This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
This article was formerly listed as a good article, but was removed from the listing because Far too unstable for GA; far too much remaining POV dispute
Race and Intelligence or Race and Behaviour?
I moved this back from the main page, as I think this would be more appropriate in an article about race and behaviour. It seems to add little to the points already made, and lengthens an already extremely long article. --Ramdrake 14:14, 21 May 2006 (UTC)
Some other psychological traits have also been found to vary significantly in distribution. Chinese-American children have been found from birth on to be calmer and more easily comforted when upset than European-American children, and when older to be significantly less active, irritable, and vocal.[1] Blonde German children have been found to exhibit significantly higher fear and behavioral inhibition than other native German children.[2] A neural dopamine gene has been found to vary substantially in worldwide frequencies, with bearers displaying greater novelty-seeking behavior and being at increased risk for attention-deficit hyperactivity disorder.[3] However, group differences in tests of cognitive ability have been the subject of more attention, as they are seen by some as one of society's most pressing problems[4].
I reinserted the previous, much shorter and concise paragraph in its stead. --Ramdrake 14:14, 21 May 2006 (UTC)
It seems like you're basing this on a categorical distinction between cognitive ability and the behavioral traits discussed here. Intelligence and these traits, like extroversion, can be approached as mental, behavioral, or physical (neural/genetic) phenomenon, which just represent different levels of analysis. Morever, the traits discussed above, such as introversion and the ability to concentrate on intellectual problems, are all likely associated with intelligence, meaning they are quite relevant. The existence of proven personality differences is, furthermore, very notable because it's strongly denied by many commentators in this area. While it can be debated whether or not some of the above examples should be mentioned in the text, there is certainly no grounds for removing these examples from a footnote, which doesn't lengthen the text of the article. (This paragraph replies to all 3 of the points you raised).--Nectar 12:52, 26 May 2006 (UTC)
You're still mixing behavioral studies with intelligence (IQ) studies. As far as I'm concerned, whether this is added to a footnote or to the main part of the text, you are still adding material to an already overburdened article. This would be great as a basis for "Race and Behavior", but as is, I don't think it is relevant to studies about race and intelligence, except maybe for drawing a portrait that looks more and more like racial profiling. And we ALL know that's the last thing we want. I think this article is better off with the previous version. And yes, intelligence can certainly be construed as an aspect of behavior. I wouldn't mind having some psychometric data in an article about behavior. But the reverse just doesn't hold - behavior is not a facet of intelligence, it's the other way around. --Ramdrake 23:16, 26 May 2006 (UTC)
That categorical distinction is quite justified - the article is race and intelligence, not race and personality. The existence of such differences is relevant, but only in so far as it relates to intelligence. If you can't substantiate "the traits discussed above...are all likely associated with intelligence" then that is not a defence. Certainly it is relevant if some personality traits vary by race, as this is analogous to intelligence varying by race, but it is an aside, not an integral part of the (overlong) article. --Coroebus 13:00, 26 May 2006 (UTC)
One could view race and intelligence as a sub-topic within race and behavior/personality, but the area that the latter is discussed under is probably most often race and intelligence - the most prominent example of the larger category - rather than being a comparable area in itself. Race and intelligence as the dominant area within race and personality/behavior can be seen in that the latter phrases combined yield 2% of the google hits for the former.
Introversion/extroversion is of interest in relation to cognitive behavior,ctrl f situational demands and introversion does correlate with academic performance,ctrl f introverted which should be of interest as an IQ-related outcome. Learning disorders' impact on education and test-taking should similarly be of interest as an IQ-related outcome.
Anyway, that's fine if this area is dealt with as an aside. --Nectar 15:46, 26 May 2006 (UTC)
[Responding to Ramdrake]
1. These traits are relevant to an article on intelligence and intelligence-related outcomes (see my above comment).
2. That intelligence differences are part of a larger picture of behavioral differences is absolutely essential to discuss. Think of this as a summary of the larger field, something which is standard practice.
3. Given this relevance, it's further noteworthy because it is strongly and nearly universally denied among critics and media accounts.
4. Given these points, discussing just 3 examples in a footnote cannot be argued to burden the article in terms of length (footnotes are optional for the reader).
5. Removing material because it constitutes statistics about races ("racial profiling") is not within Wikipedia policy. --Nectar 00:33, 27 May 2006 (UTC)
Misleading Presentations
I intend to add some remarks to the Bell Chart and the Gap Chart, including the suggestion from Rikurzhen 07:01, 21 May 2006 (UTC) above. As I said earlier, as an African American admitted to MENSA, I have no grudge against I.Q. tests. I do find it disturbing that my own I.Q. does not appear to show up in the Black Bell Curve or Black Curve in the IQ gap chart. The graphs give the impression that the maximum black IQs are respectively 135 and 118 and the minimum white IQs are respectively 55 and 67. Such obviously misleading graphs really should be removed from the article. Readers should not be required to assume that a curve line extends beyond an edge on a chart. A reader should not need a statistics background to accurately see the basic message from such a simple graph in such a controversial article.
I also intend to move the warning message to the top of the article so that readers will not have to scroll down to realize that this page is under dispute. These are obvious misrepresentations that need to be corrected. If anyone would like to undo these small changes, please explain here after you have reverted the edits. I feel strongly about this and I intend to file a dispute with the administrators if the changes are reverted rather than constructively edited. These are the first in a series of changes that I will discuss here as I have time. Elliott Small 15:26, 21 May 2006 (UTC)
After making the aforementioned edits, I have decided not to move the warning message since the edits appear to make the article more constructive and potentially worth reading, especially as additional positive conributions appear. Also, I am not sure whether the message has its current location because that is a Wikipedia convention. Elliott Small 17:08, 21 May 2006 (UTC)
The text you added had the effect of a warning label, which is inappropriate and unnecessary. I've tried to integrate your concerns into the text. However, placing an emphasis on ranges is inappropriate as thea actual ranges are unknown. I've restored the emphasis on averages, which is where the debate actually lies. I can give a fuller account if you don't get my reasoning. --Rikurzhen 18:25, 21 May 2006 (UTC)
Thanks for your time and effort. I also wondered about the fact that we do not yet have the range information. I was hoping that some persons with time might eventually provide data on the ranges. I suppose we can just wait until someone gets around to it or I have time to research it myself. I just added four words this time. In the introduction, I changed "normal" to "entire." In the graph, I added "for all groups." Elliott Small 00:27, 22 May 2006 (UTC)
Reverted to Edits by Rikurzhen and Elliott Small to Avoid Misleading Presentation
I just revert the edits made by Nectar. Rikurzhen and I went back and forth on this to reach a consensus on how to prevent the misperception that there is a limit to maximum IQ for any racial or ethnic group. Without this revision, this article is slants toward irrationally denigrating some groups. Elliott Small 05:01, 22 May 2006 (UTC)
1. People know what "average" means. 2. The article defines average anyway twice in the intro: once in the main text and once in the image caption. 3. There is no reason to repeatedly define "average" every time scores are brought up. Jahiegel makes some related comments here.[2]--Nectar 05:21, 22 May 2006 (UTC)
Jahiegel's analysis of my edits are basically apt as is Nectar comments about averages. I'll have to think some more about whether there's any merit to or grounds for talking about upper limits and ranges. For what it's worth, I did a quick simulation and found that on average you can expect that the brightest White person will score about 10 (or more) points higher than the brightest Black person given a very simple model. --Rikurzhen 06:43, 22 May 2006 (UTC)
I am going to revert these changes again and I will insert the one change (re: native americans) made after the last revert from my revert. It appears that we might have to settle this with a formal dispute. Of course, there is merit in talking about ranges. A simulation cannot yield facts about ranges. Only data can do that. It is illogical and misleading to emphasize averages without admitting ranges. For example, without ranges the article indicates that there is never a white person who scores higher than every Asian, etc. SAT correlates well with IQ and is even used for admission to MENSA. A few years ago, I interviewed a black African for admission to Harvard. English was his third language, yet he scored a perfect 800 on the verbal exam, as well as high 700's in math. I met interviewed other black African students with scores in a similar range. They are obvious examples of black persons who at mininum are brighter than nearly every white American. Leaving out ranges gives the false impression that some groups will never have the brightest person in the world. When I have time, I am going to add a formal section on ranges. Elliott Small 08:46, 22 May 2006 (UTC)
The vast majority of disputes on Wikipedia can be resolved through discussion. Reverting again would seem to just initiate an edit war. (Keep in mind WP:3RR.)--Nectar
I am aware of those rules. Please remember that Rikurzhen and I made those edits after discussion. Then you were the one who came along and reverted the edits without constructive revision. Anyway, in the interest of discussion, please see my reply to Arbor below and let me know why you think that you can continue to revert those edits on your own even though they are the work of two editors. I look forward to reaching a consensus. I know that you have all worked long and hard on this page before I saw it and reacted with my initial alarm. As the page develops, it might actually be appreciated as a tool for constructive understanding. I appreciate the willingness of this group of editors to be unafraid of facts. We just have to be sure that the facts are as complete as is possible. Elliott Small 05:12, 23 May 2006 (UTC)
I'm not sure what you are getting at, Elliot. The fact that the brightest Blacks significantly outperform the majority of Whites is obvious, and I cannot see how the article tries to be misleading about that—in fact, the introductory graph shows that very well. But you seem to want to say something different, about maximally obtainable IQs. I don't think anybody knows anything about that, so you should be very careful about what you write. In any case, you cannot simply fabricate claims about how populations are distributed at the far ends of the tails. If you can find data, please ignore my reservations and go ahead—that would be very interesting. Arbor 09:05, 22 May 2006 (UTC)
I will explain further. The bell curve and the graph chart give the impression that there is a maximum IQ for each group represented. We all know that that is not what is intended by the charts. However, we cannot expect every reader to have the time to ponder the charts like a statistician. We need new charts or additional charts that reflect the fact that every group can potentially generate the full range of scores from idiot to genius. We do not want Wikipedia to appear to lie by omission. Until we have such charts, the current charts should be explained. For example, many years ago, an organization ranked American colleges and named University of California at Berkeley number one and Harvard number two. The rankings were based on 5 areas. Berkeley was number two in all five. Harvard was number one in four but very low in engineering. If that organization had simply stated that Berkeley was number one without providing that complete information on ranges, the results would have been misleading. This is not a precise analogy, but I believe that it makes the point. Regarding claims, I did say that I intend to add a formal section on ranges. Formal implies structured. That means I will be collecting data. In the meantime, I thought that Rikurzhen and I had already reached a consensus on semantics that do not make "claims," until Nectar reverted our edits without constructive revision. Again, I do appreciate your desire to get this done right. Thanks again for your time and attention. Elliott Small 05:12, 23 May 2006 (UTC)
Talk page FAQ
I think we might be able to improve the efficiency of this page in the long term by including links to archived discussions that may be likely to recur. (along the lines of the above #FAQ:_article_name_change)--Nectar 09:01, 22 May 2006 (UTC)
Also, the search engines appear to have stopped indexing WP talk pages, so I've removed the search link that used to be there.--Nectar 09:01, 22 May 2006 (UTC)
Dumbest article
Since one person's definition may not jibe with the next person's definition of intelligence, no one knows exactly what amounts to intelligence. Religion is in the same state of unknown flux as is "intelligence." Wars have been fought and are being fought over religion. Race and intelligence may be the dumbest article in the Wikipedia because there are 1000s of different assessments of "intelligence." People are not writing about one thing in this article. It is like discussing Catholicism, Hinduism, Islamism, Buddhism, Protestantism, Shamanism, and other beliefs, all together as being one thing when they are actually all different things. People ought to find some other topics to discuss or contribute to other than Race and intelligence. Superslum 15:15, 26 May 2006 (UTC)
This article uses the commonly used scientific definition of intelligence.[3]--Nectar 00:37, 27 May 2006 (UTC)
ranges
from article: The groups have virtually equal ranges of scores. Every group has many individuals who score at both genius levels and retarded levels.
We know that their distributions are overlapping, but I don't know of any reports about ranges. "Range" can have two definitions, and I assume what's meant is a particular span of scores from lowest to highest. Because extremely high IQ individuals may exist without being detected, I imagine it's very difficult to know what the true range is.
The words "genius" and "retarded" are, AFAIK, generally not used. "Highest" and "lowest" could be substituted. --Rikurzhen 00:54, 27 May 2006 (UTC)
I just changed the word "retarded" to "lowest" after reading your remark. I used the word "virtually" since precise equality of ranges is not likely and should change every day as new testing occurs. The Reynolds data used to make the bell curves should also reveal ranges. Do you have access to that data? Anyway, we know that the low end has to be virtually equal because no group is immune to negative genetic mutations and injuries. The high end is evidenced by membership of all races in Mensa. Elliott Small 03:46, 27 May 2006 (UTC)
on the IQ tests given to nationally representative samples, the highest score anyone could get would come from getting all the questions correct. i'm not sure if there are any large studies done with tests suited for high scorers. --Rikurzhen 04:26, 27 May 2006 (UTC)
I think Rikurzhen to be correct with respect to the lack of precision in the terms of genius and retarded, and I think highest and lowest to be superior; I reverted Elliott's change, though (in ignorance, I should say, of this discussion), inasmuch as the edit summary evinced a different purpose, viz., removing an appellative with negative/pejorative connotations. Such reversion was correct inasmuch as Wikipedia is not censored; though, where meaning is equivalent, we'd always prefer to use a less divisive term (the gratuitous use of offensive terms, though only by guideline), here the term retarded appears to be the appropriate technical term (see, e.g., retarded, and thence the DSM-IV and ISM-10 IQ range enumerations, in which retarded is used) and so it should be used, notwithstanding any potential offense. In view of the other objections essayed by Rikurzhen and Elliott, though, I would think it altogether fine that we should remove, from the instant sentence, both genius and retarded; I'll not make such change myself, though, because I can't think of an acceptable formulation other than highest and lowest, and I imagine there must be something better (but perhaps there is not). Joe 04:40, 27 May 2006 (UTC)
We should keep in mind the intro already addresses this misunderstanding regarding averages and ranges that gets made by critics sometimes: "For instance, the disparity in average IQ among racial groups is sometimes mistaken for the idea that all members of one race are more intelligent than all members of another..." --Nectar 05:01, 27 May 2006 (UTC)
Nectar, thanks for pointing that out. I did see it before. However, it is best for the point to made near the beginning of the article so that readers do not start with the wrong impression and decline to read further. Elliott Small 14:51, 27 May 2006 (UTC)
ranges2
"While the distributions of IQ scores among different racial-ethnic groups overlap, being virtually equal in range from genius to retarded..."
If taking other groups into account, this statement should probably be less strong. The ranges of Australian Aborigines (IQ estimated from 29 studies to be 62 [4]) and Ashkenazi Jews (107-115) are likely significantly different.--Nectar 02:05, 29 May 2006 (UTC)
"This research based on averages is grounded in several controversial assumptions."
Isn't all research on groups "based on averages"? This seems to be covered adequately in the "while the distributions of IQ scores overlap" sentence.--Nectar 04:43, 29 May 2006 (UTC)
Nectar, do you have access to the raw data from which the ranges were computed? If so, can you tell me how to see it so that I can know the ranges. Thanks. Elliott Small 05:16, 29 May 2006 (UTC)
Rikurzhen is the one who collected the data. His user page says he's on a wiki break for a couple days.--Nectar 06:06, 29 May 2006 (UTC)
Dropped by. As I pointed out above, there's an a priori reason to doubt that the true range has been determined. Tests deployed widely enough would likely be those that are normalized for a representative population, and thus would not be able to distinguish between very high scorers. The SAT and the AFQT are probably some of most widely administered tests that will tell us about the Black-White gap, where you have 10k-100k people taking it each year. Surely there are a number of Americans of every ethnic group that get a perfect SAT of AFQT score, but that probably doesn't distinguish the highest IQs. However, Nectar's point about thinking more broadly than the four US groups is probably the real deal breaker on discussing ranges. Making a number of simplifying assumptions, the AUS Aborigine 96th percentile for IQ is the American Jewish 4th percentile. Almost certainly these groups have different ranges. --Rikurzhen 06:32, 29 May 2006 (UTC)
One way of verifying whether or not different ranges exist would be to look at the accomplishments of members of any given racial group. Nobel prize winners might be an excellent group to look at, and there have also been systematic attempts made to evaluate the IQs of the most brilliant men and woman to have ever lived. Perhaps a check for mathematicians, physicists, or other high-achievers would be a good way of falsifying or at least throwing doubt on the "different ranges hypothesis" by demonstrating that low-average-IQ populations do have members above a certain threshold. (However, this is probably more of a scientific than encyclopedic exercise, and it might be difficult to agree on a methodology which would be acceptible to most scientists.) Harkenbane 19:31, 29 May 2006 (UTC)
Resurrecting Racism
"Gil-White was fired from UPenn, but not for these views. The character may be controversial, but this is no reason to exclude this reference."
According to Gil-White, part of the reason he was fired was that he was regarded by some in the psych department as a "problem teacher",[5] which he believes centrally refers to his treatment of IQ research:
...the historical origins of racism against black people in the United States. . . especially might inconvenience the senior faculty in the psychology department. Why? Because IQ research is taught in the UPENN psychology department as if it were a legitimate science, but Gil-White documents for his students that the entire IQ literature was built on a series of frauds by psychologists who were also major propagandists for the American eugenics movement.[6]
Gil-White regards his site as a conspiracy theory site,[7] and his article is consistent with the above description, misunderstanding most of the topics in this area, with statements such as:
If there are no human biological races, then they can’t have particular traits—any traits, whether this be a level of intelligence or something else—because races, as such, do not exist.[8]
The article sections on intelligence similarly make claims that have no relation to expert opinion, whether on the pro or con sides:
"IQ-testers claim to have shown that so-called ‘black’ people are less intelligent. This is a fraud."[9]
There's no reason to include this conspiracy theory article when its treatment of the subject matter has already been rejected by Gil-White's psych department, and is written by a controversial author who is non-notable in this field.--Nectar 00:51, 31 May 2006 (UTC)
I don't know, while some of his claims (e.g. about race) are wrong, they are also widespread misconceptions, some of his other claims (e.g. about funding, or history) are not wrong. I don't think that you have established that his views have been rejected by his psychology department, and appear to know enough about the guy to cast doubt on his lack of notability. We are happy to use references to people like Lynn, someone whose work is highly dubious in my opinion, so we have to be very careful about picking and choosing who we reference.
Looking a bit harder, I can't find any evidence online that he was fired for these views, the reference above suggests they weren't happy with his focus on the history of IQ testing and eugenics (which isn't quite the same thing as arguing that he is wrong) - it seems more likely he was fired (or didn't have his contract renewed) for his truly crazy conspiracy theories about Sept 11th, Israel, Serbia etc. --Coroebus 11:03, 31 May 2006 (UTC)
RR2
1. According to Gil-White's conclusion (this is contingent upon whether or not he is a reliable source here), his department's objections over his course were indeed mainly due to his thesis - the same thesis of his online book - that the field of intelligence research is fraudulent:
"...The objection to this course must therefore be sought elsewhere. But one will not have to look far. The UPENN psychology department to this day teaches IQ research as if it were science, so Gil-White was effectively a whistleblower in his own department."[10]
2.We can say that significant portions of the book, such as the argument, constituting 5 chapters, that races can't vary in traits, *1 are confused to the point of being largely discardable. The question isn't whether some other parts of Resurrecting Racism (RR) are not inaccurate, but whether such a work as whole can be considered noteworthy and appropriate. The standards required are proportional to the size of the claim, and RR's claim is truly extraordinary: the entire expert community in intelligence research - both critics and proponents - are frauds.
For such a bold claim, RR largely avoids discussion of modern intelligence research, seemingly limiting itself to long-settled historical issues and to comparing the positions of Robert Sternberg and journalist Jon Entine. (RR appears to be under the amusing impression that Sternberg is on the pro rather than con side of intelligence testing, IQ heritability, and race and intelligence, and basically calls him a racist. *2) The modern intelligence literature in race and intelligence is certaintly not discussed, though Rushton's race and behavior book does come up. Some of the simple errors occuring throughout the online book that result from the author's lack of familiarity with the topic are amusing, such as attacking Linda Gottfredson for being a sociologist who comments on intelligence research,ctrl+f sociologist when in fact she is a long-standing expert in that sub-field (and Gil-White, in contrast, has very little business commenting on the field). Other errors, such as making a substantial mistatement on the heritability of IQ, are more serious. *3
3.Consistent with the unprofessional quality of the book, the author doesn't appear to have ever published on intelligence research, aside from his webpage and what seems to be a 1995 letter to the editor in Skeptic Magazine arguing against Vincent Sarich's support of the Bell Curve in an earlier issue. RR doesn't seem to have been reviewed or referenced in any way by professionals, except perhaps indirectly via the rejection of the author's thesis by his department, which makes it seem that Wikipedia is unduly promoting this online book. *4 Someone putting up a webpage doesn't mean we have an obligation to link to it. If RR is a poor man's The Mismeasure of Man (Gil-White acknowledges the similarities), than wiki-linking to that article would be a better alternative. --Nectar 12:49, 2 June 2006 (UTC)
1"...Showing that there is neither a black nor a white race already requires that IQ tests cannot show the ‘black race’ to be less intelligent than the ‘white’—since there is no black race, and there is no white race."[11]
2ctrl+f long. Ironically, Gil-White and Sternberg actually both state the influence on their positions their life histories have had, performing poorly when young but finding those measurements to be very unpredictive in their cases.
3 RR informs readers the IQ heritability estimate of .80 is known to be fraudulent and leaves it at that, stating in effect that IQ heritability is in any case not significant,ctrl+f .80 but this is the estimate for adults widely given in sources such as the APA reportctrl+f .75 and in the behavior genetics textbook Plomin et al. 2001 harvnb error: no target: CITEREFPlomin_et_al.2001 (help) (.80). (There are other significant errors in the linked-to RR paragraph; ask if interested.)
4 Lynn on the otherhand, has been making prominent, albeit sometimes highly controversial, contributions to intelligence research for 30 years (see, e.g., Discover mag 1982).
I don't have massively strong views on including the guy since he's clearly a bit bonkers, but I do object to the reference that his department has somehow debunked him, in particular, contrary to your point 1, the main objections to his course seem to have been "your description of how you will approach the last several topics of the course seems to have little to do with the psychological issues involved and instead to hinge on political interpretation of historical facts. The committee feels that this focus may not be appropriate for a course in psychology", which is not, as far as I can see, an assessment of his scientific view of IQ testing (he may not have a scientific view of IQ testing, but that is beside the point). Your point 2 I'm happy to accept as an argument that the guy's stuff is just poor quality. With regards to your point 3, I am concerned that this article is already over reliant on the within-community view of IQ testing, rather than that of the wider scientific community, so again, I don't think it is necessarily a valid objection that the guy doesn't publish on IQ research. (as an analogy, much EP research is considered very poor by those outside of the field, that those outside of the field don't publish in this area dos not reflect on the validity of that opinion). i.e. if you think the reference is crap, delete it, but I'd prefer your reasons were to do with quality, not dubious appeals to authority. --Coroebus 13:26, 2 June 2006 (UTC)
Re: Lynn, my point is that his scientific research is at times, not just controversial, but also highly dubious in methodology, something I've pointed out elsewhere. Relevant to my EP comparison - if you think a field is rubbish you don't work in it. --Coroebus 13:30, 2 June 2006 (UTC)
1. My point 1 does state it's contingent on him being a reliable source there, meaning we have to accept a reduction in some of his credibility in order to argue his department doesn't reject his thesis. (That's fine if we're talking about larger issues, rather than just about him.) Anyway, that quote does seem to suggest they don't view his argument as being entirely legitimate, stating it doesn't deal with the psychological issues and "instead [hinges] on political interpretation of historical facts."
3. The point about this author writing from outside of the field is really just that the authority of the work has to derive from its quality, rather than from the status of its author in the field.--Nectar 22:01, 16 June 2006 (UTC)
Actually, as was said above, the criticism leveled at him was about his political views, not his scientific ones. While some of his opinions may be wrong, I still think this reference is a worthwhile inclusion in this article as a counterpoint. I say we reinclude it. --Ramdrake 13:33, 17 June 2006 (UTC)
The link is of pretty poor quality though. Things like this, this, or this are going to present a much better counterpoint than such a paranoid rant from someone who isn't massively credible, Jay Joseph or Stephen Jay Gould do a much better job than Gil-White. --Coroebus 12:25, 23 June 2006 (UTC)
Subscores
Re: "Groups vary significantly in their IQ subtest profiles...The Asian pattern of subtest scores is found in fully assimilated third-generation Asian Americans, as well as in Inuits and Native Americans" section.
Should we not be more careful about making these bold claims based on much less evidence than the more hedged claims we're making for overall IQ, particularly lines like "Concordantly, the professions in which these populations tend to be over-represented differ", which while not actually saying 'consequently', is coming dangerously close. --Coroebus 11:03, 4 June 2006 (UTC)
This section does need some more work or referencing, including for some of the reasons you point out above[12]. The line about professions, though, seems pretty mild. It's known that people tend to go into professions related to their strengths, and individual ability scores are predictive of those outcomes.
I agree that the discussion of subscore profiles should be moved to its own section and expanded to cover more groups.. probably a subsection of the Average gaps among races section. I have some references that I'll add in a bit. --Nectar 13:52, 6 June 2006 (UTC)
here are some refs for East Asian visual skills being greater than verbal skills: --Rikurzhen 06:41, 7 June 2006 (UTC)
Georgas, J., Weiss, L. G., van der Vijver, F. J., and Saklofske, D. H. (2003). A cross-cultural analysis of the WISC-111. In J. Georgas, L. G. Weiss, F. van der Vijver and D. H. Saklofske (Eds.). Culture and Children’s Intelligence. Amsterdam: Academic Press.
Ishikuma, T., Moon, S., and Kaufman, A. S. (1988). Sequential-simultaneous analysis of Japanese children’s performance on the Japanese McCarthy scales. Perceptual and Motor Skills, 66, 355–362.
Kaufman, A. S., McLean, J. E., Ishikuma, T., and Moon, S. B. (1989). Integration of literature on the intelligence of Japanese children and analysis of the data from a sequential-simultaneous perspective. School Psychology International, 10, 173–183.
Lesser, G. S., Fifer, G., and Clark, D. H. (1965). Mental abilities of children from different social-class and cultural groups. Monographs of the Society for Research in Child Development, 30, no.4.
Lynn, R., and Hampson, S. (1986). The structure of Japanese abilities: an analysis in terms of the hierarchical model of intelligence. Current Pychological Research and Reviews, 4, 309–322.
Lynn, R., and Hampson, S. (1987). Further evidence on the cognitive abilities of the Japanese: data from the WPPSI. International Journal of Behavioral Development, 10, 23–36.
Lynn, R., and Song, J. M. (1994). General intelligence, visuospatial and verbal abilities of Korean children. Personality and Individual Differences,16, 363–364.
Lynn, R., Hampson, S., and Bingham, R. (1987). Japanese, British and American adolescents compared for Spearman’s g and for the verbal, numerical and visuo-spatial abilities. Psychologia, 30, 137–144.
Lynn, R. (1996). Racial and ethnic differences in intelligence in the United States on the Differential Ability Scale. Personality and Individual Differences, 20, 271–273.
Lynn, R., Pagliari, C., and Chan, J. (1988). Intelligence in Hong Kong measured for Spearman’s g and the visuospatial and verbal primaries. Intelligence, 12, 423–433.
Moon, S. B. (1988). A Cross Cultural Study of the Kaufman Assessment Battery for Children with Korean Children. Ph.D. thesis, University of Alabama.
Stevenson, H. W., Stigler, J. W., Lee, S., Lucker, G. W. Kitanawa, S. and Hsu, C. (1985). Cognitive performance and academic achievement of Japanese, Chinese and American children. Child Development, 56, 718–734.
United States Department of Health, Education and Welfare. (1971). Intellectual Development of Children. Washington, D.C.
Yee, L. Y., and La Forge, R. (1974). Relationship between mental abilities, social class, and exposure to English in Chinese fourth graders. Journal of Educational Psychology, 66, 826–834.
(Re)move bell curve picture?
I would like to suggest removing the bell curve picture from the top of the article, as it immediately pulls the interested reader in and suggests a connection even to the most objective viewer, an initial reaction which is then balanced throughout the rest of the article. I would suggest presenting the discussion about the data before this picture itself, if others agree I'll move the picture down into the "Average gaps among races" section.
Readers would be right to see a connection in the sense of a correlation, but beyond that, the image caption seems to adequately limit causal interpretation: "the causes and meaning of the different average scores for these groups are debated"--Nectar 17:18, 7 June 2006 (UTC)
Moreover—and this is important—the picture explains better than a thousand words that the IQ ranges overlap. That is the most important early message of this article: yes, there are average group differences between races, but there are also huge differences within races. Many, many casual visitors (on both sides of the political spectrum) do not juggle statistical concepts like "average" and "overlapping Bell curves" as well as the scientifically trained editors on this page, so the image does an imporant job very well. Arbor 17:49, 7 June 2006 (UTC)
Even if the first impression that the picture is supposed to deliver is "there *are* many Hispanic people that are smarter then many other Asian people, but on average Asian people are a whopping 20% smarter then Hispanic people", i still find that a very doubtful first impression to give... Looking at the curve for the Black population, this suggests that 50% of them have an IQ < 85, which would suggest that, on first impression, half of the black population is borderline retarded. The rest of the research in the article does not seem to corroborate those results. To me, seeing an image on the top of an article says "this is the scientific truth, below are the nuances", while after reading the article it turns out that it's really more like "this is one result, obtained under doubtful and non-academic circumstances, below are a lot of nuances". I'll readily admit I have trouble seperating my "keep it scientific" reflexes from my "keep it pc" reflexes, so my apologies if I overreact on this, but presenting people with an incorrect first impression just seems to start this article off on the wrong foot.
There are really no nuances about the distribution of IQ scores -- they may be actually a little to the left or a little to the right, they are probably a little taller/thinner for the minority groups, and they probably have fatter tails than the normal distribution would predict (espcially on the left tail). Almost all of the discussion centers on causual explanations. So to a first approximation, your impressions were accurate. However, you've actually made two errors of interpretation: (1) the IQ scale doesn't have a zero point so an IQ of 90 isn't 10% less than an IQ of 100, and (2) borderline retarded is usually said to be an IQ of 70. --Rikurzhen 16:07, 8 June 2006 (UTC)
Well, Rushton and Jensen's 2005 paper states:
Currently, the 1.1 standard deviation difference in average IQ between Blacks and Whites in the United States is not in itself a matter of empirical dispute.[13]
Nisbett's article from the same year, which has nothing but contempt for Rushton and Jensen's analysis still concedes a substantial gap.
[T]he old estimate of 1 standard deviation in ability scores no longer applies. The gap is substantially less than that at the present time, probably more like 0.6-0.7 standard deviation or approximately 10 IQ points.[14]
If we split the difference between those two estimates we get something like 14 IQ points of difference between white and black. That seems to be about the number the image in question shows so I assume it is a fair representation of current scholarship.
Disclaimer: I have no expertise in this subject, I just stumbled upon this array of articles a couple of days ago and have been idly reading some of the information. I don't know, maybe more disclaimers are needed for the image or maybe it should show slightly different data but I'm reluctant to move it down. These bell curve graphics are iconic for this debate/field-of-study so a sample seems to belong at the top of the article. Haukur 16:21, 8 June 2006 (UTC)
I'd like to point out a comment I made a while ago "I think in the text for the image we need a more explicit declaration of how these curves were made - it looks to me like these are normal distributions calculated from the mean IQs in study X and the theoretical standard distributions of test Y - is that right?" since it seems that all concerned are acknowledging that these are simulated rather than actual curves. --Coroebus 16:22, 8 June 2006 (UTC)
Scientific approach
I removed the "bell curve" picture from the top of the article because errors (not only statistical) are not evidenced and the source article is not available for all on the Web.
- In my opinion, without a computational model of the human mind is impossible to analyze such its serious and, maybe, most complex multi-attribute property as intelligence. This modeling is the current objective of intensive research in Artificial intelligence and in the systemic cognitive science, from MIT and DARPA to quasi all interdysciplinary academic and research centers. IQ tests constituted an important but the past step in this direction. At present, they are useful in neuropsychological diagnoses of mentally retarded persons but this domain is also the field of numerous pseudo- and para- scientific activities.
See Google: 88.900.000 docs on intelligence, modeling, research.
- Wikipedia is very well visible on the Web, therefore a high social responsability and mental maturity of the editors is expected. Absolute neutrality does not exist, but we may expect such their actions which are based on the Hippocrates medical rule: Primum non nocere ("first, do no harm").
- On the other hand the number of references to this article is excessive and many of them are not available for typical Wikiperdia users (?).
- I inserted the the following text:
From the recent research perspective based rigorously on the scientific paradigms/(scientific method), the application of various IQ tests to the intelligence study is seen by many scientists as a pseudoscientific activity, where not commonly accepted and ill definied model of intelligence leads to the results which can be used for different social and political manipulations. Intelligence is a very complex multi-attribute property of human brains dependent on numerous weakly measured/observed socio-cultural factors and individual experiences. Therefore some bellow presented results and opinions do not have clear scientific value related to the subject matter but are evidenced because this kind of research is a part of the human culturehistory and has strong individual and social emotional component. For example, similar studies dealing with "profession and intelligence" do not exist, see Google search but the relation race-intelligence (Google)is also widely discussed in the other Wikipedia sub-article: race and intelligence (average gaps among races).
and I hope that these lines or something similar, remain at the begining of this article.
Last comment to the reverted pages: 17:09, 8 June 2006 Rikurzhen & 18:32, 8 June 2006 Haukurth m (sp.
Intelligence is what you use when you don't know what to do. -- Jean Piaget
unfortunately ... people usually think they know what to do and ... they do it. -- G.Ridman
Dear Haukurth - see your User talk page.
and Rikurzhen - I see that you are too emotionally engaged in this page - this is not your private property - but in such cases usually rational arguments are not useful. - sorry - You have beautiful picture on your User page - my congratulation.
If you're interested in reasonable discussion, the reason for reverting that edit is that it doesn't accurately represent the literature in this area. The default stance is that drastic changes to a section that's been edited by many editors over the past year would require prior discussion.--Nectar 19:21, 8 June 2006 (UTC)
Factual accuracy and neutrality
Note that there is no agreement unless explicitly stated. Ultramarine 12:03, 11 May 2006 (UTC)
(done)
POV: Removed
"Scientist in their publications give operational definitions to the groups that they study and describe how the samples are drawn. The populations used may vary according to the purpose of the study, and there is no universally agreed upon subdivision of the races of man. For example, the very large collaboration of geneticists known as the HapMap project, which used four populations for its first extensive database of SNPs, studied not "Mongoloids" but rather Japanese from Tokyo and Han Chinese. Other parallel studies are examining other racial divisions than the four frequently mentioned groups, and there is no supposition that there is an exhaustive, unique list of the races of man. In most studies of intelligence, race and ethnicity are used interchangably, and individuals are categorized by self-identification."Ultramarine 12:51, 5 March 2006 (UTC)
Nectar, Lulu, and I agreed on a new balance of detail versus conciseness in that section. We reached a consensus. --Rikurzhen 18:12, 5 March 2006 (UTC)
Indeed I do agree. The issues of how HapMap sampling is done, and how it relates to historical racial categories is interesting, but just too far afield to need inclusion in this article. A good article for that material might be in the new Race and multilocus allele clusters. A certain skeptical position over in that article might be useful. But this is not a level of detail we need here.
I know what a bore it can be to follow the changes made to an article, so let me direct you to the edit you're referring to.[15]--Nectar 12:57, 5 March 2006 (UTC)
Hm, you link to an idiom? [16]. Anyhow, Npov certainly requires mentioning this.Ultramarine 21:33, 5 March 2006 (UTC)
The relevance of the Pioneer Fund derives from the argument Tucker makes, that this organization has had a negative influence on science through it's choice of work to fund. The fund is known to not exercise any influence on the work the grants will be used on. Tucker specificially says scientists receiving grants shouldn't be harassed. The idiom link was just in case it's not well known to editors from other countries, though I see that's unnecessary. Do you live in Sweden?--Nectar 00:30, 6 March 2006 (UTC)
The fund is certainly not known for that. Read his book. As this is often mentioned by critics, it should be included. Remember, Wikipedia should include all significant views.Ultramarine 00:35, 6 March 2006 (UTC)
Tucker's criticism of the Pioneer Fund is focused on the fund's agenda. Here's his opening summation:
"if the many grants made by Pioneer ... mask other, less laudable goals, then the fund may be hiding an oppressive political agenda behind the protection of academic freedom."
Where does he conclude that studies accepting funding changes how they should be evaluated?--Nectar 03:30, 6 March 2006 (UTC)
The source of the funding is important and should be mentioned. Again, this is often mentioned by critics and Wikipedia should include all significant views. "When Draper first founded the Fund in 1937, he was looking for "useful science." He was convinced that scientists had the answers he was looking for, but were too timid to admit the truth of race differences, Negro inferiority and the value of eugenics. From the 1960s to the 1990s the Fund has singled out individual academics whose work proved useful in the political struggles against integration, open immigration and other right wing causes. While organizations such as FAIR have received significant funding, preference has always been given to the more general purpose (or multi-purpose) scholarship supporting biological determinism, genetically based race differences, and eugenics."[17]Ultramarine
Yes, we already know that the criticism of the Pioneer Fund is that it's pursuing these political goals through it's funding. What you need to provide before you can imply it in the article is note-worthy critics arguing that studies accepting this funding changes how they should be evaluated. David Lykken argues that's not the case: "If you can find me some rich villains that want to contribute to my research - Khaddaffi, the Mafia, whoever - the worse they are, the better I'll like it. I'm doing a social good by taking their money... Any money of theirs that I spend in a legitimate and honorable way, they can't spend in a dishonorable way." (1984) --Nectar 11:52, 7 March 2006 (UTC)
Respected scientific journals usually requires disclosure of potential conflicts of interest. There is no requirement for evidence that the science is flawed, the mere potential of bias is enough. Ultramarine 06:55, 26 March 2006 (UTC)
Disclosure is already present in that section in the form of "[the pioneer fund is] the largest source of funding for proponents of the partly-genetic hypothesis." Editors on both sides have felt that it's unnecessary to change that section into ad hominems (e.g. "[Gould / Rushton] is a [Marxist / racist].")
Moreover, do we even have any sources arguing this funding creates conflicts of interest? Science writer Morton Hunt (and others) report a different picture: "One could spend hundreds of pages on the pros and cons of the case of the Pioneer Fund, but what matters to me--and should matter to my readers--is that I have been totally free to research and write as I chose. I alerted Pioneer to my political views when making the grant proposal for this book but its directors never blinked."[18]
Even Tucker went as far as to condemn ad hominems against Pioner grantees.--Nectar 06:17, 10 April 2006 (UTC)
Again, respected scientific journals usually requires disclosure of potential conflicts of interest. There is no requirement for evidence that the science is flawed, the mere potential of bias is enough. If there is no problem with the Pioneer Fund, then there should be no problem with mentioning that this is the source of funding? Ultramarine 07:29, 12 April 2006 (UTC)
1. Your point about disclosure doesn't apply because disclosure is already present in that section in the form of "[the pioneer fund is] the largest source of funding for proponents of the partly-genetic hypothesis."
2. This isn't siding with one POV, as the section doesn't name Gould as having the "conflict of interest" of "being a Marxist" (and both sides oppose these things).
3. We can't argue Pioneer funding is a potential conflict of interest without a noteworthy reference making that argument (in fact, we have plenty of references arguing against that, including Tucker), so if a reference can be provided we can continue discussing this point.--Nectar 09:04, 12 April 2006 (UTC)
1. The fact that all the major proponents have been funded from this source is remarkable and should be pointed out. 2. You can certainly add the Gould is a Marxist, if you want. I consider him an unimportant critic. 3. Again, there is no need for evidence of conflict of interest. That the Pioneer Fund may be biased is obvious considering its history, and Wikipedia allows obvious arguments. Otherwise it would just be a collection of quotes. Ultramarine 06:01, 14 April 2006 (UTC)
Your argument appears to be that Pioneer funding biases studies. That's a novel interpretation that, because it goes against the statements from defenders and critics alike, is not an obvious argument.--Nectar 13:13, 14 April 2006 (UTC)
Can you name one opponent of the genetic hypothesis funded by them? That the Pioneer Fund support a certain view is obvious considering its history. That that funding may influence research is well-known, which is why respected journals require disclosure, regardless of there has been any evidence of misconduct or no. Again, Wikipedia allows obvious arguments. Otherwise it would just be a collection of quotes. Ultramarine 13:28, 15 April 2006 (UTC)
We have many statements, even from Tucker, that the fund provides cash and then doesn't care what happens after that. On the other hand, we have your opinion that people are being bribed. The default position is to go with the published statements available, which all contradict your interpretation. That's enough to settle the question, but if we were to ignore it, your argument that disclosure is not present would still not apply, as the section already states that the fund is the largest source of funding for this research. --Nectar 13:48, 15 April 2006 (UTC)
This section is badly worded and needs cleaning up. --Coroebus 06:52, 16 May 2006 (UTC)
Are you referring to both paragraphs or just certain parts?--Nectar 19:50, 20 May 2006 (UTC)
2
POV: Counter-arguments excluded (Also Original researach) See the added graph from a subarticle. Family social-economic variables (education shown) are positively correlated with SAT scores. However, the score gap between races persists at all levels of family income and parental education. For example, in 1995 Black students from the highest income group have lower average SAT scores than White students from the lowest income group. Not mentioned is the counter-arguments from the subarticle: "IQ is correlated with economic factors. Blacks and Hispanics suffer poorer economic conditions than Whites. It has been suggested that the effects of poverty are responsible for some or all of the IQ gap. However, in the American Psychological Association report Neisser et al. 1996 harvnb error: no target: CITEREFNeisser_et_al.1996 (help) argue that economics cannot be the whole explanation. First, see the discussion in "Shared and nonshared environmental effects" below. Second, to the moderate extent that IQ and income are related, it appears that IQ determines income, and not the other way around (Murray 1998 harvnb error: no target: CITEREFMurray1998 (help)). (Note there are many other potential environmental factors beside income.) Third, there are gaps in SAT scores that are slightly smaller but still persist for individuals with similar family income and parental education. This stability has suggested alternative explanations:
Some argue that Blacks are discriminated against such that they must have a higher or at least equal intelligence in order to achieve the same socioeconomic status (SES) as Whites. One should then expect that Black children should have a higher or equal IQ compared to children from Whites with the same SES. That they score lower on SAT tests can thus be interpreted as evidence for strong adverse influence from environmental factors different from SES or from SES factors other than income and parental education, like systematic discrimination discouraging school and achievement motivation and learning or cultural differences in nutrition like duration of breastfeeding.
It is possible that Black and Hispanic parents achieve higher SES with lower intelligence; perhaps by having (on average) greater amounts of a compensating character, or through affirmative action. However, affirmative action has lts largest effect on young people newly employed with lower income.
Another alternative explanation is that by comparing the SES of parents to the intelligence of their children, the score gap shown here reflects regression towards different average racial scores from one generation to the next; a partly-genetic origin of intelligence differences would predict this effect.
SAT scores correlate fairly well with IQ scores but they are not the same and may measure different things." Ultramarine 13:03, 5 March 2006 (UTC)
(1) The observation that the B-W gap is not due to simple differences in education or income is part of the consensus statement, yet is difficult to understand if you don't see it directly. (2) The majority of these "counter-arguments" don't appear to be supported by citations, nor is a "counter-argument" needed for a consensus conclusion (that's uncontested in the scholarly literature). Moreover, it's important to understand why such exotic explanations as "being a caste-like minority" or "genetics" would even be proposed in the first place when simple SES differences would be the obvious explanation. --Rikurzhen 18:12, 5 March 2006 (UTC)
Here is another consensus statement: "what little there is fails to support the genetic hypothesis." The current text implies that SES is totally irrelevant. You simply dismiss all possible contribution from SES with the graph and "nor can they be explained by simple differences in socio-economic status." Npov requires including the above arguments and also studies supporting a partial role for SES.Ultramarine 00:12, 6 March 2006 (UTC)
Sure, if such arguments exist. I've never seen them. Those items listed above appear to be mostly original research or merely alternatives to SES. --Rikurzhen 00:19, 6 March 2006 (UTC)
The above arguments are logically obvious and should be included. The paper "Socioeconomic status modifies heritability of IQ in young children" finds that the role of the environment is more important in poorer families. Again, the article now tries imply that SES has no role which is extremely POV.Ultramarine 00:31, 6 March 2006 (UTC)
No, they're just random original research musings. I remember when you wrote them, before I was fully acquainted with the meaning of WP:NOR. Simple SES differences -- between rich and poor or between those with graduate degress and those without highschool diplomas -- do not in themselves account for the IQ gap. The gap is as large at the top as at the bottom. Whatever the heritability of IQ (be it zero or 100%), has no implications for this conclusion. Some sort of complex race-SES interaction may exist, which is not a simple SES difference. If this isn't clear, then we should make it clear. Sweeping this consensus conclusion under the carpet does a disservice to understanding all of the sophisticated theories which try to account for the gap. --Rikurzhen 00:46, 6 March 2006 (UTC)
No sources are needed for obvious logical arguments. This is the view of the arbcom. Also, I see no paper that has used your arguments regarding SAT, so this seems to be your original research. Ultramarine 00:55, 6 March 2006 (UTC)
1+1=2 is an obvious logical argument. Some argue that Blacks are discriminated against such that... is an unattributed statement. What argument about SAT? Are you asking whether the SAT is an IQ test? See Frey MC and Detterman DK. Scholastic assessment or g? The relationship between the SAT and general cognitive ability. Psychol Sci 2004. An simple editorial statement defining what simple SES differences impliles would be fine, a block of text of made up examples is inappropriate for summary style. --Rikurzhen 01:02, 6 March 2006 (UTC)
No, I want a study that uses the above statistics regarding SAT scores as evidence against a SES explanation for racial differences in IQ.Ultramarine 01:15, 6 March 2006 (UTC)
Murray 2005 harvnb error: no target: CITEREFMurray2005 (help) mentions SAT in addition to IQ I believe (or maybe it's Murray and Herrnstein 1994 harvnb error: no target: CITEREFMurray_and_Herrnstein1994 (help)), but the SAT data is a mere convenience as it is available on the internet and I can draw a GFDL licensed figure from it myself. --Rikurzhen 01:19, 6 March 2006 (UTC)
So this is actually your original reserach. Murray 2005 harvnb error: no target: CITEREFMurray2005 (help) certainly does not use SAT scores as you have done to draw sweeping conclusions regarding SES.Ultramarine 01:26, 6 March 2006 (UTC)
The conclusions of the consensus statement are that IQ differences (however you measure it, be it WAIS or the NAEP or the SAT) are not due to simple differences in SES (as they don't go away when you control for SES nor are they variant across SES). You can find such figures drawn for IQ in {AYref|Jensen|1998b}} for example. The SAT data and graph idea come from [19]. --Rikurzhen 01:31, 6 March 2006 (UTC)
The original research should be removed. You can of course use Jensen's data to redraw his figures. Using IQ data rather than indirect SAT scores would be far preferable and remove one of the objections.Ultramarine 01:38, 6 March 2006 (UTC)
The SAT is an IQ test, a fact which is veriable as per I citation I mistakenly gave you above. And as you've kindly pointed out, it's a short logical step from IQ having a property, and SAT being identical to IQ, to SAT having that property. They merely serve as examples, not as the basis of an original argument. They should stay -- and this is absolutely tangential to the point you were originally making. You're just argument shopping now. --Rikurzhen 01:45, 6 March 2006 (UTC)
The SAT is not an IQ test, but it does correlate with IQ tests (but so do lots of things). Your link doesn't provide any evidence that I can see that the SAT is an IQ test either. --Coroebus 17:45, 14 May 2006 (UTC)
This objection is far more serious than the original, although these objections will still apply if you can find data that supports you and redraws the graphs. The article now uses your own unpublished original research in order to dismiss the role of SES! Ultramarine 01:50, 6 March 2006 (UTC)
This objection is baseless. "IQ test" is not a single thing, but a class of instruments for measuring cognitive abilities. It doesn't matter if it's WAIS, WISC, RPM, WPT, GRE or SAT, it's still a measure of cognitive ability. Your objection is like saying that you can't show example data collected with tape measures if the conclusions were based on data collected with yard sticks. --Rikurzhen 02:23, 6 March 2006 (UTC)
H&M's piece in TNR has a figure which I haven't seen, but I guess would look like the one from TBC (reproduced right). While this graph is more informative about the second part of the no simple SES explanation conclusion (that the gap is somewhere between constant and increasing with increaseing SES), the use of "SES deciles" is sure to lose most readers. Education is a much easier to understand measure of SES and the HOME scores that were probably used here. In this case, it's AFQT scores rather than SAT scores, but I can't see the advantage of preferring one to the other for the sake of a graphical example. One is typlically given to 10th graders, the other to 11 and 12th graders, both are measuring crystallized g: math, reading, analogies, etc. As for the source of using SAT to make this point, as far as I can remember La Griffe was the proximal source, as I modified the figures based on those he published. La Griffe reports that Regrettably, the College Board no longer discloses these data. In 1996, they stopped publishing performance by income and parental education disaggregated by race and ethnicity, which is why we're stuck with 1995 data, which is still the newest data I could find (newer than TBC or Jensen 1998). --Rikurzhen 07:31, 6 March 2006 (UTC)
Wikipedia should not contain original research because the reader may misundertand verifiable research! And SAT scores only have a correlation fo 0.76 with IQ. High, but certainly not the same thing.Ultramarine 10:52, 7 March 2006 (UTC)
(1) The AFQT data is just harder to read. (2) This isn't a NOR violation because La Griffe published the SAT scores in this form to demonstrate the same point. (3) Are you famaliar with the range of correlations between two different IQ tests? The same IQ test given to the same person at different times? --Rikurzhen 02:04, 8 March 2006 (UTC)
You link to a personal website that does not use the same categories or racial groups that you do in your original reserach. If you are going to dismiss SES, you need at least a peer-reviewed article! Again, Wikipedia should not contain original research because verifable research is "harder to read"! Regarding your questions, I guess you are trying to state some argument. If so, state it.Ultramarine 17:54, 8 March 2006 (UTC)
The only argument presented in the graph itself is that the gap in cognitive ability test scores persists at all levels of SES. The argument in the caption, that simple differences in SES can't explain the gap, is attributed to the consensus statement, not La Griffe. As I see it, the only question then is if SAT scores can be substituted for IQ scores. It seems, based on Frey MC and Detterman DK, that SAT scores (even controlled for parental SES) can be considered in this context a sub-set of cognitive ability test scores, and they are of course regularly discussed in race and intelligence (see for example the consensus statement).--Nectar 01:38, 9 March 2006 (UTC)
To summarize, an original research graph is used to dismiss SES. SAT score only have a correlation of 0.76 with IQ, it not the same thing. The counter-arguments mentioned at the beginning are excluded.Ultramarine 16:12, 14 March 2006 (UTC)
The conclusion is that of the APA statement, not a novel conclusion drawn from this graph. The graph is an example, not the basis for the conclusion. The data was reported by the College Board with the contrasts as shown in the graph, and La Griffe pointed out that this is another example of the same pattern seen many times before. There's every reason to treat data from the SAT like any other IQ test. Two different IQ tests only tend to correlate with one another in the .7 - .8 range, which is why Frey and Detterman concluded that the SAT is a defacto IQ test.
A listing of possibly NOR-violating theories about complex explanations for the failure to see a simple connection between SES and the IQ gap are inappropriate for the main article -- there's plenty of non-original material that expands on this conclusion in the sections that follow the graph. --Rikurzhen 22:03, 14 March 2006 (UTC)
Coroebus, I didn't leave a link to the SAT/IQ paper, just the citation. Here's the link. The findings are that the correlation between SAT scores and highly g loaded tests is in the range of .72 to .86 (more than IQ correlates with academic performance, for comparison, and about the same as any two "IQ test" correlate with one another), which indicates that the SAT scores is mainly measuring g. Tests which predominately measure g are called "IQ tests" by convention as a short hand, and there's not really a better definition of "IQ test". Thus, the SAT, like RPM and AFQT which it is highly correlated with, is an "IQ test" in the sense that matters for this issue. A new paper by Jaskson and Rushton (in press) finds a g factor in the SAT scores from 1991, confirming Frey and Detterman (2004). --Rikurzhen 21:31, 14 May 2006 (UTC)
I have to disagree, that ref shows a correlation between the SAT and a factor derived from the ASVAB (which isn't an intelligence test), but the purported 'g' derived from that factor analysis only correlated at .56 with one of the included 'real' IQ tests. I'm not impressed by a correlation between a factor derived from a non-IQ "Vocational Aptitude" test and the "Scholastic Aptitude Test", both avowedly aptitude tests, not IQ tests. Similarly, the APM (an actual IQ test) correlated pretty poorly with the SAT (undocumented restriction of range correction notwithstanding), and neither regression model was particularly consistent with the other, hence the fluid/crystallised intelligence point. The SAT is not designed as an IQ test, and while it does show pretty reasonable correlations with IQ, the correlations are not good enough to make the very strong claim that the SAT is an IQ test, just as correlations between verbal and mathematical subscales are not good enough to claim that a verbal IQ subscale is a test of mathematical IQ! --Coroebus 22:39, 14 May 2006 (UTC)
You're nitpicking (which is what Frey and Detterman said in rebuttal to their critic). All that matters is that the SAT is a good measure of g. What the SAT was actually designed to measure is opaque and largely irrelevant, but the ASVAB turns out to be an excellent measure of g in the form of the AFQT, and so apparently does the SAT. Thus, it should not be surprising to find that SAT regression on SES, stratified by race/ethnicity presents a picture similar to that of the AFQT and other measures of g. --Rikurzhen 22:50, 14 May 2006 (UTC)
I am not nitpicking, I am disagreeing with your incredibly bold claim that the SAT is an IQ test. The AFQT is derived from only 4 subscores of the ASVAB, and is thus non-identical with the first factor from the ASVAB (based on 10 subscales), which was what that paper looked at. You cannot claim that the SAT is a good measure of 'g' based on that reference because it doesn't show very strong correlations between measures of g and the SAT. Indeed, the authors go on to say in their reply that "Fundamentally, we have no way of knowing what the relationship between SAT score and Raven's IQ is outside our range of scores...Indeed, applying regression equations to data beyond the range of the original data set is dangerous and never recommended". All that the study shows is that the SAT correlates to a greater or lesser extent with some IQ tests and with a factor they derive from the ASVAB and call 'g', but that also loaded on 'Auto and Shop Information', 'Mathematics Knowledge', 'Mechanical Comprehension', and 'Electronics Information' (the ASVAB is actually better modelled by three factors, rather than one, including a factor for technical knowledge). --Coroebus 23:30, 14 May 2006 (UTC)
Incidentally, I agree that the SAT scores present a picture similar to what we would expect for IQ - but that doesn't make the SAT IQ, just as educational achievement might have a similar picture, but is still not the same as IQ. --Coroebus 06:59, 16 May 2006 (UTC)
The "incidently" part is the crux of the discussion. If you have a problem with Frey and Detterman's conclusions, then you should publish them. If you think I'm misreading them in a way that's not related to the questino, of whether the SAT can be substituted for the AFQT or some other IQ test in a graph that regresses IQ on SES stratified by race/ethnicity, then I'll grant you that too so long as you don't actually have a problem with the graph. --Rikurzhen 16:24, 16 May 2006 (UTC)
See these two studies disputing the value of Sat scores: [20][21]Ultramarine 17:29, 9 June 2006 (UTC)
it wouldn't matter if the SAT is totally bogus so long as it's correlated with other IQ tests, as it has been sufficnetly demonstrated to be. the conclusion that IQ scores are only weakly correlated with parental SES is widespread, as is the corollary that SES accounts for only a small part of the IQ gap. this is usually visualized by a graph that regresses child IQ on parental SES stratifed by race, producing a chart that looks just like the SAT chart. the SAT is not the source of these conclusions, and so it's only a matter of whether the SAT can be a stand-in to visualize this. it's a sufficent stand-in if it's correlated with IQ, which is. that fact that the SAT is used in the figure is incidental. any test that measures g would have done, but the SAT data is particularly nice because it regresses over common sense measures of SES (education) and because the SAT is a familiar test which is given to very large samples of 17 year olds. --Rikurzhen 17:36, 9 June 2006 (UTC)
These studies show that the SAT is biased. Your original research graph is unacceptable. Use something from the literature. Ultramarine 17:38, 9 June 2006 (UTC)
These graphs were published by the makers of the SAT (obviously) and thus they are not original research. I could just as well use the various regression coefficients and generate a hypothetical graph that would look almost identical, but using real data is better, as it avoids any concern about original research. I granted for the sake of argument that the SAT could be biased, only to show that that wouldn't matter, but of course that it highly unlikely as it would have been replaced by an another unbiased instrument (which are widely available and easy to make) -- there's too much money in it for ETS to be so sloppy as to use a biased instrument (unless they biased the test to lessen disparate impact). --Rikurzhen 18:35, 9 June 2006 (UTC)
Source please for graphs from makers. Regardless, even if you can give one, it is still original research to claim relevance for IQ. SAT scores do not correlate very highly with IQ and may be biased as these studies indicate. Why not use arguments from the literature? Redraw a graph from a peer-reviewed study using IQ scores.Ultramarine 18:59, 9 June 2006 (UTC)
the correlation between SAT scores and highly g loaded tests is in the range of .72 to .86 [22] --Rikurzhen 19:12, 9 June 2006 (UTC)
So certainly not identical. The difference may be explained by the bias in SAT tests, for example. Anyhow, I am not interested in your personal opinons and homemade research on the subject. Read Wikipedia:No original research. Use arguments and findings from the literature.Ultramarine 19:15, 9 June 2006 (UTC)
What part of the peer-reviewed publication I linked to is original research? They conclude that the SAT is mainly a test of g. Which is the only criteria we would be looking for in a study that correlates IQ with SES stratified by race. --Rikurzhen 19:24, 9 June 2006 (UTC)
That article does not have your graph or data. Nor does it state that SAT scores is appropriate when analysing the role of genetic in racial IQ differences. Again, read Wikipedia:No original research. This is not allowed:
It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;Ultramarine 19:28, 9 June 2006 (UTC)
That's not the source of the SAT data, the makers of the SAT published that data, which was reprinted here. The paper I linked (which disproves your unsubstantiated claim to the contrary) shows that the SAT is correlated with IQ. (Which is sufficent reason to use the SAT data to draw an example graph.) Also, note that you're making a straw man argument. The caption of the figure mere claims that: Within individual countries, family social-economic variables (education shown) are positively correlated with IQ scores (SAT shown). However, the score gap between races persists at all socio-economic levels. These kinds of findings have led some to argue that that differences in socio-economic status cannot explain all the IQ gap. There's no mention of genetics. As I stated before, this figure is merely an example of the kind of analysis which is the basis a consensus scientific conclusion. Above, I provied an example of a similar analysis done with a different test, different sample, but fewer groups which looks qualitatively identical to the SAT data graph. There is no novel interpretation going on here. --Rikurzhen 23:21, 9 June 2006 (UTC)
Again, Wikipedia is not the place for your homemade research. You are using your homemade graph to dismiss the importance of SES. You cannot cite any peer-reviewed paper where this graph or this argument using SAT scores has appeared. Therefore, you are violating Wikipedia:No original research:
It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;
Publish your conclusions regarding the role of SAT scores and SES for race and IQ in a peer-reviewed paper and return. Ultramarine 00:10, 10 June 2006 (UTC)
How many ways can I say that no conclusions are being drawn from this data??? It is merely an example of a point which is the consensus view, published and discused widely. --Rikurzhen 00:22, 10 June 2006 (UTC)
The criteria that you're implicitly using would disallow the use of any figures as examples unless they are the very figures upon which a conclusion was originally drawn. One could not, for example, show a random example of the crystal structure of DNA, but only the original figure published by Watson and Crick. That's nonsense. Most original figures are copyrighted, so we have to make due with figures which merely look like the originals. --Rikurzhen 00:25, 10 June 2006 (UTC)
A. The correlation is .72 to .86 between the SAT and a factor created from test which isn't an IQ test. This factor in turn only have a .56 correlation with a 'real' IQ tests.[23] Thus, SAT scores should not be used instead of IQ scores when making claims about IQ scores and race.
B. Studies have shown that SAT scores are biased. Yet another argument for not using SAT scores when making arguments about IQ.[24][25]
C. Arguments excluded:
Some argue that Blacks are discriminated against such that they must have a higher or at least equal intelligence in order to achieve the same socioeconomic status (SES) as Whites. One should then expect that Black children should have a higher or equal IQ compared to children from Whites with the same SES. That they score lower on SAT tests can thus be interpreted as evidence for strong adverse influence from environmental factors different from SES or from SES factors other than income and parental education, like systematic discrimination discouraging school and achievement motivation and learning or cultural differences in nutrition like duration of breastfeeding.
It is possible that Black and Hispanic parents achieve higher SES with lower intelligence; perhaps by having (on average) greater amounts of a compensating character, or through affirmative action. However, affirmative action has lts largest effect on young people newly employed with lower income.
Another alternative explanation is that by comparing the SES of parents to the intelligence of their children, the score gap shown here reflects regression toward different average racial scores from one generation to the next; a partly-genetic origin of intelligence differences would predict this effect.
SAT scores correlate fairly well with IQ scores but they are not the same and may measure different things."
D. Obviously you consider this graph very important. It is an argument against the role of SES as an explanation for racial differences in IQ, one of the more important explanations. As such, it should not be based on homemade research but on peer-reviewed studies. But you have been able to show any peer-reviewed study where this graph, data, or conclusion regarding IQ and race has appeared. Thus, it is your own homemade original research. From Wikipedia:No original research:
"It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;"
E. You state that this is a consensus view, published and discussed widely". But why then cannot you cite any peer-reviewed study using these SAT scores? Use arguments from the literature instead of original research. Redraw a graph using IQ scores instead of SAT scores, if there actually is any such peer-reviewed study. Such a redrawing does not violate copyright.Ultramarine 11:50, 10 June 2006 (UTC)
In the course of the extended thread I have responded to all of the claims that you've merely repeated here. Thus, I assume you have either not read, not understood, or simply ignored my responses.
Let me go over this one more time in a single block so that it is absolutely clear.
First and foremost, you are confusing the order of relations between the conclusion regarding SES and the figure presented. The order of relations is that the conclusions about SES are a a consensus view, published and discussed widely, whereas the SAT data is merely a pretty picture that looks like that kind of data upon which these conclusions were based -- not the actual data upon which these conclusions were based. For the sake of argument, I don't see any reason why I couldn't have drawn in Photoshop a similar figure with the axes labeled "IQ" and "SES" and used that instead. Since there's no reason why I can't use a fake figure to demonstrate a point, I see no reason why I can't use actual data that looks just like the fake figure would to demonstrate the same point.
Now what is the actual conclusion? Here is the relevant text from the APA report:
the Black/White differential in test scores is not eliminated when groups or individuals are matched for SES (Loehlin et al, 1975)
excluding extreme conditions, nutrition and other biological factors that may vary with SES account for relatively little of the variance in such scores
The living conditions of children result in part from the accomplishments of their parents: if the skills measured by psychometric tests actually matter for those accomplishments. intelligence is affecting SES rather than the other way around
it is clear that no model in which 'SES" directly determines "IQ" will do
To imagine that any simple income- and education-based index can adequately describe the situation of African Americans is to ignore important categories of experience
The differential between the mean intelligence test scores of Blacks and Whites ... does not ... simply reflect differences in socio-economic status.
The conclusion described is #6. The data presented in the figure is merely the kind described in #1. This is shown in the figure because it is the one that is hardest to describe in words that will be easily understood by most people. Only a graphic will do for most. But note that the conclusion (#6) is both old and unambiguous. There's no reason why we should be limited to showing the data from 1976 when identical looking curves can be shown that look better and are easier to understand.
I believe my discussion presented so far should have completely answered your objections (D) and (E), which were based on what appears to be a misunderstanding of where the fact that the conclusions were taken as a given and example data was looked for to demonstrate the point, rather than the other way around (as implied by your criticisms).
The arguments presented in (C) are themselves original research. I very much doubt that you will find a reference to support them because I remember the day you wrote (the first two of) them, and the context in which you wrote them strongly suggests that you merely made them up yourself. I wrote the last two, but I realize now that I was also pulling them off the top of my head. Although I suspect that you will find support for the third in Rushton and Jensen's 2005 review, I realize now that I was incorrect when I wrote the fourth. But even more importantly, these claims currently still remain in the sub-article which this section is summarizing (something we should now fix). Obviously, a summary section cannot include all details, and on a subject where there is in fact no actual contention in the literature, there's no need to fill up space by scrounging for some.
Now for (A) and (B). In (A) you are confusing the vehicle (IQ) with the salient construct (g). "IQ" itself is meaningless in this context. What matters are the constructs that IQ tests measure, mostly g. Any vehicle ("IQ" test) that measures the g construct is substitutable for another. And so the finding that raw SAT scores are highly correlated with the g factor extracted from several widely used measures of g is more than sufficient to use it an the example in this context. The question of test bias on the SAT is one that I'm not fully familiar with, but the links you provided do not appear to actually answer the question sufficiently. They appear to address disparate results (not evidence of bias) and culture loading (shown not to be sufficient for bias in IQ tests). Yet even if it could be demonstrated that the SAT is biased (for example, thru predictive validity), that wouldn't matter in this context. The conclusion (#6 from the APA above) never depended on the SAT. The SAT data is merely an example of convenience. (It is convenient because it has data on Asian and Latino scores--matching the other graphs on this page--and because the data is available in a form that we can build a open-source graph from.) What matters is that this particular graph (SAT and education) matches all of the salient features of those graphs that present data from other tests:
a correlation between test score and SES in each race
a score gap between races at the medians
the score of top SES black children is at or below the score of bottom SES white children (in fact, top SES black kids perform better on the SAT in this respect than they do on other IQ tests)
There may be some aspects of the graph that do not mirror IQ data, but there are none that I can notice. I take my failure to see a problem as reason enough to believe that such problems must be minor if they do exist.
I hope you will have read and considered these comments with care. In particular--and I can't repeat this enough--the conclusions come first and the SAT graph comes second and only as an example. --Rikurzhen 20:13, 10 June 2006 (UTC)
"a consensus view, published and discussed widely," What peer-reviewed articles are you referring to? You can not name any. The graph is your original research.Ultramarine 20:29, 10 June 2006 (UTC)
you just didn't read a thing I wrote, huh? if you had read what I wrote, you wouldn't have written that response, so I can only assume that you are ignoring my responses. --Rikurzhen 20:35, 10 June 2006 (UTC)
You have still not cited any peer-reviewed study using your SAT scores. The APA report mentions IQ, not SAT. Using SAT is original reserach. Again, redraw the graph using IQ scors, if you can find a peer-reviewed study that has used them.Ultramarine 21:19, 10 June 2006 (UTC)
3 (done)
Dubious "In a 1987 survey of scholars in specialties related to IQ" The scholars were psychologists, sociologists, cognitive scientists, educators, and geneticists. This information has repeatedly been deleted. Stating "scholars in specialties related to IQ" is at least misleading, sociologists and educators may well know nothing about this. It is implied that this is the opinion of the researchers on IQ and this survey is used as a evidence against the APA consensus statement "what little there is fails to support the genetic hypothesis." Ultramarine 13:10, 5 March 2006 (UTC)
Dubious? I've read their book and at one time manually transcribed the lists of professional organizations from which they sampled experts into the talk page. There's an option for "don't know" on every question to account for a lack of knowledge on most questions. --Rikurzhen 18:12, 5 March 2006 (UTC)
People usually have opinions on lot of things they know little about. Again, the specialities should be mentioned. Let Wikipedia readers form their own opionon.Ultramarine 21:29, 5 March 2006 (UTC)
That's a different issue. We don't have room to write everything in the main article. It's my opinion that this is one of the details that adds less value than the room it takes is worth. (It appears that link in that sentence takes you directly to that list of organizations.) --Rikurzhen 23:24, 5 March 2006 (UTC)
The text misleadingly implies that this is view of IQ researchers and is used as evidence against the consensus statement. Again, the specialities should be mentioned. Let Wikipedia readers form their own opionon. Ultramarine 00:14, 6 March 2006 (UTC)
Education, Psychology, Sociology, and Cognitive Science are the disciplines listed in the sub-article. If you listed those, what people considered IQ experts would be missing? --Rikurzhen 00:21, 6 March 2006 (UTC)
If this statements are so identical, why are you so reluctant to give this more detailed information and let the readers decide for themselves? Ultramarine 00:38, 6 March 2006 (UTC)
I'm sorry, I think I've been confusing this thread with the one below it. (I should have realized this when I wrote That's a different issue.) I have no problem with stating which specialities were surveyed, but it should be pointed out that the authors chose these groups because they considered them experts. --Rikurzhen 01:08, 6 March 2006 (UTC)
POV: Hidden counter-argument Again, regarding this 1987 survery. Hidden in footnotes in one place or not mentioned at all at the other place "Whether this still applies today is unknown." Ultramarine 13:17, 5 March 2006 (UTC)
That counter-argument isn't a counter-argument. It's an editorial statement about our own ignorance of further studies, not a positive (and NOR-violating) statement about there being reason to suspect things have changed. --Rikurzhen
The claims from this survey are prominently and in length discussed at two different places in the article. Obviously opinions in the press can change greatly in twenty years. The age of the study should be pointed out in order to achieve NPOV.Ultramarine 21:24, 5 March 2006 (UTC)
Well it does say the age of the study, or rather the year of publication. It's my opinion that editorial commentary of this kind is best left to the footnotes. We can offer no citations to support the claim that "Whether this still applies today is unknown", it may in fact be known to someone else and just not us, and so it is borderline original research. --Rikurzhen 23:24, 5 March 2006 (UTC)
The text tries to imply that this still applies today. This is pov. No source is needed for an obvious logical statement, otherwise Wikipedia would just contain citations.Ultramarine 00:18, 6 March 2006 (UTC)
Um, it's three sentences written in the past tense with a big blue 1988 at the start of the first sentence. The conclusion that things may have changed is no more an a priori truth than the statement that things may have stayed the same. What is the point of making either statement sentence number 4? --Rikurzhen 00:34, 6 March 2006 (UTC)
5
Original reserach. Regarding studies arguing against the genetic explanation, this is stated without source: "Hereditarians argue that these studies are flawed, or that they do support the partly-genetic hypothesis." Ultramarine 13:19, 5 March 2006 (UTC)
This is a summary section: we don't need references for everything. We can append a partial list of references, including Jensen 1998a harvnb error: no target: CITEREFJensen1998a (help), Jensen 1998b harvnb error: no target: CITEREFJensen1998b (help), Rushton and Jensen 2005a harvnb error: no target: CITEREFRushton_and_Jensen2005a (help), etc... --Rikurzhen 18:12, 5 March 2006 (UTC)
Having look at Rushton's "arguments", I am just amazed that his article could pass peer-review. Did the reviewers fail statistics 101? What is this? "Results from some other types of studies are also consistent with that hypothesis. In her review, Shuey (1966) found that in 16 of 18 studies in which skin color could be used as a proxy for amount of admixture, Blacks with lighter skin color averaged higher scores than those with darker skin, although the magnitude of the association was quite low (r =.10)."!!! Ultramarine 21:19, 5 March 2006 (UTC)
Just so we're clear, you've changed the subject. I assume you now agree that this is not original research. That said, I agree with Charles Murray (and the heart of your own comment) that all of the so-called "direct evidence" is of little or no value at all. From my reading of the genetics literature, skin color is a poor measure of admixture among African Americans. The opinions being summarized in the sentence you quoted from the main article are about criticisms aimed at things like the WWII children born to Black American fathers and the Scarr and Weinberg interpretation of the Minnesota Transracial Adoption study, and other such admixture/adoption/etc studies. --Rikurzhen 23:24, 5 March 2006 (UTC)
Rushton is trying to frame this study as evidence for his position! Your opinion is not very interesting. If the ab~ove is Rushton's argument, then he should have stated it, not claiming that the study supported him! Again, Amazing that the peer review process allowed to pass. Ultramarine 00:24, 6 March 2006 (UTC)
What are you talking about? Perhaps this sentence needs a footnote to point out what it actually refers to. The do support the partly-genetic hypothesis was meant to refer to the Minessota transracial adoption study, not this admixture crap. --Rikurzhen 00:28, 6 March 2006 (UTC)
This is false. These statements follow immediately after one another.Ultramarine 01:54, 6 March 2006 (UTC)
You're not making any sense. The sentence says that so called "direct measures" are flawed (such as the WWII thing) or actually support genetics (such as the MTRAS). What does this have to do with Rushton's statement? --Rikurzhen 02:50, 6 March 2006 (UTC)
as far as I can tell, the conclusion of the R&J 2005 review is that Although the studies of racial hybrids are generally consistent with the genetic hypothesis, to date they are not conclusive. They certainly discuss quite a few studies, pro and con their position, and point out flaws in many of them. I don't see anything incongruent between the what we've described as their views in the main article and what they say in the review. --Rikurzhen 07:49, 6 March 2006 (UTC)
I disagree. As noted above, Rushton and Jensen just grossly misrepresent the studies opposing them. Ultramarine 16:22, 14 March 2006 (UTC)
As with 8 and 9, "I disagree" with Rushton and Jensen is not a concern for a WP article. NPOV takes that kind of concern out of our hands. --Rikurzhen 04:20, 15 March 2006 (UTC)
6
POV: Repeatedly deleted "in Northern Ireland the IQ gap between Protestants and Catholics is as large as that between Blacks and Whites in the US."
Has anyone seen this written anywhere but on that partisan website? --Rikurzhen 18:12, 5 March 2006 (UTC)
If we can include the trade book IQATHWON, then we can include this.Ultramarine 00:15, 6 March 2006 (UTC)
Partisan web sites are the primary example given by WP:RS as unreliable sources. We need some kind of corroboration or alternative. This Sowell text sounds promising and is clearly reliable. --Rikurzhen 00:26, 6 March 2006 (UTC)
It lists its sources. Partisan trade books are equally unreliable.Ultramarine 00:39, 6 March 2006 (UTC)
Non-notable web sites by definition fail the requirements of WP:RS. There are already identifiable mistakes on that page. I see The Scientific Study of General Intelligence (2003), edited by Helmuth Nyborg, has a chapter written by Lynn and refers to Lynn on 57 different pages (42 occuring in chapters by other authors), including reference to a cognitive ability study by Lynn conducted on Africans. Also note the citations of Lynn Rikurzhen listed above: Race and Intelligence: Separating Science from Myth (2001; 4 citations), Race, evolution, and behavior (2000; 61 citations), Race, IQ and Jensen (1980; 15 citations), The G Factor: the Science of Mental Ability (1998; 544 citations), or even Handbook of intelligence (2000; 55 citations). I can't see how Lynn would be a reliable source for a well-reviewed survey of the field like Nyborg's book, but not for this article. Blanket condemnations, at the very least, are out of the question.--Nectar 03:40, 6 March 2006 (UTC)
There were numerous factual errors in the trade book, including direct misreporting of IQ scores. If that can be included, then this can also.Ultramarine 13:24, 6 March 2006 (UTC)
Lynn is an academic who is reliable enough to be highly cited in his field. Your quarrel is with the psychometricians who cite him, and you're welcome to take it up with them. Regarding the known transcription errors, it doesn't seem likely that 100 studies identifying a trend all have transcription errors in the same direction, and nobody has made that claim. --Nectar 14:01, 6 March 2006 (UTC)
Here is what another researcher thinks about Lynn: "Anyone who uses Richard Lynn's national IQ values should be aware that this data is "massaged"; either that, or Lynn doesn't know elementary arithmetic."[26]Ultramarine 05:53, 15 March 2006 (UTC)
Note that Dienekes is not making the highly unlikely argument that 100 studies identifying a trend all have transcription errors in the same direction. Dienekes' argument (about Lynn's old book) is merely that there were transcription errors as high as 2.4 IQ points.
The existence of vastly more important researchers who support Lynn, such as W. D. Hamilton, means blanket condemnations are undeniably out of the question. (Hamilton calls him in a review of another of his books "brave, thick-skinned, and very persistent to swim against. . . popular antirealistic currents," and states "Lynn. . . does an excellent job with the facts."--Nectar 11:26, 15 March 2006 (UTC)
Regarding the very controversial Hamilton, see 15. Dienkes finds 5 errors out 19 original IQ scores in a single study made in Europe. Now the IQ scores in Europe are relatively well known and harder to "massage". What has happened with the IQ scores from the developing world? Ultramarine 08:34, 28 March 2006 (UTC)
[outdent]Can you substantiate your claims about Hamilton's large controversiality?
Lynn's data is just as easily checked whether the original study is from Europe or the developing world. Note that Dienekes couldn't find a pattern to the transcription errors, so "massaged" is merely inventive rhetorical flourish. Even if you add 5 point transcription errors (twice what Dienekes found), all amazingly occuring in the right direction throughout 100 studies, the general trend is very clear. --Nectar 16:20, 28 March 2006 (UTC)
It's very ironic that criticism of Lynn is being tied to this issue as Lynn is claimed to be the ultimate source of the Ireland IQ numbers. On that point however, a news story says Lynn actually finds a 3.5 point gap between Scotland/Ireland and England/Wales.
Professor Richard Lynn said the Scots average IQ of 97 was well below the England and Wales average of 100.5 and on a par with the Republic of Ireland. London and the south-east of England scored top in the UK, with an average IQ of 102. [27]
This is even more reason to want to find independent reliable verification of this 15 point gap claim. --Rikurzhen 06:17, 29 March 2006 (UTC)
7
Dubious "Indeed, even proponents of a partly genetic interpretation of the IQ gap, such as Rushton and Jensen (2005a) and Gottfredson (2005b), argue that their interpretation does not in itself demand any particular policy response" They have in other places argued against for example school integration and affirmative action.Ultramarine 00:43, 6 March 2006 (UTC)
There's a difference between a fact about the world in and of itself demanding a policy response and a fact about the world combined with an ethical preference demanding a policy response. --Rikurzhen 01:15, 6 March 2006 (UTC)
This raises some deep philosophical questions. The text should be changed since it misleadingly implies a policy that Rushton and Jensen actually do not advocate. Ultramarine 01:21, 6 March 2006 (UTC)
I don't know what they advocate, but I know what they and Gottfredson wrote in the papers cited. --Rikurzhen 01:39, 6 March 2006 (UTC)
Their real views have been documented in the literature on the Pioneer Fund, like Tucker's book. Rushton and Jensen certainly advocate very specific policies, so the above statement is incorrect or at least extremely misleadsing.Ultramarine 01:43, 6 March 2006 (UTC)
Are you offering to expand that section for us? --Rikurzhen 02:46, 6 March 2006 (UTC) Wait a second! I just read the half of that sentence that you didn't copy here. It explains exactly what I explained about alternative policy implications coming from the same fact but different policy preferences using the example of affirmative action. I missed this because I was only reading what you copied here, but surely you must have read it when you copied half of that sentence. Isn't that exactly what you claimed was missing? --Rikurzhen 07:21, 6 March 2006 (UTC)
"Indeed, even proponents of a partly genetic interpretation of the IQ gap, such as Rushton and Jensen (2005a) and Gottfredson (2005b), argue that their interpretation does not in itself demand any particular policy response: while a conservative/libertarian commentator may feel the results justify reductions in affirmative action, a liberal commentator may argue from a Rawlsian point of view (that genetic advantages are undeserved and unjust) for substantial affirmative action[80]." This is still misleading regarding what Rushton and Jensen themselves actually advocate regarding policy.Ultramarine 13:47, 6 March 2006 (UTC)
I would consider this the minimal description of this topic. I don't find the topic of individual scientists' policy proposals all that interesting, but you should go ahead and expand that section, which has yet to grow to the level of needing summary style restrictions. --Rikurzhen 19:21, 6 March 2006 (UTC)
8
Dubious "worldwide Black–White–East Asian differences in IQ, reaction time, and brain size" The evidence regarding "worldwide" differences in IQ is weak and the evidence regarding "worldwide" differences in reaction time and brain size almost nonexistent.Ultramarine 00:48, 6 March 2006 (UTC)
The sentence begins "To support these claims, they most often cite". --Rikurzhen 01:15, 6 March 2006 (UTC)
Still dubious and npov requires mentioning this.Ultramarine 13:25, 6 March 2006 (UTC)
As per (9) below, the structure of this section is: salient argument for each position in their own subsection and direct rebuttals limited to a single sentence. If the denial of this claim is a major argument cited for a cultural explanation, then it should be written in that section. I don't believe it is. --Rikurzhen 19:41, 6 March 2006 (UTC)
Ultramarine, here is a selection of studies showing worldwide race differences in IQ, brain size, and reaction time.
Jensen, A. R. and Johnson, F. W. (1994). "Race and Sex Differences in Head Size and IQ". Intelligence. 18: 309–333. doi:10.1016/0160-2896(94)90032-9. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Montie, J. E. and Fagan, J. F., III (1988). "Racial Differences in IQ: Item Analysis of the Stanford-Binet at 3 Years". Intelligence. 12: 315–332. doi:10.1016/0160-2896(88)90029-3. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Naglieri, J. A. and Jensen, A. R. (1987). "Comparison of Black-White Differences on the WISC-R and K-Abc: Spearman's Hypothesis". Intelligence. 11: 21–43. doi:10.1016/0160-2896(87)90024-9. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Osborne, R. T. (1961). "School Achievement of White and Negro Children of the Same Mental and Chronological Ages". Mankind Quarterly. 2: 26–29. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Shuey, A. M. (1958). The Testing of Negro Intelligence. Lynchburg, VA: J. P. Bell. {{cite book}}: Cite has empty unknown parameters: |coauthors= and |month= (help)
Shuey, A. M. (1966). The Testing of Negro Intelligence. New York: Social Science Press. {{cite book}}: Cite has empty unknown parameters: |coauthors= and |month= (help)
Rushton, J. P., Skuy, M. and Fridjhon, P. (2002). "Jensen Effects among African, Indian, and White Engineering Students in South Africa on Raven's Standard Progressive Matrices". Intelligence. 30: 409–423. doi:10.1016/S0160-2896(02)00093-4. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Rushton, J. P., Skuy, M. and Fridjhon, P. (2003). "Performance on Raven's Advanced Progressive Matrices by African, East Indian, and White Engineering Students in South Africa [Electronic Version]". Intelligence. 31: 123–137. doi:10.1016/S0160-2896(02)00140-X. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Lynn, R., Hampson, S. L. and Iwawaki, S. (1987). "Abstract Reasoning and Spatial Abilities among American, British and Japanese Adolescents". Mankind Quarterly. 27 (4): 397–405. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)
Lynn, R. and Shigehasa, T. (1991). "Reaction Times and Intelligence: A Comparison of Japanese and British Children". Journal of Biosocial Science. 23: 409–416. PMID1939289. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Lynn, R., Hampson, S. L. and Bingham, R. (1987). "Japanese, British and American Adolescents Compared for Spearman's g and for the Verbal, Numerical, and Visuo-Spatial Abilities". Psychologia. 30: 137–144. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Reynolds, C. R., Chastain, R. L., Kaufman, A. S. and Mclean, J. E. (1987). "Demographic Characteristics and IQ among Adults - Analysis of the WAIS-R Standardization Sample as a Function of the Stratification Variables". Journal of School Psychology. 25 (4): 323–342. doi:10.1016/0022-4405(87)90035-5. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)
Backman, M. E. (1972). "Patterns of Mental Abilities: Ethnic, Socioeconomic, and Sex Differences". American Educational Research Journal. 9: 1–12. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Buj, V. (1981). "Average IQ Values in Various European Countries". Personality and Individual Differences. 2 (2): 168–169. doi:10.1016/0191-8869(81)90013-1. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Kaniel, S. and Fisherman, S. (1991). "Level of Performance and Distribution of Errors in the Progressive Matrices Test: A Comparison of Ethiopian Immigrant and Native Israeli Adolescents". International Journal of Psychology. 26: 25–33. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Lynn, R. (1996). "Racial and Ethnic Differences in Intelligence in the United States on the Differential Ability Scale". Personality and Individual Differences. 20: 271–273. doi:10.1016/0191-8869(95)00158-1. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Lynn, R. (1977a). "The Intelligence of the Chinese and Malays in Singapore". Mankind Quarterly. 18: 125–128. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Lynn, R., Backhoff, E. and Contreras, L. A. (2005). "Ethnic and Racial Differences on the Standard Progressive Matrices in Mexico". Journal of Biosocial Science. 37 (1): 107–113. PMID15688574. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)
Willms, J. D. and Chen, M. (1989). "The Effects of Ability Grouping on the Ethnic Achievement Gap in Israeli Elementary Schools". American Journal of Education. 97 (3): 237–257. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)
Beals, K. L., Smith, C. L. and Dodd, S. M. (1984). "Brain Size, Cranial Morphology, Climate, and Time Machines". Current Anthropology. 25 (3): 301–330. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Bean, R. B. (1906). "Some Racial Peculiarities of the Negro Brain". American Journal of Anatomy. 5: 353–432. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Harvey, I., Persaud, R., Ron, M. A., Baker, G. and Murray, R. M. (1994). "Volumetric MRI Measurements in Bipolars Compared with Schizophrenics and Healthy Controls". Psychological Medicine. 24 (3): 689–699. PMID7991751. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)
Ho, K. C., Roessmann, U., Straumfjord, J. V. and Monroe, G. (1980a). "Analysis of Brain-Weight .1. Adult Brain-Weight in Relation to Sex, Race, and Age". Archives of Pathology and Laboratory Medicine. 104 (12): 635–639. PMID6893659. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Ho, K. C., Roessmann, U., Straumfjord, J. V. and Monroe, G. (1980b). "Analysis of Brain-Weight .2. Adult Brain-Weight in Relation to Body Height, Weight, and Surface-Area". Archives of Pathology and Laboratory Medicine. 104 (12): 640–645. PMID6893660. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Mall, F. P. (1909). "On Several Anatomical Characters of the Human Brain, Said to Vary According to Race and Sex, with Special Reference to the Weight of the Frontal Lobe". American Journal of Anatomy. 9: 1–32. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Pearl, R. (1934). "The Weight of the Negro Brain". Science. 80: 431–434. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Rolleston, H. D. (1888). "Description of the Cerebral Hemispheres of an Adult Australian Male". The Journal of the Anthropological Institute of Great Britain and Ireland. 17: 32–42. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Rushton, J. P. (1991). "Mongoloid-Caucasoid Differences in Brain Size from Military Samples". Intelligence. 15 (3): 351–359. doi:10.1016/0160-2896(91)90043-D. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Vint, F. W. (1934). "The Brain of the Kenya Native". Journal of Anatomy. 48: 216–223. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Jensen, A. R. and Johnson, F. W. (1994). "Race and Sex Differences in Head Size and IQ". Intelligence. 18: 309–333. doi:10.1016/0160-2896(94)90032-9. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Morton, S. G. (1839). Crania Americana; or, a Comparative View of the Skulls of Various Aboriginal Nations of North and South America. To Which Is Prefixed an Essay on the Varieties of the Human Species. Philadelphia: Simpkin Marshall & Co. {{cite book}}: Cite has empty unknown parameters: |coauthors= and |month= (help)
Rushton, J. P. (1994). "Sex and Race Differences in Cranial Capacity from International Labour Office Data". Intelligence. 19: 281–294. doi:10.1016/0160-2896(94)90002-7. {{cite journal}}: Cite has empty unknown parameter: |month= (help)
Jensen, A. R. (1993). "Spearman's Hypothesis Tested with Chronometric Information-Processing Tasks". Intelligence. 17 (1): 47–77. doi:10.1016/0160-2896(93)90039-8. {{cite journal}}: Unknown parameter |month= ignored (help)
Jensen, A. R. and Whang, P. A. (1994). "Speed of Accessing Arithmetic Facts in Long-Term Memory: A Comparison of Chinese-American and Anglo-American Children". Contemporary Educational Psychology. 19: 1–12. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Lynn, R. and Holmshaw, M. (1990). "Black-White Differences in Reaction Times and Intelligence". Social Behavior and Personality. 18: 299–308. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Lynn, R. and Shigehasa, T. (1991). "Reaction Times and Intelligence: A Comparison of Japanese and British Children". Journal of Biosocial Science. 23: 409–416. PMID1939289. {{cite journal}}: Cite has empty unknown parameter: |month= (help)CS1 maint: multiple names: authors list (link)
Again, extremelyh few studies show "worldwide" differences in reation time and brain size, the US is not the world. Misleading to show studies on IQ. Misleading to show studies done in the US when we are talking "worldwide". Ultramarine 10:48, 7 March 2006 (UTC)
Has anyone actually disputed Rushton, Jensen, et al's claim of "worldwide" differences in writing? (Not that this should stop us from reporting their claim as such.) --Rikurzhen 03:01, 8 March 2006 (UTC)
There has certainly been lots of criticism regarding the worldwide claims for IQ, although there is some evidence for this. Would you please list the studies done in the developing world on reaction time and brain size? Ultramarine 02:13, 9 March 2006 (UTC)
As an example, regarding brain size, critic Zack Cernovsky refers to a study that "showed that cranial size varies primarily with climatic zones (e.g., distance from the equator), not race."[28] (He uses this as an argument against Rushton's contention that differences in brain size are due to varying degrees of r/K selection).
Regarding cross-cultural reaction time, here are 3 example cross-cultural investigations of reactions times, together covering caucasoids, and mongoloids and negroids in their native countries:
Lynn, R. and Shigehasa, T. (1991)(above).
Lynn, Chan, and Eysenck (1991) Reaction times and intelligence in Chinese and British children. Perceptual and Motor Skills.
Lynn and Holmshaw (1990). Black-white differences in reaction times and intelligence. Social Behavior and Personality.--Nectar 11:15, 13 March 2006 (UTC)
Studies on some very limited populations in two nations is not evidence for "worldwide differences". Here is a study disputing the claims.[29]Ultramarine 16:35, 14 March 2006 (UTC)
The three reaction time studies I gave are cross-cultural studies that cover 5 nations. Given that this is the expected result in that it agrees with world-wide cognitive ability and brain size differences, this is enough to show that in the hereditarians argument there's no reason to think Africans in Africa have better reaction times than Africans in the U.S. or Britain, or that Chinese in China have worse reaction times than Chinese in the U.S. or Britain.
Your reference to Cernovsky (the author of the article you link to) appears to refer to this sentence: Jensen's recent claims about racial differences in reaction time are biased and might lack in scientific integrity (Kamin & Grant-Henry, 1987). That's fine, because there are other studies that confirm Jensen's non cross-cultural results. Note that the genetics section has a sentence summarizing the standard criticisms, just as the cultural section does: "Critics of this view, such as Robert Sternberg, argue that these studies are flawed and thus inconclusive or that they support the culture-only hypothesis." There's no reason to state "critics argue their opponents arguments are flawed" after every point in both the cultural and the genetic sections.--Nectar 17:50, 14 March 2006 (UTC)
1. It should certainly be clearly stated how weak the evidence for "worldwide" differences in reaction time and brain size is. 5 nations is cerainly not the world. 2. There is every reason to believe that blacks in Africa may be different from those in the US, the later are not a random sample of those in Africa. A very select group become slaves. Ultramarine 07:36, 12 April 2006 (UTC)
That would first depend on there being a reference to support the claim that the evidence is weak for having only looked at 5 countries (per NPOV), and second the considerations of SummaryStyle imply to me that the number of countries backing Rushton and Jensen's claim (given that it's >1) is not important for a summary section. It would be a point to consider putting somewhere if someone published the criticism that 5 does not support the claim of "worldwide", but SummaryStyle implies that the sub-article (or if it's urgent, a footnote) would receive that detail, as this article can't handle back-and-forth. --Rikurzhen 16:02, 12 April 2006 (UTC)
1. Variation in 5 African, European, and Asian countries does indeed represent global variation (that conforms to expected patterns), even if from a limited sample. These 3 studies are the ones Rushton's Race, Evolution, and Behavior references. Other works, such as Jensen's the g Factor may reference additional studies. 2. African, European, and Asian Americans are related populations to Africans, Europeans, and Asians, which can be seen in many traits.--Nectar 13:29, 14 April 2006 (UTC)
1. Obviously a very small sample from the population in 5 countries is weak evidence. 2. "which can be seen in many traits" Source please! Note that blacks in the US mainly come from a small region in Africa. And again, a very select group become slaves.Ultramarine 06:09, 18 May 2006 (UTC)
9 (Done but note how Rushton and Jensen misrepresent this!)
Dubious or Npov"race differences are most pronounced on tests that are the best measures of g, which also show the highest heritability (see Spearman's hypothesis)" Very misleading, from the subarticle "Dolan and Hamaker 2001 have reanalyzed the data from several previous studies (Jensen and Reynolds 1982; Naglieri and Jensen 1987) that used the statistical method invented by Jensen (the method of correlated vectors) with a more recent and improved method (multigroup confirmatory factor analysis). Their results statistically were consistent with the weak form of Spearman's hypothesis that black-white group differences were predominantly on the g factor. However, their analysis of the data set failed to "establish Spearman's hypothesis as an empirically established fact". They also speculate that "it is possible that the analysis of all available data sets ... will demonstrate that a model incorporating the weak version of Spearman's hypothesis provides the best description of the data."[8] This leaves the validity of Spearman's hypothesis, considered a central justification for the genetic explanation, an unresolved question." Ultramarine 00:59, 6 March 2006 (UTC)
The sentence begins "To support these claims, they most often cite". --Rikurzhen 01:15, 6 March 2006 (UTC)
Wikipedis should not contain incorrect statements on factual topics, even is someone has stated them.Ultramarine 01:18, 6 March 2006 (UTC)
Um, that's one big dose of your opinion about the meaning of the research they cite. I happen to directly disagree with your assement of the data. But that's not important because what you're suggesting is nothing like and in direct contradiction to the meaning of NPOV. --Rikurzhen 01:41, 6 March 2006 (UTC)
Only giving one view is POV. Especially when that view has been shown to be incorrect by other researchers.Ultramarine 01:52, 6 March 2006 (UTC)
This is supposed to be a section describing the arguments made in favor of a genetic explanation. The other section is supposed to describe arguments made in favor of a cultural explanation. A brief sentence of mutual disagreement was included in each section. This is the only way to cover this material in a brief manner. There's a sub-article where full descriptions are made. We can only summarize that article. --Rikurzhen 02:42, 6 March 2006 (UTC)
So the subarticles are used as a POV fork while keeping the arguments from one side in the main article. This is not allowed in Wikipedia. Dubious statements like this should not be allowed without an opposing view in the main article but should be moved the subarticles where all views are presented, if there is lack of space in the main article.Ultramarine 13:45, 6 March 2006 (UTC)
No. Look at the article! There's an entire section called Cultural explanations for positive arguments for a cultural explanation which comes before the section you pulled this quote from. In that section, a single sentence marks the disagreement of hereditarians with pro-culture arguments. The pro-genetics argument is presented in the section titled Genetic explanations, which also includes a single sentence marking disagreement by critics of hereditarianism. There's not enough room in a summary section for describing repeated reciprocal back and forth disagreement, which as Lulu described in a recent thread would lead to infinite recursion. If you think skepticism about Spearman's hypothesis is an important argument for the main article, then it belongs in the culture section. However, the culture section is already swelling in size. We need only the most salient info in the main article presented in the absolutely most concise way possible. IMHO, a paper which concludes that it can neither confirm nor deny a hypothesis and that it needs to examine a larger data set to get statistical power is not germane to the main article. Before the recent expansion, I had the entire pro-genetics argument down to a single paragraph with the added sentence of disagreement, and the entire summary section for casual hypotheses was only three paragraphs long. --Rikurzhen 19:33, 6 March 2006 (UTC)
Spearman's hypothesis is the very cornerstone of the genetic argument. Leaving out this argument is extremely pov. Most of the rest of the article should be deleted before this.Ultramarine 01:41, 10 March 2006 (UTC)
At present, the text says that critics of the partly-genetic hypothesis reject all of the claims made by supports of the hypothesis: Critics of this view, such as Robert Sternberg, argue that these studies are flawed and thus inconclusive or that they support the culture-only hypothesis. If the results of Dolan and Hamaker (2001) are so important, then you should be able to find a critic of the partly genetic hypothesis who says as much and cite them in the section of arguments for a cultural hypothesis. --Rikurzhen 01:51, 10 March 2006 (UTC)
Help! This is what Rushton's writes about this study "The results statistically confirmed the conclusion derived from the method of correlated vectors regarding a “weak form” of Spearman’s hypothesis: Black–White group differences were predominantly on the g factor" Here is what is actually stated: "On the basis of the present, as well as other results (Dolan, 2000), we are convinced that the Spearman correlation cannot be used to demonstrate the importance of g in b-w differences with any confidence." and "It is possible that the analysis of all available data sets (perhaps using an appropriate meta-analytic procedure) will demonstrate that a model incorporating the weak version of Spearman's hypothesis provides the best description of the data. However, until this work is undertaken, we cannot accept Spearman's hypothesis as an "empirically established fact" How can anyone believe anything that he writes??? Ultramarine 02:18, 10 March 2006 (UTC)
10 (done)
Dubious or POV"regression to differing means for different races (No known environmental factor can have this effect)." Even Rushton and Jensen admit that this is possible.Ultramarine 11:16, 7 March 2006 (UTC)
(No known environmental factor can have this effect) is AFAIK false. It's more like, no environmental theory would have predicted this effect, but only in the sibling regression data, the parent-child regression would be predicted I would think. --Rikurzhen 06:45, 8 March 2006 (UTC)
POV: Removal of important contradicting information The article in two different place mentions some studies showing different frequency at different geographich locations for some genes that may be involved in the brain. Repeatedly deleted is that this distribution do not follow the claimed IQ distribution for different "racial" groups and that different populations are likely to utilize different alleles to respond to similar evolutionary pressures. Ultramarine 11:52, 7 March 2006 (UTC)
Why would a reader misinterpret "occur in different frequencies in different global populations" as specifically referring to "different races"? It wouldn't be a big deal, though, to replace that phrase with "strongly biogeographic distributions between Eurasia, sub-Saharan Africa, East Asia, and (indigenous) South America."--Nectar 12:02, 7 March 2006 (UTC)
Obvious POV to exclude this distribution do not follow the claimed IQ distribution for different "racial" groups and that different populations are likely to utilize different alleles to respond to similar evolutionary pressures. Ultramarine 13:36, 9 March 2006 (UTC)
1) Re "obvious POV." The genetic explanation section currently states exactly what the distribution is, so readers can probably see for themselves what the distribution is. If you're talking about the race section, "strongly biogeographic distributions between [continents]" has zero implication of how that distribution might match various predictions.
2) This new study found a microcephaly allele that had been selected for in Africans and not Europeans and East Asians, so the section needs to be updated.--Nectar 15:38, 9 March 2006 (UTC)
For readers not familiar with the topic it will be very difficult to keep in mind and compare the distribution of both these genes and that of claimed IQ. The difference should be clearly pointed out. Ultramarine 07:39, 12 April 2006 (UTC)
In some thread that appears to be archived now, I posted by own calculation of the correlation between the frequency of ASPM or Microcephalin and average group IQ. Microcephalin was signficatly correlated, which belies the claim that they have an obviously different distribution. But more importantly, we're now faced with many dozen candidate neuronal genes which are found at differing frequencies between populations because of selection. It would not be possible (for us) to make any general claims about their distribution. --Rikurzhen 06:40, 18 May 2006 (UTC)
We have to be careful in this section, although noting that genes involved in brain development seem to be under selection pressure, we don't know that these genes actually have anything to do with IQ - so we shouldn't stress any implications for racial differences too much (in fact, a gene like COMT would probably tell us more as that has functional consequences). --Coroebus 07:01, 18 May 2006 (UTC)
I agree. The text currently emphasizes that natural selection has occured and then warns: "However, their effect, if any, on IQ is unknown." If this warning needs to be strengthened in any reasonable way, it should be. --Rikurzhen 07:42, 18 May 2006 (UTC)
Any idea what the link 'above' in ref 13 is referring to? The ref it seems to indicate is dawkins on hamilton. --Coroebus 08:02, 18 May 2006 (UTC)
Maybe that 'above' should be a 'below'. It links to Pinker's discussion of the 'Ashkenzi IQ is due to selection' hypothesis. Should be a link to the blue-eye ~ shyness paper???? --Rikurzhen 08:27, 18 May 2006 (UTC)
I've cleaned up ref 13,[30] merging it with the lower ref it referred to so the relevant studies could be discussed in one place. Is that the footnote you're referring to, Coroebus, or is it another one?--Nectar 12:09, 18 May 2006 (UTC)
12
POV: Undue weight Several of the arguments for genetic position are repeated several times in the article and outside the section where it is supposed to be. Examples include [1] the studies about some genes that may be involved in the brain and [2] the 1987 Survery. (As noted earlier above, at the same time excluding that [3] the distribution do not follow claimed racial differences in IQ and excluding that [4] the age of the survey makes it unknown if the results still apply today) This while at the same time [5] deleting information about the Pioneer Fund from the history section. Or as noted earlier [6] making the subarticles into POV-forks while keeping information from one side in the main article.Ultramarine 19:47, 9 March 2006 (UTC)
(1) It would be hard to not discuss that there are genetic differences between races in the race section.
(2)The 1987 survey is discussed in the expert opinion section and the media portrayal section. It would be hard to not note the results of the 1987 survey in the media portrayal section when comparing the results of the survey with media portrayal. Prior to your edits some time ago, the media portrayal section didn't emphasize the 1987 survey as much.
(3)The genetic explanation section currently states exactly what the distribution is, so readers can probably see for themselves what the distribution is. If you're talking about the race section, "strongly biogeographic distributions between [continents]" has zero implication of how that distribution might match various predictions.
"The Pioneer fund has supported much of the partially-genetic research and has a controversial history with critics arguing that many associated persons have supported racism."
(b) I changed that to:
"Biological perspectives on behavior in general had fallen out of favour following WWII in relation to schools such as behaviorism. The controversial Pioneer Fund played some role supporting biological perspectives until they returned to prominence in the late 20th century. (Neisser 2004 harvnb error: no target: CITEREFNeisser2004 (help)). (See also below)"
The Neisser reference is based on his quote: "Lynn's claim is exaggerated but not entirely without merit: 'Over those 60 years, the research funded by Pioneer has helped change the face of social science.'"
My edit summary was: "if pioneer is included here, this is the impact its funding has had in the history of this and related fields."
(c) I then removed the reference to Pioneer stating "the pioneer fund's impact isn't on the same level as the publication of Jensen's 1969 paper or Gould's MoM."
(6) Discussed above.--Nectar 00:35, 10 March 2006 (UTC)
(2) Note that Snyderman and Rothman had two publications on surveys of opinion about IQ: the 1987 "Survey of Expert Opinion on Intelligence and Aptitude Testing" in American Psychologist, and the 1988 book The IQ Controversy, the Media, and Public Policy. The content of the 1987 survey is re-reported in the 1988 book. It's the 1988 book which is being cited in the Media portrayal section, and the 1987 article which is cited in the Expert opinion section. (I've tried to fix the date in the Media portrayal section a few times, but it always seems to get put back to 1987.) --Rikurzhen 01:04, 10 March 2006 (UTC)
The gene findings need only be mentioned in one place, I do not understand why it is mentioned in the race section which is supposed to state a definition. Similarly the survey should be condensed, maybe to one section about media and expert opinion. Ultramarine 07:57, 12 April 2006 (UTC)
1. The race section covers the discussion about whether or not race exists biologically. That neural differences exist among races clearly has implications for the question of race in the context of race and intelligence.
2. Many of the discussions of race and intelligence in the literature begin with reference to the difficulty of knowing what's what in this area. This is probably considered necessary because most people's prior ideas of this area are based on popular misinformation, as discussed in Snyderman and Rothman 1988 and seen in Sackett et al. 2004. Thus, before discussing whether or not this research has utility, the article currently addresses the popular misinformation on this topic. Along these lines, Gregory Stock's treatment of the subject, for example, begins with:
These are highly charged matters, and they have been the object of enormous debate and misinformation. Whether it is the 1994 bestseller The Bell Curve, in which Richard Herrnstein and Charles Murray argued that genetics is responsible for most of the racial differences in IQ test scores, or the aggressive rebuttals by Stephen Jay Gould and others, the arguments are hard to evaluate and harder still to untangle from the political and social biases of their advocates. (Redesigning Humans: Our Inevitable Genetic Future, 2002, p. 44)
POV: Removed Imagine if something like this war removed when commenting on research showing no adverse health effects from smoking: "In accord with the tax regulations governing nonprofit corporations, Pioneer does not fund individuals; under the law only other nonprofit organizations are appropriate grantees. As a consequence, many of the fund's awards go not to the researchers themselves but to the universities that employ them, a standard procedure for supporting work by academically based scientists. However, in addition to these awards to the universities where its grantees are based, Pioneer has also made a number of grants to other nonprofit organizations, essentially dummy corporations created solely to channel Pioneer's resources directly to a particular academic recipient—a mechanism apparently designed to circumvent the institution where the researcher is employed [31][32].
Although the fund typically gives away more than half a million dollars per year, there is no application form or set of guidelines. Instead an applicant merely submits "a letter containing a brief description of the nature of the research and the amount of the grant requested." There is no requirement for peer review of any kind; Pioneer's board of directors—two attorneys, two engineers, and an investment broker—decides, sometimes within a day, whether a particular research proposal merits funding. Once the grant has been made, there is no requirement for an interim or final report or even for an acknowledgment by a grantee that Pioneer has been the source of support, all atypical practices in comparison to other organizations that support scientific research [33]." Ultramarine 13:59, 11 March 2006 (UTC)
These paragraphs in their present form are just many details about the fund's administrative functioning. Is there a reason to discuss them, or, in other words, are the details in support of an argument or thesis? Tucker's argument appears to be that "Pioneer's administrative procedures are as unusual as its charter." However, a fund having unusual procedures is not itself relevant to this article. An argument about those procedures, such as that they mean Pioneer is racist, would be relevant.--Nectar 14:25, 11 March 2006 (UTC)
Again, imagine that this was research showing no adverse effects from smoking and that this research had been funded by the tobacco industry in this way. Obviously this should be pointed out.Ultramarine 14:49, 11 March 2006 (UTC)
Unless we have a (non-original) argument to report about these details beyond that they're "unusual," as Tucker argues, there's no reason (argument) to include them in the article.--Nectar 15:29, 11 March 2006 (UTC)
Tucker is respectable critic. Obviously his points should be included. Again, imagine another controversial field funded this way. Ultramarine 11:47, 11 May 2006 (UTC)
Yes, but what are his points exactly? That the fund has "administrative procedures. . . as unusual as its charter" ? It appears to be an original interpretation of his writing that the researchers are being bribed. The obvious interpretation would be that the fund circumvents institutions because giving grants to universities - when they accept them - has in the past caused campus protests. The second paragraph just suggests what they care about is whether or not the research will further their cause. --Nectar
14
POV: RemovedWhen Rushton's claimed supporting references were examined they were found to include a nonscientific semipornographic book and an article in Penthouse Forum.[34]Ultramarine 16:08, 14 March 2006 (UTC)
My understanding is that those two non-scientific references agreed with the results of his and others' studies on variation in penis size among ethnicities. Variation in penis size is not discussed in this article.--Nectar 22:27, 14 March 2006 (UTC)
Well, Rushton thinks it is related to IQ. Ultramarine 07:59, 12 April 2006 (UTC)
If we ever reference the characteristics Rushton surveys that appear in the same order as IQ differences, such as primary and secondary sexual characteristics, we can rely on the hundreds of scientific studies he references, rather than on a penthouse article. Problem solved?--Nectar 10:21, 12 April 2006 (UTC)
Eh, Rushton's claims have been thoroughly debunked. For example, the latest meta-analysis find no evidence for his claim that blacks are more psychopathic. Ultramarine 06:06, 14 April 2006 (UTC)
Rushton surveyed the studies available on more than 60 characteristics. The issue is probably more complicated than blanket condemnations. For example: "The World Health Organization bases its specifications for condom width on consumer preference and penis size, citing three studies. Taken together, the studies show significant variations in penis size within all population groups, but also indicate that men of African descent on average have a slightly wider and longer penis size, Caucasian men have a medium size, and Asian men a slightly narrower and shorter size. (WHO)"Family Health International --Nectar 13:40, 14 April 2006 (UTC)
Obviously no one denies that there are physical differences. What is serious is the unscientific methods that Rushton uses, which the above is just one example of. Ultramarine 13:32, 15 April 2006 (UTC)
What exactly is the point that Cernovsky [35] was trying to make by making the claim about a nonscientific semipornographic book and an article in Penthouse Forum? He makes the claim as if it's meaning were obvious, but given the information Nectar has presented here, I'm not sure what it is supposed to demonstrate. --Rikurzhen 21:43, 15 April 2006 (UTC)
As evidence of Rushston's sources it is revealing. Especially as he is the head of the Pioneer Fund. Personally, I would not trust anything funded from that fund until duplicated by independent researchers.Ultramarine 06:18, 18 May 2006 (UTC)
Cernovsky doesn't make this point explicitly (I've never read the original piece by Rushton that includes these references) so I'm not sure it's safe to connect the dots that he doesn't connect for us given what Nectar has indicated here. --Rikurzhen 06:37, 18 May 2006 (UTC)
A more adequate discussion of this accusation can be found in Alfie Kohn's The Brighter Side of Human Nature: Altruism and Empathy in Everyday Life (1990), p. 276, in which he names the article as being Rushton, J. P., & Bogaert, A. F. (1987). Race differences in sexual behavior: Testing an evolutionary hypothesis. Journal of Research in Personality, 21, 529-551. Kohn writes that "the primary substantiation for this 'study'" was "the 1898 diary of a French Army Surgeon and an article that appeared in Penthouse Forum, a pornographic magazine." (Weizmann, Cernovsky's source, apparently considers the whole affair "anthroporn" and "ethnopornography".)
In contrast to this characterization as being based on two non-scientific sources, Rushton and Bogaert's paper reviews on the order of 100 scientific studies. The original analysis they conduct in this paper is on data from Kinsey's Institute for Sex Research at Indiana University, published in 1948 and 1953, though Kohn doesn't mention this. The 1898 ethnographic data, generally relating to the reproductive organs, was collected by a french army specialist in diseases of the reproductive and urinary systems (genitourinary diseases). Rushton and Bogaert use it as "provid[ing] early observations" of data seen in more recent studies, such as those on variation in size of testes and ovaries, angle of erection (based on Kinsey data), ovulation rates, and age of menarche, acceleration of penis growth, breast and pubic hair development, first sexual experience, and first pregnancy. Of the many data points discussed, the only claims I see based solely on the 1898 document are the following:
"In Africa, dances have been invented which emphasize undulating rhythms and mock copulation."
"the placement of female genitals (Orientals front and high; blacks back and low)"
"[variation by race in] salient buttocks, breasts, and muscularity"
variation in vagina and clitoral size
These four claims of course occur in the context of a host of related claims that refer to more recent studies. I can't find any reference to Penthouse Forum in this article.--Nectar 08:23, 19 May 2006 (UTC)
15
POV: Removed
The text states: "Some scientists, including W. D. Hamilton, [24] considered one of the greatest evolutionary theorists of the 20th century," and a long quote in the footnotes. Exluded is " (and controversial, thought that the origin of the AIDS epidemic lay in oral polio vaccines)," Ultramarine 06:19, 26 March 2006 (UTC)
Hamilton's support for the OPV AIDS hypothesis doesn't seem to deserve mention in any summary of his notability that's less than one paragraph long, and it's irrelevant to this topic.--Nectar 07:28, 26 March 2006 (UTC)
POV to have a long statement glorifying him and not the opposite view. Ultramarine 07:42, 12 April 2006 (UTC)
Hamilton is, in fact, a prestigious evolutionary theorist. If a reference discussing his overbearing controversiality can be provided, we can continue this point.--Nectar 10:26, 12 April 2006 (UTC)
His theory is certainly controversial and proved false.[36]Ultramarine 06:02, 18 May 2006 (UTC)
A scientific theory being wrong doesn't constitute a claim that one of the most prestigious recent evolutionary theorists is "controversial".--Nectar 11:48, 19 May 2006 (UTC)
16
POV: Removed
"Speculations about innate differences in group intelligence as an explanation for different cultural achievement have a long history. J. B. S. Haldane claims that the Moors who invaded Europe in the Middle Ages thought the Europeans might be congenitally incapable of abstract thought. Southern Europeans long had had their doubts about northern Europeans -- Cicero warned the Romans not to purchase the British as slaves because they were so difficult to train (Sowell, 1994, p. 156); though Caesar did feel they "had a certain value for rough work," Arguments about Northern Europeans removed, very interesting about this supposedly high IQ group. Ultramarine 07:27, 12 April 2006 (UTC)
That version looks like plagiarism, but we can't not copy Nisbett's wording because we only have the inadequate amount of information he gives us. We would need to check his references. Discussed below: #moors_2.--Nectar 10:33, 12 April 2006 (UTC)
The following copied from another discussion
(1) Let's read sources carefully. Nisbett doesn't say anything about "explanation[s] for different cultural achievement."
(2) Non-scientific speculation of this sort is likely found in most or all cultures. An occurence one thousand years ago doesn't add to contemporary data, though it does provide the insight that cultures have a tendency to make such speculation, implying this could be the case today (in which case the statement should be in the Cultural section). Note that the "British" of Cicero and Caeser's time were a different people than the substantially Anglo-Saxon etc. "British" of today.
(3) "Had their doubts" and "difficult to train" don't constitute a claim about congenital intelligence, and Nisbett doesn't state that.
1. The Arabs and the Roman republic had a more advanced culture than Europe and Northern Europe at these times. Obvious logical arguments are allowed in Wikipedia, otherwise it would be only a collection of quotes.
2. I disagree, it should be in the history section or best in the between nations sections. The inhabitants of England at this time were certainly "white".
3Looking at the context in the original article, Nisbett clearly argues that this is a similar case to that of the Moors. Ultramarine 16:18, 17 March 2006 (UTC)
1. Even if we just compare the Iberian Moors with their neighbors (that is, not counting the existence of the Eastern Roman Empire), it doesn't seem easy to say there was a significant gap in the level of the two civilizations. I'd like to rely on references if we're going to argue that there's a noteworthy history of cultures attributing differences in cultural achievement to innate differences in intelligence. (We're arguing it's a noteworthy trend by including it.) As it is right now, we wouldn't have any examples to support such an argument. (This is regarding whether to include the version at the top of this section or the version presently in the article "Speculations about innate differences in group intelligence have a long history.")
2. In order for it to be in the History section, non-scientific speculation would need to be historically noteworthy. The two examples we have didn't have any noteworthy historical impact. Otherwise, I think we mean the section World-wide Scores rather than Between Nations, because this argument doesn't have to do with the significance of national differences. However, I think the argument under discussion doesn't bear on world-wide differences themselves (the subject of that section), but rather supports the Cultural explanation (because cultures have a tendency to attribute lower innate intelligence to other cultures and which cultures those are changes over time).
3. He may imply that argument without stating it, but there's no support given for such an argument. Other peoples have been considered difficult slaves for characteristics other than innate intelligence deficits.--Nectar 02:19, 18 March 2006 (UTC)
Seeing the referenced passages in the Flynn and Sowell books could solve these problems.--Nectar 02:25, 18 March 2006 (UTC)
End copy
1. That the Arabs had an advanced culture at this time is well known. Here is something online: [37]
2. Cicero and Caesar are certainly famous and noteworthy. Nisbett is a well-known critic. However, I agree that it may be better in the cultural section.
3. Which people and who has claimed this? Ultramarine 17:23, 15 April 2006 (UTC)
1. & 2. I've tracked down references that give more satisfactory details. [added to article]
3. For historical interest, Native Americans and Irish made poor slaves in the Americas,ctrl f haughtyctrl f poor slaves as reportedly did Polynesians,ctrl f slaves and the British made poor slaves in the 1600s Barbary slave trade.ctrl f shiftful--Nectar 07:31, 12 May 2006 (UTC)
That answers none of the 3 objections. How do you know that British and Irish were not considered poor slaves due to low intelligence? Ultramarine 05:49, 18 May 2006 (UTC)
Um, so as to bring this discussion back to earth, presumptively the average IQ of the Ashkenazi increased by 15 points realtive to other Europeans during a period of centuries. It's entirely possible that North-Western Europeans were less intelligent than Greeks and Romans of the classical period. Nutritional and cultural differences alone could have caused such a gap. And OTOH, the opinions of pre-scientific figures from history are necessarily suspect for their correspondence to reality. --Rikurzhen 06:06, 18 May 2006 (UTC)
Yes, this is not evidence that genetic explanation is wrong. But it is interesting that a group that now tests high on IQ then was considered stupid, as indication of IQ or just perception of intelligence can change.Ultramarine 06:14, 18 May 2006 (UTC)
I guess I don't get the point. What was the point Nisbett was making? Do you have a link? --Rikurzhen 06:34, 18 May 2006 (UTC)
The addition I made to the article at the time of my last comment included that the intelligence of the northern Europeans of the time was disparaged by the 3 sources given.--Nectar 12:11, 19 May 2006 (UTC)
17
POV: Arguments selectively hidden in footnotes
"It is well established that within-population genetic diversity is greatest within Sub-Saharan Africa, and decreases with distance from Africa. One study estimates that only 6.3% of the total human genetic diversity is explained by race.[38]" Ultramarine 11:13, 11 May 2006 (UTC)
It's not an argument if everyone agrees on a point. Moreover, those figures are not something that anyone who doesn't already know these facts is capable of interpreting. Unless you're going to append a description of the subtleties of the Fst statistic, then there's no point in describing exactly how much variation exists where -- unless population bottlenecks and demographic history is somehow important to this topic. The actual claims made in plain language are found in the main text as on the basis of that more genetic variation exists within such races than between them and racial traits overlap without discrete boundaries. An Fst of .07-.15 is the genetic counterpart to the observation that most phenotypic variance in single complex traits occurs within familes, not between races -- skin color and a few others being the major exceptions. --Rikurzhen 16:42, 11 May 2006 (UTC)
"Moreover, those figures are not something that anyone who doesn't already know these facts is capable of interpreting." Then explain in the article. Let the reader decide for themselves. When someone starts arguing that some information must be hidden, then we are in dangerous territory. Obviously the information that within-population genetic diversity is greatest within Sub-Saharan Africa has important consequences for sweeping generalizations about IQ in Africa from a few studies in a few locations.Ultramarine 05:47, 18 May 2006 (UTC)
There's a virtually infinite collection of facts that can be claimed to be tangetially related to this topic but which are not actually important (e.g. any sentence from race). In a summary style section, the burden is on you to demonstrate that your addition is a "key" fact. I've already annotated the footnote to explain what the numbers mean -- their common sense interpretation is that on average alleles will be in a 70/30 versus 30/70 distribution between two continental-level populations. --Rikurzhen 05:54, 18 May 2006 (UTC)
Again, obviously the information that within-population genetic diversity is greatest within Sub-Saharan Africa has important consequences for sweeping generalizations about IQ in Africa from a few studies in a few locations, or from the US IQ scores.Ultramarine 06:04, 18 May 2006 (UTC)
1. When is the average IQ of Africans ever inferred from a "few" studies? For Lynn's latest book, the plurality of studies were of Africans. 2. What makes you think that greater genetic diversity indicates greater phenotypic diversity? Chimps are 2x more genetically diverse than the totality of humans while being obviously less phenotypically diverse. Europeans are in fact more phenotypically diverse for many phenotypes (e.g. hair and eye color). Moreover, the gradient of genetic diversity is not profound. (Of course, there are some isolated populations that are quite homogenous, but most Eurasian populations are not much less diverse than most African populations.) The bulk of genetic diversity has no phenotypic impact. Who has claimed that this is kind of diversity is important for race and IQ? --Rikurzhen 06:49, 18 May 2006 (UTC)
tag
8 and #9 appear to be the points still disputed; do people think we can just remove those two sentences until the points are resolved, rather than label the entire article disputed?--Admissions 09:20, 13 March 2006 (UTC)
We cannot remove those two sentences, as they are key information. I would repeat again that personal disbelief in this context is not a basis for NPOV objections. As written, the text is accurate and NPOV -- it's written as x is cited by hereditarians as support for the genetic hypothesis, not as x is evidence for the genetic hypothesis. Even if we could demonstrate on the talk page that they were mistaken, we would still have to write what we've written. Unless some has something else to object to, we appear to be done with the tag. --Rikurzhen 10:03, 13 March 2006 (UTC)
I certainly still dispute all points excepted as stated in them. Ultramarine 00:39, 14 March 2006 (UTC)
You've had time to make 200 edits since conversation stopped here. If you have any reason for the NPOV tag to remain, please bring it up in a timely manner (remember, this article is live). Cheers, Nectar 05:19, 14 March 2006 (UTC)
I tend to agree that the article-level tag is being overused. I continue to have some issues with the article, and it tends to creep in the direction of pro-hereditarian and reductionist POV relatively quickly if Ultramarine, me, or a couple others don't object strongly to many such additions. Nonetheless, slapping an NPOV or fact tag the whole article because of some relatively limited issues goes much too far. The bulk of the article has remained fairly factual and neutrally POV. I think using section-pov tags, at most, is much more appropriate. Lulu of the Lotus-Eaters 06:09, 14 March 2006 (UTC)
Serious issues remains regarding the whole article as noted above. That the opposing side have written the last word regarding some dispute do not mean I agree. However, I agree that maybe it would be better to have specific tags in each section that describes exactly what the dispute is about. Ultramarine 16:26, 14 March 2006 (UTC)
Ultramarine, you've had time to make 250 edits since the last contribution was made to the NPOV discussion. If you have anything more to add, it's welcome. Otherwise, the tag is disagreed with by editors on both sides, and not every disagreement with articles requires a tag.--Nectar 00:41, 25 March 2006 (UTC)
You have had time to correct the errors and pov but have not done so. The disagreement remains. The editors who are pushing this are mainly those who seem to spend all their time editing race related articles from a particular POV. Ultramarine 07:06, 26 March 2006 (UTC)
correct the errors and pov -- Oh come now, that's begging the question. I don't think there are errors or POV problems, and it seems that Nectar is generally of the same opinion. It's not just that I think there are POV problems but disagree with your solution, it's that what you call POV issues I think aren't even POV issues. Disagreeing with a published opinion (so long as its attributed and neutrally described) is not a POV problem, it's completely outside the pervue of WP. The other issue seems to be that you would like to see each hereditarian argument indiviudally refuted, which is simply impossible given space constraints; the sub-article where that is done is actually (slightly) bigger than this entire article. Accepting this constract, I am satisfied with seeing the best arguments for culture and the best arguments for genetics written as they are with a single instance of noting the disagreement. Anyone who wants to read the subarticle can, but otherwise they should just get the executive summary. --Rikurzhen 20:54, 26 March 2006 (UTC)
When editors on both sides of an issue disagree with a tag, it's the responsibility of the claimant to advance his or her own arguments and thereby convince other editors. If you're concerned about editors' POVS, ironically (given the disputes that sometimes have occurred), this article has been written entirely by left-of-center editors, with the exception of a couple of editors who periodically have participated on the talk page.--Nectar 22:31, 26 March 2006 (UTC)
Do not remove the disputed tags until the conflict is resolved. The claim that the group of editors trying to remove the arguments against the genetic explanation is "left-of-center" is strange and irrelevant for factual accuracy. Also, do you know each other in real life so you can make this statement? Looking at your edits, you seems to edit mostly race related articles from a particular POV.Ultramarine 11:18, 11 May 2006 (UTC)
Your understanding of these matters remains unique, and other editors seem to disagree. Maybe we can find some common ground with a The factual accuracy and neutrality of this page is disputed by User:Ultramarine and link to a separate subpage that presents your viewpoints. Clearly, the current modus operandi has come to an impasse. Without involving external arbitration (which does not seem to match the style of the current editors) I cannot see how your "disputed" tags improve the situation. Other suggestions are welcome, of course. Arbor 11:41, 11 May 2006 (UTC)
Please discuss the issues noted above. Ultramarine 11:45, 11 May 2006 (UTC)
Most of the above comment's not worth responding to, but I will say as a supporter of science you should be glad there are users representing the positions held by scientists in the so-called "realist" camp (e.g. Pinker, Hamilton, Ernst Mayr), even if you disagree with those positions. Nobody wants Wikipedia to only represent one side of science. Beyond that, note that the disputed tag is in the same place it's been since the dispute started (above the table of contents).--Nectar 12:03, 11 May 2006 (UTC)
Maybe a little less glad when said supporters of the so-called "realist" camp act as though they own the article. Several points have been raised questioning aspects of this research (the meaning of "g" as a single-number measurement, its relation to what is called intelligence in layman's terms, the representativity of the "world-wide IQ samples" used in most of the research, the very meaning and appropriateness from the viewpoint of genetics of the racial tags as they've been historically defined). These are all valid and are still hotly debated in the world of science. To say otherwise because of one or a few studies doesn't represent reality. The AAA refutes the very meaning of races as we've historically known them [39]. The APA, in its consensus statement on the Bell Curve said about the Black-White IQ difference: There is certainly no such support for a genetic interpretation. These are the facts. For these reasons, the neutrality of this article, in trying to positively substantiate a genetic cause to a measured difference in IQ numbers between racial groups remains very much in question. --Ramdrake 13:18, 11 May 2006 (UTC)
But, Ramdrake, all these points are acknowledged in the article. There is certainly more room devoted to non-hereditary positions (relative to their exposure in the scientific literature) than vice versa; the article "errs on the side" of Sternberg and others (while being very clear about labelling their position as a minority POV—as it should). In any case, almost every POV here is expressed as "X says Y, but Z says !Y" instead of "Y is the truth" or just "Y" (which is what Evolution does, and gets away with it.) So the R&I presentation is the very definition of NPOV. It certainly makes no attempt to hide any controversies—on the contrary, we have whole sections and subarticles devoted to presenting just these controversies. Arbor 13:37, 11 May 2006 (UTC)
Arbor, I am quite sure this is how you honestly see the article. I get another feel when I read the article; that's my honest opinion. You feel the article is balanced as it presents both sides equally. I feel the argument is unbalanced as I don't think the genetic side deserves as much exposure, as I think there are enough flaws with Lynn's research to make it borderline pseudoscience (this is my opinion, I'm entitled to it, and you have every right to disagree with it), and as it is central to this article, the rest logically follows. If one could devise a warning on this article which says both that it is controversial in subject and that its neutrality can also be controversial depending on one's viewpoint, I would support it. I don't know if you think that would be an improvement. --Ramdrake 14:59, 11 May 2006 (UTC)
According to WP:NPOV, NPOV is "absolute and non-negotiable". A concise description of what NPOV means is found at WP:NPoV#The_neutral_point_of_view. It describes this article precisely. There's probably always room for tweaking the language to squeeze every last bit of individual bias out of the writing, but that's not what you've suggested Ramdrake. What you've suggested is presenting one side of the controversy in less than their best light because you think they're wrong and the other side is right. That, of coures, isn't permissible. --Rikurzhen 16:58, 11 May 2006 (UTC)
pseudoscience
This is where our views diverge. Based on the statement of the AAA, and that of the APA, plus the very serious doubts that have been cast on the worlwide IQ data from several sources, I consider the overall study of "race and intelligence" as a pseudoscience. And Wikipedia says this about the policy for what is considered a pseudoscience:
If we're going to represent the sum total of human knowledge, then we must concede that we will be describing views repugnant to us without asserting that they are false. Things are not, however, as bad as that sounds. The task before us is not to describe disputes as though, for example, pseudoscience were on a par with science; rather, the task is to represent the majority (scientific) view as the majority view and the minority (sometimes pseudoscientific) view as the minority view; and, moreover, to explain how scientists have received pseudoscientific theories. This is all in the purview of the task of describing a dispute fairly.
Pseudoscience can be seen as a social phenomenon and therefore significant. However, pseudoscience should not obfuscate the description of the main views, and any mention should be proportional to the rest of the article.
There is a minority of Wikipedians who feel so strongly about this problem that they believe Wikipedia should adopt a "scientific point of view" rather than a "neutral point of view." However, it has not been established that there is really a need for such a policy, given that the scientists' view of pseudoscience can be clearly, fully, and fairly explained to believers of pseudoscience.
If I consider that "race and intelligence" is a pseudoscience as a subject, then I am within the bounds of WP:NPOV to consider both sides should not really be presented as if they were on a par, as if I go by the statements, the genetic-cause view is held by a minority, and there are sufficient issues of science with many of the more important points being presented in the "pro" position to liken this position as "pseudoscience". You don't have to agree with me, but my position is logical and within the bounds of WP:NPOV. --Ramdrake 20:09, 11 May 2006 (UTC)
One cannot begin to substantiate the claim that "race and intelligence" is pseudoscience. From either first principles or appeal to authority this kind of claim is plainly false. (Are you sure you know what pseudoscience is?) Neither the APA report nor the highly-unusual AAA statement supports such a claim. The "mainstream" statement has the greatest number of signatories of any of the various collective statements, and it's conclusions are quite clearly different than yours. Moreover, the Snyderman and Rothman surveys demonstrated that Aruthur Jensen's views are in fact the modal views among ~1000 IQ experts. --Rikurzhen 21:01, 11 May 2006 (UTC)
That being said, the claim that R&I research is pseudoscience is certainly noteworthy and has been made often enough. (For example, The Skeptics Society itself includes a sizeable number of members that believe that claim (as well as others who don't of course.) A chapter in Why people believe weird things is devoted to it and a section in Skeptic dictionary. I myself used to be in that camp, until I found out that I was being lied to and left in disgust.) I am all for having an easily identifiable section that acknowledges this viewpoint and its counterpoints. I have tried to motivate Sceptics to write such a section for more than a year now. Such a section would fit my Controversy proposal below (Accusations of pseudoscience would be a good heading.). Arbor 09:04, 12 May 2006 (UTC)
I would welcome anything that would acknowledge the size and scope of this controversy. Right now, the list of points and counterpoints looks as lively as the description of a curling match (no offense intended). And, by the way, the "mainstream" statement pales in comparison to the collective statement of the AAA, which is 10,000 strong, or the APA statement on the Bell Curve, which states that "There is certainly no such support for a genetic interpretation." Let's please remember that the APA itself has 150,000 members or so. So, fine 1000 IQ researchers say there is a difference in IQ between Blacks and Whites which is at least partly genetic. But a body representing 150,000 psychologists (who must know something about intelligence too) says the difference exists, but there is no evidence it is genetic, while another body of 10,000 experts, this time in anthropology, say that what society defines as races has little or no meaning anthropologically (which isn't to say races don't exist, just to say that if they exist, races do not correspond to ther historical-social construct most people associate them with). So, I would say, yes, maintaining that race and intelligence reseach is pseudoscience is a logically tenable position. It can be proven right or wrong yet, I'll be the first to admit it. I wouldn't mind doing a bit more research on the accusations of pseudoscience to help come up with a section. I think I have a pretty good idea where to start. --Ramdrake 13:58, 12 May 2006 (UTC)
If someone has actually made these claims, I'd love to see them presented. But I don't see them in either the APA statement (as Arbor points out below) or even the AAA statement (which if you read carefully makes very limited claims which seem to be trivially true but unrelated to the contemporary debate). Moreover, please don't mistake the fact that a panel representing the AAA or APA made a statement with the idea that the full membershp of those groups supports the statement. All survey data suggests otherwise (Snyderman and Rothman covering psychologists and Leiberman covering anthropologists), as does -- for example -- the existence of numerous rebuttals printed along the APA report. --Rikurzhen 16:30, 12 May 2006 (UTC)
Please go and re-read the statements. It's clear enough to me, for the APA statement says "there is no such support for a genetic interpretation". And while the AAA statement is more circuitous, it does state that "These facts render any attempt to establish lines of division among biological populations both arbitrary and subjective". Also, while I'm not pretending this is the opinion of the entire membership of both groups, the AAA statements is prefaced with: "We believe that it represents generally the contemporary thinking and scholarly positions of a majority of anthropologists." Basically, it states it is the majority statement.
Just one point: the claim that "R&I is pseudoscience" is not the same as the claim "The genetic explanation is wrong". Not even remotely. The current article bends over backwards in explaining that "genetic component" is a highly controversial explanation, viewed by many people as a topic for ongoing research where the "jury is still out", and at not point does the article claim that this explanation is correct. It is indeed not supported by the APA statement, neither does this article claim that. The fact that there is a gap, on the other hand, is questioned by very few people. But I had the impression that this is all laid out with utmost care in the current article already. If not, I encourage you to strengthen the presentation. Arbor 16:06, 12 May 2006 (UTC)
However, if you couple the observation that "there is a gap", with a broad interpretation of the AAA statement that human populations are not unambiguous, clearly demarcated, biologically distinct groups, i.e. well-defined races as we have historically known them, it must beg the question, what are we observing? Is it of any value, or is it some artefact of our mental perception? I think the question is legitimate, and the answer that this may well be an artefact of how we conceptualize races is a possible logical outcome, although not necessarily the only one. --Ramdrake 19:13, 12 May 2006 (UTC)
I think I understand you. I would be enclined to present the "R&I is pseudoscience" claim mostly based on things like: there is a viewpoint which holds that race does not exist, or at least is not congruent with the socio-cultural definition it has with people in general. There is also a viewpoint which holds that "intelligence" is larger an wider than the abstract construct specialists call "g". Both viewpoints would view R&I research as a pseudoscience, as for them either race as we understand it is not a sound biological construct (does not exist biologically), or intelligence cannot be distilled to a single number. The logical reasoning here is that if you study "scientifically" something that I say does not exist or cannot be measured, you are doing pseudoscience, from my viewpoint. From your viewpoint, it can be very different. Does that make sense? --Ramdrake 16:27, 12 May 2006 (UTC)
No. A scientific disagreement is different than a science/pseudoscience disagreement. Your reasoning looks like original reserach to me, having never seen such arugments in the literature before. --Rikurzhen 16:35, 12 May 2006 (UTC)
Let me find the material from the people who actually called this type of research pseudoscience. And these arguments (or something close to them) have been made time and time again in the literature. Let's start with Stephen Jay Gould, then Francisco Gil-White (psychologist and asst. Professor at U of Penn.)[40], Mark Nathan Cohen[41] (SUNY University Distinguished Professor of Anthropology), Joseph L graves, Jr (Professor of Evolutionary Biology at Arizona State University West)[42]. The last three I found with only about 15 minutes of searching the net. All have called R&I research "pseudoscience", fo reasons very close if not identical to what I was explaining above. So, with all due respect, this claim is neither unsubstantiable, nor original research.
Ah, Rik, but that is because you set the bar too high! I read lots of non-scientific rhetoric, including arguments of (say) Carroll and Shermer that are at least insinuate something like what Ramdrake is getting at. Of course, they never present the argument with such (commendable) clarity, so we cannot really attribute it to them. But maybe Ramdrake can have an extra look? The book would be Why people believe weird things by Shermer and the Skeptic dictionary is already linked from the bottom of the main article. These would be good starting points for somebody who wants to chronicle the "It's pseudoscience!" faction of the Skeptics Society. On the other side stand people like Frank Miele. Arbor 18:14, 12 May 2006 (UTC)
I see. I stand corrected. (In my defense, I don't read outside of scholarly journals, so I would have missed this.) Based on Ramdrake's links, apparently some people claim that race, psychometrics, or both to be pseudoscientific. Let me just add them to the list next to fairies and faith healing ;) --Rikurzhen 18:54, 12 May 2006 (UTC)
Look at it the way it suits you. ;) But in this debate, I am just a layman with little or no access to the scientific journals. Race in and of itself is a legitimate social construct. Psychometrics is also a legitimate endeavor of science. However, the problem may lie when trying to mix both. As one of my professors used to say "How are you supposed to quantify (i.e. measure) quality?" I never found quite the appropriate answer to this question. --Ramdrake 19:19, 12 May 2006 (UTC)
"Quality" is a moral judgment that has no place in science. --Rikurzhen 20:42, 12 May 2006 (UTC)
Phrenologists used to get in heated debates which were argued using scientific terminology. It's possible that people engaged in what they consider an application of science are in fact basing their arguments on something that is ultimately pseudoscientific. That's the argument put forth by Gould and many others which I have felt was under-represented in this article from the first time I read it. "Race and intelligence" research can be viewed as consensus science or pathological science used to reify social constructs. This article is well-cited for the POV that this is a legitimate field of scientific inquiry, but has always felt lacking in the other POV. Like Ramdrake, I don't have the background in this area to counter those who strongly believe in the legitimacy of this field, but this series of articles is presented as if this objection is considered moot by scientists (or at least the subset of "intelligence experts"). I don't believe that's neutral or accurate. Jokestress 19:49, 12 May 2006 (UTC)
There is an intersting debate regarding the precise nature of human genetic variation, and a relatively unintersting attempt to show how this variation isn't described by the word "race" as it is variously defined. (This point is well made in PMID14655871.) There's an intersting debate about the relationship of the statistical construct g to biology and the architecture of the human mind, and an uninterested debate about whehter g is really intelligence or just highly correlated with it. The interesting debates are worth writing about and none of them are discussions about pseudoscience. If there's an appreciable and important literature on the pseudoscience question that I'm not aware of, then someone should work on that. From what I can tell, the pseudoscience charge is only ever made for public consumption in the form of books like The Mismeasure of Man. --Rikurzhen 20:42, 12 May 2006 (UTC)
For those who care, I have dropped a number of references wherein Skeptics discuss R&I-related matters (including making or analysing the charge of pseudosciene) at User talk: Ramdrake. Arbor 13:41, 13 May 2006 (UTC)
I'm honestly not sure how prominent the pseudoscience argument ultimately is. Though Gould's MoM is commonly referred to by supporters as being an attack on pseudoscience, [43] Gould only uses the term twice in the book: once in the introduction to the updated version (p. 21), in which he does use it to refer to "the abstraction of intelligence as a single entity, it's location within the brain [etc.]" and once in the "Critique of the Bell Curve" chapter, in which he uses it to refer to the 19th century work of a person named Gobineau (p.383). Gould and Michael Shermer's Why People Believe Weird Things: Pseudoscience, Superstition, and Other Confusions of Our Time only mention of Jensen is an actual citation of his the g Factor to support their argument (p. 287).
At any rate, the pseudoscience argument appears to generally be based on or occur in the context of false arguments, such as that psychometrics or the work of Jensen have been unanimously rejected by experts (the Snyderman and Rothman survey showed the opposite to be true) or that if race isn't a coherent biological concept, racial differences in intelligence are impossible (as long as socially defined groups are known to be neurogenetically non-identical, cognitive differences are possible).--Nectar 19:33, 13 May 2006 (UTC)
Yes-yes. That is why I myself have never been able to summarise these writings. Originally, that was my ambition: to contribute to this very article by giving an overview of the Skeptic writings. But I have found myself unable to do so, because I have great difficulties in presenting (fairly and honestly) a viewpoint that I have come to disbelieve. Also, I feel lied to by Shermer–Gould–Carroll, whom I trusted in this matter, and that further clouds my judgement. Hence I have encouraged first Jokestress and now Ramdrake (whose viewpoints seem to harmonize with the Gould–Shermer–Carroll faction) to do so and provided them with my references. Arbor 20:59, 13 May 2006 (UTC)
If this can be done in a way that's encapsulated, then it would be a positive addition. Previous talk-page debate along these lines has focused on the removal or rewriting of current material. For example, the censoring of Lynn's IQ meta-analyses, the denial that there is reportedly widespread "acceptance" of g among those who study intelligence, or the claim that debate regarding the relationship of human population structure to "race" a priori invalidates any investigation of "racial" differences WRT biology. I'm certainly not looking to revisit these discussions, and the fact that supporting claims have been made by at least one person is not in question, only the notion that these are premises upon which we can construct this article (and thus reason to exclude material which contradict them). By encapsulated, I mean in the way that "Accusations of bias" and "Utility of research and racism" are treated, which I consider to be a success. --Rikurzhen 21:32, 13 May 2006 (UTC)
Check this paper out.PDF -- it reviews some of the discussion in the literature which closely resembles the arguments found in the 'skeptics' literaure (without actually claiming race/iq research is pseudoscience). what's noteworthy is how cogent James Flynn (by quotation) is at rebutting these claims against the basis of the research. --Rikurzhen 17:14, 16 May 2006 (UTC)
tag2
Upon review, the 17 points raised in this NPOV dispute appear to have, whether through discussion or altering the article, progressed to the point of not requiring a tag. Editors on both sides have weighed in, expressing that they consider the tag to be unnecessary in relation to the points raised. If any editors have an argument otherwise, please bring it up.--Nectar 06:36, 16 May 2006 (UTC)
Sorry, I still don't agree yet. --Ramdrake 12:14, 16 May 2006 (UTC)
Before this degenerates into a straw poll: This is not a vote. Nectarflowed is soliciting arguments. Arbor 18:54, 16 May 2006 (UTC)
You're right, I need to expound: I still think the article is slanted towards the concept that there is no serious opposition to the genetic viewpoint, whereas the opposition is there, both from the science and from the lay crowds. It gives the impression that the arguments presented herein have no serious rebuttal. Irrespective of who's right, there is still a very heated controversy over this endeavor, and my impression is that this article doesn't let that transpire. Therefore, I would argue that this article is still not "neutral". I'm working on a small section that I hope will help, though. I just would like us to avoid putting the cart in front of the oxen. --Ramdrake 19:36, 16 May 2006 (UTC)
If you're correct that the article does not portray the arguments against a genetic cause neutrally, then that's a serious problem. While you're thinking about this, try to distinguish between the topic of whether a meaningful difference exists and the question of what causes the difference. For example, James Flynn and Arthur Jensen agree on all the data and disagree on how to interpret it. For most questions, they would give the same answer except those relating to cause. I happen to think that Flynn's POV is presented well. --Rikurzhen 20:19, 16 May 2006 (UTC)
I would like to refer you to the last comment made by Jokestress a couple days ago here. It's just that I feel the same. --Ramdrake 20:53, 16 May 2006 (UTC)
That's not good enough. As editors, our only mission is to select among and present published viewpoints. Not yours. And not Jokestress'. I myself have several private opinions about much of this, but I have never lobbied for their inclusion (nor even taken them up on the talk page) — simply have no place on this page: no-one else of significance who has voiced them. So my own POV—brilliant and deep and convincing as I think it is—is underrepresented. That does not make this article !NPOV. Arbor 18:01, 17 May 2006 (UTC)
Are you saying that I need to find a published viewpoint that states this article lacks a certain degree of neutrality in its presentation? I don't think that's what you meant to say. This all started when Nectar said that all editors have expressed the view that the tag is now unnecessary. I expressed my view that it is still necessary. When told that just expressing my view was not enough, I explained the reasons for my stance. When told that I should reconsider my view, I brought forth the point that others on this page have expressed just the same opinion (i.e. not all editors think the tag is now unnecessary). Now, I'm being told that pointing to opinions similar to mine on this page is not good enough, that I need to present a published viewpoint. I will reiterate this: I do not support the removal of this tag at this point, and from reading this talk page, I can conclude that one if not several other editors feel the same. I do not think there is consensus to remove yet. --Ramdrake 18:26, 17 May 2006 (UTC)
No. I wasn't clear, perhaps. You need to be more concrete. For example, you need to point to a specific (published) POV that is under- or overrepresented. And you cannot point to Jokestress' POV. You can point to Flynn's, or Sternberg's, or Gould's. You correctly point out that Nectar should not have said "All editors think the tag is unnecessary." He should have said "No editor has been able to produce a reason for keeping the tag that a significant number of other editors agree with." Arbor 06:14, 18 May 2006 (UTC)
Just to clarify, "all editors" would be a misreading of my comment. .. Most discussions on Wikipedia don't require tagging the article as being unreliable.. and that's something that Lulu (who has played a significant role on the critical side of this article) seems to agree on.[44]--Nectar 13:04, 18 May 2006 (UTC)
How about this: using Arbor's material, and some of my own, I am writing a bit about "R&I research as pseudoscience". I've been at it for a bit less than a week now, and I expect to be done in no more than a week. The way I see it, once this part is in place, it should go a good length towards redressing what I find is the problem with neutrality. Just please leave the tag one more week, I'll take the onus of addressing the reason why I feel the tag is presently there. When I come up with the section, I will withdraw my objection to removing the tag. If I can't come up with the section in due time, you can still remove the tag as far as I'm concerned. Considering the tag's been up there for a couple of months now, I don't think one more week will kill anyone. What say you? --Ramdrake 19:53, 18 May 2006 (UTC)
That sounds great. Used like that, the NPOV tag is actually doing something constructive. Arbor 10:42, 19 May 2006 (UTC)
Certainly not agreement and certainly still require a tag. For example, the original research graph still remains.Ultramarine 05:43, 18 May 2006 (UTC)
Unless you have something new to say about #2, that thread appears to be finished. --Rikurzhen 05:51, 18 May 2006 (UTC)
Article protected
Reach a suitable consensus first before engaging in edit warring, please, all of you. NSLE(T+C) at 17:38 UTC (2006-06-09)
Please restore the >1 month old consensus version before protecting. --Rikurzhen 17:40, 9 June 2006 (UTC)
What consensus? What happened to the totally disputed tag?Ultramarine 17:46, 9 June 2006 (UTC)
We continued discussing it in your long absense and agreed that the article was NPOV and accurate. One editor made plans to expand the controversy section, but hasn't followed thru on that. --Rikurzhen 17:54, 9 June 2006 (UTC)
NSLE (or any admin), the established version of the article has been written by many editors over the past year and a half; it's reasonable to expect changes to be discussed prior to being implemented in the article. There are no grounds for instead protecting the new version, and it seems very hasty to protect an article after 2 reversions.--Nectar 00:06, 10 June 2006 (UTC)
Haukurth's change
Haukurth's change is perfectly[;)] fine with me. --Rikurzhen 20:40, 10 June 2006 (UTC)
While I don't consider it perfectly fine (it is, as he himself noted, a bit clumsy), it seems like a fair compromise. I'm for it. --Ashenai 20:42, 10 June 2006 (UTC)
(Edit conflict, I hadn't seen the above when writing the following.) I've unprotected the article but please proceed with care and try to work out a compromise. I can understand what both of you are saying - the image is based on data which was not actually used by researchers and this might be misleading. On the other hand the actual data used by the researchers, as confirmed by several citations, looks much the same so the image is illustrative anyhow. I made an amateurish compromise edit keeping the image but adding a disclaimer to the image test. Please improve (or try something completely different). Keep in mind that many people scan Wikipedia articles by looking at the images and the image texts first. We should try to avoid misleading such people, even in minor ways. Someone who doesn't read the whole article carefully but looks at the image may get the impression that SAT scores are an accepted unbiased measurement of intelligence and have been used to arrive at sweeping conclusions about the IQ gap :) Haukur 20:49, 10 June 2006 (UTC)
That's a very good point. I think the disclaimer is sufficent, but if someone can find a good-looking figure to replace the SAT data that would be fine as well. In reality, all kinds of less than perfect tests are used to measure intelligence and make conclusions about race. For example, Nisbett (2005) treats the NAEP as an IQ test, which it plainly is not (probably r=.5-.6 with IQ tests [45]). --Rikurzhen 20:54, 10 June 2006 (UTC)
Use IQ scores if there is peer-reviewed study that has used them and produced similar results as claimed. Why use inferior and possibly biased data? Ultramarine 21:25, 10 June 2006 (UTC)
Here is an actual study that shows that the gap disappers with appropriate controls.[46] Why is this not used instead of the homemade arguments using SAT scores? Ultramarine 21:29, 10 June 2006 (UTC)
UL, your suggestions is not a compromise, it's is an evasion of my arguments. You should not have simply reverted the compromise changes. Unforuntely, Fryer and Levitt's work has been highly controversial. It does not substitute for the long standing conclusions addressed in the APA report. --Rikurzhen 23:17, 10 June 2006 (UTC)
It is you who evade the arguments. You have still not cited any peer-reviewed article for your source. The APA discusses IQ, not SAT scores. This article [47] shows that the gap disappears when controlling for a small number of factors. Since this is a peer-reviewed study, why not use it instead? Ultramarine 23:27, 10 June 2006 (UTC)
(1) Your argue without references that SAT <> IQ, but according to Detterman's study the SAT is highly correlated with other tests that are good measures of g. I don't know of any other requirement to call something a de facto intelligence test. (2) Fryer and Levitt's study could be discussed, but cannot substitute for the APA conclusions. The primary problem with their study is that as you grow the list of covariates you will naturally shrink the residual gap. They just happened (natch) to choose covariates that have obvious correlates to g. Thus, their covariate is actually cutting into the effect of g. Similar studies were published in response to the Bell Curve which found that a small set of covariates could elimate the correlation between IQ and SES. Murray showed that this was because they controlled for g in his study which used siblings (who share the same SES) to show that the correlation was nearly as strong among people from the same family. Adoption studies are the equivalent to Murray's study of siblings for race comparisons. Levitt's study is comparable to adoption studies, not to the studies which look at simple differences in SES. More to the point, they proudly admit that their results are unusual compared to past studies. Any discussion of their results would have to be couched in those terms. --Rikurzhen 23:37, 10 June 2006 (UTC)
A. The correlation is .72 to .86 between the SAT and a factor created from test which isn't an IQ test. This factor in turn only have a .56 correlation with a 'real' IQ tests.[48] Thus, SAT scores should not be used instead of IQ scores when making claims about IQ scores and race.
B. Studies have shown that SAT scores are biased. Yet another argument for not using SAT scores when making arguments about IQ.[49][50]
C. Arguments excluded:
Some argue that Blacks are discriminated against such that they must have a higher or at least equal intelligence in order to achieve the same socioeconomic status (SES) as Whites. One should then expect that Black children should have a higher or equal IQ compared to children from Whites with the same SES. That they score lower on SAT tests can thus be interpreted as evidence for strong adverse influence from environmental factors different from SES or from SES factors other than income and parental education, like systematic discrimination discouraging school and achievement motivation and learning or cultural differences in nutrition like duration of breastfeeding.
It is possible that Black and Hispanic parents achieve higher SES with lower intelligence; perhaps by having (on average) greater amounts of a compensating character, or through affirmative action. However, affirmative action has lts largest effect on young people newly employed with lower income.
Another alternative explanation is that by comparing the SES of parents to the intelligence of their children, the score gap shown here reflects regression toward different average racial scores from one generation to the next; a partly-genetic origin of intelligence differences would predict this effect.
SAT scores correlate fairly well with IQ scores but they are not the same and may measure different things."
D. Obviously you consider this graph very important. It is an argument against the role of SES as an explanation for racial differences in IQ, one of the more important explanations. As such, it should not be based on homemade research but on peer-reviewed studies. But you have been able to show any peer-reviewed study where this graph, data, or conclusion regarding IQ and race has appeared. Thus, it is your own homemade original research. From Wikipedia:No original research:
"It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;"
E. You state that this is a consensus view, published and discussed widely". But why then cannot you cite any peer-reviewed study using these SAT scores? Use arguments from the literature instead of original research. Redraw a graph using IQ scores instead of SAT scores, if there actually is any such peer-reviewed study. Such a redrawing does not violate copyright.Ultramarine 11:50, 10 June 2006 (UTC)
F. This article [51] shows that the gap disappears when controlling for a small number of factors. Since this is a peer-reviewed study, why not use it instead of the original research? Ultramarine 23:42, 10 June 2006 (UTC)
don't spam the discussion page. someone else will have to play with you for a while. i think my case is complete and compelling. --Rikurzhen 23:51, 10 June 2006 (UTC)
fryer and levitt
the follow up study by fryer and levitt here fails to confirm the patterns observed in their first paper:
The racial achievement gap remains a stubborn reality. Using newly collected data on a
recent cohort from the ECLS, we document substantial Black-White test score gaps in both math
and reading that grow at approximately .10 standard deviations per year that children are in
school. The divergence in test scores relative to Whites is not apparent for either Hispanics or
Asians.
The explanation as to why Blacks are losing ground proves elusive. Fryer and Levitt
(2004) test a wide range of hypotheses, finding some empirical support for only one explanation:
differential school quality across races. When the data are extended to cover an additional two
years of schooling, however, the support for even this hypothesis weakens. We also explore
whether the growing racial test score gap could be attributed to the inherent difficulties in testing
achievement at especially young ages, or the possibility of increasing importance of home inputs
for the development of higher-order thinking, but can provide no compelling evidence
confirming these hypotheses either.
testing intelligence in young children is an important but difficult issue. most of the data discussed in the WP article focuses on adults where IQ has become largely "fixed". however, important causal inferences may be made from studies of children. someone might want to follow up on the issue of the age trajectory of the B-W gap. --Rikurzhen 00:15, 11 June 2006 (UTC)
They also state: "The divergence in test scores relative to Whites is not apparent for either Hispanics or Asians." This follow-up do not deny that the gap can initially be explained by a small number of non-genetic factors. Also, note the importance of publishing the results in peer-reviewed papers. I suggest that you try to publish your research on SAT and racial differences in IQ in a peer-reviewed paper. Wikipedia is not the place.Ultramarine 00:23, 11 June 2006 (UTC)
You realize that the test Fryer and Levitt are reporting on is a test of math and reading, right? Want to wager a guess at the correlation between the scores on a Kindergarden or First grade math test and the g-score from an IQ test? There are papers on the B-W IQ gap in children. I don't have them at hand, but review articles say the gap is ~1 sd from age 2. I addressed your claims about the SAT in the sections above. --Rikurzhen 00:36, 11 June 2006 (UTC)
As noted, their study is very large and recent. As they note, their results are different from earlier studies, possibly indicating that gap is closing. IQ correlates very highly with early school grades, so I find it at least equal in value to using SAT scores.Ultramarine 00:38, 11 June 2006 (UTC)
That the achievement gap has narrowed is the consensus view. See the NAEP graphs on that page. It was narrowed considerably by 1990. --Rikurzhen 00:44, 11 June 2006 (UTC)
simple SES differences
in the SES/IQ figure caption, i originally wrote These kinds of findings have led to the conclusion that simple differences in socio-economic status cannot explain the IQ gap (reviewed in Neisser et al. 1996 harvnb error: no target: CITEREFNeisser_et_al.1996 (help))
this has been changed to These kinds of findings suggest that simple differences in socio-economic status cannot explain all of the IQ gap (reviewed in Neisser et al. 1996).
I'm not sure that this change is appropriate. Here is the relevant text from the APA report:
the Black/White differential in test scores is not eliminated when groups or individuals are matched for SES (Loehlin et al, 1975)
excluding extreme conditions, nutrition and other biological factors that may vary with SES account for relatively little of the variance in such scores
The living conditions of children result in part from the accomplishments of their parents: if the skills measured by psychometric tests actually matter for those accomplishments. intelligence is affecting SES rather than the other way around
it is clear that no model in which 'SES" directly determines "IQ" will do
To imagine that any simple income- and education-based index can adequately describe the situation of African Americans is to ignore important categories of experience
The differential between the mean intelligence test scores of Blacks and Whites ... does not ... simply reflect differences in socio-economic status.
I know of no reports which argue against these generalization. Moreover, the observation that SES accounts for very little of the BW gap is a key motivation to the theories of Ogbu (e.g. his book on Shaker Heights, OH) and Flynn, as well as Rushton and Jensen. It is a primary impetus for a great deal of the very complicated argument found in that section. --Rikurzhen 18:59, 11 June 2006 (UTC)
The APA report is twenty years old. Here is a more recent study Socioeconomic status modifies heritability of IQ in young children[52]Ultramarine 19:05, 11 June 2006 (UTC)
You're talking about young children, whereas these conclusions are about people >=17 years old. --Rikurzhen 19:09, 11 June 2006 (UTC)
Incorrect, they mention children in your citation. Also, they do not state that SES has no role at all, only that it is not the only explanation.Ultramarine 19:13, 11 June 2006 (UTC)
it shrinks in magnitude due to the IQ-SES correlation, which is .33. they discuss children, but children is a broader category than young children. young children show a relatively low IQ heritability in any SES group. --Rikurzhen 19:16, 11 June 2006 (UTC)
there are many things about young children that aren't true for older children and adults. what text in the paper you linked to argues that SES accounts for the BW gap? --Rikurzhen 19:14, 11 June 2006 (UTC)
Here is another article mentioning the first article.[53] Again, the twenty years old APA report does not state that SES has no role at all, only that it is not the only explanation.Ultramarine 19:23, 11 June 2006 (UTC)
size of the BW gap in the 2000s
UL changed the text relating to the size of the BW gap. the BW gap in g was previously (70s-80s) estimated at 1.1 to 1.2 sd (e.g. roth 2001)
the very latest findings on the BW gap copied from here:
If you take all the black-white differences on IQ tests from the first ones in World War I up to the present, there is no statistically significant downward trend.
the 1997 re-norming of the Armed Forces Qualification Test (AFQT), which showed a black-white difference of .97 standard deviations ... the 1997 results represent noticeable improvement
2003 standardization sample for the Wechsler Intelligence Scale for Children (WISC-IV), which showed a difference of .78 standard deviations ... The black and white means on the WISC-IV’s measure of full-scale IQ were 91.7 and 103.2 respectively (Prifitera, Weiss, Saklofske et al. 2005: 24) [this is the strongest reduction of any of the recent reports]
2001 standardization of the Woodcock-Johnson intelligence test ... showed an overall black-white difference of 1.05 standard deviations and, for youths aged six to eighteen, a difference of .99 standard deviations ... The results from the WJ-III are noteworthy because the WJ-III provides the best known statistical estimate of g. Uniquely among the major standardized tests, the scoring system for the WJ-III uses principal-components analysis to find the best weighted combination of subtests instead of treating all subtests equally (Schrank, McGrew, and Woodcock 2001).
The Kaufman Assessment Battery for Children (K-ABC) is a test that has consistently shown smaller black-white differences than other IQ tests. ... for the original standardization published in 1983, the means on the “Mental Processing Composite” ( K-ABC’s version of an IQ score) for the white and black samples were 102.0 and 95.0 respectively (A. S. Kaufman and Kaufman 1983: 152). Twenty-one years later, those means were both within a point of their 1983 values—102.7 and 94.8 respectively (A.S. Kaufman and N.L. Kaufman 2004: 96).
that is, the BW gap may have shrunk from 1.1 -1.2 sd to about 1 sd --Rikurzhen 19:12, 11 June 2006 (UTC)
achievement tests
Here are different views.[54] Tha gap is today 0.6-0.7 SD.Ultramarine 19:14, 11 June 2006 (UTC)
That's the NAEP, not an IQ test. That's about the achievement gap, not the IQ gap. --Rikurzhen 19:18, 11 June 2006 (UTC)
here's the summary of achievement tests: --Rikurzhen 19:21, 11 June 2006 (UTC)
Among students ages nine, thirteen, and seventeen, the black-white differences in math as of the first NAEP test in 1973 were 1.03, 1.29, and 1.24 standard deviations respectively. For nine-year-olds, the difference hit its all-time low of .73 standard deviations in 2004, a drop of .30 standard deviations. But almost all of that convergence had been reached by 1986, when the gap was .78 standard deviations. For thirteen-year-olds, the gap dropped by .45 standard deviations, reaching its low in 1986. For seventeen-year-olds, the gap dropped by .52 standard deviations, reaching its low in 1990.
In the reading test, the comparable gaps for ages nine, thirteen, and seventeen as of the first NAEP test in 1971 were 1.12, 1.17, and 1.25 standard deviations. Those gaps had shrunk by .38, .62, and .68 standard deviations respectively at their lowest points in 1988
comparing six large data bases from different time periods from 1965 to 1992. The black-white difference on a combined measure of math, vocabulary, and reading fell from 1.18 to .82 standard deviations in that time, a reduction of .36 standard deviations -- Hedges and Nowell (1998): 154.
Here are different views.[55] The IQ gap is today 0.6-0.7 SD.Ultramarine 19:24, 11 June 2006 (UTC)
Again, you haven't read what I've written. My reference and yours are citing the same numbers. There is no doubt that achievement tests measure a gap that's signficantly reduced, and smaller than the IQ gap. --Rikurzhen 19:27, 11 June 2006 (UTC)
There are no nationwide recent iQ tests. I am citing a peer-reviewed study about what these other tests imply. You are not. As such, your conclusions are original research.Ultramarine 19:31, 11 June 2006 (UTC)
Above I listed the AFQT, WISC-IV, WJ-III, and K-ABC test scores. AFAIK, all are nationally representative. --Rikurzhen 19:35, 11 June 2006 (UTC)
Names of peer-reviewed articles please.Ultramarine 19:38, 11 June 2006 (UTC)
Read the text I copied. See [56] for the full citations. --Rikurzhen 19:40, 11 June 2006 (UTC)
Eh, 1, Murray cites his own unpublished results. 2. "The second positive result comes from the 2003 standardization sample for the Wechsler Intelligence Scale for Children (WISC-IV), which showed a difference of .78 standard deviations, as against the 1.0 difference that has been typical for individually administered IQ tests.42" 3. "The first is from the 1997 re-norming of the Armed Forces Qualification Test (AFQT), which showed a black-white difference of .97 standard deviations.40 Since the typical difference on paper-and-pencil IQ tests like the AFQT has been about 1.10 standard deviations, the 1997 results represent noticeable improvement"
So the gap is narrowing.Ultramarine 19:47, 11 June 2006 (UTC)
It's published if you just read it. The "results" are his doing the math of a previously published data set which is collected by the U.S. gov't. Unless you think he's bad at math, then there's no question about his results. --Rikurzhen 19:54, 11 June 2006 (UTC)
From 1.1-1.2 sd to around 1 sd. That's what I said, no? Actually, you can't technically say that the gap is "narrowing" since there is no statically signficant trend yet, but you can say that recent studies find a gap of around 1 sd. --Rikurzhen 19:50, 11 June 2006 (UTC)
Most of these results are uninteresting, they are Murray's unpublished opinions. However, even so I note that one result showed only .78 SD. More importantly, I cite an actual peer-reviewed study that states the IQ gap today is 0.6-0.7 SD.[57]Ultramarine 19:53, 11 June 2006 (UTC)
You're misreading. I don't have time to repeat myself any more right now. Take the time to actually read my comments instead of engaging in a POV push. --Rikurzhen 19:57, 11 June 2006 (UTC)
p.s. you think Nisbett's reply to Rushton and Jensen was peer-reviewed? Did you bother to read Gottfredson's work on the NAEP also? the NAEP is an achiement test. --Rikurzhen 19:59, 11 June 2006 (UTC)
Murray also uses achievement tests when stating his unpublished opinions. I find his decriptions very POV. For example, "There is more to be said on both sides of this issue, but nothing conclusive" followed by a very long fotenote where he hides a long list of opposing evidence.Ultramarine 20:06, 11 June 2006 (UTC)
Murray discusses both achievement tests and IQ test. The gap in achievement tests has clearly decreased by a substantial amount. You must not have read him correctly or fully because he goes on at length about the differences between the tests (correlation with g) and that the a priori expectation is a reduction in the achievement gap followed by a reduction in the IQ gap. That the IQ gap has shrunk so little and so late is, he claims, surprising and disheartening. Go ahead and take the time to actually read what he wrote, rather than arguing with me. --Rikurzhen 20:10, 11 June 2006 (UTC)
Probably beyond the scope of this page, but the discussion of whether the improvements were at the bottom or top of the distribution in interesting (i.e. in relation to the Flynn effect). --Rikurzhen 20:12, 11 June 2006 (UTC)
Again, I have given a reply published in peer-reviewed journal. Exactly what peer-reviewed study are you refering to? I can only see Murray's unpublished opinions. Full name and authors please. Ultramarine 20:14, 11 June 2006 (UTC)
Um, Murray's article is published in the magazine Commentary, which is an reliable source. The Nisbett piece is in a peer-reviewed journal, but is a reply to another article and so is most certainly not subject to peer-review (it is, itself, an instance of "open" peer review). However, Gottfredson has two 2005 papers (one in the same issue as Nisbett, here is the other one) that say that the NAEP is not evidence of a shrinking IQ gap. Murray's is the only article I've seen that does provide evidece from IQ tests that the gap is less than it was in the 70s, however, the gap is clearly still ~1sd based on the IQ tests he cites. --Rikurzhen 20:19, 11 June 2006 (UTC)
Murray himself states that is has narrowed and gives .78 SD for one of his resutls. Commentary is not scientific journal. Nisbett text is published in a peer-reviewed journal. At the very least, there is disagreement regarding what the current gap is. As such, graphs based on twenty-five year old data should not be in the article.Ultramarine 20:31, 11 June 2006 (UTC)
The Reynolds data gives a gap of 1.00 sd (precisely, according to my math). The AFQT and WJ-III gives gaps of 1.0 sd (rounded, according to Murray's math). It's the children's IQ test, WISC-IV, which gives a gap of .78 sd. However, there's a histogram in TBC IQ studies prior to 1994?, which has a range from ~2 to ~0 sd gaps which centers around ~1 sd. That's why there is (at least not yet) a signficiant downward trend in the IQ data. Perhaps future IQ tests will give a value consistently less than 1 sd, but it hasn't happened yet. The largest meta-analysis of IQ data is Roth et al (2001) which finds an IQ gap of 1.10 sd (obviously on tests adminstered before 2001). --Rikurzhen 22:37, 11 June 2006 (UTC)
A. From a peer-reviewed journal. Nisbett states that the IQ gap is today 0.6-0.7 SD, much lower than in the graphs.[58]
B. Murray gives several different results. They are based on his unpublished research and thus not of much value.[59]
C. Even, so one of his results show a SD of 0.78 SD.
D. Murray uses achievement tests, just as Nisbett. There are no recent peer-reviewed nationwide IQ tests available.
E. Roth (2001) uses a variety of tests, mostly achievement as Murray points out. There are no recent peer-reviewed nationwide IQ tests available. Also, "In many instances, standardized group differences are less than 1 standard deviation."[60]
F. Roth (2003) "For Black–White comparisons, the overall results show a standardized ethnic group difference for job performance ratings of approximately one third of a standard deviation" [61]Ultramarine 23:40, 11 June 2006 (UTC)
You're misreading. I don't have time to repeat myself any more right now. Take the time to actually read my comments. Perhaps someone else will chime in to correct you. --Rikurzhen 23:50, 11 June 2006 (UTC)
1995 SAT data
I haven't forgot about it. The data was first published in: College Board, The. (1995). 1995 profile of SAT and achievement test takers. New York. AFAIK, this was the last year they broke out the statistics by race * SES. --Rikurzhen 00:23, 12 June 2006 (UTC)
Are you trying to make a point? What? Ultramarine 00:26, 12 June 2006 (UTC)
You've claimed, erroneous, at several times that the 1995 SAT data is some-how made up. I explained the source of the data, but you persisted. Above is the very first publication of the data. --Rikurzhen 00:38, 12 June 2006 (UTC)
I have never claimed you invented the data. Your original research is that they combined with SES are relevent to IQ and race.Ultramarine 00:45, 12 June 2006 (UTC)
So you don't think that SAT scores stratified by race are relevant to IQ and race? Or just when they take SES into account also? --Rikurzhen 00:52, 12 June 2006 (UTC)
That is your theory. I suggest you try to publish it in a journal outside Wikipedia.Ultramarine 00:53, 12 June 2006 (UTC)
Which is my theory? [emphasis added] --Rikurzhen 00:55, 12 June 2006 (UTC)
That a combination of the variables SAT and SES are related to the varibles IQ and race. Your synthesis violates Wikipedia:No original research:
It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;Ultramarine 00:59, 12 June 2006 (UTC)
Do you agree that the question of IQ v SES v race is published?
Do you agree that the question of SAT as a measure of intelligence is published?
Do you agree that the question of SAT as a measure of BW differences in intelligence is published?
Do you agree that the question of SES in causing BW differences in IQ is published?
Do you agree that the SAT graphs were published here -- an article title "Standardized Tests: The Interpretation of Racial and Ethnic Gaps" -- which was the proximal impetus for including them in the article?
At which step in that sequence did I introduce my own original work? --Rikurzhen 01:09, 12 June 2006 (UTC)
A. The correlation is .72 to .86 between the SAT and a factor created from test which isn't an IQ test. This factor in turn only have a .56 correlation with a 'real' IQ tests.[62] Thus, SAT scores should not be used instead of IQ scores when making claims about IQ scores and race.
B. Studies have shown that SAT scores are biased. Yet another argument for not using SAT scores when making arguments about IQ.[63][64]
C. Arguments excluded:
Some argue that Blacks are discriminated against such that they must have a higher or at least equal intelligence in order to achieve the same socioeconomic status (SES) as Whites. One should then expect that Black children should have a higher or equal IQ compared to children from Whites with the same SES. That they score lower on SAT tests can thus be interpreted as evidence for strong adverse influence from environmental factors different from SES or from SES factors other than income and parental education, like systematic discrimination discouraging school and achievement motivation and learning or cultural differences in nutrition like duration of breastfeeding.
It is possible that Black and Hispanic parents achieve higher SES with lower intelligence; perhaps by having (on average) greater amounts of a compensating character, or through affirmative action. However, affirmative action has lts largest effect on young people newly employed with lower income.
Another alternative explanation is that by comparing the SES of parents to the intelligence of their children, the score gap shown here reflects regression toward different average racial scores from one generation to the next; a partly-genetic origin of intelligence differences would predict this effect.
SAT scores correlate fairly well with IQ scores but they are not the same and may measure different things."
D. Obviously you consider this graph very important. It is an argument against the role of SES as an explanation for racial differences in IQ, one of the more important explanations. As such, it should not be based on homemade research but on peer-reviewed studies. But you have been able to show any peer-reviewed study where this graph, data, or conclusion regarding IQ and race has appeared. Thus, it is your own homemade original research. From Wikipedia:No original research:
"It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;"
E. You state that this is a consensus view, published and discussed widely". But why then cannot you cite any peer-reviewed study using these SAT scores? Use arguments from the literature instead of original research. Redraw a graph using IQ scores instead of SAT scores, if there actually is any such peer-reviewed study. Such a redrawing does not violate copyright.Ultramarine 11:50, 10 June 2006 (UTC)
F. This article [65] shows that the gap disappears when controlling for a small number of factors. Since this is a peer-reviewed study, why not use it instead of the original research? Ultramarine 23:42, 10 June 2006 (UTC)
G. [66] is a personal webpage. Which is forbidden as a source in Wikipedia. Ultramarine 01:12, 12 June 2006 (UTC)
is that an answer? posting large blocks of text will only serve to keep other editors from commenting. --Rikurzhen 01:16, 12 June 2006 (UTC)
Yes. You are doing an original synthesis of many variables, something not allowed. Do not try to publish your personal theory here. Do it in a real peer-reviewed journal.Ultramarine 01:20, 12 June 2006 (UTC)
Did you answer my question? --Rikurzhen 01:22, 12 June 2006 (UTC)
Your questions miss the point. Regardless of if there has been some research on the associations of some these variables, it is still new research to do new combinations of them. Your are making an original combination of variables and claiming an original result. That is original research. Ultramarine 01:27, 12 June 2006 (UTC)
You seem to have agreed that the data (i.e. the SAT*SES*race graph) is itself not original. You seem to be claiming that merely the idea that an SAT*SES*race graph would be worth mentioning in an article about "race and intelligence" is an original contribution. Clearly, it is so obviously salient that it's already been done at least once before. Graphs of the form IQ*SES*race have been pushised in discussions of R&I, etc. --Rikurzhen 06:17, 12 June 2006 (UTC)
Again, what peer-reviewed article has used these SAT and SES data to argue about race and IQ? Ultramarine 10:42, 12 June 2006 (UTC)
Ultramarine: Rik's justification is more than enough to warrant inclusion of the graph. It is clearly a honest, relevant, non-original, and unbiased presentation of a significant POV in this debate. Please try to present your arguments as cogent syllogisms instead of rhetorical questions, pointers, quotations, and appeals to your own version of WP policies. Especially, stop insisting on the use of peer-reviewed articles as the only valid source for Wikipedia content. Your understanding of WP policies in this matter remains unique, and the small set of articles that you have chosen to lavish your attention on may not be the best platform for your programme. Arbor 11:57, 12 June 2006 (UTC)
Wikipedia is not the place for homemade research. Again, what peer-reviewed article has used these SAT and SES data to argue about race and IQ? Ultramarine 12:00, 12 June 2006 (UTC)
Please stop deleting opposing view. NPOV requires inclusion of all, not just those of one side.
A. From a peer-reviewed journal. Nisbett states that the IQ gap is today 0.6-0.7 SD, much lower than in the graphs.[67]
B. Murray gives several different results. They are based on his unpublished research and thus not of much value.[68]
C. Even, so one of his results show a SD of 0.78 SD.
D. Murray uses achievement tests, just as Nisbett. There are no recent peer-reviewed nationwide IQ tests available.
E. Roth (2001) uses a variety of tests, mostly achievement as Murray points out. There are no recent peer-reviewed nationwide IQ tests available. Also, "In many instances, standardized group differences are less than 1 standard deviation."[69]
F. Roth (2003) in a new new meta-analysis: "For Black–White comparisons, the overall results show a standardized ethnic group difference for job performance ratings of approximately one third of a standard deviation" [70]Ultramarine 23:40, 11 June 2006 (UTC)Ultramarine 11:48, 12 June 2006 (UTC)
Nisbett's review/article is nice but it's principally concerned with criticing the hereditist position, we probably shouldn't rely heavily on it for other things than its main focus. As for the 0.6-0.7 and 0.78 std. numbers it seems to me (as a rank amateur, though a somewhat scientifically literate one) that they're not really that inconsistent with a gap of "about 1 std." and considering the high stability of IQ scores for adult individuals it seems that swift reductions in the IQ gap are not to be expected. Even the optimistic Nisbett thinks it would take "between 20 and 60 years" for the gap to be eliminated, that shows fairly clearly that a study isn't obsolete even if it's a few years old. Also note that the article treats the subject in a historical light and the graphs don't necessarily have to show the best estimate of the current gap as long as the age of the data is clearly stated. Imagine when we'll be sitting here in 40 years discussing the article and, assuming Nisbett proves correct, the IQ gap is gone. Historical graphs showing the IQ gap as it existed in, say, the 1980's, would still be of interest to readers fascinated by this barbaric time. Haukur 18:58, 12 June 2006 (UTC)
The graphs show data that is 25 years old. An increase in IQ of 5 points fits well with the Flynn effect. Anyhow, we at wikipedia should not decide which researcher is right. NPOV requires that all significant views should be included. Nisbett and the many tests he reports is a significant view. As such, it violates NPOV to exclude it and instead include 25 years old data.Ultramarine 19:25, 12 June 2006 (UTC)
[shh... i'm on vacation] First, Haukur's comment are directly on target. Second, UL is treating Nisbett's paper as if it were the consensus view. Gottfredson (2005) and Rushton and Jensen (2005) explicitly reject Nisbett's conclusions in the same issue of that journal, and Murray's (2005) independent analysis draws a different conclusion than Nisbett. Thus, there is a difference of opinions. My last edits to the main article incorporated a description of this debate (remember, we describe debates, we don't take sides in them). What we can say w/o dispute is that the IQ gap was around 1 sd during the 20th century; and the debate is about what it is in the 21st century. That should be more than enough to settle this. -- Later --Rikurzhen 20:37, 12 June 2006 (UTC)
I am not treating it as consensus view. It is a view that is excluded from the main text, violating NPOV. The 25 years old data is described as the current situation. It should be clearly pointed out that some argue that the gap is much smaller today. However, we can keep the graphs with such a clear disclaimer. That is, until Black Americans reduce the racial IQ gap: evidence from standardization samples by Flynn and Dickens, currently in press, is published.Ultramarine 20:47, 12 June 2006 (UTC)
Until??? What are the chances that a paper with such a title will be published without an accompanying commentary? Fortunately for us, the Reynolds data already showed only a 1 sd BW gap. --Rikurzhen 04:17, 13 June 2006 (UTC)
Sorry if I'm asking a question which betrays my ignorance, but doesn't a narrowing of the IQ gap, at just about any rate, strongly point to SES (or other environmental factors) as the cause for the gap, and away from a hereditary explanation? Just asking... --Ramdrake 20:57, 12 June 2006 (UTC)
Of course. That is why Rushton and the other Pioneer fund grantees so strongly deny it.Ultramarine 21:00, 12 June 2006 (UTC)
Ramdrake, as you stated it, it does not. It depends entirely on the magnitude of the reduction. If the BW gap were to disappear, then it would become highly likely that there was no genetic effect. However, no one thinks the BW gap is 100% due to genes. Rushton and Jensen guessed a 1-1.2 sd gap @ 80% heredity. Under their model, a reduction to .8-.96 would be expected if environmental conditions were improved. As an aside, I find it noteworthy that Murray's review found a range of .78-1 sd among recent IQ tests. --Rikurzhen 04:17, 13 June 2006 (UTC)
Professional language
It's unnecessary to use 3 sentences when one sentence will do.
3 sentences: "Whether the IQ gaps are shrinking and the causes and meaning of the different average scores for these groups are debated. Note that the data is 25 years old. Some researchers find much smaller gaps today."
1 sentence: "Whether the IQ gaps have shrunk, and the causes and meaning of the different average scores for these groups are debated."
Ultramarine, that accusation of bias edit is already discussed here: Talk:Race_and_intelligence#1. (Continue the discussion there if you're still interested.)--Nectar 02:33, 14 June 2006 (UTC)
My prior arguments still remain.Ultramarine 02:43, 14 June 2006 (UTC)
Here, I'll paste in your last comment from that thread:
. . .That the Pioneer Fund support a certain view is obvious considering its history. That that funding may influence research is well-known, which is why respected journals require disclosure, regardless of there has been any evidence of misconduct or no. . . Ultramarine 13:28, 15 April 2006 (UTC)
This is a moot point, as blanket disclosure is already present: "The largest source of funding for proponents of the partly genetic interpretation, the Pioneer Fund, has been criticized by some critics as. . ."--Nectar 09:26, 14 June 2006 (UTC)
Gene studies
Five recent studies have shown that some genes involved in brain development and other neuronal functions have variants that have spread to high frequencies under selective pressure and now occur in different frequencies in different global populations. The novel variants of microcephalin and ASPM, genes previously implicated in the evolution of human brain size, were found at low frequencies in sub-Saharan African populations, at 70% and 50% in Europeans, 70% and less than 50% in East Asians, and microcephalin at 100% in the tested South American Indian populations. ASPM is estimated to have arisen 500-14,100 years ago, with researchers favouring 5,800 as most likely. Different populations may utilize different variants to respond to similar evolutionary pressures.
A. For some reason this is repeated twice. This is an overview with many subarticles. Why is this repeated? Once is enough
B. It should be mentioned that these distributions do not follow those claimed for IQ.
C. Have anyone actually argued that these genes are related to IQ? They could be anything, from reading emotions to production of speech.Ultramarine 02:43, 14 June 2006 (UTC)
If you want to be taken seriously you're going to have to read the discussions you participate in on this page.
A. The "repeat" of this topic is a single sentence in the race section. I've previously responded: The race section covers the discussion about whether or not race exists biologically. That neural differences exist among races clearly has implications for the question of race in the context of race and intelligence.
A. Here you are confusing yourself. First you state that the section is about whether races exist. Then you state that section is about racial differences in IQ. Clearly a section about whether races exist at all should look at numerous genes, not just a few. The material belongs in the pro-genetic section.
B. I see no consensus there.Ultramarine 20:21, 14 June 2006 (UTC)
C. That is only one of the genes. What about the others? Ultramarine 20:13, 14 June 2006 (UTC)
A. The race section currently summarizes the debate about whether or not race exists ("numerous genes"), and then addresses this debate as it relates to intelligence (are races neurogenetically identical?). Under the terms of your argument, the argument that race doesn't exist would only belong in the cultural explanations section.
B. This is a mute point, as the article already states "[some of these genes] now occur in different frequencies in different global populations." This sentence means what it says and doesn't mean things it doesn't say. If a claim was being made that the allele distributions are consistent with IQ distribution, then the sentence would make that claim. For a better understanding of why this question misunderstands genetics, see the last two comments in the first above link.
C. They discuss a number of candidate genes for biogeographic IQ diversity.--Nectar 21:49, 14 June 2006 (UTC)
The same references appear in two places because they make two different points (with the same references). The titles of the ASPM and MCPH1 papers are "Ongoing adaptive evolution of ASPM, a brain size determinant in Homo sapiens" and "Microcephalin, a gene regulating brain size, continues to evolve adaptively in humans". These are most relevant to the genetic cause section, but also speak to the race section. (You can learn more about these genes here and here.) The titles of the papers which demonstrate massive, genome-wide signatures of recent selection (which happens to include a perponderance of genes involved in neuronal function) are "A Map of Recent Positive Selection in the Human Genome" and "Global landscape of recent inferred Darwinian selection for Homo sapiens". These are most relevant to the race section, but also speak to the genetic cause section. --Rikurzhen 03:47, 15 June 2006 (UTC)
I included the quotes directly from the researchers involved in the ASPM and MCPH1 papers, indicating their caution about interpreting their results as a genetic cause for intelligence differences, as well as their admonition against assuming anything about what kind of distribution the next two genes they pick may have around geographic populations. I'm getting the feeling that a lot of these studies are being quoted out of context, and simplistic conclusions are being drawn not from what the studies or the authors stated, but by interpretation by laymen. --JereKrischel 04:17, 15 June 2006 (UTC)
The text already said "their function is not yet known, and their relationship to intelligence unclear." This is more suscinct that the quotation you appended. You also appear to have missed that footnote #14 contains the references. --Rikurzhen 05:17, 15 June 2006 (UTC)
It seems important to note that the researchers involved explicitly cautioned against the interpretation being implied by the succinct note. Also, the footnote #14 doesn't contain references to the specific citation for microcephalin and ASPM distributions, nor the timeline for the ASPM gene (it seems to refer to a dopamine gene). Perhaps you are referring to the last footnote in footnote #14, in which case it deserves its own cite for that specific info. --JereKrischel 05:38, 15 June 2006 (UTC)
You missed them: Mekel-Bobrov et al. 2005, Evans et al. 2005. What Lahn wrote is better conveyed by the fact that unclear." This is a more basic/fundamental caveat, restricting interpretation more than Lahn's quotation does. More importantly, it achieves the same end without block quotes. This is a sub-section of a summary section, which means it should ideally be a short as possible. (In theory, three paragraphs for the entire causal explanations section.) I'm a geneticist, not a layman, so I can explain this more if you'd like. --Rikurzhen 05:43, 15 June 2006 (UTC)
Can we break up the #14 footnote to the specific assertions presented? It seems confusing to have several assertions by different studies, all cited as one - it makes things unclear to the reader. I'm just afraid that what is being stated is being taken out of context - i guess if we believe their function is really unknown, and their relationship to intelligence unclear, why does it serve as any proof of race-based intelligence differences? --JereKrischel 05:48, 15 June 2006 (UTC)
It doesn't prove anything, but it does make the hypothesis plausible. We can certainly break up the footnote. I'll do that if you'd like. --Rikurzhen 05:49, 15 June 2006 (UTC)
We can also do with or without the percentages if that's confusing. --Rikurzhen 05:50, 15 June 2006 (UTC)
I'm not sure how something whose function is unknown, and relationship to intelligence is unclear, makes a hypothesis plausible. I removed it for now...maybe we can figure out how best to present it before putting it back in? BTW, thanks for your patience! --JereKrischel 05:52, 15 June 2006 (UTC)
Plausible hypothesis
Okay, Rikurzhen, here's my question - how does a gene whose function is unknown, and relationship to intelligence unclear, make a case for the race/intelligence genetic explanation?
It seems that one could take any gene of unknown function, that had anything to do with anything neurological, and come up with a frequency distribution that might vary across sample populations - are we merely trying to assert that there are frequency distributions amongst genes in various populations? It seems that there are better studies which already assert that. Are we trying to imply that even though the relationship to intelligence is unclear, it is probable? It doesn't seem like the authors of those studies were willing to take that leap...
Care to elaborate on your POV? --JereKrischel 06:01, 15 June 2006 (UTC)
The only known functions of these genes are to regulate brain size, and evolutionary change in them has been previously associated with brain size increases in the primate lineage. Brain size increases in the human lineage are thought to correspond with increases in intelligence. --Rikurzhen 05:54, 15 June 2006 (UTC)
Maybe if we specifically mentioned these two steps - that it is the combination of regulating brain size, and the belief of some (refuted by others) that brain size corelates to intelligence. We'll have to put in the obligatory cite regarding male/female brain size differences, but IQ similarities, which seems to blow the basic hypothesis out of the water, but maybe if we made it clear what chain of thought is building the case, it would be helpful. --JereKrischel 06:01, 15 June 2006 (UTC)
BTW, When we write "their function is not yet known, and their relationship to intelligence unclear" that refers to the dervied alleles, rather than the genes themselves. Clearly the genes regulate brain size, but what specific effects these recently derived alleles have is unknown. --Rikurzhen 06:00, 15 June 2006 (UTC)
It is also not clear that brain size corelates to intelligence in this case. Although generally between species of primates this may be true, that does not make it true within members of a single species (of any primate). Sounds like a logical stretch, but I'm willing to put it out there if it is accompanied by the obligatory evidence against such a hypothesis. --JereKrischel 06:03, 15 June 2006 (UTC)
the question of brain size - intelligence relationships are much less contentious or ambiguous on the scale of mammalian evolution where the observations about intraspecific variation in these genes were made. rodents have small brains, primates have big brains, apes have bigger brains than monkeys, etc. humans with loss of function mutations in these genes have tiny brains and are retarded. not really any ambiguity about that.
Whoa, I'm sorry, maybe I missed something - did you just assert that the genes in question, regulate brain size to the degree that having or not having them is the difference between being normal and retarded? Certainly if that is the case, I can understand your point, but is that really being asserted by any of these studies? --JereKrischel 06:22, 15 June 2006 (UTC)
what's unknown is what these derived variants do. clearly they had a selective advantage in the last 50-5k years, and they now vary in their geographic distribution. thus, one guess at the function of these derived alleles is cognitive ability (as per the papers) but it could logically be anything, and nothing has been proven yet. --Rikurzhen 06:13, 15 June 2006 (UTC)
So essentially, when it comes to relating it to intelligence, we've got some ambiguity there - unless of course we're talking about genes that undeniably regulate brain size to a degree which causes retardation. Can you clarify your point on that matter? That is to say, are any of the 5 studies in question asserting a biogeographical distribution of genes that cause unambiguous retardation? --JereKrischel 06:22, 15 June 2006 (UTC)
None of the alleles looked at in these studies cause retardation. Other mutations in these genes cause microcephaly, which is associated with retardation. A particular variant of a gene is called an allele. Retardation is an extreme consequence of these genes not functioning properly, which is generally taken as evidence that their normal function is to regulation brain size. The selective signature in primates corroborates this belief. What's missing is a study of the effect of the "normal" variants, including the newly derived alleles, on IQ. --Rikurzhen 06:28, 15 June 2006 (UTC)
Proposal
How does this sound, Rikurzhen?
Five recent studies have shown that some genes involved in brain development and other neuronal functions have variants that occur in different frequencies in different global populations. For example, the novel variants of microcephalin and ASPM - genes previously implicated in the evolution of human brain size - were found at varying frequencies in sub-Saharan African, European, East Asian, and South American Indian sample populations. Those who believe that human brain size in normal ranges strongly corelate with intelligence look at this data and see a possible link between genetic variation and intelligence. However, very little is known about the actual impact of these variants - the researchers caution that they may not have anything to do with cognition or intelligence, despite any specific affect on brain size.[5]
^Bernstein and Nash 1999 harvnb error: no target: CITEREFBernstein_and_Nash1999 (help), referencing Kagan 1994 harvnb error: no target: CITEREFKagan1994 (help) and Smith & Freedman 1983.[1] Kagan found the Chinese-American children were less talkative, less prone to laughter, less likely to flare up over toy disputes, and had less variable heart rates. The European-American children consisted of samples in Boston and Dublin, with the Boston children exhibiting the highest level of reactivity.
^Moehler et al. 2006 harvnb error: no target: CITEREFMoehler_et_al.2006 (help); also discusses previous findings of slightly higher inhibition in American caucasian children with blue eyes, compared with a statistically insignificant increase in inhibition in German children with blue eyes.
^Harpending and Cochran 2002 harvnb error: no target: CITEREFHarpending_and_Cochran2002 (help). Harpending and Cochran suggest this gene "may be a model system for understanding the relationship between genetic variation and human cultural diversity," noting high frequencies in South American Indians, such as the Yanomamo (sometimes referred to as "the Fierce People"), intermediate frequencies in Europeans and Africans, and rare to nonexistent frequencies in East Asians and !Kung Bushmen (sometimes referred to as "the Harmless People").
I'm not sure about Those who believe that human brain size in normal ranges strongly corelate with intelligence look at this data and see a possible link between genetic variation and intelligence and the text that follows. The consensus view seems to be that IQ-brain size correlation is r=.4, which is not strong. Also, I think it's better to say that nothing is known about the link between these variants and intelligence, setting brain size aside. --Rikurzhen 06:56, 15 June 2006 (UTC)
I removed the "strongly" part - but I think it's important we make clear the logical steps required to believe that these study affirm racial differences in intelligence. 1) we need to believe that these genes regulate brain size in some direct way, 2) we need to believe that brain size correlates to intelligence in some direct way. And I guess the authors themselves indicated a 3) we need to believe that there are no other genes regulating brain size in some direct way that have inverse population distribution to these. Do you have suggestions on how to make those three steps clear? --JereKrischel 07:10, 15 June 2006 (UTC)
Not all of those steps are necessary. Given the modest IQ-brain size correlation, these alleles could signficantly affect one and not the other (or both or neither). The fraction of total IQ variation they explain is likewise an unknown; but some estimates put the total number of IQ QTLs at over 100. The previous caveat was "their function is not yet known, and their relationship to intelligence unclear." That's why these findings can only really make a genetic explanation plausible, rather than being proof of a genetic cause. Even if we found that the D allele add +3 IQ points, that wouldn't prove much. --Rikurzhen 07:23, 15 June 2006 (UTC)
I'm sorry, what steps do you think are the bare minimum? Are you asserting that all we need to believe is that these genes are biogeographically distributed, and therefore can assert that intelligence is also (although in an indeterminate way)? Sounds like at the very least you need to get #3, to account for inverse distributions of other regulating genes that may counterbalance the currently observed distributions...I guess, can't it be possible that the D allele add +0 IQ points, and thus be a wash? Maybe if you could step through the implications for me one by one I could understand what steps you think can be omitted, and still assert that these studies support the race/intelligence hypothesis. --JereKrischel 07:28, 15 June 2006 (UTC)
Maybe I've been awake for too long. I'll get back to you after getting some sleep. Nectar may be able to fill in for me here. --Rikurzhen 07:30, 15 June 2006 (UTC)
Here's a shot -- These papers are of interest b/c the genetic hypothesis predicts the existence of alleles which affect IQ that have an unequal biogeographic distribution, but doesn't predict much more than that, but the environmental hypothesis is also compatible with differential distribution of alleles, so long as the net effects is 0 (an unlikely outcome). However, the genetic hypothesis would be falisifed if there were no IQ QTLs with such a distribution. Brain size is a side-issue. Any genes that affect neuronal fuction are possible candidates, but brain size modifies are particularly strong candidates. -- later --Rikurzhen 07:38, 15 June 2006 (UTC)
Okay, so at the very least, you need to believe that ALL genes that affect neuronal function are distributed in the same unequal ways - that is to say, you need to believe that studying a couple of genes that affect neuronal function and seeing a certain distribution will not be counterbalanced by inverse distribution of other genes which may act in similar ways. The idea being that biogeographic parallel development and adaptation can occur in different ways, to the same end (i.e., protection of intelligence, or other neuronal functions). The researchers specifically caution against that, but I think if that's what you're trying to assert, it makes sense to work that into the section. --JereKrischel 15:57, 15 June 2006 (UTC)
There are many other important points as well. First, they are a concrete counterexample to the claim that human evolution has not had enough time to produce functionally significant genetic variation between populations. That claim, now, is simply false (or at least much more difficult to maintain with a straight face), and it used to be important.
You're taking that too far - not knowing the function or relation to intelligence of these genes, there is no way to assert that these two genes represent functionally significant genetic variation. Again, the idea that 2 of the many genes that affect brain size have a specific distribution does not preclude inverse distribution of other genes that duplicate that function in different ways. --JereKrischel 15:57, 15 June 2006 (UTC)
From a broader perspective, these findings elevate the genetic hypothesis to the same level as the environmental one. The environmental hypothesis says that (A) yes, there are differences in how (say) black and white children are fed/raised/treated/taught and (B) we also have good reason to believe that these differences are functionally significant within groups, even though the precise relationship between nurture and outcome is complex and little understood . Hence, it is not-stupid to speculate from A and B that (C) these differences (ill-understood as they are) contribute to the between-group variance (That conclusion, after all, is routinely made in the standard social science model). By analogy, we now know that (A') yes, there are known allele frequency differences, and (B') the alleles/genes/combinations in question are know to affect how the brain is built, even though this is all very little understood. So, from (A') and (B') we infer the non-stupid speculation (C'), that genetic variation does explain part of the variance between groups. (That conclusion, after all, is routinely made in the rest of the sciences.) We all understand that A+B does not imply C. It is merely very plausible. Similarly, we all understand that A'+B' does not imply C'. It is merely very plausible. But imagine if we would treat all the social-science explanations on this page like this: (hyperbole for clarity : ) "The precise relationship between not giving children an adequate education and their later socio-economic status is not understood, and critics have cautioned that Black children may not experience any detrimental effects at all from not learning to read in school. Many researchers who support the environmental hypothesis are also communists." Arbor 08:28, 15 June 2006 (UTC)
In agreement with Arbor, the presence of broad recent neuronal selection (Wang et al.) and substantial differences in neuronal allele frequencies (ASPM etc.) represent a line of converging data in the argument that genes could be involved in the IQ gap. It's a secondary point that some of these neuronal genes may actually represent candidate genes for that argument (Harpending and Cochran). Nobody's making assertions about a relationship.--Nectar 09:02, 15 June 2006 (UTC)
I think you'd want to be extraordinarily careful making claims that any of these genes are involved in IQ, or even brain size (e.g.), you fuck up a developmental gene you get developmental abnormalities, doesn't necessarily tell you what the gene does anymore than the intellectual deficits in PKU tell you what phenylalanine hydroxylase does. --Coroebus 12:39, 15 June 2006 (UTC)
I haven't seen any "claims that any of these genes are involved in IQ," though statements of a probabilistic nature ("candidate genes") have been made.--Nectar 19:18, 15 June 2006 (UTC)
But since they don't seem to be involved in brain size either (see my e.g.) singling them out as even candidate genes is unwise at this juncture (i.e. they're genes that have some vague neuronal role, like half the genes in the genome, but no obvious connection to intelligence). --Coroebus 21:31, 15 June 2006 (UTC)
Be careful about equating a negative result with evidence against a hypothesis. I can fail to find anything if I don't have a good sample and/or don't use the right methods. ASPM and MCHPH1 don't need to be the examples used, but they are the ones most discussed (e.g., Harpending and Cochran). --Rikurzhen 22:01, 15 June 2006 (UTC)
I think you mean that you can't fail to not find anything (double negatives are a bitch). Actually that negative result is very much evidence against the hypothesis. The reason is that the only evidence for a role in microcephaly genes in brain size is that mutations in them cause microcephaly. This might suggest that they are determinants of brain size in normal individuals, but is not necessarily the case. Now someone has actually gone and had a look at whether they determine brain size in normal individuals, and they don't seem to. Now obviously it is possible that another study, perhaps larger, will come along and show that they do, but since the original evidence they were involved in brain size was indirect, the newer better evidence undermines that original contention. I looked at the Harpending and Cochran thing and was surprised at their suggestion that there was evidence of ethinic genes associated with IQ, so I looked at some of Cochran's work, it seems his ideas about things like microcephaly genes and BRCA genes are entirely indirect speculations based on assumptions of genetic group differences and working back to genes that vary in these groups, not any direct evidence. In fact the evidence for any individual gene having a role in intelligence is currently pretty limited (e.g.), particularly for IQ in the non-old. --Coroebus 08:35, 16 June 2006 (UTC)
Selecting a bad sample or a bad technique can fail to find a (small) significant difference: a false negative. They used a sample of convenience of their study, whereas I would have used same-sex siblings. Thus, they have to control for race/ethnicity and sex in their study, reducing power while not totally controling for population stratification. OTOH, I wouldn't be surprised to find that these variants affect disease resistance or reproduction because the selection is so strong. --Rikurzhen 18:54, 16 June 2006 (UTC)
Obviously a study can fail to find something that is there, but it is still prima facie evidence against a role in brain size (any study can fail to find something that is there, if the effect is small enough, tangentially, I was reading an interesting article the other day about how blinding doesn't work if what you want to find is no difference, thought provoking). I seem to recall there's a study suggesting the selection is actually on their role in sperm. --Coroebus 19:01, 16 June 2006 (UTC)
Arbor has summarized the point very well. There's no need to focus the discussion only on ASPM and MCPH1, as Wang et al has a list of neuronal genes.
The text previously stated:
Five recent studies have shown that some genes involved in brain development and other neuronal functions have variants that have spread to high frequencies under selective pressure and now occur in different frequencies in different global populations. ... [about ASPM and MCHP1] ... Different populations may utilize different variants to respond to similar evolutionary pressures.
This summarizes the main point, and summarizes the caveat that was offered in the press reporting. --Rikurzhen 19:46, 15 June 2006 (UTC)
My problem, Rikurzhen, is that it is not made clear by what logical steps we can assert that those distributions for those genes as indicated by those studies can have an affect on intelligence. It seems like simply stating it in that manner does not indicate *why* it is considered supportive of the genetic explanation hypothesis. We need to clearly indicate that the reason why this is believed to be supportive is because we ASSUME that the genes studied are representative of all the genes related to brain size, and therefore related to intelligence. Without a clear indication as to the intended line of reasoning, I think we are leaving the article in a confused state. --JereKrischel 20:02, 15 June 2006 (UTC)
Oh. That's a very good point; I would have no problem with such a caveat. Arbor 20:06, 15 June 2006 (UTC)
Something along the lines of Arbor's description could work. JereKrischel, the word "representative" is almost correct, but it's not that they are representative of all "neuronal genes" but that they are representative of a subset of them (some neuronal genes show regions of strong purifying selection, for example FOXP2 -- although it too has evolved in the hominid lineage). I'm going to offer a medical excuse for not being very helpful on this one. --Rikurzhen 20:14, 15 June 2006 (UTC)
The primary value to the article is that the recent finding of a degree of neurogenetic diversity strengthens the plausibility of phenotypic diversity. I think we can probably just write the section clearly so that we don't need a series of caveats. I've restored more of a concise version of the paragraph; if it looks like it covers all the bases, maybe we can look at what form is best for any additions, such as the distributions of ASPM and MCPH1. I believe their primary value here is the unexpectedly strong variation in their distributions (ranging from rare to 100%).--Nectar 23:11, 15 June 2006 (UTC)
Okay, let me see if I've got your rationale correct, Nectar - so long as the genes in question have "unexpectedly strong variation", somehow we can assert that biogeographic associations to intelligence are plausible? It seems the researchers in question made it very clear that taking these genes as representative of all other distributions of genes which affect intelligence is a stretch - and isn't that required before asserting that this research supports the biogeographic genetic explanation for intelligence?
I guess without the additional explanation of rationale, it looks just like a statement that has no particular bearing on the question as to whether there is a genetic explanation for any observed differences. --JereKrischel 00:13, 16 June 2006 (UTC)
The conversation may have moved on, so I'll just respond briefly. I think the reason things may appear to not quite add up is probably because you responded to arguments that weren't present in my post. I'll break down the argument here anyway. The partly genetic hypothesis is impossible if groups are neurogenetically identical, and some commentators have argued this to be the case. These recent findings, in contrast, suggest a degree of neurogenetic diversity, strengthening the possibility of a degree of diversity in the phenotypes associated with this genetic area (the brain). (If you convert these probabilistic statements to blanket declarations then this argument will not add up.)--Nectar 06:32, 17 June 2006 (UTC)
The two "genome-wide" papers provides the other half of the picture. They identify a great many genes with signatures of recent selection, including genes implicated in neuronal function. I've tried to refocus the emphasis on evolution, which is what each of these papers actually addresses. --Rikurzhen 01:51, 16 June 2006 (UTC)
Is it possible to explain how recent selection invalidates the concept of tying it to race? That is to say, such a recent selection points towards environmental selection based not on the archtypical Negroid/Mongoloid/Caucasoid classifications (asserted to be due to isolations started hundreds of thousands of years ago). It seems that the recent selection evidence, coupled with data regarding nearest common ancestors, indicates that our social categories of race simply don't match the data, don't you agree? --JereKrischel 02:25, 16 June 2006 (UTC)
our social categories of race simply don't match the data. Hmm? No. Evolution invalidates the essentialist concept of race (and the essentialist concept of species). Instead, recent selection is actually evidence for the strength of selective forces that have been acting on humans in the last 50k years, since the northern migrations out of Africa. There's much more that can be said on this topic, but none of it speaks against the idea that current human populations are genetically (geographically) differentiated. --Rikurzhen 03:50, 16 June 2006 (UTC)
Really? I would think that the strength of selective forces in such recent times (within 5k years), would indicate that the categories of race, based on geographic origin hundreds of thousands of years ago, are inherently invalid. That is to say, if selective forces are causing changes as recently as 5k years ago, then our genetic racial categories would be defined by geographic distribution circa 3000 BC - considering the incredible migrations, intermixings, and of course continued selective pressure since then, it's hard to imagine this staying still for long as well.
Especially given the evidence of genetic variation within these arbitrary groups far outweighing the genetic variation between these groups on average, it seems that current human populations are far more diverse within their geographic boundaries, than without. Perhaps I'm misunderstanding you though - as a geneticist, by what mechanism would you propose geographic differentiations, especially considering the recent history (2k-5k years) of global migrations and intermixings? --JereKrischel 05:30, 16 June 2006 (UTC)
Short answer to that last question: Selection, varying geographically. That's clearly the reason for skin color differences -- for example -- between groups. A great deal of the total variation in skin color between Europeans and Africans has been accounted for by recent studies. The article on race used to be a fairly good review of these issues, but it's been fragmented recently, so I can no longer recommend it as strongly. You're quite right to point out that gene flow would destroy genetic differentiation, ceteris paribus; but we do in fact observe quite signficant genetic differentiation between geographically distant populations, which strongly points towards differing selective pressures. OTOH, geographic boundaries (e.g. mountains, deserts, oceans) also play a role in keeping down gene flow. --Rikurzhen 05:54, 16 June 2006 (UTC)
Regarding "significant genetic differentiation between geographically distant populations", haven't the studies shown more significant genetic differentiation between geographically concordant populations? I found this quote:
Third, how much genetic variation exists? Overall, humans share about 99.8% of their genes in common. The other .2% accounts for all human variation. This happens because 67% of all human genetic loci are invariant. The remaining 33% are polymorphic, that is capable of genetic variation and this is where the discussion of genetic variation within and between groups must begin. For forty years, we have known that 85% of polymorphic human genetic variation occurs at the individual level. What that means is that any two people in this room, without regard to the geographic origin of their ancestors, by chance alone shares 85% of their different genes in common. Nei and Roychoudhury in 82 showed that this could be as high as 93-97% of the genetic variation within groups and only 3-7% between them. Mitochondrial DNA shows the same picture. There is about 5-7% genetic variation between populations on the same continent but only about 10% between different continents.
I guess I mean to ask, if there is even more significant genetic differentiation between members of a given geographic group, isn't the proper genetic grouping based on the major difference, rather than the minor difference? For example, if you had a set of blue socks filled with coins, and a set of green socks filled with coins, wouldn't it make more sense to group the socks based on their contents (put all the <25 cents together, all the >75 cents together), rather than based on their color, if the average difference between the two groups was only 2 cents, but in each group there were large clusters 70cents or more apart? --JereKrischel 06:16, 16 June 2006 (UTC)
If race weren't in such bad shape, you could find the answer there. The crux of this issue is discussed in Lewontin's Fallacy and Race and multilocus allele clusters. Basically, for any single loci, on average, most variation is within rather than between groups. However, the variation is correlated, such that if you take whole genomes, individuals are strongly similar to people from the same population (race) and there's little chance of misclassification (one study found <<1% misclassfication). Conflating these two senses of genetic differentiation (i.e. single gene versus whole genome) was Lewontin's mistake. --Rikurzhen 06:33, 16 June 2006 (UTC)
I'm assuming the misclassification study referred to self-identification of people matching observed genetic clusters - that is to say, given 3000 or so people, all of those why identify as Martian (pick your ethnic label here), will have certain markers closer together than with those who identify as Venutian. I haven't found a copy of the study online, but I'd be interested in how they chose their sample, and how they chose their labels. I guess the other question is the significance of those particular markers chosen for such clustering, and whether or not those have significant practical effect on any expressed differences.
Has anyone or is anyone doing further research like this, with a broader sample size, different markers, and self-identification of multiple ethnicities? --JereKrischel 06:57, 16 June 2006 (UTC)
A quick pubmed search just turned up lots more, so the study you describe may have been done. --Rikurzhen 07:16, 16 June 2006 (UTC)
Jewish IQ
"Ashkenazi Jews, for example, demonstrate verbal and mathematical scores more than one standard deviation above average, but visuospatial scores roughly one half standard deviation lower than the White average..."
This suggests that Jewish IQ means are well-established. Actually, though, they are estimated from small, unrepresentative studies. While almost every study shows this pattern of abilities, exact values differ considerably from study to study. Thus I have changed this:
"Ashkenazi Jews, for example, demonstrate high verbal and mathematical scores, but average or below average visuospatial scores..."
"The curves extend across the entire X-axis (from <55 to >145) for all groups and do not indicate a maximum or minimum IQ for any group."
These tails clearly taper off, indicating descreasing frequency, rather than indicating a "maximum" or "minimum". Including every pet comment someone has makes the article unreadable. If any other editors disagree, can you weigh in?--Nectar 19:40, 15 June 2006 (UTC)
The tails aren't very clear - perhaps maybe we could create a new graph that made the asymptotic relationship more obvious, and lose the wording? Is there a source for the data used to create the graph? --JereKrischel 20:06, 15 June 2006 (UTC)
That might be a good idea. Until then, explaining that there are intelligent people from all minorities is not a "pet comment". Obviously this is important to point out.Ultramarine 20:08, 15 June 2006 (UTC)
There are very few people at +/- 3 SD. That's a limiting factor. But no one thought it was a problem for a long time until someone else chimed in with that misunderstanding. --Rikurzhen 20:57, 15 June 2006 (UTC)
Images for environmental explanations
Since many people read articles my reading the introduction and then only the image captions, maybe we need some sexy graphs for environmental explanations. I suggest we include Bell curves for the Flynn effect, following Neisser 1997. See [71], look for figure 59, for what I have in mind. I would like the upper graph (with the intersection Bell curves) rendered in the same style that we render the Race vs. IQ Bell chart in the introduction, with a brief caption, maybe like this: Environmental effects: If children from 1997 were to take a 1932 IQ test, their average would be around 120. Since the genetic markup of these two populations is largely unchanged, the difference between these populations must due to environmental causes. This phenomenon is known as the Flynn effect, and its causes are very little understood. It is possible that the same unknown environmental effects that could explain the Flynn effect also explain racial differences in IQ. Or something like that. This would get an important example of non-genetic reasoning across quickly, and give the article some graphical consistency as well. And it's much more attractive than a picture of Flynn. Arbor 20:03, 15 June 2006 (UTC)
clancy blair's paper finally published
moved from main text:
A recent theory hypothesizes that fluid cognition (gF') may be separable from general intelligence, and that gF' may be very susceptible to environmental factors, in particular early childhood stress. Some IQ tests, especially those used with children, are poor measures of gF', which means that the effect of the environment on intelligence regarding racial differences, the Flynn effect, early childhood intervention, and life outcomes may have been underestimated in many studies. [72]
It is accompanied by a dozen (more or less?) responses. the relation of his theory to the BW gap doesn't appear to have been discussed, but assertions about Gf, about the ability to measure Gf, etc are all discussed. Prima facie these commentaries have questioned most of the assertions made in the text I copied above. We should tread carefully in describing this new paper; definitely read the commentaries before putting this material back. --Rikurzhen 02:24, 16 June 2006 (UTC)
Can't get it online, busy at the mo', but will read + commentaries soonish and get back to you. Normally wouldn't accept the existence of opposing views as acceptable reasons not to include a scientific view, but BBS have deliberately controversial articles with open responses. --Coroebus 22:57, 17 June 2006 (UTC)
We need to be able to describe both Blair's view and any reported problems with it. There are so many commentaries that I don't have the time to go thru them all. Blair's responses to critics would be a place to start. There's also the issue of importance. The Flynn effect is important; I'm not sure that Blair's theory rises to front page importance on this topic, but maybe it should be described in the articles more directly related to his thesis. --Rikurzhen 17:36, 18 June 2006 (UTC)
In recent decades, the standard response to claims of genetic differences has been to deny the existence of intelligence, to deny the existence of races and other genetic groupings, and to subject proponents to vilification, censorship, and at times physical intimidation. Aside from its effects on liberal discourse, the response is problematic. Reality is what refuses to go away when you do not believe in it, and progress in neuroscience and genomics has made these politically comforting shibboleths (such as the non-existence of intelligence and the non-existence of race) untenable.
In theory, we have the intellectual and moral tools to defuse the dangers. "Is" does not imply "ought." Group differences, when they exist, pertain to averages, not to individual men and women. There are geniuses and dullards, saints and sinners, in every race, ethnicity, and gender. Political equality is a commitment to universal human rights, and to policies that treat people as individuals rather than as representatives of groups; it is not an empirical claim that people are indistinguishable. Many commentators seem unwilling to grasp these points.
I think the more problematic issue has been the gross oversimplification and misinterpretation of the data in support of modern eugenics theory (such as Rushton). For example, the genetic averaging idea (that "Mongloid" +"Negroid" = "Caucasoid"), is a completely baseless assertion, requiring one to believe that if your cousins from your mother's side mated with your cousins from your father's side, you'd have someone genetically equivalent to your sibling.
Clearly, biogeographic differences are observable, but casting these on the social concept of "race" is fallacious logic. Similarly, the idea of three major "races" diverging, and then treating them as if they were species evolving in some ascending order is unsupportable - certainly environmental selective pressures did not stop in place for any particular group when they were isolated from each other, and environmental pressures are not value based - developing darker or lighter skin depending on your environment is not a genetic "advance", it is a genetic adaptation.
As a geneticist, I think you would agree with these problematic usage of genetic science in the manners I've illustrated - but perhaps you have a different perspective? --JereKrischel 22:13, 17 June 2006 (UTC)
The only part of this that I recognize is "such as Rushton". I haven't read any of Rushton's books, only his articles in scholarly journals, so I suppose I could be missing something. If you're looking for input on a particular question or claim, I could help with that. --Rikurzhen 05:23, 19 June 2006 (UTC)
Other variance
Some other psychological traits have also been found to vary significantly in distribution. [49] However, group differences in tests of cognitive ability have been the subject of more attention, as they are seen by some as one of society's most pressing problems.
This was removed with this argument "[1] No evidence of genetic role, [2] unrelated to discussion about IQ and race."
1. I don't know if you read the article before making edits, but the sentence doesn't speculate on environmental or genetic causes, and the article is very far from being exclusively about genetic differences. To make this argument more silly, the footnote explicitly covers variance in a dopamine gene associated with ADHD and other behaviors.
2. It is relevant that IQ variance doesn't exist in a vacuum, and other traits also vary. ADHD, for example, is also related to the measurements used in the description of "intelligence". A section on the topic would be debatable, but a sentence or two are not.--Nectar 22:18, 18 June 2006 (UTC)
I've combined the two sentences into one, as this probably works better.--Nectar 22:31, 18 June 2006 (UTC)
Misleading, there were two edit summaries. In more detail:
A. "However, group differences in tests of cognitive ability have been the subject of more attention, as they are seen by some as one of society's most pressing problems" The quoted source does not state this. It says " Subgroup differences in performance on high-stakes tests represent one of American society’s most pressing social problems" There is nothing in that source indicating that research in IQ has prevented other research. Thus, orginal research.
B. "Some other psychological traits have also been found to vary significantly in distribution." Nothing in the source say that it is related to IQ and race. Thus, again original research to claim a connection.Ultramarine 23:02, 18 June 2006 (UTC)
A. The edit you reverted no longer contains the sentence you're referring to, as mentioned above. The new sentence is "Some other psychological traits have also been found to vary significantly in distribution,[49] but variance in cognitive ability is seen as a more pressing phenomenon.[50]" It's very clear that variation in intelligence is seen as more important than variation in, say, childhood temperament, or hearing[73]. Sackett et al. is a reference for how important variation in intelligence is seen.
Sackett et al. says does not say that cognitive ability is a more pressing concern than other racial differences. False use of source. Ultramarine 01:44, 19 June 2006 (UTC)
B. Here is an example in the race and intelligence literature of the context of racial variation being discussed:
Out of all the interesting and intractable differences that may eventually be identified, one in particular remains a hot button like no other: the IQ difference between blacks and whites. . . all sorts of group differences [exist] . . . Let us talk about . . . differences between Russians and Chinese that might affect their adoption of capitalism. About differences between Arabs and Europeans that might affect the assimilation of Arab immigrants into European democracies." Charles Murray 2005, Commentary Magazine.[74]
This is certainly enough to justify inclusion of a single sentence.--Nectar 01:25, 19 June 2006 (UTC)
You have provided no reliable source. Commentary magazine is not scientific journal. None of you other references states what you claim in the text. Thus, original research.Ultramarine 01:38, 19 June 2006 (UTC)
That Commentary is a not a peer-reviewed scientific journal is not important for the kind of source it is being used for here. (Moreover, the WP:RS guideline says nothing of the sort.) The material should be fine in the form Nectar last presented. --Rikurzhen 05:10, 19 June 2006 (UTC)
There is still no source for that cognitive ability is a more pressing concern than other racial differences. Please include Commentary magazine if you claim it as a source. Ultramarine 12:16, 19 June 2006 (UTC)
Let me break things down. Cognitive ability is associated with a host of outcomes considered important, such as job success, academic performance, and longevity. This is why Sackett et al. state it is important. Childhood temperament is not associated with anything. Hearing[75] is not associated with anything. If we weigh these associations, job success etc. for cognitive ability and nothing for other traits, it is more than obvious that job success etc. is considered more important than nothing. If you can show us otherwise through argumentation, you always have that choice.
The sentence (Some other psychological traits have also been found to vary significantly in distribution,[49] but variance in cognitive ability is seen as a more pressing phenomenon.[50]) says what it says and doesn't say what it doesn't say. It does say that X is researched more than Y because X is considered more important. It does not say that "IQ reseach is preventing other research." As per standard practice, the footnotes both refer to the preceding clauses, rather than the entire sentence.--Nectar 19:30, 19 June 2006 (UTC)
Again, Sackett et al does not say what you claim. It is your own original research that cognitive differences are more important and has prevented other reserach. Again, if you want to include Commentary magazine, do so, and state that it is not a scientific journal.Ultramarine 19:36, 19 June 2006 (UTC)
I have added the Murray reference.--Nectar 20:07, 19 June 2006 (UTC)
1. Re:"Not shown that has nany releance race and IQ"
It is relevant that IQ variance doesn't exist in a vacuum, and other traits also vary. ADHD, for example, is also related to the measurements used in the description of "intelligence" such as test performance and academic performance. Other differences are discussed in the literature, meaning they can be discussed in this article. We all like to have everything go exactly the way of our POV, but you may need to compromise occasionally.
2. Re:"reference does not state what claimed"
When a sentence is divided into 2 clauses and each has its own footnote, each clause cites the footnote that follows it. What you mean then, is that it's original research to combine these citations in this way. Cognitive ability is seen as very important by researchers, but less important differences, such as hearing, are seen as less important. Obvious arguments are allowed, or Wikipedia would just be a string of quotes. If you can convince us otherwise, you are free to make your edit.--Nectar 06:05, 21 June 2006 (UTC)
dolan's work with MGCFA
dolan et al (2004) are criticizing the statistical technique called the "method of correlated vectors" not spearman's hypothesis (jensen developed the method of correlated vectors to test spearman's hypothesis). the data set they examined was not large enough to reject all of the alternative hypothesis about the BW gap, which left spearman's hypothesis as one of several alternatives. i know of no studies which reject spearnman's hypothesis, nor of any which claim that it is necessary for a genetic cause of the BW gap. as an example, the BW gap could theoretically be due to differences on a highly g loaded second-level factor (Gv or something like it), rather than g itself, and yet the gap could largely be due to genetic differences. --Rikurzhen 20:24, 19 June 2006 (UTC)
But the Pioneer fund grantees has argued that the "method of correlated vectors" proves the hypothesis true.Ultramarine 20:27, 19 June 2006 (UTC)
To quote Dolan et al "This leaves the validity of Spearman's hypothesis, considered a central justification for the genetic explanation, an unresolved question.""Ultramarine 20:30, 19 June 2006 (UTC)
Note how Rushton and Jensen describe this study "The results statistically confirmed the conclusion derived from the method of correlated vectors regarding a “weak form” of Spearman’s hypothesis: Black–White group differences were predominantly on the g factor" Here is what is actually stated: "On the basis of the present, as well as other results (Dolan, 2000), we are convinced that the Spearman correlation cannot be used to demonstrate the importance of g in b-w differences with any confidence." and "It is possible that the analysis of all available data sets (perhaps using an appropriate meta-analytic procedure) will demonstrate that a model incorporating the weak version of Spearman's hypothesis provides the best description of the data. However, until this work is undertaken, we cannot accept Spearman's hypothesis as an "empirically established fact"
I will give Rushton and Jensen the benefit of doubt. This grave misrepresentation may have been due to incompetence, rather than being deliberate.Ultramarine 20:34, 19 June 2006 (UTC)
I don't know what Rushton and Jensen are thinking, but if you understand Dolan's work you'll see that everything I wrote was based on the assumption that Dolan is correct and that there are important flaws in the method of correlated vectors: that it lacks specificity. Dolan's findings are limited with respect to Spearman's hypothesis because his data set lacks the power to rule it in or out. Because it also cannot rule out alternatives, it is incorrect to say that MCV proves spearman's hypothesis is correct. OTOH, the failure to find any data that contradicts it is certainly suggestive that when the proper test is done, spearman's hypothesis will survive. Perhaps R&J were thinking along these lines. --Rikurzhen 20:41, 19 June 2006 (UTC)
Rushton and Jensen give a grave misrepresentation. You may be right, they may have been mentally referencing an unpublished and hypothetical future study. Thus, just incompetence rather than a deliberate misrepresentation.Ultramarine 20:46, 19 June 2006 (UTC)
The more I'm learning about J. Philippe Rushton, the more difficult it is to take his work seriously. Time and time again, he references research by others as supporting his conclusions, but then these researchers (experts on their own research, of course) come back and say they've been completely misinterpreted.
For example Meheler, speaking about Rushton's cite of Allen Wilson:
"Rushton's theories are a bizarre mélange of nineteenth century anthro-pometrism and twentieth century eugenics. Although there is no evidence showing different cranial sizes between races, Rushton has cited the genetic distance studies of Allen Wilson of the University of California to claim that Africans have smaller brains and are more primitive than whites and orientals, who evolved to cope with the more demanding northern climes. Wilson commented: 'He is misrepresenting our findings'. These 'show that Asians are as closely related to modern Africans as Europeans are'. When asked if he was aware of any anthropological evidence at all that might support Rushton's claim, he replied, 'I'm not aware of any such evidence. The claim shocks and dismays me'.Foundation for Fascism: the New Eugenics Movement in the United States, Patterns of Prejudice by Dr. Barry Mehler
Whether or not you give the benefit of the doubt on intention, I think it is critical to note when citing Rushton's studies both his controversial methodology ("aggregation" of studies without critical analysis), as well as the direct disclaiming of his conclusions by the authors he cites (note: AFAIK, the vast majority of all his work is NOT original research). --JereKrischel 20:44, 19 June 2006 (UTC)
AFAIK, there are three references in the main article where Rushton is an author. (1) Rushton 1999 is a primary research paper on g and the Flynn effect, (2) his 2000 book as a citation for the claim that he supports a hereditarian explanation for the BW gap, and (3) he and Jensen's 2005 review paper. UL and I are not the only people I know who noticed that he glossed over Dolan's conclusions in the 2005 review paper. But I don't know that there's much in this article that can be said to be questionable because it relies on Rushton. --Rikurzhen 20:53, 19 June 2006 (UTC)
"Glossed over" implies that Rushton and Jensen omitted some details when they in fact stated the opposite of what Dolan et al found. Rushton and Jensen are cited more than any other researcher in this wikipedia article. Here is another amazing bit from their 2005 paper "In the Minnesota Transracial Adoption Study, the IQs of the mixed-race (Black/White) adoptees averaged between those of the “nonmixed” White and the “nonmixed” Black adoptees, as expected under a genetic hypothesis (see Table 2). Results from some other types of studies are also consistent with that hypothesis. In her review, Shuey (1966) found that in 16 of 18 studies in which skin color could be used as a proxy for amount of admixture, Blacks with lighter skin color averaged higher scores than those with darker skin, although the magnitude of the association was quite low '(r =.10)'"Ultramarine 21:03, 19 June 2006 (UTC)
What is the correlation between skin color and African/European admixture? AFAIK, it too is quite low. --Rikurzhen 21:31, 19 June 2006 (UTC)
Spare me the original research. Shuey certaly did not support the genetic explanation like Rushton and Jensen states.Ultramarine 23:07, 19 June 2006 (UTC)
It seems like the entire "Genetic Explanation" section heavily relies on Rushton - and as we've seen is subject to the general criticism of invalid interpretation of other's work. Where else do you see anyone making the assertions RJ make about the genetic explanation? Maybe those should be more prominent, given the controversey around R&J's interpretations of other people's research? --JereKrischel 21:13, 19 June 2006 (UTC)
Arthur Jensen --Rikurzhen 21:29, 19 June 2006 (UTC)
As noted, he is also the author the 2005 study with its deliberate misrepresentation/gross incompetence. He has also received very large sums of money from the Pioneer fund.Ultramarine 21:42, 19 June 2006 (UTC)
Well, I'm not going to try and categorize anyone as deliberately misrepresenting something, or being grossly incompetent - HOWEVER, Ultramarine is right in pointing out that Rushton and Jensen are closely associated both by funding and philosophy, and have been criticized in the same manner for their work. Do we have any non-Pioneer Fund folks whose research supports the "Genetic explanation"? It seems that the research with the least controversey around it should be used to give the strongest support to a particular explanation - otherwise, we're putting up a straw man that is not NPOV (i.e., easily knocked down due to the vast amount of criticism). --JereKrischel 21:50, 19 June 2006 (UTC)
This discussion is leading in a direction that is inappropriate for WP. Suffice to say that Jensen is highly regarded, even by this scholars who disagree with him (e.g. Thomas Sowell and James Flynn). --Rikurzhen 22:02, 19 June 2006 (UTC)
I agree with JereKrischel. Jensen is severly criticzed in for example The Funding of Scientific Racism. Or for example, "Jensen claimed the problem with black children was that they had an average IQ of only 85 and that no amount of social engineering would improve their performance. Jensen urged "eugenic foresight" as the only solution"[76]
If we rely entirely on descriptions from a sub-section of one side of the debate, we may end up with a skewed perspective (right?). More can be read about how Jensen is in reality regarded in his field in Intelligence's 1998 special issue on Jensen, headlined "Kings of Men".[77] Note even leading critic Sternberg ends his article with criticism of the frequent ad hominems and with praise for Jensen.
If editors don't have access to that site, most of that issue is apparently reprinted at a third party website.[78] (I see Gottfredson's article, for example, was one of the few not reprinted there.)--Nectar 22:28, 19 June 2006 (UTC)
Regardless, Rushton and Jensen has received very large sums of money from the Pioneer fund. Their 2005 study is contradicted by its own references. So I again ask, like, JereKrischel, "do we have any non-Pioneer Fund folks whose research supports the "Genetic explanation"? Ultramarine 22:37, 19 June 2006 (UTC)
OMG. What does this have to do with editing this WP aritcle? --Rikurzhen 22:43, 19 June 2006 (UTC)
As noted, Rushton and Jensen are quoted more than any other researchers for this article. Their 2005 paper is contradicted by its own sources. Rushton has received much similar criticism for his earlier work. All prominent pro-genetic researchers have received very large sums of money. Imagine of there was much research about the benefits of tobacco sponsored by the industry. Do you agree that it would be reasonable that the claims should be duplicated by independent researchers? Ultramarine 22:52, 19 June 2006 (UTC)
Problem solved: unlike researchers who argue tobacco is benevolent, the position lead by Jensen is supported by the majority of experts in the field.[79] --Nectar 23:06, 19 June 2006 (UTC)
That is a 1987 survey of people like educators and sociologists. They may know nothing about IQ. Ultramarine 22:59, 19 June 2006 (UTC)
If you know nothing about the field and which sub-fields experts derive from that's fine, but don't make us repeat the answers to your questions every couple of weeks.--Nectar 23:05, 19 June 2006 (UTC)
Again, that survey was not of IQ researchers. It included many who probably knows nothing about IQ. Again, imagine of there was much research about the benefits of tobacco sponsored by the industry. Also, just assume that a majority of the researchers had recieved large sums of money and stated that it is beneficial. Do you agree that it would be reasonable that the claims should be duplicated by independent researchers?
Yup, this discussion now has nothing to do with writing this article. What is the point of repeating these talking points here? --Rikurzhen 23:49, 19 June 2006 (UTC)
As noted, Rushton and Jensen are quoted more than any other researchers for this article. Their 2005 paper is contradicted by its own sources. Rushton has received much similar criticism for his earlier work. All prominent pro-genetic researchers have received very large sums of money. Imagine of there was much research about the benefits of tobacco sponsored by the industry. Do you agree that it would be reasonable that the claims should be duplicated by independent researchers? Thus, if there any independent researchers arguing for the pro-genetic position they should be mentioned. Are there any? Ultramarine 23:53, 19 June 2006 (UTC)
Yes, there are many independent researchers supporting the partly genetic position.[80] Your argument is with the expert community, that they shouldn't support a position because of its funding.
(If you look at the Snyderman and Rothman study you'll see they only surveyed experts in intelligence research. These experts derived from the relevant subfields within a number of related fields, such as cognitive science and educational psychology (e.g. Jensen). This is all in the archives from the past year.)--Nectar 00:02, 20 June 2006 (UTC)
They did not survey IQ and race researchers. There are not 1020 IQ and race researchers in the world. Ultramarine 00:14, 20 June 2006 (UTC)
Rushton & Jensen - are they the best advocates for a genetic explanation?
I think maybe my point has been lost on both sides - let me try and be a bit more clear.
Both Rushton & Jensen are caught up in a maelstrom of criticism about their intentions, research methdologies, the Pioneer Fund, and the new eugenics movement. Regardless of whether or not their research is accurate or not, there is a cloud over it. Let me repeat - not using them as primary advocates for a genetic explanation is NOT a matter of asserting they are wrong.
What I propose is that less controversial researchers, not connected to the arguably facist Pioneer Fund, and not directly contradicted by people they cite, who argue for the genetic explanation, be given top billing - we don't need to censor mention of Rushton/Jensen, but they should not be the lynchpin upon which the argument is made.
In doing so, we preserve NPOV by giving the BEST argument possible for the Genetic explanation, instead of one that is already commonly attacked for a host of reasons, including easily demonstrated contradictions in methodology and reporting (that is to say, authors they cite who directly refute their interpretation of the original data).
I won't even entertain for a moment that there isn't any other research that supports the genetic explanation - nor will I assert that the motivations of the Pioneer Fund are the best argument against R&J. I'm SURE there is research out there we can prominently cite, not associated with the Pioneer Fund, that will do a better job of making the case for the genetic explanation.
As much as I may disagree with the genetic explanation, I want to give its section the best treatment possible, and the best argument possible. I think the argument is better made by research that is less controversial. --JereKrischel 23:58, 19 June 2006 (UTC)
The correlation between publically advocating for the hereditarian position and using Pioneer Funds is probably too great to satisfy your criteria. Jensen is (now approaching the end of his life) the "best" hereditarian in terms of productivity and prestige. The correlation between advocating for the hereditarian position and having bad things said about you must be nearly r=1. There are many who have merely described the hereditarian hypothesis without contributing original research: Charles Murray, Detterman, Snyderman and Rothman, etc... --Rikurzhen 00:41, 20 June 2006 (UTC)
If you say bad things about a whole group of people, things repeated by other people encouraging racial violence, you may expect criticism. Jensen has received even more money than Rushton. Ultramarine 00:49, 20 June 2006 (UTC)
I believe the phrase I used was "bad things said about you", i.e. ad hominem, which is usually taken to be a fallacy. I've yet to be convinced, and expect I never will, that Pioneer funds contaminate a researcher; but to your point, Jensen has been around longer, and probably does more expensive research (chronometrics). --Rikurzhen 00:53, 20 June 2006 (UTC)
The problem is that the Pioneer fund has such contaminated history that many people, like me, want independent research. Especially considering how Rushton and Jensen use sources. Ultramarine 00:57, 20 June 2006 (UTC)
Clearly the IQ data that comes from the U.S. gov't and testing companies is free of this issue. Was the Minessota Transracial Adoption Study paid for with Pioneer funds? [i don't know] What about the non-race IQ research? It's all be replicated to death. Since most of the scholarly disagreement is about interpretation, not the veracity of the data, I don't see how there's much of an issue in this dimension. --Rikurzhen 01:05, 20 June 2006 (UTC)
There is no dispute that there are differences in average IQ scores. But I, and probably many others, want the major studies used as evidence for the genetic explanation replicated by independent researchers. (I should just note that I do not consider it impossible that that there are genetic differences in the US. Those that become slaves were probably not the most intelligent. It is also well-known that many domestic animals are less intelligent than those free.)
With current technology, there are experiments which can be done to directly answer the question of a genetic contribution to IQ differences between admixed populations (e.g. U.S. Whites, Blacks, and Latinos). They are neither too expensive nor too time consuming to accomplish with a normal sized research grant. At least one pro-hereditarian has made a semi-public offer to anti-hereditarians to jointly conduct such a study. Thus far, there have been no takers. --Rikurzhen 01:31, 20 June 2006 (UTC)
So you say. If the Pioneer fund wanted to been seen as more legitimate, it should transform itself into an independent trust, free from its earlier history and members, with clear and open procedures for giving grants to good researchers from all sides. There should be no suspicion that the grants go to the researchers themselves by using dummy corporations. But guess this will never happen, since then Rushton would never get any money, considering his prior dubious research.Ultramarine 02:24, 20 June 2006 (UTC)
We are not seeking to stimulate research on potential race differences in intelligence. [But if you wanted to do it, here's a five step program. ] ... (i) all participants contributing to any group comparisons should be fully debriefed about the study’s aims and predictions ... Another way to preserve freedom of inquiry and reduce the potential for mischief (or worse) might be to require any studies of race and intelligence to be conducted by adversarial collaboration140. This framework offers important advantages over exchanges in the peer-reviewed literature, because investigators on both sides agree beforehand on the precise research question and the methods for testing it. They then commit themselves before data is gathered to specific interpretations of the possible outcomes, and a neutral third party conducts the work.
we suggest approaches and guidelines for assessing the contribution of genetic factors to between-group phenotypic differences, including studies of candidate genes and analyses of recently admixed populations ... [group differences in intelligence]
Okay, so I think so far, the answer to my question is that Jensen is the "best" hereditarian in terms of productivity and prestige - and we don't know of any not connected with the Pioneer Fund. I find it difficult to believe that a single group could dominate that area of the field...but I accept your expertise in that area as being greater than mine, Rikurzhen.
Do you think maybe we can separate some of the non-controversial claims from the controversial ones? Maybe in the genetic explanation, we can all accept the evidence of biogeographic diversity, but clearly indicate the "tri-level race hiearchy" proposed by Rushton is completely untenable to the vast majority of scientists in the field...maybe some strong discussion as to what we mean by "race" - as per the biological clear demarcation, the social bucket, etc, and how *that* part of their assertions is what drives the controversey, not the simple claim that genetics can have a biogeographic affect on intelligence, but that somehow, they've conflated the social view of race as prototypical...Gil-White has lots of nice stuff on this, maybe we can work it in? --JereKrischel 05:32, 20 June 2006 (UTC)
I don't see much evidence that debate about what race is is at the center of the controversy. It's really the claim that a genetic (translated to imutable by most) cause is partly responsible that elicits criticism. Rushton's work is ostenisbly the most controversial, but that may be as much a consequence of his style as his content. --Rikurzhen 05:44, 20 June 2006 (UTC)
I think you're right about that - Rushton definitely does himself no favors with his style. And part of that style is conflating pre-Nazi ideas of "Mongoloid/Caucuasoid/Negroid", with modern biogeographical genetic diversity. I mean, anytime you report your results in terms of "higher/intermediate/lower", on an vague variable like "sociability", for three distinct "races" that in fact, don't match genetic studies of diversity, you're asking to be taken to the cleaners. Not only that, but good god, to be chastized by the very researchers he cites for taking them out of context, or misinterpreting their data/results? That would be like someone claiming Bush supports gay marriage because he's stated, "marriage strenghens our society"...it's just hard to take someone seriously after that.
The way ordinary Americans separate people into supposed races is based on external appearance, for example, skin color, hair texture and color, shape of nose, thickness of lips, etc. In How Humans Evolved, Robert Boyd and Joan Silk point out that these criteria do not yield crisp categories. For example, regarding skin color, they note that if you take a plane in Scandinavia and then parachute down over Nairobi, you will notice stark differences between the people at your places of departure and arrival. However, they say, what if instead of flying and parachuting down you went on a bike tour from Scandinavia all the way to South Africa (taking a ferry across the strait of Gibraltar).[2] Where exactly would you start seeing ‘black’ people? Before you saw any ‘blacks’ you would see ‘dark browns,’ and before that you would see ‘light browns.’ At no point, actually, could you say that you jumped from the ‘white’ race to the ‘black,’ so how will you answer the question of where ‘blacks’ begin? What will you do with all those ‘intermediates’?
There are easy answers for a lot of these things. The reason why Pioneer dominates this area is because researchers (such Bouchard and Lykken) often see no reason why they should turn down the free funding that is available from Pioneer if they should choose to apply (they did). Researchers have written about how the standard sources of funding often won't touch this area because of protests, villification, etc. --Nectar 05:52, 20 June 2006 (UTC)
There are also other "easy" answers, that are not so flattering. Pioneer Fund has a dark history in the field of eugenics. One might argue that standard sources of funding won't touch this area because it is pseudo-science. Again, not trying to make an assertion one way or the other, but I think that putting up research that is so easily torn apart, both based on funding bias and particular methodology controversey doesn't give the genetic explanation its best argument. --JereKrischel 06:53, 20 June 2006 (UTC)
Although the fund typically gives away more than half a million dollars per year, there is no application form or set of guidelines. Instead an applicant merely submits "a letter containing a brief description of the nature of the research and the amount of the grant requested." There is no requirement for peer review of any kind; Pioneer's board of directors—two attorneys, two engineers, and an investment broker—decides, sometimes within a day, whether a particular research proposal merits funding. Once the grant has been made, there is no requirement for an interim or final report or even for an acknowledgment by a grantee that Pioneer has been the source of support, all atypical practices in comparison to other organizations that support scientific research (http://www.press.uillinois.edu/epub/books/tucker/intro.html)
Even if they are trying to conduct serious research, it seems that they've fallen afoul of the best practices of the industry. --JereKrischel 06:59, 20 June 2006 (UTC)
So.. you're saying they trust the researchers and don't have any strings attached to the grants. What was Tucker's comment on it, or what significance did he attach to it?
You realize in arguing this is pseudoscience, your argument is with the majority of the intelligence research community, right? That's fine; in fact, it makes our job easier, as we just need to report what the main views in the professional community are.--Nectar 07:12, 20 June 2006 (UTC)
I think you're making a conclusion that is not necessarily justified by the opinion study on the matter. Asserting that genetics influence intelligence, and then trying to combine that with the idea that specific genetic variation across geographic in some discrete ways (rather than gradual varying clines in many directions), in a 1-2-3 stepwise fashion representing blacks, whites and orientals is *not* the main view in the professional community. The Pioneer Fund has some serious credibility issues, not matter how you cut it - regardless of whether you believe those issues have any substantive basis. --JereKrischel 03:22, 21 June 2006 (UTC)
The assumption the partly genetic hypothesis in general is based on is that socially defined races are genetically non-identical.
The Pioneer Fund does have credibility issues, but luckily they don't conduct research. The credibility of researchers derives from their work and their position in their field. If the grants are hands-off and no-strings-attached, as has been indicated by Tucker's description and a number of other sources, I don't see why the source of the funding would transform research, and I'm not sure if that part has ever been explained.--Nectar 04:38, 21 June 2006 (UTC)
Ha! That's a novel interpretation - "no strings attached"...I kind of read it as "no accountability required" :). Sorry, don't mean to be chuckling at you...but that touched me!
It seems to me that the grants are doled out based on ideology, have no requirement for reporting, and are funneled to specific researchers through fronts - it almost seems like a perfect recipe for ideological payola. YMMV, of course. I just hoped we could find people with less of a cloud over their research, but apparently the Pioneer Fund is omnipresent in this area... --JereKrischel 06:11, 21 June 2006 (UTC)
Genes and red herrings
You guys are doing an interesting thing. You have an argument against brain size, so you're trying to make the section about brain size so that you can then refute it. That's good if you're pushing a POV, but bad if you care about the quality of the article.
The value of neurogenetic variance that has an unknown effect to the discussion is that it increases the probability of phenotypic variance in that area (the brain). This is a prominent point because it's long been argued that groups are neurogenetically identical. This is still the reason why the section is in the article, and Woods et al. do conclude their study "suggests that the selective pressure on these genes may be related to subtle neurobiological effects or to their expression outside the brain."--Nectar 03:00, 21 June 2006 (UTC)
Wait a second...part of the question I started proposing was *how* did the citations imply a link between race and intelligence, and the reply was this:
The only known functions of these genes are to regulate brain size, and evolutionary change in them has been previously associated with brain size increases in the primate lineage. Brain size increases in the human lineage are thought to correspond with increases in intelligence.
Now, if this is not the connection, then you're right, brain size doesn't have anything to do with it. If you're just saying that genetic variance on geographical variation somehow implies a connection between race and intelligence, then you've got to contend with the clear statement by Lahn, et al., that these two particular genes could very well have an inverse relationship to other neurological genes which produce a zero variance in effective intelligence.
Remember, neurogenetically identical is one thing, but functionally indentical is another. Once you find a genetic variance, you need to be able to discern all other genes that may compensate in one way or another. --JereKrischel 03:17, 21 June 2006 (UTC)
Brain size was the inital guess at why ASPM and MCPH1 were rapidly evolving, but the actual part of the finding that was salient was neurogenetics. The Moyzis paper is actually more salient than the Lahn papers as it gets at the magnitude of the neurogenetic differences -- they are large. Nectar is correct. Brain size is a red herring. --Rikurzhen 04:25, 21 June 2006 (UTC)
Sorry, I think you're losing me again - are we back to the assertion that ASPM and MCPH1 are somehow representative of all intelligence based neurogenetic genes, and that their biogeographic distribution directly implies a biogeographic distribution of intelligence? Step me through your POV here in baby steps - conclusions you're able to bridge without thinking about it seem obscure to me. --JereKrischel 05:01, 21 June 2006 (UTC)
Read the abstract of the Moyzis paper. It says genes involved in neuronal function (and other functions) have been evolving recently. ASPM and MCPH1 (Lahn's findings) are just two (high profile) additional examples. The important point for this section is that Several recent studies have found some neuronal genes have variants that have spread to high frequencies under selective pressure and now occur in different frequencies in different global populations. This is evidence of (1) neurogenetic diversity and (2) selective pressure that affects neuronal genes in "recent" times. --Rikurzhen 05:07, 21 June 2006 (UTC)
Yeah, I read the abstract - but how do we get from neurogenetic diversity (especially if the effects are recent), to an implication that the "classic" races of Mongoloid/Negroid/Caucasoid (purported by Rushton/Jensen) have IQ differences? We have to assume that neurogenetic diversity is *functional*, right? That is to say, all genes related to IQ from the pool of all neurogenetic genes distribute in a skewed manner, right? And even more salient, we have to assume that these "recent" changes are representative of any changes since Rushton/Jensen assert the 3 "main races" separated - that is to say, if evolution is occuring on such a recent scale, we have no basis to believe that the asserted linear Negroid->Caucasoid->Mongloid sequence has remained static throughout a long period of time where adaptations could occur. Specifically, populations, for example, who migrated from upper latitudes to lower latitudes or vice versa, and it is asserted that Caucasoid "evolutionary pressures" were caused by cold weather, any migrations within recent times that haven't been long enough to produce morphological differences (skin color/hair type), would still have exerted selective pressures on neurological genes. --JereKrischel 05:36, 21 June 2006 (UTC)
This data provides indirect evidence of the plausibility of a genetic cause to currently observed race differences. It confirms a prediction of the hereditarian hypothesis, which is that there should be neurogenetic diversity. The sample populations in the data set on which the conclusions are based happen to be of unambiguous European, East Asian, and African descent; touching the populations we're interested in. Don't read too much into this. Also, don't confuse Rushton and Lynn's evolutionary hypotheses with the more general view put forward by Jensen. The existence of recent changes doesn't tell us anything about older changes. A total examination of the genomic data is still not available, and we can expect more details to come. At the moment, the conclusions are limited and specific. --Rikurzhen 06:20, 21 June 2006 (UTC)
(Edit conflict)Re:"We have to assume that neurogenetic diversity is *functional*, right?"
No, the only statement here is that neurogenetic diversity increases the chances of functional diversity. (How could that statement not be true, right?)
Re:"That is to say, all genes related to IQ from the pool of all neurogenetic genes distribute in a skewed manner, right?"
No, if some genes related to IQ distribute in a non-identical manner than that implies these groups are non-identical in IQ.
The last part of that post wasn't quite clear, but I'll respond that some scientists, such as Cochran, focus on selection in recent times as the explanation of population differences.--Nectar 06:30, 21 June 2006 (UTC)
Okay, I think I understand the point you're making, and I gave it top billing in the genetic explanation section. I went through some of the Jensen stuff to make sure we're asserting beliefs, not conclusive facts, and changed the "four main lines" of data into more sections - they really were more than 4 of them, and the wording ended up being confusing (hard to tell where one sentence fragment regarding one set of data left off, and the other one began). Help in making sure we keep NPOV throughout that section is appreciated! --JereKrischel 08:13, 21 June 2006 (UTC)
Regarding No, if some genes related to IQ distribute in a non-identical manner than that implies these groups are non-identical in IQ.:
I'm not sure if you can make that assertion - if *all* genes related to IQ distribute in the *same* non-identical manner, then that implies that those groups are non-identical in IQ. Maybe even if *most* genes related to IQ distribute in the *same* non-identical manner, you could make that assumption. But that's like taking three bags of currency, finding out that pennies distribute differently amongst them, and then asserting that because of that, they 1) must contain different amounts of money and 2) you can infer the distribution of money by the distribution of the pennies. At best, I think that you can say that if some genes related to IQ distribute in a non-identical manner than that implies that those groups *could* be non-identical in IQ. It hardly comes close to implying anything - it only leaves the possibility open. --JereKrischel 08:21, 21 June 2006 (UTC)
Rushton/Jensen argument
I've reverted the changes. The Ruhston/Jensen argument needs top billing as the most important. --Rikurzhen 08:17, 21 June 2006 (UTC)
I've got to disagree here - top billing should be given to the least controversial, not the most. And the edits to their "four main lines" make things more readable. If we're just quoting them directly, let's quote them...but otherwise, their list of four is just completely incomprehensible. Furthermore, the data they cite and Rushton's "aggregation" methodology without critical analysis is *highly* contested - we need to make clear that their beliefs are not incontrovertible fact. --JereKrischel 08:26, 21 June 2006 (UTC)
Jensen is the preeminent researcher within the partly genetic hypothesis and is one of the biggest figures in intelligence research. [81] The list of four points and the section as a whole have been carefully written by a number of editors over the past year. I'm not sure what you mean about "aggregation" methodology.--Nectar 08:59, 21 June 2006 (UTC)
Primarily a Rushton critique, but it is relevant to the basis on which Jensen asserts these "converging data":
Rushton claims to use a methodology he calls the "aggregation" of evidence, in which he averages hundreds of studies, modern and historical, with equal weight to demonstrate the patterns he asserts. He claims that by averaging many studies, which may include those that used poor research methods, the results one gets can be very accurate. He claims that measurement error typically cancels out when multiple studies are averaged, and asserts that his approach is less biased than the work of researchers who selectively pick and choose from the world-wide literature based on critical analysis.
Douglas Wahlsten, who in a review of Rushton's book wrote: averaging does no thing to reduce bias in sampling and measurement, and such flaws are abundant in the cited literature. For example, among the 38 reports on brain weight, all but two gave figures for only one group, with most cases being people living in the nation of their ancestors, such as an article on Japanese living in Japan and another on Kenyans living in Kenya. The obvious differences in environment make all of these data of dubious worth for testing hypotheses about genetic causes of group differences.
I'm simply concerned that controversial assertions regarding data are being presented as incontrovertible fact. The list of four points is muddied as it stands. Let me address them each in turn:
# Worldwide Black–White–East Asian differences in IQ, reaction time, and brain size, with Black-White IQ differences observable in the U.S. at every age above 3 years, within every occupation or socioeconomic level, in every region of the country, and at every time since the invention of ability tests.
This states flatly that there are "Worldwide black-white-east asian differences", which is highly debatable both on the arbitrary categorization into three "races". It then uses the word "with" to introduce the second line of data conflated into this single sentence.
This second line is making such incredibly sweeping claims, it's hard to take seriously - "within every occupation or socioeconomic level"? Without discussing what categories of occupation or socioeconomic level they used, the superlative is hardly justified. Similarly "in every region of the country" - what regions? "At every time since the invention of ability tests" - this asserts that at any arbitrary point in time, in any subregion, in any occupation, and in any socioeconomic level these IQ differences exist? This is also where the critique of Rushton's "aggregation" comes in - if they collected their data regardless of the critical analysis of its worth, averaging it does not improve the data quality. --JereKrischel 09:15, 21 June 2006 (UTC)
# US race differences are most pronounced on tests that are the best measures of g[85](see Spearman's hypothesis), which also show the highest heritability, show the highest inbreeding depression scores[86], and on tests which are the least culturally loaded
Again, the wild use of superlatives..."least culturally loaded", "best measure of g" - these are not objective statements being made, and they shouldn't be presented as incontrovertible fact. --JereKrischel 09:20, 21 June 2006 (UTC)
# The rising heritability of IQ with age (within races; on average in the developed world heritability starts at 20% in infants, rises to 40% in middle childhood, and peaks at 80% in adulthood) and the disappearance (~0.0) by adulthood of shared environmental effects on IQ (for example, family income, education, and home environment), making adopted siblings no more similar in IQ than strangers[87]
This one is mostly just another conflation - there are two data lines here, and they should be separated. --JereKrischel 09:20, 21 June 2006 (UTC)
#US comparisons of both parents to children and siblings to each other finds regression to differing means for different races (85 for Blacks and 100 for Whites) across the entire range of IQs[88], despite the fact that siblings are matched for shared environment and genetic heritage, with regression unaffected by family socioeconomic status and generation examined[89]
Again, controversy as to the quality of the data, the viability of arbitrary racial categorization, etc...perhaps if it was preceded by "Some...", and instead of asserting "despite the fact", asserting "despite". --JereKrischel 09:25, 21 June 2006 (UTC)
I've made the most minor changes - hopefully we can address the first 2 lines of data mentioned, and the inappropriate assertion of incontroverible fact they make. --JereKrischel 09:28, 21 June 2006 (UTC)
Your comments suggest that you do not understand this data, where it came from, or how it is treated by critics of hereditarianism. You should leave the original version while we discuss this on the talk page. I looked at all your edits and I believe them to be very bad. To begin, you should read papers cited from the June issue of PPPL http://www.udel.edu/educ/gottfredson/30years/
I'm going to revert again to restore the very carefully crafted consensus version. Let's not change it until the talk page discussion is complete. --Rikurzhen 17:11, 21 June 2006 (UTC)
Please be more specific - can you address my issues directly? The fact that the current list is not clearly understandable seems to indicate a need for change, don't you think? --JereKrischel 17:15, 21 June 2006 (UTC)
Change for clarity is always a good idea. The criticisms you've listed appear to be your own. They also appear to express misunderstanding. The best place to start is to read the source(s) and I can explain what you don't understand. For example: there's no doubt that there are world-wide B-W-EA differences in IQ, but the magnitude of that gap (not discussed) is controversial. The phrase "best measure of g" means highest g-loading, which is a mathematical calculation. Heritability and shared-family effects are measured simultaenously and go hand-in-hand such that (heritability + shared effects + non-shared effects) = 1. The fact that people can and do reliably describe themselves as "White", "Black" and "East Asian" in many circumstances, and that those categories can be used from social science and psychology research is non-controversial. Data quality has never been an issue AFAIK with these conclusions as they are in many cases data collected by U.S. government agencies. --Rikurzhen 17:37, 21 June 2006 (UTC)
Perhaps you're thinking of criticisms of Rushton's attempts to measure many "life history variables" which are not enumerated here and don't affect these data. --Rikurzhen 17:39, 21 June 2006 (UTC)
Getting #3 to have sub-bullets was a neat trick. --Rikurzhen 17:45, 21 June 2006 (UTC)
Thanks, I think it needs to be done for bullet #1 too...please feel free to make that change!
Regarding your other reponse - I'm sorry, isn't there great doubt about even the utility of B-W-EA categories? Before you even get to any observed gaps, there is a pretty high bar that needs to be overcome to assert applicability - particularly in light of things like Cavalli-Sforza, which instead of 3 clusters found 4 (at least)...
Regarding "highest g-loading", it is more than just a mathematical calculation - it is an assertion that a specific mathematical calculation represents the best model for general intelligence, right?
Regardless of your particular faith in the data quality, and the applicability of limited social labels to any imagined biogeographical discontinuities, it seems harmless to make explicit that these trend lines are not universally interpreted as incontrovertible fact. Is there a particular reason why disclaiming the data observations bothers you? Is there specific compromise disclaimer wording you would accept? --JereKrischel 17:50, 21 June 2006 (UTC)
(1) most of the data being discussed is about B-W differences in the U.S. There's no doubt that there are people labeled as "Black" and "White" in the U.S.
(2) there's already a disclaimer that researchers discusing this topic take for granted that that g exists
(3) There's already a disclaimer at the top of the article about race, a discussion of race in the first main section, and the disclaimer that critics do not interpret this data as indicating
You really should read the 2005 PPPL papers to get an idea of where the controversy actually exists. It's not about whether these data exist or what they mean in simple terms but what they imply for the cause of the IQ gap. --Rikurzhen 18:07, 21 June 2006 (UTC)
Hey Rikurzhen, thanks for helping out on the section. It's starting to look much better. Let me chew on it for a while, and see if I can propose some alternate text that might fit both of our concerns. Thanks again! --JereKrischel 21:23, 21 June 2006 (UTC)
Total red herring, but it made me think
There's a line in there from Lynn:
Lying about race differences in achievement is harmful because it foments mutual recrimination. Because the untruth insists that differences cannot be natural, they must be artificial, manmade, manufactured. Someone must be at fault. Someone must be refusing to do the right thing. It therefore sustains unwarranted, divisive, and ever-escalating mutual accusations of moral culpability, such as Whites are racist and Blacks are lazy.[1]
If we can assert that there are biogeographic distributions of intelligence, why not assert biogeographic distributions of racism, or laziness, or any other brain-based function? Boy, wouldn't it be funny if a study was done that correlated slave owning societies and discriminatory societies and found that there was a 1-2-3 stepwise ladder with Whites being the most racist, and White scientists being most prone to misinterpret data? :) --JereKrischel 21:21, 21 June 2006 (UTC)
(The quote isn't from Lynn.)--Nectar 21:30, 21 June 2006 (UTC)
You're right, it's Linda Gottfredson (Lynn...Linda...jumbled them up for a sec). Still, can you imagine the uproar if a series of studies indicated that white scientists are more likely to misinterpret data and be racist than black or east asian scientists? :) --JereKrischel 21:45, 21 June 2006 (UTC)
Yeah, there's probably more humorous stuff than we might sometimes expect.--Nectar 22:42, 21 June 2006 (UTC)
