV problem. This article relies heavily on experimental data. That is fine -- and obviously relevant. But scientists employ other methods; Gould is using a historical-critical approach, and his arguments are still relevant. Slrubenstein | Talk 19:52, 1 Mar 2005 (UTC)

Gould is mainly criticzing the statistical methods used in experiments. Current research has made his critique irrelevant. At least that part should be eliminated. Ultramarine 20:00, 1 Mar 2005 (UTC)

Well, insofar as those critiques have led psychometricians to develop other methods, I think there still may be a place in the article for them. Some of his points perhaps can be kept as "concerns" or questions worth remembering, and not just as specific arguments against specific claims. In any case, granting your point, he is also making conceptual/theoretical claims about what can and cannot be learned from statistics, and making a historical point about the culture of Western scientists, which should definitely stay. Slrubenstein | Talk 20:05, 1 Mar 2005 (UTC)

Exactly what are those points? Ultramarine 20:19, 1 Mar 2005 (UTC)

Well, as general caveat's there is the warning about inferring too much from corelations. The questioning about "g" is very important -- his point is not that it is so easy to falsify "g", but that it is too easy to use statistics to demonstrate "ge." I think the point that he makes which is still valid today because many scientists today continue to make this mistake, concerns the misuse of heritability. Finally, the general point that Western scientists have for a long time strongly privileged quantitative measurement and biological explanations, and that these values are rooted in out culture. This does not mean that quantitative methods are never justified, nor does it mean that biological explanations are always wrong. It does, though, call on people to be extremely suspicious when quantitative data and biological explanations conform to (or even support) popular prejudice. Slrubenstein | Talk 20:32, 1 Mar 2005 (UTC)

Your "thinking" about g or current researchers is uninteresting and original research. Show evidence that the researchers today do not apply statistics correctly. Ultramarine 20:49, 1 Mar 2005 (UTC)

(1) How is the stuff on "ge" my "original research?" (2) you seem to miss Gould's point that the matter is not whether or not researchers use statistics correctly, but the limitations of their research when they do use statistics correctly. (3) it does not matter whether you personally judge Gould's arguments interesting or uninteresting. Our NPOV policy is designed specifically to get us to go beyond our own judgement. Slrubenstein | Talk 20:57, 1 Mar 2005 (UTC)

Gould's critique is irrelevant to much modern research. Your claim that it is relevant to modern research after the publication of his book is original research unless you have some sources. Ultramarine 21:04, 1 Mar 2005 (UTC)

First, you are ignorant of our No original research policy. My claim does not in any way constitute original research. Second, your own claim about "modern research" seems not to have any basis. The debate concerning race and I.Q. is of interest to different kinds of researchers. The fact that you are not interested in the book in no way means that "all researchers" are uninterested in the book. Currently, the article is dominated by research by psychologists. Whether psychologists agree with this or not, the fact is that discussions of race and IG are also of interest to physical and cultural anthropologists, and perhaps evolutionary biologists. The book is still assigned in college courses, and is still very well-considered by people in various fields. Right now the article cites as sources Murray's book on IQ and income disparity; I think that book is out of print. It also cites Jensen's book on the G factor. That book has an Amazon.com sales rank of 416,383. It mentions Rushton's book, Race, Evolution, and Behavior, by Rushton. That book circulates more widely than Jensen; it has an Amazon.com sales rank of 244,952. Gould's book however has a sales rank of 42,858.

I have no idea why you are so sensitive and reactionary to any inclusion of Gould. My advice is to bracket your own emotions, and focus on writing a good article. If some researchers consider Gould's work irrelevant, include that in the article (with proof). But the fact that some researchers are uninterested in the book does not make the book irrelevant to this article. Slrubenstein | Talk 21:39, 1 Mar 2005 (UTC)

You seen to have little knowledge of current research in this area. Rushton's and Jensen's book are only mentioned briefly, as they should be. They are all outdated by new research. Citing Gould is as relevant as citing Silent Spring in DDT research. It should certainly be mentioned for historical reasons but research has advanced far beyond that. Rankings in Amazon is not evidence for anything. Evolutinary evolutionists and anthropologists are represented among those doing research, contrary to what you think. Note that none of knowledgable critics of Jensen use Gould. Ultramarine 21:47, 1 Mar 2005 (UTC)

Excuse me, "evolutionary evolutionists?" Let's look at one section, on brain size. This section doesn't mention any of the physical anthropological research that, along with points made by Gould, seriously call into question the assumption in this section that there is a meaningful relationship between brain size and evolution. Slrubenstein | Talk 21:57, 1 Mar 2005 (UTC)

You say the article mentions anthropologists, but I could find only 2: Morton, who wrote in the 1830s when anthropology did not really exist as a modern science, and Ruth Benedict, who -- important though she is -- was a cultural anthropologist writing in the 1940s. Gould is certainly more recent and more relevant than Morton and Benedict. Slrubenstein | Talk 22:03, 1 Mar 2005 (UTC)

Regarding anthropologists, read the brain size section again and see all the refernces. Even Boas is cited indirectly.
Regarding Gould, here is a critique that Gould never answered. [ http://www.prometheism.net/articles/gould03.html] The best example is of course Jensen's "The G factor", but I do not think there is a free copy on the net. Gould never answered that critique either. I will shorten the Gould section unless you have a cited response to this critque. Note that many of the authors points has since in turn been criticzed but no by Gould or using Gould's arguments. Ultramarine 22:21, 1 Mar 2005 (UTC)

---

Can't we just have a link that says see the Mismeasure article for details on Gould's criticisms? p.s. Ultramarine please answer my Flynn effect criticisms before making more changes to the article. --Rikurzhen 22:27, Mar 1, 2005 (UTC)


As I stated above, I believe that the Gould critique has to be represented in this article to ensure NPOV. Concerning Ultramarine's ultimatum above, the summary of Gould's points here are not related to Rushton's criticisms. Also, you keep misunderstanding how NPOV is ensured: do not delete content. add content. Slrubenstein | Talk 22:33, 1 Mar 2005 (UTC)

Also, Ultramarine is being disingenuous. I read the Brain size section. It ignores the mainstream views among physical anthropologists. It ignores Boas's point and doesn't even cite him. No mention of Holloway, who is the leading expert on the matter. Slrubenstein | Talk 22:37, 1 Mar 2005 (UTC)

The article is already too long. And Gould's view are already elsewhere in Wikipedia. Anyone interested can just click a link to see them. Articles cannot contain all historical research done earlier. Again, please anwer the critique against Gould.
You call this ignoring Boas?

Cranial vault size and shape have changed greatly during the last 150 years in the US. These changes must occur by early childhood because of the early development of the vault. One explanation is the Flynn effect. [11][12][13][14] Ultramarine 22:42, 1 Mar 2005 (UTC)

Yes I do. The point Boas is making (supported by Gravlee et. al.) is not that "early development of the vault changes brain size," the point is that cranial capacity is too heavily affected by the environment to be of any use in identifying biological populations (e.g. races). And again, Boas lived a long time before Gould -- I do believe that Boas is still important; my point is, this is evidence of the fact that the year in which an article or book was written is not necessarily indicative of its current importance. And again, why no mention of Holloway? Slrubenstein | Talk 22:50, 1 Mar 2005 (UTC)

Please mention Holloway if there is some relevant peer-reviewed articles. Again, this has nothing do with Gould. Boas is important which is shown by the fact that some of his research has been replicated by 2 independent researchers. Gould on the other hand has been severely ciritczed and no one has replied to that critique, not even himself. Again, please show cited replies to the critique against Gould. Otherwise he has only a historical interest in this already too long article and can be reduced to link. Especially as no material will be lost. Ultramarine 23:11, 1 Mar 2005 (UTC)

I wouldn't pin your entire critique on a pseudo-scientist like Rushton. As I said, the points concerning Gould here are not ones Rushton criticized, so his criticisms in the review you mention are irrelevant (especially since Rushton doesn't understand the biology behind Gould's claims). Slrubenstein | Talk 23:16, 1 Mar 2005 (UTC)

I do not think that Rushton views are correct, due to peer-reviewed critique. But no one has replied to his critique of Gould and as such one must assume that it is correct. Furhtermore, the main critique is in the "G factor". Neither Gould nor anyone else has answered to that critique. Others have criticzed the G factor, but not using Gould's arguments. 23:24, 1 Mar 2005 (UTC)

Wicherts et al 2004 on Flynn effect vs BW gap

Paragraph in question:

Others think that a simpler explanation is that there are no genetic differences and that the Flynn effect will eliminate differences in IQ test scores in the future. They argue that the Flynn effect started and will end sooner for the more affluent parts of society and that blacks will or have started to close the gap.

I ammended: However, this explanation is contested; for example, one report concludes "that the nature of :the Flynn effect is qualitatively different from the nature of B-W differences in the United States" (Wicherts et al., 2004).

Ultramarine removed: "The context is discussing measurment invariance, not if the BW gap is genetic"

I added it back b/c I think comparing the Flynn effect to the BW gap is very releveant--the comparison is commonly made--no matter which section of the article it appears in. Wicherts finds that Flynn effect is a qualitatively different phenomena than the BW gap (based on previous analyses of the BW gap) and his analysis of the Flynn effect.

If we wanted, we could move the entire paragraph to the "other explanations" section. --Rikurzhen 18:30, Mar 1, 2005 (UTC)

The paper draws no conclusion about the whether it BW gap is genetic or not. That the BW gap has measurment invariance and the Flynn effect do not says nothing about genetics. Ultramarine 18:40, 1 Mar 2005 (UTC)
The link to genetics is created, not by me, but by the earlier text and the position of this paragraph in the genetics section. The Wicherts quote needs to stay b/c it counters the assertion that the Flynn effect can explain the BW gap better than genetics. Either they both stay or they both go -- I believe this is SLR's point about NPOV above. --Rikurzhen 19:09, Mar 1, 2005 (UTC)
Don't be ridiculous. The abstract says nothing to support your interpretation. It the study had claimed what you say, then that should certainly have been mentioned prominently there. Again, your quote refer to measurment invariance, not to genetics. Ultramarine 19:16, 1 Mar 2005 (UTC)
Give me some credit, I didn't fabricate that quotation. You can find the PDF online here [15]. The papers says exactly what I wrote: "that the nature of the Flynn effect is qualitatively different from the nature of B-W differences in the United States" (Wicherts et al., 2004). He makes this conclusion in the end of the paper b/c it involves integrating his results with previous publications. --Rikurzhen 19:24, Mar 1, 2005 (UTC)
Again, your interpretation of what this quote means should certainly have been mentioned in the abstract, if it was correct. What you are quoting refers to measurment invariance, not to genetics. Ultramarine 19:28, 1 Mar 2005 (UTC)
(1) You have the right to be incredulous, but then the responsibility falls to you to explain what the quote really means. Sometimes authors don't put all their conclusions in the abstract. From my reading of the paper, that quote is straightfoward: Flynn effect is a different kind of phenomena than the BW gap. (2) There's a fundamental problem in our communication here. I think what you consider to be evidence for genetics, is not what I consider evidence for it. But getting over that might be difficult. The point about Flynn effect vs BW gap being qualitatively different stands. I suggest, then, that we move the paragraph to the "other explanations" section along with the Wicherts quotation, so that we can move past this genetics thing. --Rikurzhen 19:56, Mar 1, 2005 (UTC)
That there are differences between measuring the same group at different times and measuring different groups at the same time cannot be used as evidence for genetics. The quote refers to differences in measurment invariance, not to differences in genetics. Ultramarine 20:24, 1 Mar 2005 (UTC)
Right, except that's not what I wrote. I wrote that the Flynn effect is qualitativley different than the BW gap, and so offer the Flynn effect as evdience against the genetic hypothesis is contested. I didn't write that this is evidence for anything. --Rikurzhen 21:03, Mar 1, 2005 (UTC)
I fail to see your point. What has differences in measurment invariance to do with the role of genetics? Your interpretation of the paper is original research, the authors make no such interpretation in abstract or anywhere in the text. Please cite all text in your evidence. Ultramarine 21:12, 1 Mar 2005 (UTC)
What I wrote was: the Flynn effect is a poor explanantion of the BW gap as per the quotation from Wicherts. Here's the exact text: However, this explanation is contested; for example, one report concludes "that the nature of :the Flynn effect is qualitatively different from the nature of B-W differences in the United States" That's it. Nothing about genetics. The part about genetics was already there. Indeed, I think it's a mistake to juxtapose the Flynn effect and genetics as per my complaint below. --Rikurzhen 21:30, Mar 1, 2005 (UTC)
Cite this part about genetics that the authors apparently failed to mention in the abstract. Ultramarine 22:06, 1 Mar 2005 (UTC)
I'm sorry I don't understand. Are you being sarcastic? --Rikurzhen 22:17, Mar 1, 2005 (UTC)

All of this is taken of Wicherts 2004:

The recent applications of method of correlated vectors to intelligence score gains (e.g., Colom et al., 2001; Flynn, 2000b; Must et al., 2003) followed Flynn’s critique on the conclusions that Jensen and, particularly, Rushton (2000) based on this method (Flynn, 1999c, 2000a, 2000b). From its beginning, the Flynn effect has been regarded to have large implications for the comparison of these B–W differences (e.g., Flynn, 1987, 1999c). Because the current approach (MGCFA) was previously applied in U.S. B–W comparisons, we have the opportunity to compare those B–W analyses to the current analyses of different cohorts. Here, we use results from Dolan (2000) and Dolan and Hamaker (2001), who investigated the nature of racial differences on the WISC-R and the K-ABC scales. We standardized the AIC values of Models 1 to 4a within each of the seven data sets to compare the results of the tests of factorial invariance on the Flynn effects and the racial groups. These standardized AIC values are reported in Fig. 2.

As can be seen, the relative AIC values of the five Flynn comparisons show a strikingly similar pattern. In these cohort comparisons, Models 1 and 2 have approximately similar standardized AICs, which indicates that the equality of factor loadings is generally tenable. A small increase is seen in the third step, which indicates that residual variances are not always equal over cohorts. However, a large increase in AICs is seen in the step to Model 4a, the model in which measurement intercepts are cohort invariant (i.e., the strict factorial invariance model). The two lines representing the standardized AICs from both B–W studies clearly do not fit this pattern. More importantly, in both B–W studies, it is concluded that the measurement invariance between Blacks and Whites is tenable because the lowest AIC values are found with the factorial invariance models (Dolan, 2000; Dolan & Hamaker, 2001). This clearly contrasts with our current findings on the Flynn effect. It appears therefore that the nature of the Flynn effect is qualitatively different from the nature of B–W differences in the United States. Each comparison of groups should be investigated separately. IQ gaps between cohorts do not teach us anything about IQ gaps between contemporary groups, except that each IQ gap should not be confused with real (i.e., latent) differences in intelligence. Only after a proper analysis of measurement invariance of these IQ gaps is conducted can anything be concluded concerning true differences between groups.

Whereas implications of the Flynn effect for B–W differences appear small, the implications for intelligence testing, in general, are large.

I think this is perfectly clear. --Rikurzhen 17:41, Mar 3, 2005 (UTC)

Again, measurement invariance is different. This says nothing about whether the BW gap is genetic or not. Ultramarine 18:34, 3 Mar 2005 (UTC)
I don't understand your insistance that this has something/nothing to do with genetics. I never tried to make that assertion. The text in the article is:
However, comparing the Flynn effect (IQ differences within races over time) to contemporary IQ differences between races is contested. Some claim that the following quote from a recent paper is relevant: "the nature of the Flynn effect is qualitatively different from the nature of B-W differences in the United States" (Wicherts et al., 2004). Others note that this is not mentioned in the abstract, refers to "measurement invariance", and is not a statement about the role of genetics in the B-W gap.
Which doesn't mention anything about genetics. It's about the Flynn effect being different than the BW gap and so talking about one doesn't tell you much about the other. I don't understand the need for the qualificatin of the quotation. It seems straightforward to me. --Rikurzhen 19:41, Mar 3, 2005 (UTC)

Dolan's papers

Ultramarine, can you provide a quotation to support this claim:

However, reanalyzing of the data from several previous studies that used the statistical method invented by Jensen with a more recent and improved method shows no clear evidence for genetics [16].

I believe we talked about these papers/conclusions before and they don't apply to US IQ tests. That is, they don't tell us about the US BW gap. --Rikurzhen 19:24, Mar 1, 2005 (UTC)

Is this the quotation you're thinking of?

Applications of MGCFA to study B-W differences in psychometric intelligence are given by Dolan (2000), Dolan and Hamaker (2001), Gustafsson (1992), and Lubke et al. (2003). Dolan and Dolan and Hamaker investigated B-W differences in WISC-R and the K-ABC test scores in U.S. samples1 and concluded that the tests are unbiased with respect to group. This implies that the same constructs are measured in the groups. However, because various competing models fitted the data approximately equally well, it remained difficult to establish the central role of g in B-W differences with any confidence.

If so, I think we need to flesh out the meaning of "a more recent and improved method shows no clear evidence for genetics". Jensen's original work was not clear evidence for genetics, so failing that test doesn't say much. (Clear evidence would probably require molecular biology.) They basically arrive at Speaman's hypothesis but don't have enough significance to say that g is what's central to the BW gap. --Rikurzhen 19:37, Mar 1, 2005 (UTC)

Jensen and other researchers certainly previously claimed to have proven Spearman's hypothesis.

It is therefore disconcerting that Jensen’s method is gaining widespread acceptance (Colom et al., 2000, 2001; Flynn, 2000; Helms-Lorenz et al., 2003; Must et al., 2003; Nyborg & Jensen, 2000; Rushton, 2001; Te Nijenhuis et al., 2000) and that results obtained with this method are taken at face value (e.g., Roth, Bevier, Bobko, Switzer, & Tyler, 2001, p. 305). In view of its insensitivity, this method cannot trusted to establish the role of g (or of any other latent variable; e.g., see Helms-Lorenz et al., 2003) in group differences. [17] Ultramarine 19:44, 1 Mar 2005 (UTC)

But even proving Spearman's hypothesis doesn't prove that the BW gap is genetic. We should be as precise as possible. The papers in question failed to prove Spearman's hypothesis but also failed to falsify it. We've got to say something about that, not about clear evidence for genetics. --Rikurzhen 19:50, Mar 1, 2005 (UTC)
Changed to text to something hopefully better.
I think the text is now so long and complex with so many opposed views that most people not doing research will just be confused by it. :) Ultramarine 20:13, 1 Mar 2005 (UTC)

the Flynn effect

The Flynn effect should be discussed separately from the genetics section. The mechanism of the effect is unknown. Amongst the possible mechanisms are nutrition, heterosis, and any variety of testing artifacts. These are not evidence against genetics; but rather are orthogonal to it. That's why I think it would be best to give the Flynn effect it's own section where it can be fleshed out -- rather than trying to put it up directly against genetics. --Rikurzhen 21:07, Mar 1, 2005 (UTC)

Will try to add a section. Ultramarine 22:44, 1 Mar 2005 (UTC)

Brain size

Concerning the increase in brain size during the course of human evolution, you might find this paper interesting: [18] Slrubenstein | Talk 22:04, 1 Mar 2005 (UTC)

That is intersting, but on reflection not that surprising from a cognitive science/brain module pov. --Rikurzhen 22:19, Mar 1, 2005 (UTC)

Agreed -- but still a POV that should be represented. This too: [19] Slrubenstein | Talk 22:28, 1 Mar 2005 (UTC)

Peer-reviewed relevant articles are just fine. This has nothing to do with Gould. Ultramarine 22:45, 1 Mar 2005 (UTC)

So? Slrubenstein | Talk 22:58, 1 Mar 2005 (UTC)

Oh, I see what you're getting at. I don't believe these findings are applicable to the race/IQ discussion, unless there is also evidence for differences in expertise capacity between races, and I think that is unlikely. However, it would make a good addition to the Brain size and intelligence article. --Rikurzhen 23:15, Mar 1, 2005 (UTC)

Put another way; comparisons among modern humans seem to be different that comparisons between modern and pre-modern humans. --Rikurzhen 23:19, Mar 1, 2005 (UTC)

Well, the fact is, I am suspicious of the entire section on brain size and intelligence within this article. The point (perhaps not held by you, but really held by many) that comparisons among humans is different than comparisons between humans and other hominoids or hominids does not mean that the comparative data is irrelevant, but that the way some scientists interpret data from humans is unscientific. Holloway and others have argued that since brain size (adjusted for body size) does not correlate wish differences in intelligence (however it may be defined) between different species (whether humans and whales or humans an H. habilus), differences in brain size don't explain variation in intelligence among humans. For one thing, difference in brain size (adjusted for differences in body size) and differences in intelligence may both be determined by pre- and neo-natal nutrition (thus, they correlate but there is no ausal relation). Holloway has also provided examples of people with relatively small brains and superior intelligence. (Ultimately, he argues that differences in brain structure are more important than differences in brain size; Gould argues that differences in intelligence are more often explained by difference in culture). In any event, perhaps one could argue that the relationship between intelligence and brain size in humans is utterly unlike the relationship in other species -- but a scientist should have a strong basis before accepting this kind of human exceptionalism. Slrubenstein | Talk 23:30, 1 Mar 2005 (UTC)

I see where you're going; I just don't think it'll work out. Before you get too invested in this idea, check out this abstract [20]. --Rikurzhen 23:47, Mar 1, 2005 (UTC)
Here's an interesting one [21] --Rikurzhen 00:09, Mar 2, 2005 (UTC)

Stop taking everything personally

It seems now like Ultramarine is going to take issue with anything I do, no matter what. The article on culture, gender, and intelligence is relevant to the article on race and IQ because it provides additional support for a claim made by opponents of Jensen and Rushton, namely, that culture is the primary determinant of variations in intelligence among humans, rather than heredity. That finding calls into question much of the reaserch being done by some (e.g. Rushton). Slrubenstein | Talk 23:20, 1 Mar 2005 (UTC)

I read the article. There was no mention of Rushton, Jensen, IQ or race. Is this you idea of a replay to Rushton's or Jensen's critique? Also, avoid ad hominem attack and show some real replies to the critique against Gould. Not an article that do not even mention his nameUltramarine 23:33, 1 Mar 2005 (UTC)

The article on culture, gender, and intelligence is relevant to the article on race and IQ because it provides additional support for a claim made by opponents of Jensen and Rushton, namely, that culture is the primary determinant of variations in intelligence among humans, rather than heredity. That finding calls into question much of the reaserch being done by some (e.g. Rushton). Slrubenstein | Talk 23:20, 1 Mar 2005 (UTC)

How can an article do not even mention possible race differences in intelligence be relevent to this article? Or relevant to Gould's ideas when it does not mention his name or book? This is original research. Ultramarine 23:42, 1 Mar 2005 (UTC)

Hispanic?

Is "Hispanic" a race? Surely it is a cultural rather than racial designation. Are the Spanish a different race to the French? Andy G 23:39, 1 Mar 2005 (UTC)

Right. That's covered in the race article pretty well. That doesn't seem to be a big problem for this article though, right? --Rikurzhen 00:11, Mar 2, 2005 (UTC)

IQ table

A while ago we had a table of average IQ scores per group. We couldn't agree on a standard for constructing it. I found that The Bell Curve has most of the data we wanted: African Americans 85, Latino 89, White 103, Asian 106, and Jewish Americans 115 (The Bell Curve pp. 273-278). Should be put in a table with that data and reference The Bell Curve? --Rikurzhen 00:34, Mar 2, 2005 (UTC)

More data (copyrighted text) --Rikurzhen 00:38, Mar 2, 2005 (UTC)

Lynn (1991, 1996) showed that, on average, Orientals score higher on tests of mental ability than do Whites, both within the U.S.A. and in Asia, whereas Africans and Caribbeans score lower. Oriental populations in East Asia and North America typically have mean IQs falling between 101 to 111. White populations in Europe, South Africa, Australasia, and North America have mean IQs of from 85 to 115, with an overall mean of 100. Black populations living south of the Sahara, in the Caribbean, in Britain, and in North America, average IQs of from 70 to 90.
Here are others table [22] [23]. I have read some studies that Jews have much lower IQ than 115. Might be other disagreements between the data also. Ultramarine 00:49, 2 Mar 2005 (UTC)

I've tried to find a definitive way of tabulating this data, but there are so many values nothing short of a histogram of results would accurately capture the information. I'll leave it to someone else to try to solve this. --Rikurzhen 23:53, Mar 2, 2005 (UTC)

Survey data

In the 1980s Mark Snyderman, a psychologist, and Stanley Rothman, a political scientist, sent a questionnaire to a broad sample of 1,020 scholars, mostly academicians, whose specialties give them reason to be knowledgeable about I.Q. Among other questions, they asked, "Which of the following best characterizes your opinion of the heritability of the black-white difference in I.Q.?" The answers were divided as follows: The difference is entirely due to environmental variation: 15 percent. The difference is entirely due to genetic variation: 1 percent. The difference is a product of both genetic and environmental variation: 45 percent. The data are insufficient to support any reasonable opinion: 24 percent. No response: 14 percent. from [24] --Rikurzhen 18:38, Mar 2, 2005 (UTC)

Old and unclear who they asked. Ultramarine 21:29, 2 Mar 2005 (UTC)
That's a good reason to prefer one survey over another. Please post any alternatives you can find. But it's not a good enough reason to exclude this data. See the survey on the race page for an example of how this data is helpful in Wikipedia articles. --Rikurzhen 22:11, Mar 2, 2005 (UTC)
I see little need to quote old views. If we do not quote Gould, we do not need to quote surveys about opinions much older than his book. Ultramarine 23:35, 2 Mar 2005 (UTC)

Save for a current-events piece, I know of no article that restricts itself to sources only from the last year, or five years. Moreover, I know of many articles that have accounts of the history of debate over an issue. Put the survey in, but make the context clear. Slrubenstein | Talk 23:42, 2 Mar 2005 (UTC)

Yeah. The survey is also in The Bell Curve, with a nice analysis of what it means: that no single opinion is in the majority and (with the exception of the all genetic view) most views have significant support. I don't think that's a controversial analysis -- then or now. --Rikurzhen 23:51, Mar 2, 2005 (UTC)
Such surveys are essentially useless without knowledge of who they asked. Ask sociologists and there will be no support for genes. Ask genetic researchers and they will convinced that genes are of major importance. Ultramarine 00:00, 3 Mar 2005 (UTC)


Obviously, it is not useless from the point of view of those who did and published the survey. It does not matter whether any of us editors like or do not like the survey, it is a valid point of view that should be represented. Slrubenstein | Talk 17:11, 3 Mar 2005 (UTC)

I agree. Ultramarine should not delete this addition. Replace it with a better survey if you can find one. --Rikurzhen 17:42, Mar 3, 2005 (UTC)
Ok. Added back with some modifications. Ultramarine 17:55, 3 Mar 2005 (UTC)


I found the survey paper. The survey made use of membership in professional societies to choose samples. It looks like 12 different groups with 60-120 people in each. --Rikurzhen 03:14, Mar 7, 2005 (UTC)

Stevertigo's new intro

Stevertigo's new intro, while laudable in its goal, is highly biased to one POV in its presentation (especially use of words like subjective to descrive race and intelligence). The point about sociology is contradicted by the fact that many studies, e.g. Wicherts 2004, are interested in IQ and mathematics more than sociology. Claim about what can be extrapolated to "race" are similarly POV. I think a reversion is the only way to rescue it. --Rikurzhen 17:47, Mar 3, 2005 (UTC)

I think that he may be refering to the fact that most IQ studies are done in the US. However, this is mentioned elsewhere in the article. Maybe we should state that the article assumes that IQ tests capture at least some intersting aspect of intelligence. Those arguing otherwise should argue in different articles. Ultramarine 18:05, 3 Mar 2005 (UTC)
Is it not a valid aspect to deal with the inherently local nature of "studies", and distinguish sharply between sociological use, popular interpretations, and scientific intent for these studies? What can be generally said about any aspect of pseudoscientific interpretation which cannot be labelled POV? -==SV 19:14, 3 Mar 2005 (UTC)
You are right that many wrongly generalize from the US results to the world as a whole. That may need to mentioned. However, I think critique like EQ or other critique of IQ itself belong in IQ or intelligence. Ultramarine 19:24, 3 Mar 2005 (UTC)

And you are right to point out where things written may be POV or otherwise do not fit. But as a common practice here on wikipedia, the argument that arguments "belong somewhere else" is typically frowned on as "Un-wiki", meaning either not in the spirit of collegiality or is otherwise attempting to sweep a pov under the rug rather that give it adequate or due represenation. In any case many good writers of content fail to see the importance of writing good introductory context. The idea of making any assertion between race and intelligence without sufficient context cannot be recognized as valid. The whole point of the article (at this general level) is to treat the issue with some decent level of respect for the intelligence of the reader —to understand the nature of debate, and of where scientific research, inconclusiveness, and utter fallibility meet. Certainly there are some people in the scientific community who's work is substandard enough that they may allow issues of context (race in the United States) to extrapolate to a meaning on a much wider scale - even so far as to attempt to tie concepts of genetics with notions of intelligence. In anycase, and putting aside that both of you have conceded me a point, the wholesale reversion of edits not vandalism is niether justified nor socially congruent on Wikipedia. --==SV 19:39, 3 Mar 2005 (UTC)

Ultramarine - you reverted my edits entirely, labeling them as a 'minor edit'? You also use the term 'no consensus' - perhaps referring to my edit which just took place a few minutes ago... Im sure you havent taken any "concensus" with anyone but yourself in just a few minutes. -==SV 18:37, 3 Mar 2005 (UTC)

The minor edit and "no consensus" refer to another edit, not the introduction or your writing. Ultramarine 18:43, 3 Mar 2005 (UTC)
I, for one, applaud Ultramarine's edit. That there is a consensus around the Introduction (Ultramarine largely restored it) is indicated by the past months of active edits on the page with few to the Intro; I was the last to give it a significant upgrade, I think. The present version acknowledges the many debates; the SV version seems to try to bold, underline and italicize the debates and add further murk. When an article on a research topic begins with statements questioning the validity and reliability of the research, the motives of the researchers, and the very definitions of the two topics under consideration, enough qualifying has been done. --DAD 20:45, 3 Mar 2005 (UTC)

Your opinions on the various issues are noted, but your further explanation of how this version is materially better than one written weeks, months or years ago (in terms of its actual balance) would be useful. Simply claiming "When an article begins with [some general disclaimers] enough qualifying has been done" is not IMHO a decent enough defense of such a mediocre writeup. Im not saying "contrary to the faith, the article has been edited into mediocrity" [25] but by not treating more seriously the nature of the perceptual debate (ie adressing the actual meaning of the use of "meaning" as applied to a race-intelligence connection) to some more substantive depth, the article is wide open to the perception of an intrinsic bias in its presentation. "Adding further murk" is not really a valid issue where dealing with a controversial topic on a introductory level article on a public encyclopedia. Neither science nor "see other articles" is valid enough context. And, I chose to be liberal with my use of actual links to other articles, rather than italics. -[[==SV]]

Very well. Take, for example, this sentence:
The basic intent and purpose of these studies is moreoften a sociological study within a local environment, (typically European-industrialized countries).
First, it's not a proper sentence, and moreoften is not a word. Setting aside simple quality-control issues, the content is POV, unsubstantiated, and incorrect. The psychometric community does not focus on sociological studies (that is, studies attempting to reveal the reasons for various sociological issues such as poverty); rather, studies are typically interested in the causes, reliability, and validity of measured cognitive ability. All the studies in the U.S., which are at least a plurality if not the majority, are not European. It is your POV that the intent of these studies is sociological and local.
Or this sentence:
This controversy may concern the reliability of the studies and the motives of their authors, and extend to include the validity and fairness of intelligence tests in general — the extent to which measured intelligence is determined by biological and social factors. (See nature versus nurture for a related topic.)
While similar to the previous version, the dash both makes the sentence grammatically incorrect and implies that "validity and fairness" are related to "biological and social factors", when these are in fact distinct questions. That is, an IQ test can be both valid and fair, or not, regardless of the cause of IQ differences.
Or this sentence:
Furthermore there are open questions as to the validity of all IQ testing —for example the importance of abstract IQ relative to emotional intelligence or "EQ."
First, what is "abstract" IQ, as opposed to IQ? The word is POV and not used by researchers. Why not use a more mainstream proposed alternative to IQ, such as Sternberg's triarchic theory, rather than one that appears to have little if any distinct predictive validity (see [26])? Finally, space is wasted defining the abbreviation "EQ", though the acronym is never used in the remainder of the article.
These sorts of inaccuracies, POVs and quality control issues undergird my opinion that the present version (U's restored version) is superior. --DAD 22:02, 3 Mar 2005 (UTC)
Though you might have saved me the trouble by linking the term "psychometric community," you nevertheless include it, and therefore make my point by merely doing so. By claiming this article functions as a "research topic" within a field of science which itself barely exists, and whose "prominent" members include Karl Pearson (who 's "scientific view of a nation" was one "kept up to a high pitch of internal efficiency by insuring that its numbers are substantially recruited from the better stocks, and kept up to a high pitch of external efficiency by contest, chiefly by way of war with inferior races.") the prominent Arthur Jensen (known for the g factor popular among eugenicist circles) and the little known (L. L. Thurstone (who "introduced statistical analysis" to such scientifically measurable topics as attitudes toward religion, etc.)... one does not need to but define "psychometrics" for its accurate meaning to demolish this basis for a treatment of this article's topic, let alone to allow such bounds to be its "scientific" basis.
Putting all of that aside for the moment, you still mistate my writing as claiming that the "intent was sociological" - this is not what I wrote. I wrote that the very use of data from a local population to be extracted to have meaning in terms of genetics, (and further to have bearing on the concept of racial distinctions) is fundamentally flawed. But flawed only by legitimate and accepted scientific measures. -==SV 23:05, 3 Mar 2005 (UTC) PS. My "abstract IQ" might have been better said as "the abstract concept of IQ."
Your claim about local populations, genetics, and racial distinctions being fundamentally flawed is not NPOV. The race article covers this question in detail, and does not arrive at an answer that supports any such consensus. Indeed, there are many recently published studies that directly contradict your claim. The intro section of this article is not the place to rehash this complex debate. --Rikurzhen 23:33, Mar 3, 2005 (UTC)
The intro section of this article is not the place to rehash this complex debate. Nor is this the place to avoid it. I could just as well claim that such avoidance represents a POV. -==SV 02:04, 4 Mar 2005 (UTC)
Please avoid ad hominem attacks. They will not help your case. Physics have had researchers who believed that Earth is flat, designed weapons of mass destruction, and participated in genocide. Medical researchers in many countries have committed various atrocities in designing new weapons. Physics and Medicine are still valid branches of science. Attack the arguments, not the researchers. Ultramarine 23:43, 3 Mar 2005 (UTC)
I made no ad hominem attacks. And I agree that while science (being a component of (typically Western) society) has any number of skeletons in closet, this does nothing to validate claims of an honor by association between the (questionable enough) science of psychology with psychometrics —which by certian criterial appears to be entirely dubious in origin. Again, the essential criticism of a more broad treatment appears to be the claim (entirely contrary to principle) that such material belongs elsewhere - where else does it belong, and why is that not appropriately linked? In order to make the claim that x is elsewhere, one must ensure that links to x are prominent, and well addressed. It is this very elemental debate that presents enough substance to justify keeping this article under this title. Otherwise, give articles to each independent study and treat them as a separate context - independently they dont create enough context to form a view of race and intelligence. Keep in mind that race is not genetics, perhaps a treatment of intelligence and genetics is a more appropriate title for a more limited discussion. Genetics can at least be well defined; choosing two very subjective terms to connect is not the way to title a reportedly scientific article. -==SV 02:04, 4 Mar 2005 (UTC)
Your disparagement of Jensen, Thurstone and Pearson, particularly for their political views or even (in the case of Jensen) their admirers, is the very definition of ad hominem. --DAD 02:53, 4 Mar 2005 (UTC)
Moreover, your claim that psychometrics is a small field, small enough in particular to warrant scorn, is pure bunkum. The field has its own specialty journals (e.g. Intelligence), with hundreds of authors a year. Most articles in psychometrics are published in other journals, such as Nature, Science, PNAS, Psychological Bulletin, NeuroImage, Psychological Science, and so on. While those who study the measurement of IQ exclusively are a subset of psychologists -- though not a small subset, until you define small for us -- those who carry out research examining the nature and validity of IQ and g in various contexts (the field of psychometrics relevant here) is enormous. For example, it includes vast numbers of industrial organizational (IO) psychologists; Chapter 1 of the Blackwell Handbook of Principles of Organizational Behavior [27] is, "Select on Intelligence", written by John Hunter, among the most respected of this breed of psychologists and a primary researcher, but not a psychometrician. Most of the military studies on IQ, some of which are drawn on for the race and intelligence data, were carried out by military psychologists, not psychometricians, but all of these studies are aspects of psychometry. Studies linking educational aptitude testing to IQ are typically done by educational psychologists. Now, by any rational measure (number of researchers, number of papers, number of journals, number of study subjects, length of time field has existed), the field of psychometrics (or more precisely, IQ research) is supposed to "barely exist" and not even be science? You'll have to provide more than italics to back that claim. --DAD 02:53, 4 Mar 2005 (UTC)

BTW, this kind of disclaimer is highly POV: "The following article assumes that IQ tests measure some interesting aspect of intelligence and that some interesting information may be gained by studying racial group differences. For a critique against these assumptions, please see the previously mentioned articles." An article here cannot claim such wide assumptions, and must deal with criticisms in parallel - from intro to end as balanced. More detailed treatments of individual studies (as there is no limit to number of detailed articles...) can refer here to this overview. -==SV 21:05, 3 Mar 2005 (UTC)

Are you arguing that all the discussions from race, IQ, and intelligence should be copied here? Ultramarine 21:08, 3 Mar 2005 (UTC)
No. Merely that its current introduction is inadequate, and that aside from the separate aspects, the the very idea of connecting notions of race to those of intelligence needs much more introductory clarification. -==SV 21:18, 3 Mar 2005 (UTC)
As per my criticism of Ste's edits, I don't believe the current intro can be made any content-richer without violating NPOV or copying large amounts of the linked articles. --Rikurzhen 21:26, Mar 3, 2005 (UTC)
This still does not actuall deal with my comments in substance, but rather claims that there is no space to treat it with any interest. -==SV 21:31, 3 Mar 2005 (UTC)
I would like to avoid duplicate material in two articles. This only makes reading, updating and discussion confusing. I think EQ as a critique against IQ belongs in the IQ or intelligence article. Is there some other specific critique you have that is not mentioned in other articles? Ultramarine 21:33, 3 Mar 2005 (UTC)
This article is already a detailed sub-article that addresses the intersection of the race and the intelligence debates. Stevertigo, your logic would seem to entail an infinite regress of ever larger articles that repeat all the content of those above before adding a new deatil. A practical solution is required. I believe the current intro is a satisfactory solution. --Rikurzhen 23:41, Mar 3, 2005 (UTC)

adoption studies

Ultramarine, are you sure about these adoption studies?

  • Clark, E. A., & Hanisee, J. (1982). Intellectual and adaptive performance of Asian children in adoptive American settings. Developmental Psychology, 18, 595-599.
  • Frydman, M., & Lynn, R. (1989). The intelligence of Korean children adopted in Belgium. Personality and Individual Differences, 12, 1323-1325.

I've looked at the meta-analysis paper you cited, but a 1996 paper from Rushton seems to contradict their reporting:

Transracial adoption studies also reveal genetic influence. There have been at least three studies of Korean and Vietnamese children adopted into White American and White Belgian homes (Clark & Hanisee, 1982; Frydman & Lynn, 1989; Winick, Meyer, & Harris, 1975). As babies, many of these children had been hospitalized for malnutrition. Nonetheless, they excelled in academic ability with IQs 10 or more points higher than national norms. In contrast, Weinberg, Scarr, and Waldman (1992) found that at age 17, Black and mixed-race children adopted into White middle-class families performed at a lower level than the White siblings with whom they were raised. Adopted White children bad an average IQ of 106, an aptitude based on national norms at the 59th percentile, and a class rank at the 54th percentile; mixed-race children had an average IQ of 99, an aptitude at the 53rd percentile, and a class rank at the 40th percentile; and Black children had an average IQ of 89, an aptitude at the 42nd percentile, and a class rank at the 36th percentile. (Race, genetics, and human reproductive strategies by J. Philippe Rushton Genetic, Social & General Psychology Monographs, Vol. 122 02-01-1996.) [28]

I haven't been able to find the original papers online. --Rikurzhen 06:41, Apr 11, 2005 (UTC)

Rushton can claim to be correct by a very twisted use of words. The 1975 study had malnourished children but this study did not find higher IQ scores. The other studies two had no malnutrition and higher IQ scores. According to what I can see from the meta-analysis. Should make anyone suspicious of anything written by Rushton.
On the other hand, he has a point regarding the adopted black children where the prior high IQ scores seem to disappear in adolescence. Environmental influences could still have a negative effect both before adoption and in adolescence. I also note that this seems to invalidate almost all of the studies in IQ and the Wealth of Nations since they almost all are done on children younger than 17. Ultramarine 07:33, 11 Apr 2005 (UTC)

regression

Ultramarine: Cannot use regresseion towards a racial mean since the race has been divided into different groups. Give an explanation why average IQ for similar income groups shoudl be lower for blacks.

That's just being obstructive. Clearly we're talking about any pair-wise comparison between minorities and whites as a point of comparison. I'm reverting back to my addition. --Rikurzhen 17:44, Apr 18, 2005 (UTC)

No. We are talking about comparison groups between blacks and white adjusted for SES. Are you arguing that blacks with the same income as whites have a lower IQ? Then you can argue that regression towards the mean would apply. But that requires an explantion why they would have lower IQ in this situation, in face of discrimination. Ultramarine 17:49, 18 Apr 2005 (UTC)
No, that's yet another possible explanation -- affirmative action perhaps -- but I'm simplying talking about generational regression. Compare parents to children, as we have here, and you'll see that (use height as an example) tall parents have (on average) less tall (but still relatively tall) children and short parents have (on average) taller children. Now assume for the sake of argument that racial genetic background affects average height. You regress child height on parental height broken down by race and you'll expect to see children of each race regress towards a different means -- that is, the shorter (on average) race shows shorter children (on average) for any particular parental height. Comparing generations is the original use of regression -- from Galton. --Rikurzhen 17:56, Apr 18, 2005 (UTC)
Yes, I know. But you have divided the races into different groups, you are not anymore comparing all of the races against each other. Taking height, if you divided the races into different groups with different means and compare the groups with the same mean, you are not going to get any differences. And by adjusting for SES you have arguably divided the races into groups with similar mean intelligence which are then compared. Ultramarine 18:14, 18 Apr 2005 (UTC)
No, no. Seriously, you will. The equivalent would be if 6' tall Chinese parents had shorter children on average than 6' British parents who in turn had shorter children than 6' Zulu parents. (Assume they all live comfortably in the U.S.) The genetic explanation there is that the 6' tall Chineese parent has a rarer phenotype/genotype than the 6' tall Zulu parent; and when their genes recombine (ususally with someone of their own ethnicity) in a new genearation, you'll get different children's heights on average. --Rikurzhen 18:24, Apr 18, 2005 (UTC)

Regression among siblings to children matched for IQ

I do not understand the arguement here. The white and black children can have different environments which give different mean IQ even if the genetically determined intelligence is the same. Matching for IQ and looking at siblings only shows regression to the mean IQ for the races which may be environmentally affected. That the siblings have a similar environment do not change that the races can have different environments. Ultramarine 16:43, 19 Apr 2005 (UTC)

The culture-only theory predicts that culture depresses the IQ of blacks. But it should be doing so uniformly! All blacks should be affected equally or at least proportionately to their IQ. For example, if the negative effect causes an average 15 point depression of IQ among blacks -- if they had been white, then they would have been 15 IQ points brighter -- then you predict the effect would affect siblings to the same extent. Why then do the siblings of smart blacks have lower average IQ than the siblings of IQ matched whites? The effect occurs for any set IQ matching point; it works at all IQ levels. It would require that the environmental effects "imitate" the effects of genetics. --Rikurzhen 17:16, Apr 19, 2005 (UTC)

W is a group of white children, B is group of black chilren. They have the same genetic intelligence, environment lower IQ for the blacks to 85. You pick out some of those who score 110 in both groups and look at the IQ of their siblings. These IQ will regress to means of the races which will be lower for the blacks, but this is entirely due to environment. Ultramarine 17:28, 19 Apr 2005 (UTC)

Regression to mean means the genetic + environmenal mean, not the genetic mean. Ultramarine 17:31, 19 Apr 2005 (UTC)

sorry ... edit conflict ... oh, i misunderstood your question. the regression data is a corroborated prediction of the hereditarian model, but it is essentially neutral with respect to the culture only model. it could be an odd culture effect, but it certainly isn't the prima facia prediction of a culture-only theory. supports genetics, but neutral for culture. --Rikurzhen 17:32, Apr 19, 2005 (UTC)

maybe we should break the other evidence section down into corroborted predictions that are neutral vs contrary to environment -- or put the point-counterpoint into a table for comparison -- something to clean it up --Rikurzhen 17:55, Apr 19, 2005 (UTC)

A table would be interesting if you have the time. Or simply moving the bullet points into ordinary paragraph form. Ultramarine 18:04, 19 Apr 2005 (UTC)

Don't use one test to attack all tests

The present text of the article makes no sense to me. It says that since one test from Long Island was not graded reliably, this somehow proves that no progress has been made in reducing cultural bias. That is not a logical or rational argument. Are all IQ tests, SATs, SAT IIs, and other tests somehow racist or culturally biased because one group on Long Island made a test that wasn't graded correctly? RK 17:38, Apr 20, 2005 (UTC)

RK, I believe that section has deteriorated since that reference was added. Check this PDF on page 24 if you want to fix it. --Rikurzhen 17:54, Apr 20, 2005 (UTC)

I don't have a problem with you changing the text or making additional claims. My problem is solely in the fact that we have a section that seems to show a solid trend, from many sources, that is then contradicted from a single exam! Ugh! This is not logical. While I'm going to bed now (already past my bedtime) I will take a look at the link you mention tomorrow. In the meantime, please do not think that I have a problem with you adding information or context! RK 02:45, Apr 21, 2005 (UTC)
RK, I think you misunderstand; the Gottfredson paper discusses an attempt to reduce selection bias that failed. It is an example of the trend. I think an editor misunderstood that point, and I think I've restored it now. I think that you and I actually agree. --Rikurzhen 04:06, Apr 21, 2005 (UTC)

Many studies of cognitive ability have shown that racial groups differ in measured intelligence.

I'm not unduely attached to that intro sentence, but it is at the very least accurate (note "measured") and is well endorsed by people on both sides of the debate described in this article, including the American Psychological Association, etc. --Rikurzhen 05:05, Apr 21, 2005 (UTC)

The role IQ gradually becomes weaker with age and after school. ???

That seems like an over-statement (and happens to not be written from NPOV). Would it not be more straight-forward to say that personality and other factors than IQ also play a substantial role in life outcomes? There is a large section discussing the practical imporatance of intelligence in that article. --Rikurzhen 21:39, Apr 22, 2005 (UTC)

Well, try changing it to something better. Ultramarine 21:48, 22 Apr 2005 (UTC)
ah, okay. thought i might have missed something. --Rikurzhen 21:59, Apr 22, 2005 (UTC)