Talk:Race and intelligence/Archive 72

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 65 ← Archive 70 Archive 71 Archive 72 Archive 73 Archive 74 Archive 75

Lead

I need to run for a bit, but I want to mention the actual lit reviews in the lead, as these are peer reviewed versions of this article. Neisser et al 1996 and Sternberg et al 2005 look very important. Any other suggestions? T34CH (talk) 02:56, 12 October 2009 (UTC)

I think peer reviewed survey content should probably be covered in the first section after the lead instead of the current "History/Contemporary" stuff that's there. I'll look at the reviews in my spare time. Aprock (talk) 05:23, 12 October 2009 (UTC)
Aprock, I like that idea, though I think that such reviews should be used to source any consensus or lack there-of in the lead. The section you suggest could be titled something like "Academic consensus", and discuss what we actually know more in depth. T34CH (talk) 15:24, 12 October 2009 (UTC)
Yes, "Academic Consensus" sounds much better than "Truth from on High", especially as it allows changes over time in the consensus to be mentioned, illustrating that this is an evolving field. Aprock (talk) 16:14, 12 October 2009 (UTC)

heritability of intelligence within races (added break)

  • "the removal of properly sourced information (cited from peer-reviewed journals, not the NYT or Medical Hypothesis) by him"

Actually I didn't remove cited information from peer-reviewed journals. I kept the information in the article, I simply changed what it said. This is the reason, the source states:

It is clear (Section III) that genes make a substantial contribution to individual differences in intelligence test scores, at least in the white population. The fact is, however, that the high heritability of a trait within a given group has no necessary implications for the source of a difference between groups (Loehlin et al, 1975). This is now generally understood (e.g., Herrnstein & Murray, 1994). But even though no such implication is necessary, some have argued that a high value of h2 makes a genetic hypothesis more plausible. Does it?[1]

This is what the article stated:

The consensus among intelligence researchers is that IQ differences between individuals of the same racial-ethnic group reflect real, functionally and socially significant, and substantially heritable differences in intelligence.[2]

These are not saying the same thing at all. As I understand it, when the heritability of IQ is measured within poor and black communities it is significantly lower, so the high heritability of white middle class populations may well be a product of the homogeneous environment, and not attributable to genetic causation. So I changed the text to more accurately reflect what the source says

IQ differences between individuals from within any "white" test group reflect a substantial genetic contribution to intelligence.

That's much closer to what Neisser et al actually say. I also removed some citations from books, that's because they are not cited properly, book citations should at least give page numbers. Cheers, Alun (talk) 07:08, 13 October 2009 (UTC)

Whether page numbers should be included depends on whether the entire book is being cited. If the entire book is devoted to the topic of IQ being heritable, specific page numbers aren't necessary.
Since I don't have all nine of the originally cited sources on hand, I can't easily determine whether you're right that none of them support what this sentence said before you edited it, but I would ask that you wait for there to be a consensus about this before removing this sentence again. The previous consensus was that it was acceptable, and it will require a new consensus to overturn that. --Captain Occam (talk) 07:26, 13 October 2009 (UTC)
No, you need to cite page numbers if you are going to cite a book, that's standard on Wikipedia, and in academia. I didn't say that none of then support what's claimed, what I said is that the claim here is overstated. What I have done is change the text so it is more in line with what Neisser et al. actually say. That doesn't require a consensus for the change. By that logic no one would ever be able to edit without getting consensus first. I have some literature at home that address this, and there is literature that shows that for poor and black groups heritability may well be much lower than it has been measured for middle class white people. I'll take a look, but I see no problem with making the claim more similar to what Neisser et al. say. I even quote what Neisser et al. say in my above post, and link to their paper so anyone can go and read it. I don't see the problem here, it's not like I'm trying to remove the claim, only make it more accurate. Alun (talk) 07:34, 13 October 2009 (UTC)
Is that really a standard on Wikipedia? It most certainly is not in the academic world, that is, it is perfectly acceptable to refer to an entire work with the author's name followed by the year if the work in question is being referred to as a whole. --Aryaman (talk) 07:37, 13 October 2009 (UTC)
Wikipedia:CS#Including_page_numbers:If you are quoting from, paraphrasing, or referring to a specific passage of a book or article, you should if possible also cite the page number(s) of that passage.
This is a specific claim about heritability, so we should say exactly where in a book this claim is made. It is not a broad thrust argument that a whole book might cover. It states that heritability is always high within groups. But at least one of the sources cited doesn't make that claim at all. I'd like to see exactly what the other sources say, and where they make this claim. It's a controversial claim, and exceptional claims require exceptional sources. Alun (talk) 07:48, 13 October 2009 (UTC)
As I said, I don’t have most of these sources on hand at the moment, but I already know that at the “Mainstream science on Intelligence” statement supports this claim as it was originally phrased. This report says, “Individuals differ in intelligence due to differences in both their environments and genetic heritage. Heritability estimates range from 0.4 to 0.8 (on a scale from 0 to 1), most thereby indicating that genetics plays a bigger role than does environment in creating IQ differences among individuals.” It does not say anywhere that this within-group heritability is limited to whites.
So now we see that there's one source that supports this sentence in the article as it was originally phrased, and one that doesn’t. We ought to look at some of the remaining seven sources now. --Captain Occam (talk) 07:56, 13 October 2009 (UTC)
I just looked up Sternberg's book. Here:
"It is now widely recognized that intelligence in a heritable trait." [Brody (2000:27)] in [Nathan Brody (2000). "History of the Theories and Measurements of Intelligence" in: Sternberg, Robert J. Handbook of Intelligence. Cambridge: Cambridge University Press. ISBN: 0-521-59371-9.]
Does that help? --Aryaman (talk) 08:05, 13 October 2009 (UTC)
PS: Let me add this:
"Binet, Galton, and Spearman are the progenitors of the contemporary study of intelligence. Their contributions resonate in our contemporary discussions of research on intelligence. Galton's belief in genetic influences on intelligence is widely accepted. His belief in a genetic hierarchy of ability defined by ancient Greeks, British people, and Africans is analogous to contemporary discussions of a racial hierarchy of Asians, Whites, and African (see Jensen, 1998; Lynn, 1987; Rushton, 1995). Although belief in a racially defined, genetically influenced hierarchy of intelligence is not accepted by many contemporary students of intelligence, such beliefs continue to be expressed. A majority of psychologists assumed to be knowledgeable about intelligence surveyed by Snyderman and Rothman in 1990 believed that racial differences in scores on intelligence tests were partially determined by genetic differences between individuals with different racial identities. Galton's belief in a relationship between intelligence and elementary cognitive processes is related to an active program of contemporary research. Spearman's belief in the overwhelming importance of g is present in many contemporary discussions, and attempts to measure intelligence similar to those initiated by Binet are common." [Brody (2000:30)]
I'm still looking for info on "IQ differences between individuals of the same racial-ethnic group" --Aryaman (talk) 08:34, 13 October 2009 (UTC)
PPS: You can add Loehlin, John C. Group Differences in Intelligence (2000:182) from the Sternberg volume to the list of citations for this particular claim. He gives a detailed summary of Plomin, DeFries, McCartney, Harris, and Bernieri (1997); McGue et al. (1993); McCartney, Harris, and Bernieri (1990) and Neisser et al. (1996) which supports the statement above. --Aryaman (talk) 08:43, 13 October 2009 (UTC)
You misunderstand my problem. My problem is not whether IQ is a heritable trait, it is whether the sources support this statement

The consensus among intelligence researchers is that IQ differences between individuals of the same racial-ethnic group reflect real, functionally and socially significant, and substantially heritable differences in intelligence

So I know that Neisser state that

It is clear (Section III) that genes make a substantial contribution to individual differences in intelligence test scores, at least in the white population.

Which is not the same thing at all as saying "individuals of the same racial/ethnic group". So I repeat, which sources claim that heritability is always high in all racial/ethic groups? My understanding is that heritability is known to be high in middle class white groups, but that little work has been done on other groups. I also have some recollection of a paper that claimed that in low ses groups heritability was more like 40, which is not that high. Some suggest that this is due to heterogeneous environments in low ses groups. This argument has also been used to explain the lower heritability of IQ in children compared to adults, i.e. that adults have more control over their environments, and so environment has less of an effect on IQ. So let's stick to the problem, which is not that IQ is heritable, but that it is equally highly heritable in all groups, especially the claim about "races". Thanks. Alun (talk) 11:17, 13 October 2009 (UTC)
Based solely on a quote comparison (I haven't poured through the source), Alun appears to be correct. The "at least in the white population" part of the source is left out of our text, and the source seems to make a point of qualifying the preceeeding statement with it. To be true to the source, something like that needs to be added. --Nealparr (talk to me) 11:48, 13 October 2009 (UTC)
The same-race results are primarily derived from studies done on twins as far as I can tell. That's why I mentioned Loehlin (2000:182) above. It does support the statement you're pointing to, and does so while summarizing 4 other pieces of literature. I don't know whether they've done research outside of "middle-class white" groups, but I'm a bit perplexed by the suggestion that it matters. Do they need to state "we've done studies on all extant racial groups" before you'd be satisfied with the statement? Either way, I'd think that Plomin, DeFries, McCartney, Harris, and Bernieri (1997), McGue et al. (1993), and McCartney, Harris, and Bernieri (1990) would need to be consulted if you'd like to find out. --Aryaman (talk) 12:31, 13 October 2009 (UTC)
No Aryaman they don't need to state "we've done studies on all extant racial groups", but they do need to say that intelligence is highly heritable within ethnic/racial groups, because that's what we say in the article. A blanket statement about heritability being high is not the same thing. But there's another problem which I'm going to address below. Alun (talk) 06:21, 14 October 2009 (UTC)

"Race and intelligence" within the scope of "Intelligence"-related articles

I'd like to open a discussion on the scope of this article in relation to Intelligence and what should arguably be considered its child articles:

These are all factors with arguable relevance to intelligence, and it would seem to me that all of them need to be taken into consideration in any discussion regarding the scope of one of them. In my opinion, these articles should be largely comparable. Right now, Race and intelligence would appear to be serving as a theater for one aspect of the nature vs. nurture debate, with the result that the article is on the brink of becoming an unreadable mass of caveats and couched warnings. I think much of the conflict over this article could be circumvented if more effort was put into adequately developing the discussion at Intelligence on these factors and bringing the child articles into line with that. There is, as far as I know, no need to attempt to develop each of these articles to the point that they could stand alone. Rather, each one should expand upon a particular factor and leave the discussion regarding the relative importance and/or acceptance of said factor up to Intelligence. Comments? --Aryaman (talk) 07:30, 13 October 2009 (UTC)

Agreed. This is what I was trying to get at with my suggestion above about splitting the article into two parts. One was to attack the issue is to think in terms of three levels. At the highest would be Intelligence. At the second level would be two overview articles that split up all the children articles into two (?) categories. The first of these would be "Intelligence and Human Groupings." (Perhaps intelligence should always be the first word in each title?) This would cover articles like Sex and intelligence, Religiosity and intelligence, Fertility and intelligence and Race and intelligence. Each of these articles might be thought of as the child of the "Intelligence and Human Groupings" article and the article about that specific human grouping. All the debates about that grouping (is it meaningful? is it socially constructed?) belong in the article about that grouping. Each child article could then be much smaller.
The second second level article would be something like "Intelligence and Its Covariates." (Lousy title, I know.) The child articles in this branch would be stuff like Environment and intelligence, Health and intelligence, Height and intelligence and so on. Again, each article is viewed as the child of the main Intelligence article and the associated covariate article. Debates about the meaning, definition and so on of terms like "health" or "height" belong in those articles.
Needless to say, the above is not the only reasonable plan. Thinking of "height" as a covariate and "religiosity" as a grouping is arbitrary. The overall goal is to make each of the lowest level articles much shorter and more concise. It is silly to refight debates over the meaning of intelligence in each location. David.Kane (talk) 13:36, 13 October 2009 (UTC)
Agreed. As far as the titles go (i.e. having "intelligence" first), we might need to consult with the folks at the relevant projects related to the topics with which these intersect (e.g. Health and intelligence intersects heavily with medicine, etc.). I'm happy either way. Regarding your distinction between the three "levels", I think it's something worth developing, especially as - and I really liked this part - it would help reduce the 3rd level articles considerably. Is there some venue where this discussion would be more appropriate, considering that it spans quite a few articles? If not, I'm fine with continuing it here. Thanks for the sane response, btw. :) --Aryaman (talk) 14:51, 13 October 2009 (UTC)
@Varoon Arya. In general I agree, but in particular I don't think there's any point in waiting to make changes, especially given the poor state of the article as it stands. If you are interested in pursuing this line of development, there should probably be an "intelligence project" started so that all the articles in the series can be identified, and cross linked. My view is that work can be generally independent to cleaning up of this article until all the infrastructure is in place. Aprock (talk) 16:23, 13 October 2009 (UTC)
Agree with Aprock that work on this article should continue independent of work on a wikiproject. T34CH (talk) 16:32, 13 October 2009 (UTC)
(ec)
Aryaman, this needs to become a wp:WikiProject or a centralized discussion. Preferably a wikiproject as that is a group of people working together, while a centralized discussion is a group of people arguing. Eventually, a wp:Portal could be developed. I think this is a good idea as well, allowing this article to focus on being a summary of the academic literature. The above suggested outline would still be appropriate as a reader would want a complete picture when reading this article (again, in summary with links to the most appropriate main article). I suggest starting work at Wikipedia:WikiProject Intelligence (currently a redirect to military intel.), with notifications at the relevant psychology, sociology, biology and education portals, as well as the village pump. Using this page for the discussion would slow down the improvement of this article. T34CH (talk) 16:32, 13 October 2009 (UTC)
I would just like to add to the list a topic that covers intelligence from a different perspective.
Wapondaponda (talk) 16:34, 13 October 2009 (UTC)

Outline

I think the outline of the article is very convoluted. As suggested above by Aprock, I think one great addition would be to discuss current consensus up front as described by the most appropriate lit reviews. The current outline is very unclear (e.g. are "test score differences" and "genetic and environmental factors" not contemporary issues???), mislabeled (e.g. the 1970s is "history") and misleading (e.g. the majority of academia seems to be criticizing those who hold the strongest hereditarian beliefs, but having a "criticism" section suggests the opposite). Even the "external links" section needs a lot of attention (e.g. the "review papers" are not review papers).

I suggest a format roughly as follows:

There are other issues that might be worth mentioning, such as Scientific racism and Racialism. Much of the 3rd level headers are just examples and sources of information, not a full outline. How does this look to others? T34CH (talk) 16:48, 12 October 2009 (UTC)

I agree that redoing the outline is important, but I worry that it will require a lot of meta effort. One way to break down the effort is to agree to a future outline here on the talk page, and then work incrementally on the article page, adding missing sections, rewriting required sections and eliminating redundant, or non-relevant sections. In building a future outline, I think we should just start at the highest level where consensus should be easier to build. So I might suggest:
  • Academic consensus
  • Discussion of the major issues
    • Issues in discussing race
    • Issues in discussing intelligence and IQ
    • Statistics issues
  • Evidence of environmental factors affecting population IQ
    • Effects of health
    • Effects of culture, education, ses
    • Institutional racism, castes
  • Evidence of genetic factors affecting intelligence
  • History
    • prior history
    • recent history
    • current viewpoints
Even that may be too fine grained to achieve a starting consensus, but I think coarse to fine is the way to go here. Aprock (talk) 16:59, 12 October 2009 (UTC)
Agree. Nice outline. Alun (talk) 17:10, 12 October 2009 (UTC)
Also agree, with the caveat that it may be changed by consensus as appropriate.--Ramdrake (talk) 17:23, 12 October 2009 (UTC)
Nice outline. I like the discussion of major issues right upfront because that seems to be what appears most talked about, the social issues surrounding this kind of thing. --Nealparr (talk to me) 18:21, 12 October 2009 (UTC)
One concern about this outline is that it doesn't leave much room for history. Putting it at the end of the article may be reasonable, or possibly creating a separate article and referring to it here. It seems like a large enough topic to warrant it's own subject, and different enough from current understanding that we probably want to separate the two topics. Aprock (talk) 22:14, 12 October 2009 (UTC)
I was just about to say the same thing regarding History. I think an inherently controversial topic such as this one needs to be presented in its historical context. Also, the history of a controversy should be something editors can agree upon and work together to improve, as it mostly revolves around more or less concrete events with little room for commentary. This might help to repair some of the damage that has been done recently in the collegiality department. However, I think it would be best to expand the History section considerably before considering a split-w/-summary-type move. --Aryaman (talk) 22:23, 12 October 2009 (UTC)
I agree about the article needing some space to discuss history. There are also a couple other elements that I think it ought to include:
1: Somewhere, we ought to mention what the actual average tests scores are. Perhaps it could go in the "academic consensus" section. (Everyone agrees on what the test score difference is, even if they don't agree on its cause.) Since the explanation of this data is the central point in this controversy, the data itself ought to be somewhere.
2: Somewhere we need to mention the viewpoints of various researchers about this. The article currently does this in two parts: the "genetic hypothesis" section mentions Jensen, Eysenck and Cattell as supporting this viewpoint, and the "viewpoints" section lists 15 researchers who support this viewpoint (although oddly omitting Cattell) and seven who support the environmental hypothesis. At the very least, the two lists of researchers who support the genetic hypothesis should be combined into one. --Captain Occam (talk) 22:34, 12 October 2009 (UTC)
I agree that there should be a discussion of actual test results. One might put a broad description in results in the Academic Consensus section, with more detailed information in the issues area, probably in intelligence/IQ area, but maybe it's own subsection. With respect to viewpoints, I think that should generally be wrapped up into a History section. I've added that to the outline above, keeping in mind that the outline is currently fluid. Aprock (talk) 22:39, 12 October 2009 (UTC)
Agree to the history. Test results, no opinion. Viewpoints of individuals, I disagree that it should be in here. The reason I'm opposed to listing of individual's view points is because this is an overview article, meant to be written in WP:SUMMARY style. The articles devoted to those view points are the proper place for a list of individuals who agree with it and their views. We also run several risks related to WP:UNDUE, that is, giving non-notable or less notable view points more weight than they deserve. --Nealparr (talk to me) 22:47, 12 October 2009 (UTC)
If folks are generally in agreement with the idea that more needs to be done on the History section, would it be in order to suggest that editors discuss literature which could be used for such a section, i.e. identify solid, neutral, secondary reviews of the history of this controversy? The neutrality of such sources might be an issue which requires careful consideration, but identifying possible sources should be a relatively straightforward matter, should it not? Perhaps a new talk section could be created for listing candidates for inclusion. --Aryaman (talk) 23:00, 12 October 2009 (UTC)
I agree that the history of the debate might be useful, but it may be difficult to get right without overshadowing the rest of the article (as it does now). Putting it at the end is a decent stop-gap. The test scores of course should be mentioned because that's what all the fuss is over in the first place. That should have gone without saying that you have to talk about the scores in the IQ section. Aprock makes an excellent appoint above by suggesting that we start with a spare outline and work from there. As for a list... that could become very misleading as not every researcher has explicitly published their personal view. There may be selection bias in compiling such a list. T34CH (talk) 23:06, 12 October 2009 (UTC)

Setting aside for a moment the issue below of naming names, can we assume we have consensus to begin reformatting the existing text to fit the suggested outline? That much does seem uncontroversial to me given the positive comments from everyone thus far. I do think we should wait a day or two so that we're proceeding with an abundance of caution... so that there may be no doubt that we have consensus. T34CH (talk) 00:35, 13 October 2009 (UTC)

I think so. --Nealparr (talk to me) 00:36, 13 October 2009 (UTC)
Although I approve of the suggestion to clean up the article, I would recommend that we create a draft for the new version in someone’s userspace, rather than trying to immediately make these changes to the article directly. Individual sentences in this article have been so contentious, it seems like it could easily be a nightmare to try and edit the entire article at once. --Captain Occam (talk) 00:50, 13 October 2009 (UTC)
You may have misunderstood Occam. I suggest moving sentences around to conform with the proposed outline, not writing new sentences. Writing an entirely new draft was one of the old suggestions. I don't think that getting consensus to get consensus about consensus is productive. T34CH (talk) 00:58, 13 October 2009 (UTC)
Moving sentences around would be fine, but if there was any content that you were thinking of actually removing, it might be prudent to obtain consensus about that first. --Captain Occam (talk) 01:52, 13 October 2009 (UTC)
Yes, I would very much like to see outline editing be separate from content editing as well. Maybe we should all try to set aside content issues over the next couple of days and work on getting something that is in the general shape of the outline (although certain sections would be missing). Aprock (talk) 03:03, 13 October 2009 (UTC)

List of supporters

"Selection bias", that's the point I was trying to make with the potential WP:UNDUE problems. Thanks. I was going to say, if you list off supporters of position A, and list off supporters of position B, you get into pissing contests of which list is longer. Say you listed off five supporters of position A and five supporters of position B (because that's all you could find that wrote a paper), but an entire scientific body agrees with position A and states that position B is unproven, you run the risk of giving position B greater weight than it deserves. Fortunately, summarizing the positions avoids all of that. --Nealparr (talk to me) 23:13, 12 October 2009 (UTC)
I humbly disagree. If the article states that "most experts agree that X is Y", it becomes irrelevant who is listed in favor of that view, as you've already said that the majority hold it. At the same time, listing those who disagree with the majority view by name becomes even more important, for these individuals stand out from the majority, and - assuming they are not crackpots - must do so for some very good reason, which is exactly what the reader needs to know. There is no need for such a "pissing contest". Stating the majority view as the majority view and providing one or two experts explaining the majority view is enough. But the qualified "deviant" experts need more detailed coverage. If that conflicts with WP Policy, so be it. That's common sense and good editorial practice as opposed to writing academic propaganda. --Aryaman (talk) 23:35, 12 October 2009 (UTC)
I'm not sure how you can say "If that conflicts with WP Policy, so be it." This is exactly why wp:UNDUE exists. There can be a separate List of people who think that white people are innately smarter than black people, but in this article it is entirely inappropriate and in fact amounts to nothing less than "propaganda". T34CH (talk) 23:41, 12 October 2009 (UTC)
Right. Listing the "qualified 'deviant' experts [with] more detailed coverage", in this article, would be directly conflicting with NPOV policy. A list of that type, itself, would conflict with guidlines for how overview articles should be written (ie. summary style). There's nothing barring naming names and giving details about their views in spinoff articles. It's a matter of where such things are appropriate. --Nealparr (talk to me) 23:50, 12 October 2009 (UTC)
Nothing I suggested goes against WP:UNDUE as far as I can tell. If is appears to do so, then I have failed in communicating my position clearly, and I apologize. I'm not going to quote WP:UNDUE, as it's there for anyone to read. But qualified minority views need to be explained in enough depth so that readers can understand exactly how and where they differ from majority views. Shuffling them off to their own "spin-off" articles is not in the spirit of NPOV. Reading the phrase "in proportion" to mean we sit down and start counting experts is childish - not that I'm accusing anyone here of childish behavior (though I cringe at having to add that caveat). --Aryaman (talk) 00:09, 13 October 2009 (UTC)
I think minority viewpoints should generally be mentioned in the text of the article where appropriate. Aprock (talk) 00:14, 13 October 2009 (UTC)
Perhaps I was unclear in the use of the phrase "more detailed". I meant for the emphasis to be on the word "detailed" as opposed to "more", i.e. I'm not talking about the quantity of coverage, I'm talking about the detail of the coverage of the minority views. I'll try to be more precise in the future. Thanks, --Aryaman (talk) 00:18, 13 October 2009 (UTC)
The NPOV policy is pretty clear about both the depth of coverage and details in regards to minority views. --Nealparr (talk to me) 00:31, 13 October 2009 (UTC)
I'm not sure that we need an actual "list" per se, but the identifies of some of the especially prominent researchers who hold the minority viewpoint about this (that is, the hereditarian view) probably ought to be mentioned because they're one of the main reasons why this hypothesis is taken as seriously as it is. If you just look at raw numbers, the researchers who believe the IQ difference is purely environmental probably outnumber those who take the hereditarian view by at least three to one. But on the other hand, the hereditarian theory's supporters include some of the most highly-respected psychologists of the past 40 years, such as Jensen, Gottfredson, Eysenck and Cattell. Omitting this fact seems like it would be omitting an important piece of information. --Captain Occam (talk) 00:04, 13 October 2009 (UTC)
This is what I'm talking about. Say hereditarian is position B. You list off these people because they're "highly-respected psychologists", not because the view itself is widespread, it's an appeal to the credentials of supporters, an appeal that doesn't have much to do with how widespread the view itself actually is. You line up 5 experts with great credentials, just because they have great credentials (nothing whatsoever to do with the merits and acceptance of the view itself) and you run the risk of giving that view (the view itself) greater weight than it actually deserves. An appeal to credentials is ultimately meaningless because "OMG, the guy's a nobel laureate, it must be true!" when the view may be discounted completely by the scientific community at large. Trust me. I edit a lot of parapsychology related articles. You have actual nobel laureates who support parapsychology, but the view of parapsychology is almost completely dismissed by the scientific community at large. It's undue weight to the view itself, given by an appearance of acceptance based only on the PhD following the supporter's name. --Nealparr (talk to me) 00:31, 13 October 2009 (UTC)
If you agree that it's important to attribute the research in this area to the people responsible for it, this may be a moot point, but I'm not sure the analogy you're using is a good one. Do the supporters of parapsychology who have impressive credentials actually have those credentials in the relevant fields? The four psychologists whom I mentioned are all experts in psychometrics, which is very relevant to the topic of race and intelligence. But it makes much less of a difference if parapsychology is supported by someone who's received a Nobel Prize in economics. --Captain Occam (talk) 00:44, 13 October 2009 (UTC)
Yes on the parapsychology question (physicists, psychologists). An appeal to their credentials would make it look like parapsychology enjoys more acceptance than it actually does. But let's step back to my earlier point. List off the four psychologists with impressive credentials who support that view, what about the ones with impressive credentials who don't? Say there's hundreds of them? We're back at selection bias. An appeal to credentials of supporters, rather than the merits of the view itself, or having to pick and choose who to list in a selection bias, either way, it has problems. Fortunately (again) summary style avoids those problems. --Nealparr (talk to me) 01:00, 13 October 2009 (UTC)
Well, an argumentum ad populum isn’t exactly a stronger argument for any viewpoint than an appeal to authority is.
I still think there might be a way to include the names of some of the hereditarian view’s supporters, while still avoiding the problem of making it sounds like it has more support than it does. We could say something like “While the genetic hypothesis is considered a minority view among psychologists, it enjoys the support of several prominent researchers in this area, such Arthur Jensen, Linda Gottfredson, Hans Eysenck, and Raymond Cattell.”
However, I also agree with what T34CH said, that it probably isn’t necessary to resolve this right now. --Captain Occam (talk) 02:02, 13 October 2009 (UTC)
For better or worse, Wikipedia does, in fact, appeal to popularity in the WP:UNDUE clause of NPOV. It doesn't claim truth through popularity, but it does rank ideas according to their popularity (Wikipedia isn't interested in "truth"). It's just the nature of the Wikipedia environment. Your sentence I would object to, by the way, but I'll wait to see if it ends up in the article, per T34CH, because until you put it in it's moot. --Nealparr (talk to me) 03:04, 13 October 2009 (UTC)
Keeping in mind general balance, I think the work of most, if not all, of those researchers have some place in the article. Aprock (talk) 00:17, 13 October 2009 (UTC)
I agree, but it's important to actually mention them by name. Most readers can't be expected to look carefully at the authors of the material being cited, so if it's significant that a particular piece of research was performed by Jensen or Cattell, the article should specifically mention who performed it. --Captain Occam (talk) 00:22, 13 October 2009 (UTC)
Naming them by name really depends on the context. I expect naming prominent researchers like Rushton and Jensen will occur naturally. For the others, it'll really depend on the content. Specifically, I don't think there's much need for a laundry list of position results. Aprock (talk) 00:26, 13 October 2009 (UTC)
If material is notable enough to include, it is appropriate to attribute this work to the researcher. That sort of summary style is appropriate, as opposed to just listing supporters off. --Nealparr (talk to me) 00:31, 13 October 2009 (UTC)
I'm not sure this discussion is going to end productively. It boils down to different interpretations of NPOV, but at the same time the end result everyone is mentioning sounds remarkably similar. In any case, this is a very small issue when considering the whole of the article and a bridge we should cross when the time is more appropriate. I'd prefer if we let smaller details be mentioned and recorded, and perhaps discussed on the side, but our efforts should be concentrated on working together on big picture issues. It is noted that at least two editors want to retain mention of the prominent supporters of the "strong" hereditarian hypothesis, and that at least three other editors support the mention of such researchers in the context of their significant contributions to the field. T34CH (talk) 01:07, 13 October 2009 (UTC)
That's fair. For the record, I'm only opposed to an actual list-list. Including supporters in-line along with their work (if notable), I am not opposed to. But, yes, definitely, it's only an issue if it becomes an issue so we can move on. --Nealparr (talk to me) 01:27, 13 October 2009 (UTC)
With regard to that section of Aryaman's statement(23:35, 12 October 2009)that follows, Stating the majority view as the majority view and providing one or two experts explaining the majority view is enough. But the qualified "deviant" experts need more detailed coverage. If that conflicts with WP Policy, so be it. That's common sense and good editorial practice as opposed to writing academic propaganda. I disagree. An over-representation (dominance) of "qualified "deviant" experts need more detailed coverage" is what got this article into an unreadable state in the first place along with under-representation of majority views. The naming of the article, Race and intelligence presumes differences that the majority disavows. This matter should be settled in accordance with and not in conflict with WP Policy. It is academic propaganda to give greater space and more detailed explanation to the equivalent of Flat Earth theories.Skywriter (talk) 10:55, 16 October 2009 (UTC)

Split article into two parts?

I am in favor of this new outline, but wanted to mention a (vaguely) related idea. Would it be useful to split this article into two parts. The first would be something like "Human groups and intelligence." This would, one would hope, be much less contentious. It would allow us to thresh out many of the underlying difficulties in a somewhat less charged atmosphere. We would still have to deal with how to define intelligence, how to determine "groups" and so on, but with, I would hope, less drama. Then, subsets of this article could cover topics like race and intelligence, sex and intelligence, other-groupings and intelligence and so on. David.Kane (talk) 04:19, 13 October 2009 (UTC)

That's interesting... though it might be hard to manage without too much syn. I do like the idea though. Are there lit reviews on the concept that would help guide us? Certainly there must be in the psychometrics literature... as dry as it is and lacking in pop-notability, I'm sure the general concept of group differences in intelligence over-shadows (and even encompasses) the arguments extant here in terms of practical importance within testing theory and practical application of the results of studies. This (reframing the article as Observed between-group IQ differences) also deals with something that's really been bothering me for quite a while: the present article isn't about any sort of thing. It's simply functioning as a place-holder for the thing Controversy over the link between ethnicity and intelligence. T34CH (talk) 04:55, 13 October 2009 (UTC)
Why not simply frame this article as a spin-off of the Intelligence article? In other words, limit its content to that which actually discusses the connection between race and intelligence - naturally, with the all-important caveat that this connection is contested. There are already articles on Health and intelligence, Environment and intelligence, Height and intelligence, Sex and intelligence, etc. The Intelligence article should summarize all of these as potential factors influencing intelligence. Maybe we could defuse the nature vs. nurture argument by letting Environment and intelligence and Race and intelligence present these views individually, and have Intelligence carry the burden of refereeing between them? --Aryaman (talk) 05:29, 13 October 2009 (UTC)
I like the idea of having an article that deals with the variation in IQ in a general sense. Though the issue of group differences has been the most thorny issue, there is significant variation in IQ scores between individuals within a population, since all populations have a normal distribution of IQ scores. The controversial book The Bell Curve, is most famous for its discussion on race. However, only two chapters discuss race, the rest of the book is devoted to the general variation in test scores. Wapondaponda (talk) 05:44, 13 October 2009 (UTC)
As far as I can tell, this is best covered here. I don't see why that section cannot be properly expanded and become a better summary of several of the spin-off articles I mentioned. --Aryaman (talk) 05:51, 13 October 2009 (UTC)

RFC: The usage of a blog source in the article

Resolved

Pending that the dispute was not resolved after the post at WP:RSN, I would like everyone to get going here on an RFC. What I would like to see first here is one concise, neutral sentence explaining the dispute going on in which all parties can agree on. That sentence can be written below here. Once we get agreement on paper as to what the dispute is, we can get the RFC going. MuZemike 18:59, 12 October 2009 (UTC)

I think we can agree that the blog is no longer being used as a source. That issue has been resolved. T34CH (talk) 19:05, 12 October 2009 (UTC)
I agree. I think it's been resolved. --Nealparr (talk to me) 19:22, 12 October 2009 (UTC)
OK. I don't think I saw the latest edits. However, I am still concerned about the ANI post. MuZemike 19:32, 12 October 2009 (UTC)
Ah, well that's a different matter, and one which has only just now come to anyone's attention here. Might want to suggest to Occam that they let involved parties know about such postings. T34CH (talk) 19:43, 12 October 2009 (UTC)
Anyone interested in discussing the trolls here and the sudden resurgence-of-interest-in and reincarnations-of-certain race related articles should go to Wikipedia:Ani#Trolling and POV-pushing in race-related topics. T34CH (talk) 19:46, 12 October 2009 (UTC)
Good grief, it was resolved at WP:RSN - as was the use of pisspoor journals such as Medical Hypotheses. Verbal chat 20:15, 12 October 2009 (UTC)
This isn't the same issue that was discussed at RSN. What I'm talking about there is Alun's personal attacks on talk pages (including talk pages for other articles), and the removal of properly sourced information (cited from peer-reviewed journals, not the NYT or Medical Hypothesis) by him and Muntuwandi without any explanation. Bringing up this issue at ANI was suggested to me by the admin user:Dbachmann. --Captain Occam (talk) 22:17, 12 October 2009 (UTC)
Occam, please try not to spread discussions all over the place, but also keep them in the appropriate place. Discussing content disputes is not appropriate at RSN or ANI, and should be confined to this talk page. Discussing behavior is not appropriate here, and specific issues should be confined to specific forums (ANI, 3rr, etc.). Complex issues may be combined, but try to work out disagreements on user talk pages first. Sorry to post this here as it should actually be on your talk page, but I want to stop this thread now as the issue is resolved. T34CH (talk) 23:11, 12 October 2009 (UTC)
And yet you call me a troll above. Alun (talk) 06:53, 13 October 2009 (UTC)

CHRM2 is mentioned as a possible candidate gene for intelligence. A recent study

  • Lind (2009). "No Association Between Cholinergic Muscarinic Receptor 2(CHRM2) Genetic Variation and Cognitive Abilities in Three Independent Samples" (PDF). Behavioral Genetics. doi:10.1007/s10519-009-9274-z.

states:

In summary, although CHRM2 does appear to be related to dementia in animal models, and initial genetic work suggested it was associated with human intelligence our findings in three independent cohorts differing in age from early adulthood thru middle age to early-old age must cast doubt on the involvement of CHRM2 in intelligence, at least in the normal population.

I therefore suggest that there is probably no need to mention candidate genes whose association with cognition is only speculative. Wapondaponda (talk) 13:31, 16 October 2009 (UTC)

heritability of intelligence within races (2nd break)

Ok I'm really uncomfortable with the current wording of this sentence. I tried to change it, but obviously many here thought my change went too far, I'm sure we can come up with something better. Here's what it currently says (again).

The consensus among intelligence researchers is that IQ differences between individuals of the same racial-ethnic group reflect real, functionally and socially significant, and substantially heritable differences in intelligence.

Here are the problems as I see them.

  • Confusion between the words hertability and heritable, these can mean substantively different things.
able to be inherited, passed from parents to their children
(uncountable) The condition of being inheritable
(countable, genetics) The ratio of the genetic variance of a population to its phenotypic variance; i.e. the proportion of variability that is genetic in origin

So what are we trying to say? I think we are trying to say that intelligence is highly heritable, ie passed from parents to offspring (and that doesn't necessarily have to mean genes obviously), and that is currently what the statement says. But then it goes on to discuss within group variation, which is heritability right? So is the statement about how heritable the trait it, or is the sentence about how much genes affect within group variation (heritability)? I think we need clarification, if we're saying that intelligence is highly heritable (i.e. transmissable from parent to offspring, a property of the trait, in this case intelligence), then we don't need to mention within "race"/ethnic group estimates of genetic contribution to variance (heritability) at all, although we might want to mention, as Neisser et al do that Vocabulary size, for example, is very substantially heritable... although every word in an individual's vocabulary is learned., to emphasize that heritable traits are not always genetic in origin.

On the other hand if we are trying to say that the contribution of genes to variation within group is always very high (a property of a population that can vary between populations), then we need to use the word heritability and nor heritable. I also think we need to make sure that our sources really do say that this is true generally for "racial"/ethnic groups if we are going to use them to support this claim (e.g. a blanket claim of high heritability in a source may apply to the global human population, but does not necessarily apply to sub-populations). I also think that, because heritability is quantifiable, we need to give numbers, and if they are different between groups we need to clearly say that. On that note I went to have a look for the source I was talking about above, there are several:

  • The general pattern of results showed consistently higher measures of heritability for White than for Black families and for advantaged as against disadvantaged groups of both races. Indeed, the observed genetic effects in the lower socioeconomic groups were so small as to prompt Scarr-Salapatek to conclude that "genetic factors cannot be seen as strong determinants of the aptitude scores in the disadvantaged groups of either race"(p.1292) From "Race and IQ Expanded Edition" (1975, 1999) Ashley Montagu (ed.) in Nature with Nurture: A Reinterpretation of the Evidence by Urie Bronfenbrenner p. 173.
  • Briefly, their argument is that heritability is not a measure of the magnitude of the genetic contribution to a given trait (as Jensen leads us to believe) but a measure o the extent to which the variability of a trait is due to genetic factors relative to environmental factors. For example, in a uniform environment the heritability estimate will be high. Correspondingly, in a population which is uniform genetically (i.e. homozygous for the trait in question), the environmental estimate will be high. (this is important because if heritability estimates are taken only fro, say the "White" population, then it cannot be claimed int he article that they apply to all human sub-populations because of the possible environmental differences). From "Race and IQ Expanded Edition" (1975, 1999) Ashley Montagu (ed.) in On the causes of IQ differences between groups and implications for social policy by Peggy R. Sanday p. pp 279-280
  • The rational for rejecting the hypothesis of genetic equality between the samples is based on the ubiquitous heritability estimate of 0.80 for the white population. as above p.281
  • A second interesting result that can be noted in these data is that most of the within-family variances (e2) are larger for black twins than for white twins, and most of these values for the black twins are larger than the 0.20 figure suggested by Jensen for the total amount of environmental variance. as above put p. 286. (this is particularly interesting, the within family environmental influence on intelligence between black twins is often greater than the 0.2 it is for the whole of the white population as measured by Jensen. That implies that the family environments of black twins are more heterogeneous than the environments of the whole "white" population that Jensen measured).

I think there are three possible things that can be done.

  1. Only talk about how heritable the trait is, but to avoid confusion I'd use the word inheritable.
  2. Only talk about heritability, but be sure we're more precise about the differences in estimates between groups.
  3. Split the sentence into two, with one discussing inheritance of the trait, and the other discussing the heritability of the trait within different groups.

How does that sound? Cheers, Alun (talk) 06:21, 14 October 2009 (UTC)

First, let me say that I am certainly no expert in this field, so I'm just going off of common sense and what I've read. As far as I can tell, "heritability" in this context is being used in a very specific way, i.e. the proportion of inter-individual phenotypic variation within a population that can be attributed to genotypic differences among the individuals. Thus, heritability is specifically referring to the genetic component, and not to a non-genetic environmental influence. I think these sources really do intend to apply their claim across racial boundaries, i.e. that what holds true in the case of a White population should hold true in the case of Black or Asian populations. Whether or not they are justified in doing this, however, is clearly open to criticism. Now, in light of the results in the studies you mention, this statement might need to be qualified by stating that some research has pointed out the weakness of a trans-group claim. However - and this is just me talking - I'm really wondering if there isn't some confusion as to environmental vs. genetic homogeneity/heterogeneity here. The studies developed for and performed on White groups seem to assume a relatively high level of genetic homogeneity, and subtract a high mean accordingly. Fine so far. But - and this question has been raised, if I remember correctly, in qualified sources - it seems that, in light of the high degree of heterogeneity in the "Black" population (i.e. the high degree of admixture resulting in higher degrees of expressed variation, even within children of the same parents, due to the unique heritage of African Americans), we should only use these tests with great caution outside the "White" population. In other words, the results which seem to indicate lower levels of heritability in "Blacks" might be due to factors which the tests were not designed to account for. But, I digress. To the point: I think the statement as given is correct, though I also think this difference in heritability should be made mention of. But, I'd like to wait a few more hours and see what Captain Occam has to say about this, as he's done far more research and has a far better grasp of this issue than I do. Thanks, --Aryaman (talk) 11:25, 14 October 2009 (UTC)
I think being more clear about what is being said is important here. I find the sentence very misleading. Specifically the use of the fragment "of the same racial-ethnic group" isn't clear to me. I fully support any editing which makes the lead to that section clearer and more straightforward. Aprock (talk) 15:30, 14 October 2009 (UTC)
Apparently there are differences in how scholars use the term "heritability". Rushton gives the most easily understandable definition I have run across (though other have criticized this definition for being too simplistic), which is this: Heritability tells us to what degree the variations in the expression of a trait are determined by genes, as well as to what degree they are - presumably - determined by the environment. A heritability of 1.00 means that the variations are determined purely by genes, while a heritability of 0.00 means that the genes do not determine the variations, the assumption being that they are determined by the environment. What this means in the context of the statement under discussion is that most intelligence researchers have found that variations in IQ between individuals of the same racial-ethnic group are closer to 1.00 than they are to 0.00, i.e. something greater than 0.51 (my assumption, not a real number). As far as we're using something close to Rushton's definition, the statement seems clear enough. The concerns Alun raised (assuming I've understood them properly) is whether this conclusion is based upon studies which have tested outside a white population. I personally do not know the answer to that question. However, I also think that we're not supposed to second-guess the accuracy of that statement provided that it's properly sourced, and I believe it is (based on the limited research I've done). If, however, there can be found credible and reliable sources which explicitly make that criticism, i.e. that the conclusion reached by the above mentioned experts is to be understood as misleading as it is based upon studies conducted on white-only test subjects, then we should also mention that. Thus, rather than change the statement as it now reads, I think we should focus on sourcing possible criticism of it. Though, like I said, I'm really not the one to be asking about the details here. --Aryaman (talk) 18:36, 14 October 2009 (UTC)
attempted explanation
Aryman, there seems to be some confusion here. Heritability is an estimate of the genetic contribution to variation within any group. It is a proportion. That part of the variation that is not estimated to be of genetic origin is of environmental origin. That's clear and undisputed. It therefore follows that if one environment is heterogeneous, heritability estimates will be low, whereas is one environment is homonegeous, heritability estimates will be high. In fact this is the whole rationale for using identical twin studies, because with identical twins the amount of variation due to genes is zero, so all variation is environmental. I have tried to explain how a heterogeneous environment affects genetic contribution to variation in the accompanying image file that I made. I'm not sure how well it uploaded here though.
Also I don't understand why you say "I think the statement as given is correct". The question is not about the correctness of the statement, it is about the precision of the statement. Currently it is unclear what the statement is trying to say. I have covered the reasons why I think it is unclear. But you didn't really respond to that part of my post. Furthermore, whether you personally think the statement is correct is neither here nor there, what's important is whether the statement accurately reflects what reliable sources say, not whether editors believe it to be true.
"stating that some research has pointed out the weakness of a trans-group claim." I'm not entirely sure what you mean by this. It's not about any trans-group claim, it's about whether the sources we are citing explicitly say that heritability is substantial across "racial"/ethnic groups. As I said, a blanket statement that heritability is high is not the same as claiming that it is high in all human sub-populations.
These are not easy ideas to get to grips with, and I may not be explaining myself very well. But I am clear in my mind about what I mean. So I repeat, we need to
  • Remove the confusion between inheritance and heritability, by not using the word heritable.
  • Have a comment about the fact that intelligence is highly inheritable. (i.e. that clever people have clever children)
  • Have a more precise couple of sentenses about heritability wrt within group magnitude.
Cheers Alun (talk) 18:18, 14 October 2009 (UTC)
At a fundamental level, I think the complex jargon and overloaded meaning is part of the problem. The lead paragraph of the section should strive to explain things as clearly as possible. Further details and subtleties can be covered in the body of the section. As it stands now, I agree with User:Wobble that it need to be updated to clearly reflect the underlying sources. Aprock (talk) 19:39, 14 October 2009 (UTC)
When I said "correct", I meant that I think it correctly reflects what is being reported in the sources, not that I think its contention is correct. ;) I don't see removing the word "heritable" as an option given its importance in the literature. But, like I said, I'd really prefer to wait for Occam to get back his editing powers before making a change, as I'm sure he'd have an informed and intelligent opinion on this. I was just reading Sternberg & Grigorenko (2002:314-315), and - though this is based on research (Wadsworth, 2000; Olson, 1999; Wechsler, 1974) limited to reading disabilities and word-recognition - it seems that heritability is intelligence-dependent. Specifically, that when the base IQ>100, then the studies produce a heritability estimate of 0.72, but when the base IQ<100, the heritability was estimated at 0.45. That makes good sense to me, but I don't see how it squares with what Alun was posting. Thus, I really don't know enough to say one way or the other, lol. (Unless this means that the low heritability in non-white groups has to do with a low base IQ?) Occam? --Aryaman (talk) 20:15, 14 October 2009 (UTC)
All right, here I am again. VA, I largely agree with you that we need to make our sentence consistent with what the source material says, and not try to second-guess it unless some of the sources explicitly dispute this conclusion. That also means that if the sources use the word “heritable”, we should use it also. However, there’s also something else I’d like to point out:
Although the distinction between within-group heritability and between-group heritability is definitely worth making, I’ve never seen any source draw the distinction between within-group heritability and within-group parent-to-child inheritability. I know they have slightly different definitions, but trying to distinguish them really seems to be splitting hairs. And the reason why I don’t think this distinction matters is because if the parent-to-child inheritability of a trait is 80% (or any other percentage) for all members of a single population, then as long as that number is approximately the same for all members of the population, the amount of variation of the trait in that population which is due to genetic factors will inevitably be 80% also. Unless we’re specifically discussing something where this extremely subtle distinction matters, I don’t think it’s one that needs to be made.
I know that Wikipedia articles are supposed to be as precise as possible, but a line needs to be drawn somewhere, and that line probably ought to be at trying to be more precise than the source material is. In addition to it simply being unnecessary, it runs the risk of us misrepresenting what the source material was intending to say. --Captain Occam (talk) 00:25, 15 October 2009 (UTC)
  • "if the parent-to-child inheritability of a trait is 80%"
What does that mean? What is being measured here? Heritability is a population based measure. You seem to be giving a measure for a trait. How would one calculate this figure? Heritability and inheritability are totally different. Clearly they do not have "slightly different meanings". One is the transmission of something from parent to child, it is a property of the trait, the other is a measure of the genetic contribution to the variation of a trait within a population, it is a property of the population. All reliable sources say this clearly, it is this confusion between inheritance and heritability that is the root of this subject. As my quote above from Peggy Sanday states clearly "Briefly, their argument is that heritability is not a measure of the magnitude of the genetic contribution to a given trait (as Jensen leads us to believe) but a measure o the extent to which the variability of a trait is due to genetic factors relative to environmental factors." You are trying to say that heritability is a measure of the contribution of genes to intelligence within a population. This is incorrect, it is an estimate of the contribution of genes to the variance within that population, and not a measure of the contribution of genes to the magnitude of the trait, even within a population. You seem to be as confused about this as everyone else, which is why we need to change this sentence. Let me have a think about some alternatives, I'll suggest something and we can discuss the best way to go about making the necessary changes. Cheers Alun (talk) 03:38, 15 October 2009 (UTC)
"What does that mean? What is being measured here?"
What I meant by parent-to-child heritability is the extent to which the IQ of the child is the result of the IQ of his or her parents, and the extent to which it’s the result of the environment in which he or she was raised. This can’t be measured for individuals, at least not as far as I know, but when looking at several parents and children it’s possible to measure it by examining the extent of regression toward the mean, via Galton’s Law.
Let’s take a primarily genetic trait, such as height, as an example. According to Galton’s Law, if one compares the height of fathers who are above or below this height to that of their sons, their sons will be an average of halfway between the height of their fathers and the average male height. The same is also true in reverse, with people of above-average adult height tending to have fathers with a height halfways between their own and the population average. On the other hand, for a trait that has virtually no genetic component, the number will be much less than 50%. For example, if you were to compare hair length between fathers and sons, you would not find any meaningful correlation. For all quantifiable traits in sexually reproducing organisms, the extent of regression towards the mean varies directly with the extent to which the trait is genetic, from 50% for purely genetic traits to 0% for purely environmental traits. Since you’ve said that you’ve studied genetics, I’m assuming you’re already familiar with Galton’s Law.
Incidentally, the parent-to-child heritability for IQ among blacks can be measured via Galton’s Law. According to Jensen’s 1973 study about this, the degree of regression towards the mean in IQ for whites in the United States is 40%, and for blacks it’s 38%. (This was actually based on siblings rather than parents and children, but since siblings share 50% of their genes also, it’s another way of measuring the same thing.) Since the maximum is 50% for purely genetic traits, this means that more than half of the within-population variation for both groups is genetic.
Now, back to the point I was making: you seem to be wanting to second-guess what the source material about this says, and state something that isn’t explicitly supported by it. Unless it’s only a minor change of wording that you’re suggesting, this sounds an awful lot like WP:SYNTH. --Captain Occam (talk) 04:35, 15 October 2009 (UTC)
Captain Occam, everything you just wrote was synth. When you have a reliable source, by all means share it. In the meantime, Alun and Muntuwandi seem to have things well in hand here. Why not leave the work here to people who actually know what they are talking about? Slrubenstein | Talk 05:19, 15 October 2009 (UTC)
This isn’t synthesis. Everything I just wrote is explained in pages 467-472 of Jensen’s book The g Factor.
In any case, why does it matter whether this is synthesis? This information isn’t actually going in the article. Alun was claiming the article shouldn’t say that IQ has high within-group heritability in general, because he thinks the only evidence of this is for whites, and now we’re having a discussion to see whether what he suspects about this is actually the case. There’s no rule against engaging in synthesis in a discussion on a talk page about what the data does or doesn’t say. --Captain Occam (talk) 06:03, 15 October 2009 (UTC)
Not true Occam, why do you constantly misrepresent what I say? What I said was that the sentence in question does not appear to accurately reflect at least one of the sources. I also think the sentence is badly written and confusing. I want to simplify it and make it more precise. Alun (talk) 06:24, 15 October 2009 (UTC)
  • "What I meant by parent-to-child heritability is the extent to which the IQ of the child is the result of the IQ of his or her parents, and the extent to which it’s the result of the environment in which he or she was raised."
Yes, that's what I thought you meant, but you're wrong, heritability estimates don't measure that at all. Heritability estimates measure the contribution of genes to the variation of the trait within a population not the contribution of genes to the trait as passed from parent to child. So when we say it's 80% heritable in "white" populations we don't mean that in those populations 80% of intelligence is inherited from parent's genes. What we are saying is that 80% of the differences in intelligence between individuals are caused by genes. Or to put it another way, if we take two people from this "white" group, and one has an IQ of 110 and the other an IQ of 100, an 80% heritability means that 80% of the ten IQ point difference between the two is due to genes, i.e. 8 IQ points out of the 10 difference is genetic in causation.
"the parent-to-child heritability" again, this makes no sense, heritability is a measure of variation, not the contribution of genes to a trait. Heritability is a population level trait. See also my response above.
Regression towards the mean (what you call Galton's law) simply states that the progeny of outliers in a trait (e.g. the extremely tall, the extremely intelligent) will tend towards the mean. That is irrelevant to heritability estimates. Can you provide a reliable source that states that regression towards the mean can be used to estimate heritability? I don't think that's right and would like to see a source that claim this, you quote Jensen's analysis of regression towards the mean, but this is not a measure of heritability, or of populations, it is a property of outliers. I think we should stick to the point.
I think you have to do better than accuse me of synthesis. Indeed you haven't addressed any of the salient points I am making. I'm not suggesting a synthesis, I am suggesting that the sources don't support what the sentence currently states. Especially I don't think the sources specifically make reference to "the same racial-ethnic group". I've noticed that the word "heritable" is often used as the adjective of heritability, so I don't have a problem with this word now, it is clear that we are talking about heritability estimates and not heredity. But I think that we should be clear that heritability varies from group to group depending upon environment.
I also don't like the round about language of the sentence, we should say what we mean in a straightforward manner.
  • "The consensus"- who says it's a consensus?
  • "IQ differences between individuals"- i.e. variation
  • "of the same racial-ethnic group"- at least one of the sources doesn't say this, "Intelligence knowns and unknowns" doesn't say this, it says "It is clear (Section III) that genes make a substantial contribution to individual differences in intelligence test scores, at least in the white population." which is not the same at all.[3] BTW the article also states, as I do above, that "So defined, heritability (h2) can and does vary from one population to another. In the case of IQ, h2 is markedly lower for children (about .45) than for adults (about .75)" and "The value of h2 can change if the distribution of environments (or genes) in the population is substantially altered."
  • "reflect real, functionally and socially significant"- I don't know what this part is supposed to be saying. I don't see why it can't just say "reflect".
  • "and substantially heritable differences in intelligence"- this is just a convoluted way of saying that the differences mentioned earlier in the sentence have a substantial genetic component, right? But it's written in a confusing manner.
I'd suggest this:

It has been estimated that genes contribute substantially (40-80%) to IQ differences between individuals within some human groups.

I think it says pretty much the same thing, but in a simpler and easier to understand way. I also think it more accurately reflects what at least one source says.
Cheers Alun (talk) 05:58, 15 October 2009 (UTC)
“Regression towards the mean (what you call Galton's law) simply states that the progeny of outliers in a trait (e.g. the extremely tall, the extremely intelligent) will tend towards the mean. That is irrelevant to heritability estimates. Can you provide a reliable source that states that regression towards the mean can be used to estimate heritability? I don't think that's right and would like to see a source that claim this, you quote Jensen's analysis of regression towards the mean, but this is not a measure of heritability, or of populations, it is a property of outliers. I think we should stick to the point.”
Galton’s Law involves more than just regression toward the mean, although regression toward the mean is one of the principles that it involves. Its full name is “Galton's law of ancestral heredity”. I’m kind of puzzled that you aren’t familiar with this principle, claiming to have studied genetics, but if you Google for it you’ll be able to find more information about it. It’s a principle that’s quite specific to heredity.
I guess I’ll quote Jensen about this, although I can explain in more detail about how this works. This is on page 468 of The g Factor:
In quantitative terms, Galton’s law predicts that the more that variation in a trait is determined by genetic factors, the closer the degree of regression (from one parent to one child), on average, approximates one-half. This is because an offspring receives exactly one-half of its genes from each parent, and therefore the parent-offspring genetic correlation equals .50.
Jensen isn’t making this up; he’s just summarizing a basic principle of genetics. If a trait is 100% genetic and a child has 50% of his parent’s genes for it, the child will tend to regress to halfway between their parent and the mean. If the trait has no genetic component, then the child will tend to be at the mean of whatever the trait is, regardless of what the trait was like in their parent. For a partially genetic trait, it will be somewhere between 0% and 50%. The way in which the degree of regression towards the mean corresponds to the degree of heritability of a trait was established by Ronald Fisher in his 1918 paper, "The Correlation between Relatives on the Supposition of Mendelian Inheritance".
Based on Fisher’s analysis of this, we can determine what portion of black Americans’ IQ is inherited from their parents. Since the degree of parent-to child IQ correlation for them is 38%, this means the degree to which their IQ is based on that of their parents is approximately twice that, or 76%. (Remember, 50% is what it would be if their IQ were 100% genetic.) So I think this shows that the within-group heritability of IQ among blacks is fairly high also.
Now, if any of the sources currently being used specifically reject this idea, then the sentence in question in the article should be changed to reflect what its sources say. However, the only one of them that draws any kind of distinction in this respect merely says “It is clear (Section III) that genes make a substantial contribution to individual differences in intelligence test scores, at least in the white population.” In other words, this particular author is uncertain as to whether this also applies to more than whites. And it seems like most of the other sources being cited just state that the within-group heritability of IQ is high in general, without even adding that qualitication.
I know you’ve been able to find sources disputing the heritability of IQ among blacks, but how much of an “academic consensus” is this? I could find sources disputing the same thing for whites, if I wanted to, as well as sources claiming that IQ doesn’t mean anything significant at all. (Gould’s Mismeasure of Man comes to mind.) That doesn’t mean it's any kind of consensus. If you want to change this sentence, what you’ll need to demonstrate is not only that this sentence of the article as it stands is unsupported by the current source material, but also that the academic consensus goes against what it says. --Captain Occam (talk) 07:33, 15 October 2009 (UTC)
  • "we can determine what portion of black Americans’ IQ is inherited from their parents. Since the degree of parent-to child IQ correlation for them is 38%, this means the degree to which their IQ is based on that of their parents is approximately twice that, or 76%."
Even if one accepts this claim (and it looks like OR to me, if it were that simple to estimate the contribution of genes to a trait, then we wouldn't be having this discussion at all, it would have been resolved long ago), you are not attempting to measure heritability, you are attempting to measure the contribution of genes to a trait. That is not heritability, heritability is an estimate of the contribution of genes to the difference (variation) in a trait within a population. You don't seem to understand this difference, it is important and I've repeated it four or five times now, without an understanding that heritability is a measure of differences (variation) and not causation (i.e. contribution of genes to a trait), then it's impossible to have a sensible conversation about this. Alun (talk) 08:24, 15 October 2009 (UTC)
I want to stick to the point. I want to change this sentence, I've made a proposal, do you think you could comment on my proposed change? I'm not interested in discussing Galton, or regression towards the mean. I'm interested in the fact that this statement, as it stands, does not accurately reflect what the sources say. This is about heritability and not Galton. You appear to be trying to move the conversation towards a discussion of Galton and regression towards the mean. We are discussing a specific sentence in the article that deals with heritability. You are not talking about heritability, you are talking about transmission of a trait from parent to offspring, that is not heritability. Heritability is a measure of genetic contribution to variation within a population, it is a population level measure, what you are saying is irrelevant to heritability. The sentence is about heritability and makes claims about "race/ethic groups" that are not made in at least one of the sources. As I say, this sentence has several other structural problems. I have outlined these above, maybe you could respond to my comments on those? I think that would be more constructive. If you don't have specific comments about my suggestion, then I'll assume you're happy with it.
I also notice that you don't comment about my observation about heritability being an estimate of genetic contribution to variation, and not about genetic contribution to an inherited trait. I assume that this means that you accept what I say.
Cheers, Alun (talk) 08:08, 15 October 2009 (UTC)
I already said earlier, the genetic contribution to a trait in any population is a direct result of the degree to which it’s transferred from parent to child. The percentage for one is the same as the percentage of the other, so most sources don’t bother to distinguish between the two. For that reason, I don’t think it’s appropriate for our article to make this distinction.
You’ve presented one source that states IQ is highly heritable “at least for whites”, while not saying either way about other groups, but it seems that all of the other sources currently being used state that the within-group heritability of IQ is high in general, regardless of what race is being discussed. If you think that one source doesn’t support this sentence, then we can just stop citing it, and have this sentence cited to the remaining eight sources. Other than possibly needing this one source removed, I don’t believe you’ve demonstrated that there’s anything wrong with this sentence. You’ve complained about it’s conflation of inheritance and heritability, but I’ve explained why I don’t think this distinction matters, and you’ve also claimed that there’s no evidence of high within-group heritability of IQ in blacks, but I’ve presented one example of that based on Galton's law of ancestral heredity. So in my opinion the sentence is clear, it’s factually accurate, and it’s supported by eight of its nine sources, so in my opinion it can be left the way it is.
As I side note, you ought to follow the advice from the admin noticeboard about being less antagonistic. The wording of your comment makes it pretty clear you aren’t happy with shown to be wrong about regression towards the mean having nothing to do with how heritability can be estimated, even though this was only marginally related to what’s actually going in the article. You shouldn’t take things this personally, especially when they matter this little. I notice that for two days, your userpage said that you were retiring from Wikipedia this month, until you removed that notice today; doing so might be worth considering if this sort of thing tends to make you upset. --Captain Occam (talk) 10:15, 15 October 2009 (UTC)
  • "the genetic contribution to a trait in any population is a direct result of the degree to which it’s transferred from parent to child"- no geneticist would agree with that comment, I know of no geneticist who would claim that there is no such thing as a gene x environment interaction, besides it is irrelevant to heritability, which is a measure of variation and not genetic contribution.
  • "all of the other sources currently being used state that the within-group heritability of IQ is high in general"- but do they? But it doesn't matter if they say that heritability is high in all groups, I want them to specifically state "racial"/ethnic groups, because that's what our sentence says. Unless they specifically say "racial" groups, then we need to change our sentence.
  • "I don’t believe you’ve demonstrated that there’s anything wrong with this sentence."- Oh well, we'll see what the consensus is. I think this is a very poor sentence that is confusing, poorly written and very imprecise. You actually haven't actually made any arguments as to why this sentence is any good. Maybe you should spend some time saying why tyu think the sentence is any good, that would be more constructive than your previous deviation.
Cheers Alun (talk) 11:38, 15 October 2009 (UTC)

(outdent) If I may, I'd like to point something out. I pointed it out long ago, but either it was overlooked or simply ignored: There is not 100% agreement on what heritability means. Jensen and everyone who follows Jensen's line of reasoning thinks it means one thing. Jensen's critics think it means something else. Specifically, the issue is about "heritability being an estimate of genetic contribution to variation, and not about genetic contribution to an inherited trait", which Alun keeps harping on like it's some incontrovertible fact. It's not. It's - politely put - one interpretation of heritability. (Put another way, it's an example of "Let's keep changing the method until we get the results the sociopolitical climate dictates we should get" - but that's my commentary.) Case in point: Take a look at Heritability of IQ. You'll notice that the reference given for the definition in the header links to this article. If you read that article, you should notice that, instead of being a general interest article, it's actually a highly contentious, non-neutral opinion piece which has the "correct" definition of heritability as one of its main points of contention. (Whether this article should be used to source the definition of "heritability" on the Heritability of IQ article is another matter.) In regards to the matter at hand, it boils down to this: Do the references cited as being in support of the statement in question understand heritability as Jensen does, or rather as Lewontin does? That alone would tell us what these sources are really trying to say, and how far we can use them to support such a statement. Is anyone following me here? --Aryaman (talk) 10:04, 15 October 2009 (UTC)

I think you are wrong, there is no dispute about what heritability measures. Heritability measures the contribution of genes to variation within a group. I don't know any reliable scientist who would deny that. Analysis of variance is a well know statistical tool, but it is analysis of variance and no one, not even Jensen, pretends that they are measuring anything else. You don't provide any sources to support your claim that "It's - politely put - one interpretation of heritability." So what's the other? Where are the sources that say that heritability estimates are not measures of the contribution of genes to within group variation? You don't provide any sources to support your contention. You only provide one source, which actually disagrees with what you say, and then say this source is biased and unreliable without providing any evidence that this is the case. If what you say is correct, then you should be able to find a reliable source that says that heritability is not a measure of variation. Then you could use that source to change the Heritability of IQ article. I also suggest you go and change the Heritability article, which currently states Heritability is the proportion of phenotypic variation in a population that is attributable to genetic variation among individuals. Variation among individuals may be due to genetic and/or environmental factors. Heritability analyses estimate the relative contributions of differences in genetic and non-genetic factors to the total phenotypic variance in a population You should definately change that definition if what I am saying is only one point of view. In fact I suggest you take it to the genetics wikiproject and discuss it on the talk page there, clearly you believe that the article about heritability is unbalanced. Alun (talk) 11:38, 15 October 2009 (UTC)
Are you saying that there is, to your knowledge, no disagreement or dispute in the academic world regarding Jensen's understanding of the meaning of heritability? This was and continues to be one of the major areas of discussion of Jensen's work, no? --Aryaman (talk) 12:16, 15 October 2009 (UTC)
Whether there is a debate about how Jensen applies the concept of heritability in his research is one thing. But the meaning of heritability is very straightforward. There is no wriggle-room. Just like other concepts (genetic drift, for example) the word heritability has one definition that all geneticists, all evolutionary biologists, all physical anthropologists, agree on, there is just nothing controverial about this. Slrubenstein | Talk 13:09, 15 October 2009 (UTC)
  • "Are you saying that there is, to your knowledge, no disagreement or dispute in the academic world regarding Jensen's understanding of the meaning of heritability?"
Yes, that is exactly what I'm saying. Jensen has never claimed to be measuring anything other than genetic contribution to variation within a group. Where Jensen has been criticised is his use of a heritability estimate calculated for a "white" American test group of 80% and claiming that this high estimate applies to "black" Americans. The fact is he couldn't measure the heritability of African Americans because there just hadn't been enough twin studies on African Americans, so he used the statistic from the "white" group, and got rightly and soundly criticised for being less than scientifically rigorous. His claim that this "high" heritability statistic proves that some groups are "ineducable" was also correctly roundly criticised because it ignored the fact that heritability does not mean that environment cannot affect the average group score when the environment changes (e.g. if education becomes better). But of course those claims of his have been debunked time and again and I don't want to go into all that now. If you want to read about criticism of Jensen's interpretation of heritability results, then there is plenty of material out there. I can recommend Ashley Montagu's "Race and IQ", it's really cheap to get a second hand copy from amazon. Cheers. Alun (talk) 17:32, 15 October 2009 (UTC)
Yes, I've read some of that critical material. I've also read Flynn's treatment of it, in which he guardedly defends Jensen's approach over and above the dozens of critics (particularly pointing out the critical flaw in Lewontin's corn analogy, which Jensen's critics apparently love to recite without, according to Flynn, really understanding it as far as its true implications go). Flynn comes to a different conclusion than Jensen, for his own reasons which are worth considering (particularly in regard to "factor X"). But, as you're quite well-read, I feel safe in the assumption that you know all this already. You must also know that Jensen tried repeatedly to make the environment responsible, but eventually considered it impossible in light of the extreme and very peculiar requirements of "factor X". But, then, are you intentionally ignoring that fact in order to push one particular view? No, I must assume good faith here. Then, do you honestly do believe that Jensen and Lynn are quacks, and that this issue was decided long ago? Then why on earth is there still controversy in the academic world? Or do you think there is no controversy? --Aryaman (talk) 19:03, 15 October 2009 (UTC)
I think this discussion was specifically about the meaning of heritability. About that, I think there is no controversy. Aprock (talk) 19:33, 15 October 2009 (UTC)

Outline restructuring

I've made several cosmetic changes to the article to move it towards conforming with the proposed outline above. Some minor content changes, mostly heading titles, were made. Aprock (talk) 20:24, 14 October 2009 (UTC)

I notice you've changed the title of the "1990s debate" section to "The Bell Curve debates". I'm still of the opinion that the original title was better, both because it was consistent with the "1970s debate" title for the preceding section, and because the debates in the 1990s involved more than just The Bell Curve and its commentaries. I know you don't agree with this, though, so I'd like other editors to express their opinion about it. If most other people here agree that the original title was better, it ought to be changed back. --Captain Occam (talk) 00:32, 15 October 2009 (UTC)
I agree with Occam on the grounds that more was discussed in the 90's than Bell Curve, and unless the material relating specifically to Bell Curve is to be moved to its own sub-section, "1990s debate" is a superior title. At the same time, seeing as the higher-level heading is already "History of debate", I'm wondering if "debate" is superfluous in those sub-headings. Also, I'm questioning the propriety of having "Policy debates" and "Viewpoints" listed under the same heading, i.e. "History of debate". Unless they are to be rewritten in a distinctly historical/chronological way, they probably need to be grouped differently. (Mind you, this is a cursory criticism of the current outline. Give me time to mull it over. :) --Aryaman (talk) 00:45, 15 October 2009 (UTC)
Every paragraph in that section relates directly to the debate generated by the publication of The Bell Curve. Aprock (talk) 01:31, 15 October 2009 (UTC)
Made the same observation. The Bell Curve debates is more descriptive.--Ramdrake (talk) 01:41, 15 October 2009 (UTC)
Aprock, we already discussed this here. The Bell Curve was the catalyst that caused this debate, just like Jensen’s 1969 paper in Harvard Educational Review was the catalyst that caused the 1970s debates. But that doesn’t mean either section of the article needs to be named after these catalysts, and it especially doesn’t mean we should do this for one section but not the other. --Captain Occam (talk) 01:46, 15 October 2009 (UTC)
Let's see what the editors want. If most of them opt for The Bell Curve debates, then the title should stay.--Ramdrake (talk) 01:52, 15 October 2009 (UTC)
You know, guys, it can be both. The 1990s Bell Curve debates (no reason to not call it Bell Curve without the addition of more material, and no reason not to add the decade to identify the era). --Nealparr (talk to me) 10:56, 16 October 2009 (UTC)

What is heritability... (and other OR issues)

... is not a question we should be addressing here. This is an article about the consensus reflected in journal articles, best defined by the wording of the major lit reviews. Specific points brought up by the lit reviews can then be addressed by other sources. There is too much bickering here about personal opinions, when in reality that is total OR.

Let's ask: What are the real controversies? Even Jensen and Rushton say it's not a question of 100-to-0% vs 0-to-100% nature/nature... they define the "strong heriditarian position" as 50-50%.

How are Jensen and Rushton viewed by the general scientific community? and answer honestly. If they're viewed as fringe (or some adjective similar to that... let's not get bogged in semantics again), then that's how they get described here.

Let's ask: are there sources that say race and intelligence are linked? Even the Bell Curve doesn't state that the question has been answered. The controversy is that some people even suggest they could be linked.

Until we start focusing on the big picture problems with this article, we'll go nowhere. T34CH (talk) 18:58, 16 October 2009 (UTC)

Well, as has been pointed out before, describing the hereditarian position (Jensen, Rushton, Lynn, Eysenck, Shockley, etc.) as "fringe" is probably a violation of current policy, regardless of how much it offends others. Is the hereditarian position a "popular" one? Absolutely not. Has the heriditarian position had a significant impact on academia. Yes, undeniably so. Is there a mountain of literature criticizing the hereditarian position. Yes, there is. It is our place to decide who is right in this? No, it's not. It's our job to report the findings, report who holds what views, what those views mean, what criticism has been made, etc. It's not our job to beat our readers over the head with how morally despicable the conclusions of the hereditarian position may or may not be. Does that answer your question regarding the "big picture"? Or, perhaps you should enlighten us as to what the "big picture" is? (Though, even suggesting that the "big picture" could be anything other than discussing this issue in a neutral manner bothers me, personally.) --Aryaman (talk) 19:44, 16 October 2009 (UTC)
I agree wholeheartedly, which is why I focused on restructuring instead of minutiae. The one big step that could be made now would be begin development of an "Academic Consensus" section which summarizes our current understanding, and which includes much of what you discuss above. One reason this article is so poor is that there isn't any kind of overall summary, even in the lede. Aprock (talk) 19:04, 16 October 2009 (UTC)

Academic consensus

Three cheers for NPOV. But I'm wondering if we can't do one better. How about we put a big red banner on the page that reads:

"WARNING: Before we discuss this topic, we feel required to inform you, gentle reader, that it is our sincerest hope that none of the information presented herein may lead you to draw any conclusion other than the one we intend to spoon feed you. In fact, that you've found your way to this article may indicate you have suppressed racist sentiments. Please turn off your computer and seek professional counseling immediately."

That would help in terms of the "big picture", would it not? :) --Aryaman (talk) 05:37, 17 October 2009 (UTC)

Once again, the problem you have is with the way NPOV policy dictates that articles be written. If that is the battle you want to take up, the place to do so is talk:NPOV. T34CH (talk) 06:18, 17 October 2009 (UTC)
Actually, this section was proposed days ago in the new outline section AND the lead section just above it, and you didn't seem to have a problem with it then. I'm sorry that you've changed your mind. Feel free to suggest alternatives rather than complaining in a sarcastic way. T34CH (talk) 06:23, 17 October 2009 (UTC)
T34CH, this isn't NPOV. Let's look at this sentence from the "academic consensus" section:
"Race and ethnicity are socially defined groups--rather than biological observations--whose members are not homogeneous; races and ethnicities are often defined by affiliation with very large geographical areas (Asian) or common language (Hispanic). For these reasons, discussions of correlations between race and intelligence which extrapolate a genetic causation are fundamentally flawed."
The article presents this statement as fact, not opinion. I don't think the "academic consensus" section is necessary to begin with, since it doesn't provide any information that isn't already provided by the rest of the article, but at the very least we shouldn't be trying to make up readers' minds for them about which theory to believe. This is pretty basic.
I'm tempted to just revert all of your new edits to the article, but I'm near my revert limit for today, and I suppose we might as well discuss whether any of them are worth keeping.
Looking at the overall structure of the article in its current state, it has a pretty blatant balance problem. We've got a large section specifically devoted to criticism of the genetic hypothesis, with nothing equivalent for criticism of the environmental view, and the "criticism of hereditarian positions" section is longer than the explanation of the hereditarian perspective itself. What's more, the "Evidence for genetic factors" section lists almost no actual evidence for this; it's mostly just a general explanation of what the hereditarian hypothesis is.
WP:UNDUE states that we should cover each viewpoint in proportion to the degree of coverage it receives in the scientific community, not that the entire article should be structured around convincing readers to accept the most popular view. If your goal is to convince readers that the hereditarian hypothesis is false, the current state of the article actually works against that goal, because anyone who reads it can see pretty easily that it's essentially nothing more than propaganda. --Captain Occam (talk) 07:10, 17 October 2009 (UTC)
I fully support structuring the article in relation to the prominence of views. That's the spirt and letter of NPOV policy. Articles are not structured this way to convince readers of anything. They are structured this way to report on how views actually are received in the academic community. Balance doesn't mean 'giving views equal validity', and that's expressly written into policy. In a neutral article, views compete on their own merits as demonstrated through reliable sources, and in no way should a minority view be inflated more than it deserves to achieve a false sense of equal validity.
That said, claims of consensus require reliable sources Wikipedia:Rs#Academic_consensus. The source needs to be both reliable, and explicitly state that there is a consensus for us to have the title "Academic consensus". The claim of consensus should also be substantiated with details about the consensus. For lack of a better example, check out how intelligent design demonstrates a consensus by citing the academic bodies and court rulings relative to the topic. Our article here, currently, doesn't give us the details that support that a consensus has been reached. --Nealparr (talk to me) 11:53, 17 October 2009 (UTC)
I renamed it to "Overview" (mostly as a place holder that you guys can change to whatever is appropriate). --Nealparr (talk to me) 12:03, 17 October 2009 (UTC)
"I fully support structuring the article in relation to the prominence of views."
So do I, but I think it's kind of obvious that the article goes overboard with this. We've got three sub-sections devoted to criticism of the hereditarian view, and another five devoted to non-hereditarian hypotheses. (Without any criticism presented.) The title "evidence for genetic factors" notwithstanding, how much space does the article actually give to describing the evidence that proponents of the hereditarian view present in support of it? The answer is around two sentences, towards the end of that section. The rest of that section just provides background information about the hereditarian view, including some more criticism that didn't make it into the "criticism" section.
If the article were weighted in proportion to the proportions of these views in academic literature, the evidence for the hereditarian hypothesis would be given around one-half as much space as the evidence for environmental causes. In the article's current state, it's given no more than 5% as much space as the environmental view. This was something of a problem even before T34CH's recent edits, although his edits made it considerably more obvious. In order for the article to be NPOV, this needs to be changed.
And as I said before, trying to make up the reader's mind for them about what to believe (as the sentence I quoted does) is not acceptable under any circumstances. --Captain Occam (talk) 13:13, 17 October 2009 (UTC)
I think I can predict where this discussion is likely to go from here, so I’m going to head it off before it does. Someone is going to claim that the hereditarian hypothesis is “fringe”, and that for this reason it doesn’t deserve any more coverage in the article than something like the opinion that the world is flat. So before anyone claims that, I advise anyone who's thinking of doing so to read the earlier discussion about this here, if they haven't already.
As I said in my last post there (to which no one replied), whether or not this viewpoint is “fringe” is not a matter of semantics. When a peer-reviewed journal is being careful to present this topic in a balanced way, what term is being used for this balance doesn’t even matter; the only thing that matters is that Wikipedia ought to present this topic with a similar type of balance to the one that it receives in the professional literature. --Captain Occam (talk) 13:25, 17 October 2009 (UTC)
I see no problems with T34CH's edits, which have returned the article to a more balanced state, firmly rooted in academic references. It's not quite clear why there has been a recent increase in editing activity for articles related to Rushton and Lynn and their theories. This matter has been discussed several times recently on noticeboards. Mathsci (talk) 13:44, 17 October 2009 (UTC)
Stating that one of the main viewpoints about this is "fundamentally flawed" is not balanced. I don't see how this matter can even be a subject of dispute. It rather explicitly violates WP:NPOV. --Captain Occam (talk) 14:01, 17 October 2009 (UTC)
I'm using the only review articles which we have agreed to use for the purpose of evaluating consensus. They make it very clear that there's no justification at all for causative statements such as, "blacks score lower than whites because of genetic differences." They make the point that such statements are flawed on several points (conceptions of black/white are flawed, environmental issues cause lower scores, tests are biased, genetics are not well enough understood). Thus, I see absolutely no issue with pointing out that the fundamental basis for the statement "blacks score lower than whites because of genetic differences" is flawed reasoning. Fundamentally flawed. We can say it some other way if it really hurts your feelings, but there's no reason to sugar coat reality. T34CH (talk) 15:45, 17 October 2009 (UTC)

(outdent) In the absence of intellectual honesty and integrity, NPOV policy is worthless. NPOV assumes that the editors can take an objective, uninvolved stance regarding their subject matter. This is not a case of pushing a "fringe" view. This is a case of pushing a majority view as the only correct view. Unfortunately, in its effort to neutralize the real "fringe"-pushers, Wikipedia often turns a collective blind eye to this kind of POV-pushing.

If the hereditarian position was as "fringe" as the editors would like to lead readers to believe, one wouldn't find it being discussed in every major review of the issue in such depth; it would simply be ignored. I'd like to quote a passage from a colleague and occasional critic of Jensen who - and please note the intellectual honesty - comments upon this very issue:

The massive, and vituperous, attacks on hereditarian findings clearly signal how seriously the environmental program is challenged by [the] evidence. The hereditarian program is the only one that has generated a theoretical structure that explains, in a consistent (non-ad hoc) fashion, the observed correlations in IQ between relatives (reared apart or together, related by ancestry or not). A review of specific criticism of hereditarian research, with monozygotic twins reared apart (MZA), will show clearly that the critics view these data very seriously. That is, they behave as though, if the data were true, it would be telling. I understand this to mean that 'in their heart of hearts', they recognize that studies of twins reared apart, adoption studies, and classical twin studies are just what their proponents claim they are: 'experiments of nature' in the most fundamental sense of the term 'experiment'. —Thomas J. Bouchard

Another passage, this time from Jensen himself:

There is simply no doubt about it: There is a double standard among journal editors, referees, book review editors, textbook writers, and reviewers of research proposals when it comes to criticizing and evaluating articles that appear to support what the readers may interpret as either 'hereditarian' or 'environmentalist' conclusions. I have had plenty of experience with this, for I have published many articles that range widely on this spectrum. I approve the thorough critical scrutiny to which 'hereditarian' articles are subjected but deplore the fact that many 'environmentalist' articles receive much more lax reviews. There is unquestionably much more editorial bias favoring 'environmentalist' findings and interpretations. For example, I was recently told by a journal editor that one of my articles - which took all of seven months to be reviewed - had to be sent to seven reviewers in order to obtain two reviews of the article itself; the rest were merely diatribes against 'Jensenism'; the editor apologized that they were too insulting to pass on to me. —Arthur Jensen

What we have here is simply an extension of that very same bias. Wikipedia is supposed to rise above such bias and present views in a fair, neutral and objective fashion. This article fails miserably, and to invoke the policy of NPOV as a defense is to make a mockery of the same. --Aryaman (talk) 16:07, 17 October 2009 (UTC)

The issue with that is that those quotes reinforce a marginalization approach to hereditarian coverage. Ours is to reflect academia, not right great wrongs. If we were to make our article different than academia actually does, it would violate a number of Wikipedia policies. --Nealparr (talk to me) 17:45, 17 October 2009 (UTC)
This space is for suggesting changes to the article, not random quotes with no source and complaints about fundamental wikipedia policies. T34CH (talk) 16:16, 17 October 2009 (UTC)
(1) There's probably a misunderstanding here. Those quotations are probably references to the heritability of IQ in general, rather than race differences. There used to be considerably more controversy surrounding that topic. (2) The section currently says "The majority of intelligence researchers argue that studies which may appear to support a hereditarian hypothesis of intelligence are either flawed and thus inconclusive, or else that they actually support a primarily environmental (<20% genetic) hypothesis", citing a series of review papers from 2005. It would appear to me that what's being expressed in the writing in this section is the opinion that the conclusion that there is no genetic contribution and the conclusion that there's not enough evidence to support a conclusion are actually one and the same conclusion. I believe that may very well be a working assumption of the cited authors (could we provide full citations). From what I've read elsewhere, I'm not sure that that view is actually held by a majority of intelligence researchers. The majority appear to adopt the view that all of the studies (pro and con) are too weak to support either minority conclusion. --Distributivejustice (talk) 19:41, 17 October 2009 (UTC)

Reliably sourcing consensus statements

Regarding my earlier point that consensus statements must be reliably sourced, [1] is in fact a consensus statement from the American Psychological Association. However, it would have to be used in the article something like this: Iq#The view of the American Psychological Association, giving context to the consensus statement and not synthesizing other sources/statements to the consensus statement. --Nealparr (talk to me) 14:45, 17 October 2009 (UTC)

I think that's very appropriate. It should be possible to lay out the different opinions. --Distributivejustice (talk) 19:55, 17 October 2009 (UTC)

Regarding the nature-nurture debate, there doesn't seem to be an academic consensus among people who publish on the topic. (I assume without any documentation that a random poll of all academics would find support for an entirely environmental source of black-white IQ differences. I likewise assume a poll of the general public would show the same.) I tried to do a literature review for citations supporting a consensus. I found that there were roughly two minority and one majority conclusions about the nature-nurture. The two minority conclusions were that (a) there's definitely no genetic contribute and (b) there's definitely a non-trivial contribute. The majority conclusion was that (c) there's not enough evidence to convincingly accept either (a) or (b). Here are the most commonly cited sources I found regarding academic opinion:

  • Loehlin, J.C., Lindzey, G., & Spuhler, J.N. (1975). Race differences in intelligence. San Francisco: Freeman.
  • Snyderman, M. & Rothman, S. The IQ Controversy, the Media and Publication(Transaction, New Brunswick, NJ, 1988).
  • Neisser, U., Boodoo, G., Bouchard, T.J., Boykin, A.W., Brady, N., Ceci, S.J., Halpern, D.F., Loehlin, J.C., Perloff, R., Sternberg, R.J., & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77- 101.
  • Brody, Nathan (1992), Intelligence. San Diego: Academic Press.
  • Gottfredson. (1997). Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography. Intelligence, 24, 13-23.

I found Earl Hunt's opinion (Intelligence (2009) pp. 1-2) to be the most succinct summary of what appears to be the majority opinion (in the context of a book review of Nisbett 2009):

  • Nisbett provides substantial documentation of Black--White environmental differences that probably influence the gap. But then he goes too far. On page 118 he states "Genes account for none of the differences in IQ between Blacks and Whites." In order to show that environment effects account for the entire gap requires showing that the range of environmental effects in the population is wide enough to do so. I can think of studies that could address the issue, but they are wildly impractical and certainly have not been done. I wish that the loud voices speaking about the genetics of the Black--White gap would write "We do not know" on the blackboard at least 500 times.

I seem to recall Nathan Brody saying the same thing in a review of Jensen's work, but I can't find the citation now. --Distributivejustice (talk) 19:31, 17 October 2009 (UTC)

I forgot to include in that list Lohlin's book chapter titled "Group differences in intelligence" in the "Handbook of intelligence" edited by Robert J. Sternberg. His conclusion regarding the cause of black-white differences: "So we are left with the usual conclusion: More research is needed." If you read carefully, he seems to lean towards a gene-environment interaction model as an avenue of future research, but appears to be sticking with the conclusion that the available data is contradictory and weak. --Distributivejustice (talk) 19:53, 17 October 2009 (UTC)

Tags

Now that the outline has been restructured, and chunks of extra information (like the two long lists of names) have been removed, I think we can revisit the question of the NPOV and unbalanced tags. What do folks think? I suggest moving tags to relevant sections now if there are specific problems. As a whole the article looks much better to me (though I still think it should be named after an actual thing rather than a nebulous correlation of two nebulous concepts. T34CH (talk) 18:10, 17 October 2009 (UTC)

I think that the tags should be removed. David.Kane (talk) 21:28, 17 October 2009 (UTC)
I agree with that. Obviously the entire article isn't in dispute. I recommend removing the tags and encouraging editors to tag individual passages or sections they have issue with and then fixing those. --Nealparr (talk to me) 21:36, 17 October 2009 (UTC)

Overview section and nature/nurture debate size

I'd like to help with the overview section. The content is generally correct, but the reasons stated for the conclusions are not. A few examples from the first paragraph:

  • There are important caveats about conclusions made about intelligence from IQ scores, but intelligence rather than IQ is a construct. IQ is a measurement. g is a more precise construct, being mathematically defined. Whether group differences in IQ are also differences in g is thus an important question.
  • Racial and ethnic groups are socially defined and not homogeneous, but it's inaccurate to say that they are no biological differences. No one knows if there are biological differences related to intelligence -- that's just begging the nature/nurture question.
  • The major criticism of Jensen's hypothesis is that the research supporting his conclusions are flawed and not compelling, not that he simply infers a genetic contribution in the absence of evidence.

I'm not certain about the best approach to fixing this.

One issue to consider is whether the nature/nurture debate should be allowed to consume so much space. The authors of this article appear to have been trying their best to muster support for competing hypotheses. There seems to have been particular emphasis put on providing primary literature references to evidence for an environmental cause for the differences. Yet the existence of many environmental differences doesn't seem to be in dispute. Overall, this seems like a flawed approach to writing about this topic. The details of the debate and why a particular piece of research is considered flawed or not is usually very esoteric. That kind of detail seems out of place here. If these sections were rewritten to put less emphasis on data and more on reasoning and conclusions, it would probably be more readable and accessible. That would also shrink the nature/nurture sections to a reasonable size and give room for other topics.

If those sections were made less dense, then should nature/nurture be the focus on an overview section? It would seem that the focus should then be on providing a broader overview.

Tentatively, I'd suggest that focusing on a few key figures, such as Flynn and Jensen, would help focus the nature/nurture section to a readable form. --Distributivejustice (talk) 20:36, 17 October 2009 (UTC)

You bring up some good points. First it is important to state that most of the current text was written for a completely different formulation of the article. A lot of the text has simply been moved around to conform to our newly agreed upon outline (see above). Quite a bit of work remains to better integrate the paragraphs into their new homes, cut out dead weight, and write new text to bridge wandering lines of reasoning. It is probably the case that a lot of the language came to be as reaction to statements which have now been downgraded in prominence or removed, so keep this in mind while analyzing the current text.
I think you are quite right to suggest that more attention needs to be paid to other issues (such as problems with measuring intelligence and quantifying race). As for your statement that "it's inaccurate to say that they are no biological differences", the sources used in that section support that language. I'm sure we'll be able to find sources which show that skin color and facial topography have a high degree of genetic basis... but that doesn't negate the problem that "race" is difficult to quantify reliably because it is not a well defined construct. T34CH (talk) 20:59, 17 October 2009 (UTC)
Thanks. Okay, so taking that as an accurate summary of the cited sources, I'd suggest that the summary is simply incomplete. Assuming that what you've summarized is a common enough argument, then it should be encapsulated as being the opinion of the cited authors and then extended with other alternative views. Along those line, I did some more digging and I'd propose the following as a complete-enough discussion of the core nature-nurture debate over the last 30 year: (1) Start with Jensen's argument. (2) Then give Lewontin's response to Jensen. (3) Then give Flynn's responses to both Jensen and Lewontin. It would seem that there are also other parallel threads, like the ones you've summarized currently. As these different threads seem incompatible with one another, I'd suggest not trying to merge them into a single narrative synthesis, but laying them out individually without connecting them together. --Distributivejustice (talk) 21:08, 17 October 2009 (UTC)
That would be a good outline for the history section of the article, but we examined that as a possibility for outlining the whole article and I believe we abandoned the idea. Neisser (1996) is basically the statement of the APA, and Sternberg (2005) is the only other source even suggested so far to be used as a basis for discerning academic consensus. If there's a source we've missed we should work it in, but I think it would be much better for this article to try to fill in the gaps rather than turn it back into a dialogue of opposing viewpoints. T34CH (talk) 21:20, 17 October 2009 (UTC)
In particular, it seems that the Sternberg et al paper supports the summary you gave but the Neisser et al paper gives the alternative and very different argument I've seen elsewhere. --Distributivejustice (talk) 21:16, 17 October 2009 (UTC)
Can you quickly state which summary you're talking about here? I just got lost because I read the two sources as mostly agreeing with each other. T34CH (talk) 21:24, 17 October 2009 (UTC)
Sorry I really wasn't clear :). The overview section of this article seems to summarize Sternberg only. It seems to me that Sternberg's argument is entirely different than what's present in Neisser (hence my objections above about why this argument doesn't match what I've seen before). It seems that those two (Sternberg and Neisser) only agree on the most superficial level. Their reasons are entirely different.
I recognize the Neisser article as being a recap of Lewontin and a little bit of Flynn. There's more and newer material from Flynn. (And more and newer material from Jensen.) The Sternberg article seems to be an entirely different 4th view. I'm not suggesting a necessarily historical article, but rather that the focus should be on these 3 (4?) main arguments which I see repeated in the secondary literature. So, I think I see the distinction between history and dialogue you're suggesting, but I'm not sure that it means that there's only a single dialogue happening. --Distributivejustice (talk) 21:39, 17 October 2009 (UTC)

[outdent] So concrete next steps would be (1) attribute "consensus" views to their authors directly (e.g., Sternberg et al). (2) Focus on the interlaced arguments/views made by at least (a) Jensen, (b) Lewontin and (c) Flynn, using them as exemplars rather than as historical figures. This should be like describing utilitarianism and categorical moral philosophies by referring to Kant and Bentham. --Distributivejustice (talk) 21:45, 17 October 2009 (UTC)

I disagree that the overview only summarizes Sternberg. Neisser et al states:
  • "These groups [Chinese and Japanese Americans, Hispanic Americans ("Latinos"), Native Americans ("Indians"), and African Americans ("Blacks")] (we avoid the term "race") are defined and selfdefined by social conventions based on ethnic origin as well as on observable physical characteristics such as skin color. None of them are internally homogeneous. Asian Americans, for example, may have roots in many different cultures: not only China and Japan but also Korea, Laos, Vietnam, the Philippines, India, and Pakistan. Hispanic Americans, who share a common linguistic tradition, actually differ along many cultural dimensions."
  • "It is clear, however, that these differences--whatever their origin--are well within the range of effect sizes that can be produced by environmental factors."
  • "Because there are many ways to be intelligent, there are also many conceptualizations of intelligence... Standardized intelligence test scores ("IQs"), which reflect a person's standing in relation to his or her age cohort, are based on tests that tap a number of different abilities. Recent studies have found that these scores are also correlated with information processing speed in certain experimental paradigms (choice reaction time, inspection time, evoked brain potentials, etc.), but the meaning of those correlations is far from clear."
  • "Because ethnic differences in intelligence reflect complex patterns, no overall generalization about them is appropriate."
  • "Explanations based on factors of caste and culture may be appropriate, but so far have little direct empirical support. There is certainly no such support for a genetic interpretation."
  • "It is widely agreed that standardized tests do not sample all forms of intelligence."
This seems to completely agree with the first paragraph of the Overview. If you feel something is missing feel free to add it, but as you can see below I think the article is too broad in scope. T34CH (talk) 22:00, 17 October 2009 (UTC)
The overview (this article) and Sternberg are arguing that it's impossible for there to be a genetic cause of race differences in intelligence because (a) the very concept is too imprecise to be meaningfully examined and (b) races are not genetically different enough to allow such an effect. That's entirely different than arguing that the question is intelligible, that the empirical evidence allows for an entirely environmental cause, and that no firm conclusions can be made because of the weakness of the empirical evidence. The difference is in the structure of the arguments. --Distributivejustice (talk) 22:09, 17 October 2009 (UTC)
I think you're misreading the overview. It says:
Intelligence (often approximated using IQ) is not a well defined construct, and IQ tests do not provide definitive measures of intelligence. Race and ethnicity are socially defined groups—rather than biological observations—whose members are not homogeneous; races and ethnicities are often defined by affiliation with very large geographical areas (Asian) or common language (Hispanic). For these reasons, discussions of correlations between race and intelligence which extrapolate a genetic causation are fundamentally flawed.
The first sentence is supported by "Because there are many ways to be intelligent, there are also many conceptualizations of intelligence... Standardized intelligence test scores ("IQs"), which reflect a person's standing in relation to his or her age cohort, are based on tests that tap a number of different abilities. Recent studies have found that these scores are also correlated with information processing speed in certain experimental paradigms (choice reaction time, inspection time, evoked brain potentials, etc.), but the meaning of those correlations is far from clear." and "It is widely agreed that standardized tests do not sample all forms of intelligence.", as well as "Such a summary merely acknowledges that performance levels on different tests are correlated; it is consistent with, but does not prove, the hypothesis that a common factor such as g underlies those correlations. Different specialized abilities might also be correlated for other reasons, such as the effects of education. Thus while the g-based factor hierarchy is the most widely accepted current view of the structure of abilities, some theorists regard it as misleading (Ceci, 1990). Moreover, as noted in Section 1, a wide range of human abilities--including many that seem to have intellectual componentsmare outside the domain of standard psychometric tests." and "Finally, we do not yet know the direction of causation that underlies such correlations. Do high levels of "neural efficiency" promote the development of intelligence, or do more intelligent people simply find faster ways to carry out perceptual tasks? Or both? These questions are still open."
The second sentence is supported by "These groups [Chinese and Japanese Americans, Hispanic Americans ("Latinos"), Native Americans ("Indians"), and African Americans ("Blacks")] (we avoid the term "race") are defined and selfdefined by social conventions based on ethnic origin as well as on observable physical characteristics such as skin color. None of them are internally homogeneous. Asian Americans, for example, may have roots in many different cultures: not only China and Japan but also Korea, Laos, Vietnam, the Philippines, India, and Pakistan. Hispanic Americans, who share a common linguistic tradition, actually differ along many cultural dimensions."
The third sentence follows ipso facto, as well as being supported by "It is clear, however, that these differences--whatever their origin--are well within the range of effect sizes that can be produced by environmental factors." and "Because ethnic differences in intelligence reflect complex patterns, no overall generalization about them is appropriate." and "Explanations based on factors of caste and culture may be appropriate, but so far have little direct empirical support. There is certainly no such support for a genetic interpretation."
Please let me know if I'm missing something here. T34CH (talk) 22:31, 17 October 2009 (UTC)
They are drawing the conclusions they draw for very different reasons than the first two sentences. The third sentence is Sternberg's conclusion, which he draws from the first two. "It is clear, however, that these differences--whatever their origin--are well within the range of effect sizes that can be produced by environmental factors." -- This is referring to the Flynn effect. "Because ethnic differences in intelligence reflect complex patterns, no overall generalization about them is appropriate. -- This is referring to g vs non-g factors in group differences. "Explanations based on factors of caste and culture may be appropriate, but so far have little direct empirical support. There is certainly no such support for a genetic interpretation." -- This is saying there's no scientific answer at present, although caste and culture are directions to look for an answer.
I drafted an outline below. I hope that helps. I'll try to flesh it out further and include Sternberg et al. --Distributivejustice (talk) 22:43, 17 October 2009 (UTC)
Consensus views, by definition, would not be attributed to individual authors, but rather to communities (for example the APA). Neisser's article, by example, would be the view of the APA itself rather than his own, as that was the purpose of the paper and it was published as such. --Nealparr (talk to me) 21:49, 17 October 2009 (UTC)
Right, yes. More directly -- (a) Sternberg's article isn't a consensus view but rather his own and (b) Neisser's article has a little bit of consensus ("At present, this question has no scientific answer.") and a lot of summarizing Jensen/Lewontin/Flynn. (Wikipedia:Rs#Academic_consensus) --Distributivejustice (talk) 21:55, 17 October 2009 (UTC)
I asked for more suggestions to asses consensus, but none were given and no objection was made to Sternberg. Please feel free to add some sources for this purpose, or suggest a method of assessing consensus. T34CH (talk) 22:02, 17 October 2009 (UTC)
I don't know about Sternberg's article (not familiar with it), but the Neisser article was commissioned by the APA for the purpose of stating the view of the APA, and it was published by the APA as their view. As such, it is not the view of Neisser, Jensen, Lewontin, or Flynn, but the view of the APA as a whole. Attributing it to the authors, or the author's sources, rather than to the APA, greatly diminishes the support the view actually has. The paper is a golden example of how one would reliably source a consensus statement per WP:RS#Academic consensus. --Nealparr (talk to me) 22:04, 17 October 2009 (UTC)
I'm sorry for being unclear with those names. I think we aren't disagreeing. Neisser is the lead author of the task force. Neither Jensen, Flynn nor Lewtonin are authors of the report. Rather, I'm saying that they are the primary subjects of the portion of the report related to nature/nurture of group differences. Sternberg authored an entirely different review article which is being cite. It seems to me that the overview summarizes Sternberg but not the APA. --Distributivejustice (talk) 22:12, 17 October 2009 (UTC)
I see. I thought you meant that the stuff that is APA you wanted attributed to Neisser et al. versus the APA. Sorry if I got that wrong. --Nealparr (talk to me) 22:30, 17 October 2009 (UTC)

Focus of article

"I think you are quite right to suggest that more attention needs to be paid to other issues (such as problems with measuring intelligence and quantifying race)." I disagree. Problems with measuring intelligence and quantifying race are extremely well-covered in their respective Wikipedia articles. There is no need to rehash those debate here, beyond proving a brief acknowledgment that there are debates and links to the appropriate parts in Wikipedia where those debates are hashed out in detail. This article would be better if it were shorter and more focused. Or am I the only one who thinks that? David.Kane (talk) 21:33, 17 October 2009 (UTC)

I would agree, if the article were called Hereditarian hypothesis of intelligence. But under the current title I don't think we can split off too much body without running afoul of wp:UNDUE. T34CH (talk) 21:41, 17 October 2009 (UTC)
I certainly agree that the focus is poor. But that's one reason for having a clear Overview section at the beginning of the article which summarizes the current understanding of how race and intelligence relate to each other. Aprock (talk) 22:26, 17 October 2009 (UTC)
David, in my opinion, you're right. I think this article would benefit greatly from letting the main articles deal with the larger controversies, and focusing more on the central issue. See WP:MNA. --Aryaman (talk) 01:27, 18 October 2009 (UTC)
I think this is a good approach as well. I've been including "see also" and "main" links in hopes the overall article could move in that direction. Aprock (talk) 01:55, 18 October 2009 (UTC)
T34CH, can you explain the connection to wp:UNDUE here? Consider race. Obviously, if race is a meaningless concept, then an article on race and intelligence makes little sense. Some people do, indeed, argue that race is a meaningless concept. But that discussion can (and does!) occur in the race article. There is no need (and much harm) to rehashing that here. Instead, we should just reference that debate. Or do you think that the entire debate about the meaningfulness (or not) race that occurs at race needs to be repeated in this article? David.Kane (talk) 01:33, 18 October 2009 (UTC)
"can you explain the connection to wp:UNDUE here": It's simple Dave, "In general, articles should not give minority views as much or as detailed a description as more popular views... In articles specifically about a minority viewpoint, the views may receive more attention and space." If this article is about "Race and intelligence", then the majority of description/attention/space has to be about the majority academic opinions. This is because it is framed as an overview article. If this article is about the "Heritability hypothesis of intelligence," then it is very easy to cut out over half the article as obviously unnecessary (and even a violation of UNDUE), as long as it is made clear that the hypothesis is a minority one and why.
"do you think that the entire debate about the meaningfulness (or not) race that occurs at race needs to be repeated in this article": Uh, no. Never said that. Never will. Show me, however, what section in the article rehashes Social interpretations of race so much that's doing all this harm you mentioned above.T34CH (talk) 02:18, 18 October 2009 (UTC)
Well, I think David has raised a very important issue for discussion, and WP:MNA supports it. David and I both have made mention of this before, but no one seemed to see that it could represent a solution to some of the difficulties with this article. I'd like to see other editors comment on this before rejecting it offhand. --Aryaman (talk) 02:32, 18 October 2009 (UTC)
"there is probably not a good reason to discuss some assumption on a given page, if an assumption is best discussed in depth on some other page." No one has argued with this or dismissed it. This section in particular came about in reference to a statement I made that the article needed to outline some issues specifically not outlined in the text, but are discussed at length by the sources. Nowhere have I said we needed to re-hash some other article. I also don't think you will find discussion in other articles of intelligence researchers failing to quantify race. T34CH (talk) 02:50, 18 October 2009 (UTC)
If there are sections which you feel are duplicates of text elsewhere in Wikipedia, please inform us. T34CH (talk) 02:51, 18 October 2009 (UTC)
I do not think that currently there are duplicate sections. I was reacting to your claim that "more attention needs to be paid to other issues (such as problems with measuring intelligence and quantifying race)." More attention does not need to be paid to these issues in this article. The amount that we currently pay attention to these issues (mostly by referencing other Wikipedia articles) is enough. Do you disagree? How much more attention (a sentence, a paragraph, 5,000 words) do we need to devote to these topics in this article? David.Kane (talk) 14:46, 18 October 2009 (UTC)
Since you're asking, I think the whole tone of the article, from the sections down to the individual sentences, needs to be examined in light of WP:MNA. Take a sentence such as "Self defined black and white United States citizens have been the subjects of the greatest number of studies." Why "self defined"? Is this sentence really the place to stress - yet again - that "race is a social construct"? The article makes mention of this, and the issue is discussed at length at race. Our reader is intelligent enough to make a mental note of this fact, and to understand the simple, everyday language of "Black" and "White" in light of it. No need to beat him/her over the head with it at every turn. Such writing reeks of either paranoia or pedantry, and in this case, both. The article is saturated with this kind of thing. But really, I'd prefer to let other editors comment on this. I think we've expressed our opinions sufficiently. Let's shut up and find out what other editors think. --Aryaman (talk) 03:15, 18 October 2009 (UTC)
In this specific article extra care must be made to ensure that discussions of race match the underlying sources. The issue is that the article is really about what determines intelligence, with all researchers listing various social and environmental factors (including social race) and with a few researchers listing genetic race as a factor. Because both genetic race and social race are significant topics, the article needs to be clear which is being discussed in any given section. Aprock (talk) 03:28, 18 October 2009 (UTC)
If you can't cite an example of a chunk of text we should remove, then we are not even talking about anything. Your example is a fallacy because it refers to a specific type of study: the study of IQs of people who self-identify as black. This is different from studies of IQ in general populations, studies of spacial memory in people identified by the researcher as Inuit, or studies of GPA in people identified by their best friends as mole people. It is a specific type of study, and that is the specific point the sentence is trying to make. I feel you are grasping at straws to assert a point with no actual basis in the existing text. T34CH (talk) 03:49, 18 October 2009 (UTC)
Sorry to post here again, and I'm loathe to make a point, but since when is it not self-identification? As someone who has had the privilege of enjoying close ties to Yupik and Inupiat people, I'd like to state for the record that no one "assumes" their identity for them - no government agency, no social worker, no police officer, and certainly no one administering an IQ test. Perhaps it would surprise you to learn that they're quite conscious of their own identity, and are generally rather articulate in expressing it. Please, be careful with what you're asserting, as it can easily come across in a way I'm assuming you do not intend. As for the rest, if others feel the idea is worth discussing, I'd be happy to bring up numerous examples of where I feel this article could be improved. Thanks, --Aryaman (talk) 04:17, 18 October 2009 (UTC)

what's nature-nurture consensus?

There was a question above regarding what the APA 1996 report said on the nature-nurture question and race. I thought it best to open a new section on that.

This is the relevant text from the conclusions section (literally their final word on the topic) from Neisser et al: "Explanations based on factors of caste and culture may be appropriate, but so far have little direct empirical support. There is certainly no such support for a genetic interpretation. At present, no one knows what causes this differential."

Their other formulation of this conclusion was that "In short, no adequate explanation of the differential between the IQ means of Blacks and Whites is presently available."

Earlier in the text they wrote: "If group differences in test performance do not result from the simple forms of bias reviewed above, what is responsible for them? The fact is that we do not know. Various explanations have been proposed, but none is generally accepted."

That makes their conclusion pretty clear. If anything they are hinting towards preferring Flynn's view, which is the same thing you can sort-of read between the lines in Loehlin 2000, but neither are ready to affirm that hypothesis.

Indeed, the same conclusion was stated in Loehlin 2000: "So we are left with the usual conclusion: More research is needed." and Gottfredson's article: "There is no definitive answer to why IQ bell curves differ across racial-ethnic groups." and my favorite for style from Hunt 2009: "I wish that the loud voices speaking about the genetics of the Black--White gap would write "We do not know" on the blackboard at least 500 times." [emphasis added in all above]

I've read other authors claim affirmatively that there's definitely no genetic contribution and that there definitely is a genetic contribution, but none were claiming to speak for a consensus. --Distributivejustice (talk) 07:13, 18 October 2009 (UTC)

This set of quotes seems a bit selective to me. For example, a more complete quote of one of your selections (with your selection bolded):

Several culturally based explanations of the Black/ White IQ differential have been proposed; some are plausible, but so far none has been conclusively supported. There is even less empirical support for a genetic interpretation. In short, no adequate explanation of the differential between the IQ means of Blacks and Whites is presently available.

So yes, no adequate explanation has surfaced. But of the explanations that were available then, the environmental ones had more empirical support. The academic consensus then, and I believe the academic consensus now, is well represented by those three sentences. At the fine grain data level, there is a lot more nuance, but at a high level almost all sources I've read conform to this general view. The empirical support for plausible environmental explanations is generally better than for plausible genetic explanations. Aprock (talk) 08:56, 18 October 2009 (UTC)
Ah. You have to look at what they are referring to specifically as the plausible explanation -- the Flynn effect. They are saying that this is an explanation for which no contrary data was presently available. It also happened to be the newest explanation and one for which there was also no affirmative evidence. That is why they ultimately conclude more generally that there is no answer. --Distributivejustice (talk) 09:00, 18 October 2009 (UTC)
Two related things worth noting. (1) The other multi-author survey (published by Gottfredson and quoted above) says the same thing ("There is no definitive answer"). (2) Flynn has a recent book: "What Is Intelligence?: Beyond the Flynn Effect", which fleshes the new material since 1996. --Distributivejustice (talk) 09:04, 18 October 2009 (UTC)
[interjecting here to correct myself] Actually, to be more precise the conclusion is that the Flynn effect makes an environmental explanation plausible (unstated: as a work around to Jensen's argument). The one or more environmental factors aren't known (nor is the cause of the Flynn effect itself), but they suggest that caste and/or culture could fit that bill. --Distributivejustice (talk) 09:17, 18 October 2009 (UTC)
[Another interjection] In a parallel manner, they also say that genetic explanations are plausible depending on the answer to the question: "How different are the relevant life experiences of Whites and Blacks in the United States today?" They answer: "At present, this question has no scientific answer." -- The question is a rephrasing of Lewontin's argument. That squares with my initial impression on reading Neisser that it was mostly a restatement of the Jensen/Lewontin/Flynn narrative. --Distributivejustice (talk) 09:53, 18 October 2009 (UTC)
Why don't we just get rid of the "overview" section entirely, at least for now? As I said before, at present it doesn't add anything to the article that isn't already better-explained by other parts of it, particularly the lead section, which is itself an overview also. Aprock, you and I spent quite a while coming up with an accurate summary of the consensus to put in the lead section, and the explanation in the "overview" is both redundant to this and appears to contradict it.
This isn't to say we ought to never have a section of the article along these lines, if Distributivejustice can come up with a new version of it that's acceptable. But as for what we currently have, I think the article is better off without it than with it. --Captain Occam (talk) 09:12, 18 October 2009 (UTC)

[outdent] Above I asked whether the overview was a background primer. Aprock said: No. It is to give an overview of our current understanding of the relationship between race and intelligence. Simply said, what are the current conclusions.

So then I would agree that we only need one such summary, above the TOC. --Distributivejustice (talk) 09:20, 18 October 2009 (UTC)

Obviously there's no consensus to remove it yet, Captain, so stop deleting it. A few hours and a two editors chiming in does not a consensus make. --Nealparr (talk to me) 11:26, 18 October 2009 (UTC)
Let me reiterate that point because it's a recurring problem on this article. You said why don't we get rid of it at 09:12. Exactly one person responded at 09:20. You go ahead and delete it, without waiting to see what anyone else thinks, despite ongoing multiple conversations occuring on the talk page about the very section. There is no hurry. There is no deadline. There is no reason not to wait for support on contentious edits (like removing an entire section). We all likely live in multiple time zones and have different schedules. This is the kind of editing behavior that leads to edit wars. --Nealparr (talk to me) 11:49, 18 October 2009 (UTC)
My removing of it wasn’t just based on that one exchange with Distributivejustice. More than anything, it’s based on the fact that the addition of this section seems to have been an erroneous edit to begin with, and the exchanges on the relevant discussion pages support that conclusion. T34CH added it to the article yesterday without any prior consensus or discussion, and it wasn't even mentioned on the talk page until Varoon Arya brought up the problems with it. VA tried to talk to T34CH about these problems on his userpage, and T34CH replied with a flippant response that did not even attempt to address VA's concerns. The most that can be said for T34CH's decision to add this section is that he's failed to justify it, and his utter indifference about whether it violates NPOV (when VA was carefully pointing this out) suggests that he may just be trying to push a POV without regard for Wikipedia's policy.
It bothers me that this edit has been allowed to remain for as long as it has. Over the past day, Distributivejustice has explained how it does not reflect the academic consensus on this topic; VA has explained how it violates NPOV; I've explained how it's redundant with the lead section, which it contradicts, and none of our objections to it have been answered in any substantial way. Should an edit that was made without discussion or consensus, and is this clearly erroneous, be allowed to remain in the article for over a day just because not enough users have weighed in on this fact yet? --Captain Occam (talk) 12:09, 18 October 2009 (UTC)
The Overview (formerly "Academic consensus") was part of the restructering outline (above) that was proposed and generally accepted several days ago. It's been mentioned that with the restructuring, several passages of text may seem disjointed or out of place and that they'll need to be reworded. I support an overview/consensus section. I'm not convinced that the text of our current overview is proper, but it's been repeatedly stated that the article is in flux because of the restructuring and that now is the time to discuss changes to the text. I agree. Now is not the time to off-hand remove sections that were agreed to earlier, but rather take the time to patiently discuss what the text of that section will say. If consensus builds to remove it, that's another matter, but the prior consensus was that some sort of academic overview belongs in the article, and belongs at the top. If you disagree, make the case for removing it, as the case was made to restructure the article according to the proposed outline from several days ago. Most importantly, though, WP:AGF and give editors time to respond to your complaints. It creates a better editing environment. --Nealparr (talk to me) 12:27, 18 October 2009 (UTC)

(reset indent) For the record: I support T34CH's summary over DistributiveJustice's. I strongly disagree with DJ's interpretation that the APA's statement reference to plausible environmental explanations limits itself to the Flynn effect. I believe in this respect, DJ is overinterpreting.--Ramdrake (talk) 12:31, 18 October 2009 (UTC)

@DJ, If you have a citation which describes genetic explanations for the racial gap as having more empirical support than the environmental explanations, by all means share it. Regarding the overview, The article is about many things, not just the academic consensus. Replacing the lede with such a summary would probably be considered undue weight. Aprock (talk) 14:35, 18 October 2009 (UTC)
  1. ^ Cite error: The named reference Neisser, U. 1996 was invoked but never defined (see the help page).
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:Race_and_intelligence/Archive_72&oldid=1074822621"