Talk:Race and intelligence/Archive 43

OR images
Removed several OR images, compiled and interpreted from data sets in violation of WP:NOR. For example, the GRE image, showing "average GRE score by race", failed to include the data in the cited table regarding number of examinees and standard deviation data, both highly relevant to the utility of the "mean" column in the table.

Without direct citation of the data tables involved, such constructed images are clearly OR, and must be removed. --JereKrischel 09:04, 14 January 2007 (UTC)


 * You can make a case for included the SD and N, but you can't claim the graphical presentation of a published data table is OR -- not without a new reading of NOR. I'm not sure what the difference between "direct citation" and mere "citation" is. --W. D. Hamilton 17:31, 14 January 2007 (UTC)


 * regarding edit summaries - they suggest that the data is "compiled" from more than one source or otherwise manipulated before presentation. this is not the case. they are direct plots of data from data tables or figures from single soures, without any intermediate manipulation. the original use of the data is preserved in the exiting presentations. in several cases, the graphs are near-identical reproductions in all but the trivial details of the figures. For example, the world-wide map is based on a nearly identical figure in from Lynn 2006 but with the number of colors/classes reduced as per a figure from Rushton. The top figure is created from a consensus of various nearly-identical graphs published by Gottfredson, but with different color schemes -- the message of the figure is the same as Gottfredson's (showing gap and overlap). --W. D. Hamilton 17:47, 14 January 2007 (UTC)


 * "Nearly identical" figure should simply be scanned from Lynn 2006, instead of created in a novel way. Graphs from Gottfredson should be scanned and cited directly.  Images are clearly violating WP:NOR. --JereKrischel 19:15, 14 January 2007 (UTC)


 * You can't violate NOR at the same time that you have the possibility of "simply" scanning. That in itself is proof that NOR is not violated. WP cannot possibly have a policy that editor-created diagrams and figures from acceptable source materials are OR, else there would be no editor-created diagrams or figures possible. --W. D. Hamilton 20:12, 14 January 2007 (UTC)
 * In further support of the impossibility of the suggestion, I highlight that accepted practice of creating charts, maps, etc. by linking the figures created by fellow editor Ultramarine --W. D. Hamilton 20:27, 14 January 2007 (UTC)
 * Ultramarine you can defend your opinions here. In the meantime, if you two continue to insist that simple figures are OR, then I'll be forced to seek escalation. I think that would be a colossal waste of a lot of people's time (mostly my own), but this move is too much a bully-move to simply let go. It reflects the worst of WP, and is part of the process that has led so many others to give up hope and leave the project. --W. D. Hamilton 21:10, 14 January 2007 (UTC)
 * My figures contain no original interpretation since they simply depict data already presented as a table. Do your figures similarly represent repsent data already present as prefabricated tables in articles, or do you make an original synthesis by for example combining two data sets? Ultramarine 23:40, 14 January 2007 (UTC)


 * I mostly agree with the distinction you draw and these graphs satisfy that criteria. They do not combine data in a way to draw any new conclusions. Most are re-creations of existing graphs. Those that are based on only on data tables draw from a single source (Lynn 2006) and single data tables are used for each *series* presented. As JK does at one point, you could argue that additional or less data should be put into each graph (I used by judgment for what data was important -- averages -- and the trivial details of colors, etc), but there's no compilation or synthesis issues. --W. D. Hamilton 00:24, 15 January 2007 (UTC)


 * If, as you say, the recreated graphs you've made are identical representations of the graphs elsewhere, I have no objection. However, selectively creating a graph out of data that has not been otherwise graphed (for example the GRE scores), is definitely OR.  If you'd like to include graphs, please include the data tables you've built them from as well, with proper citation/qualification of the data. --JereKrischel 09:59, 15 January 2007 (UTC)


 * The comment you have made here is not concordant with what you did to the article. Graphing data from a singe table is, as per Ultramarine's comment, a perfectly acceptable practice. You can discuss the GRE image here -- as that's the only one for which you've made a specific objection that isn't mooted by the fact that the original image being replicated also has the features you object to. I have no intention to try to copy and paste massive amounts of data into the talk page. Follow the various sources or simply assume good faith that I know how to make a graph. --W. D. Hamilton 17:29, 15 January 2007 (UTC)


 * It would be sufficient for you to paste the data into the description page of the image itself, with appropriate references. If you have the original image, please use it instead, and give a direct page/author citation, and avoid POV characterizations of the data as fact rather than as interpretation. --JereKrischel 10:08, 16 January 2007 (UTC)

It's not possible to put copyrighted images into Wikipedia for this purpose. I've done it previously on talk pages and they are always removed. Most data tables are illegible as plain text. All sources are reasonably available. The RDI tables have been posted at various places on the internet. (Don't worry, I have a copy of the book.) Tangent: Check NPOV for the definition of a fact. It is a fact that a certain piece of data was found/reported. --W. D. Hamilton 17:56, 16 January 2007 (UTC)


 * Although it may be a fact that a certain piece of data was reported, it is inappropriate to assert that data as factual with no qualification. So you may say that "the B-W gap in the U.S. was reported by X as being Y in 1986", but it is inappropriate to say "the B-W gap in the U.S. was Y in 1986".  You seem to have a particularly difficult time seeing that difference. --JereKrischel 02:21, 17 January 2007 (UTC)


 * That's simply not how people write about research -- citations go at the end of sentence / in footnotes for the reason that those details are generally not of interest unless there is a specific reason for it to be. Look for precedents and/or bring up specific details for discussion. I hope we've moved beyond "OR images" now. --W. D. Hamilton 05:29, 17 January 2007 (UTC)


 * I'm still very concerned at your continued insistence of including your own OR images, and find your claim that "how people write about research" must inherently violate the Wikipedia protocol on how to report "fact". Please reconsider for a moment the objections that have been raised.  I believe on further reflection you may better understand the concerns with your contributions, which have obviously taken a great deal of effort on your part.  I'm afraid that you've been treating these articles on R&I as your own personal research papers, rather than an encyclopedic take on the topic.  You may find it more fulfilling to expand on your other online presences such as your gnxp blog, and create your own wiki for collecting and synthesizing your copious original research on the subject.  It would certainly help avoid the WP:NPOV problems strewn throughout this series of articles. --JereKrischel 08:06, 17 January 2007 (UTC)


 * Nothing in your suggestion is concordant with my understanding of WP:NPOV. While the community standards for neutrality, verifiability, etc. continue to change (improve i would say), I don't see them diverging from that practiced in the scholarly literature. (For an example of a high-quality scholarly review article on the topic of intelligence, see .) Such a state of affairs would be both surprising and highly noteworthy. I see no evidence for it. To the main topic: a figure cannot be OR simply because it isn't a scanned image from a publication. It cannot be OR if it presents material from a single source as the source intended it to be presented. The inclusion of helpful images is one of the featured article criteria, and thus a marker of a good WP article. You have offered no good reasons for the removal of images that illustrate the points made in the text. Surely this article will present a constant challenge to meet the highest standards of WP. However, mass removal of material is very unlikely to be a move towards meeting those highest standards. All good articles are a combination of extensive research and skillful writing (really rhetoric). Deletion is seldom a solution per WP:NPOV. If we're offering suggestions - I would suggest that you can improve this article most by finding the references I've missed and seeking neutrality through careful use of balancing rhetoric, rather than by trying to silence a publicly unpopular POV by removing descriptions of it. --W. D. Hamilton 02:13, 18 January 2007 (UTC)


 * Whether or not your POV is popular does not prevent your excessive weight on that POV from becoming problematic in a neutral article. Your example of practices in scholarly literature highlight your fatal flaw in contributing here - you are treating this set of articles as your own personal research papers, and you've done a wonderful job of presenting something worthy of publication in a journal.  You have not, however, managed to construct a WP:NPOV article.  You have failed to give any sort of balance of rhetoric or data, and to assert that my caution to you about WP:NOR represents any sort of "silence" of your POV is to ignore the still unbalanced article we have here. --JereKrischel 05:44, 18 January 2007 (UTC)

The part of NPOV you're looking for is WP:NPOV. The solution to a perceived imbalance of weight (note carefully the criteria of proportionality to the literature) is to add material for balance and use summary style where appropriate. The wholesale deletion of material is rarely the solution. --W. D. Hamilton 05:49, 18 January 2007 (UTC)


 * I'm not sure if "undue weight" is what we're talking about here, although I suppose it is a close approximation. My suggestion is to prune much of what you've added here in copious edits, and use a "summary style".  We're hardly doing wholesale deletes of material here, simply pruning excessive OR on your part. --JereKrischel 06:04, 18 January 2007 (UTC)

WP:NOR -- there's nothing OR about turning data tables into figures or duplicating copyrighted figures. It's a common practice in WP. Show me the part of NOR that says otherwise, and then we'll have a lot of WP figures to have removed. --W. D. Hamilton 05:54, 18 January 2007 (UTC)


 * I'm arguing that your selective filtering of the data tables into figures represents OR. --JereKrischel 06:04, 18 January 2007 (UTC)


 * Are you talking about the GRE image. The one originally created by someone else, which I cleaned up? The original author of the figure clearly used good editorial judgment in what data to present from the table. For the sake of the argument, I'll add the error bars. The product will be unfortunately less useful, and I hesitate to do it given that the only reason you can offer is a mistaken conception of NOR, but if it will placate your anger... --W. D. Hamilton 06:17, 18 January 2007 (UTC)


 * My apologies if you are under the impression that I am angry - far from it, I am quite eager to improve the article and the images with your help. You obviously have skills and tools for creating graphs that I have not acquired, and your contributions are greatly appreciated.  I'm simply concerned that your technical expertise and personal point of view has made it difficult to see when you are performing OR, and when you are pushing POV.  By no means are you particularly obvious about inserting bias, and I truly believe that you may not be aware of it yourself, but as a neutral observer, I must advocate for balance. --JereKrischel 21:48, 18 January 2007 (UTC)

The OR / NPOV distinction is crucial for how we treat material. You can write neutral original research, and you can have biased descriptions of published research. Neutral original research should be removed, baised but not OR material should be improved. Your claims about OR fail to convince me of anything. I'm highly sympathetic and open to concerns about WP:NPOV. OR is clear cut in most cases (especially the cases at hand) but NPOV is subtle and difficult to meet, especially the weight issue. However, simply removing entire concepts from the article is not within the purview of a weight balance. Neutrality is explicitly defined in terms of a correspondence between the published literature on a subject and the content of an article. (Similar to the correspondence theory of truth, but we're talking about what's verifiable rather than what's true.) So for example, Lynn's collation of IQ test results is undeniably the largest and most comprehensive ever done. As stated below the very low scores are controversial, but the scores for minority groups in developed countries most definitely are not. There's nothing OR about representing Lynn's synthesis in the form of a graph. I would argue it's a great help to show the range of values that have been reported, rather than just the overall average. In the controversy section, your attempts to blend the "utility of race" with other topics is probably an NOR:synthesis problem. These are difficult to diagnose, but case is pretty clear to me. Simply give sections titles that reflect their content in the most proximal way, rather than trying to build your own synthesis of ideas to avoid the issue. Likewise, you cannot put your own analysis of data into the article's main space. Only published conclusions can be included in an article. I write for this article exclusively by finding review papers that discuss some aspect of the topic and translating their content into WP content. Based on a strict reading of NOR, this is the only way to go. However, it doesn't assure that appropriate weight is given. The big 3 can help with that a little, but it mostly takes an iterative process of finding new review articles. --W. D. Hamilton 23:08, 18 January 2007 (UTC)


 * I'm hardly advocating removing entire concepts from the article - I'm certainly advocating for balancing the level of detail used to build the POV case on either side. For example, Lynn's collation of IQ test results may be "undeniably the largest", but given his extreme lack of scrutiny of the data, is hardly "comprehensive" in the sense of being a complete analysis of the data.


 * You write for the article by exclusively finding review papers that support your POV, and have done a wonderful job of making your POV case. This is a blessing, and a curse in disguise.  It seems obvious you have copious amounts of time to spend on the topic (perhaps from a Pioneer Fund grant? :) ), and there doesn't seem to be an equivalent to you on the other side.  Perhaps you could try and "write for the enemy", and translate content from published articles that do not match your POV - it may help to give appropriate balance, if you were to attempt it. --JereKrischel 03:06, 19 January 2007 (UTC)

GRE image
on the suggestion that the gre image needs to present the SD and N, rather than just the mean: --W. D. Hamilton 22:24, 15 January 2007 (UTC)
 * the only way to show both SD and N at the same time is to show SEM, which is near zero for such large samples
 * if you show just SD, it will only look like the current graph but with approximately equal length bars sticking out of each. there's terribly little difference in the SDs:
 * the other category is more variable, naturally
 * the scores of Blacks are less variable across the board, so they actually come out looking worse if bars are added
 * the verbal scores of asians are highly variable, also to be expected
 * there is no discussion of SDs anywhere in the section (article?) and there isn't enough to really say much about them if you were to try to build such a section. mostly what you'd find is the variance in the black population (for IQ) is smaller than variance in the white population (but the mean is lower too, so the SD/mean is about the same)


 * By interpreting the data in a novel way, you change the message expressed by the GRE table provided in the original source. I'm not arguing that the graph should look "better" or "worse" than it does, I'm arguing that your graph is OR and misleading because it selectively removes data given in the original table.  If you were to add bars, and make a note of the sample sizes in the description page of the image, I would be happy with it as a fair representation of the data table referenced. --JereKrischel 10:08, 16 January 2007 (UTC)


 * No. Your interpretation of WP:NOR is totally unorthodox. Not reporting all of the details is something that encyclopedias do. If excluding essentially irrelevant details is WP:NOR then the encyclopedia couldn't function. You will not find your redaction theory of NOR in the NOR page. It's very simple. NOR means arriving at novel conclusions. You can't arrive at a novel conclusion by simply not presenting all of the details of something. --W. D. Hamilton 17:51, 16 January 2007 (UTC)


 * You are arriving at a novel conclusion by selectively omitting important data. If I were to assert that the average human height was 7'3", because I omitted all humans except for NBA players, it would certainly count as a novel conclusion. --JereKrischel 02:18, 17 January 2007 (UTC)


 * That makes no sense in the current context. Are you familiar with means and variances? The mean is the issue discussed by the literature, and so we show the mean. --W. D. Hamilton 05:22, 17 January 2007 (UTC)


 * Deciding to take the discussion from the literature and use it as a selective filter on the data being quoted is clearly OR. --JereKrischel 08:02, 17 January 2007 (UTC)


 * Showing a graph of the data that's being discussed in the literature is not OR. The suggestion that not showing the SDs from a data set constitutes OR is equally nonsensical. There is no content to your argument. It appears to be based merely on a false understanding of what NOR means. --W. D. Hamilton 01:57, 18 January 2007 (UTC)


 * Showing a graph of the data that's being discussed, but failing to include relevant data, is OR. There is a simple answer here - adjust the graph, please. --JereKrischel 05:39, 18 January 2007 (UTC)


 * I outlined the reasons why it's impractical to show both N and SD in a graph where N is large. I can easily update the graph, but without looking closely you wouldn't be able to tell the difference (error bars of ~0 height). If, OTOH, you accept that we just can't show the content of the data table in the graph, then I stand by the original claim that we should show what's been discussed in the literature and not add details that are irrelevant. --W. D. Hamilton 05:44, 18 January 2007 (UTC)


 * Simply show SD, and list N. --JereKrischel 06:00, 18 January 2007 (UTC)



Now there are approximately equal length bars sticking out of each column, which only serves to make the graph more difficult for the average person to understand. --W. D. Hamilton 06:24, 18 January 2007 (UTC)


 * Nice edit, please put bars over the colored areas, so that the lower bound can be seen as well. It is much more informative, and satisfies my WP:NOR concerns.  Thank you! --JereKrischel 19:21, 18 January 2007 (UTC)


 * I find it hard to believe that the irrelevance of equal length bars is lost on you. Per the 3rd bullet point in my initial post on this topic, there's a reason that within group variance isn't discussed. These are not "error" bars but descriptions of within group variance. When N is massive, even small differences in mean become statistically significant. --W. D. Hamilton 20:39, 18 January 2007 (UTC)


 * The bars coming out of each column should not be equal length (they certainly aren't so in the data table)...I haven't measured pixel by pixel, but I assume you captured the data as presented. --JereKrischel 21:54, 18 January 2007 (UTC)
 * As I pointed out above, the SDs are so similar that they are visually indistinguishable (at least to me). The absence of major differences in within group variance is a feature of IQ data that is sometimes but rarely noted. Variance just isn't a topic of interest AFAIK, and so the bars are not doing anything but making a complex graph hard to read. --W. D. Hamilton 22:48, 18 January 2007 (UTC)

Equal heights
as i said, the original image uses equal heights. that would be enough to close discussion, but i have to point this out -- to have unequal heights you would need a data source for the relative population sizes. such a data source does not exist side by side with any break down of IQ scores that I know of. by the strict NOR criteria suggested by ultramarine, using two data sources in the same figure would be NOR. i actually think that wouldn't be a big deal -- not NOR -- but because of the massively different population sizes, the reason that the original image used equal heights becomes clear: the figure would be unintelligible without equal heights. --W. D. Hamilton 18:09, 16 January 2007 (UTC)


 * Use the original image if you wish, and cite it, but to reconstruct an idealized image, rather than having an accurate graph showing the actual distributions of test scores and the relative sizes of the populations is simply misleading. If in fact the original image looks very similar, it may be inappropriate to include it because of its misleading nature. --JereKrischel 02:16, 17 January 2007 (UTC)


 * The original image looks almost identical (colors differ). You're opinions about it being "misleading" are quite irrelevant (and wrong for reasons that we need not go into). Take up your objections with the scientific community. --W. D. Hamilton 05:21, 17 January 2007 (UTC)


 * Whether or not the image is misleading is quite relevant. The constructed image clearly fails to demonstrate the vast difference in sample sizes of the various groups, and provides an illusion of simplicity that does not exist in the actual data. --JereKrischel 08:01, 17 January 2007 (UTC)


 * The graph is exactly the kind of graph that you'd want to explain that there are gaps but there is overlap. That was the content it was originally used for and that's why we replicate it here. The graph is not a histogram, does not claim to be, and I see no reason why anyone would be confused that it is. You cannot show the gap and the overlap while showing the relative proportions of each group in the U.S. population. Doing so, you only manage to show those proportions. The original figure most likely shows equal heights for these important reasons and your judgments are thus both flawed and contradicted by the literature. --W. D. Hamilton 02:01, 18 January 2007 (UTC)


 * Asserting that a clear overlap exists, when a proper graph would show the difficulty of discerning such clean breaks due to the relative sample sizes, is exactly what is objectionable. --JereKrischel 05:38, 18 January 2007 (UTC)


 * That's nonsense. A fabrication from your own mind, unique and (I hate to use the word) OR. Basic stats shows how you are mistaken. Don't make things up -- seek sources and represent their opinions -- leave your own opinions the door. --W. D. Hamilton 05:41, 18 January 2007 (UTC)


 * I'm sorry, but seeking sources which are misleading, by misrepresenting data in simplistic graphs, you do a grave disservice to the article. A proper graph is much preferable to a POV pushing one presented as fact. --JereKrischel 05:57, 18 January 2007 (UTC)


 * That again is nonsense. Stop removing cited materials. The graph is an ideal way to open the article. It educates on the subject while clearing up common misconceptions. --W. D. Hamilton 06:14, 18 January 2007 (UTC)


 * A graph is an ideal way to open the article. The particular graph you have chosen is not.  It is misleading, and perpetuates common misconceptions. --JereKrischel 19:19, 18 January 2007 (UTC)


 * You have not identified any criticisms of this graph that have a basis in published opinion. Your one suggestion implies that it is not established that there are statistically significant differences in average scores; this of course is false. You have also removed graphs which for which you have offered no criticism. --20:32, 18 January 2007 (UTC)


 * Since I'm asserting your graph is OR, it would be hard for anyone to have already critiqued it in published literature. --JereKrischel 21:56, 18 January 2007 (UTC)


 * Because the graph is a near duplication of one published many times now, you're claims about OR are baseless. --W. D. Hamilton 22:41, 18 January 2007 (UTC)


 * Then why the aversion to actually using an exact duplication of one published, with proper attribution? --JereKrischel 03:00, 19 January 2007 (UTC)


 * That's a copyright violation; there's no fair use rationale, and publishers won't release their figures under the GFDL. The current figures have the advantage of shading to show the overlap more clearly. --W. D. Hamilton 03:10, 19 January 2007 (UTC)


 * There certainly is a fair use rationale - a single image from a book would certainly count. Would you like me to email the publisher for permission? --JereKrischel 03:25, 19 January 2007 (UTC)


 * INAL, but my understanding of fair use in the content of WP says that won't work. For example, we couldn't use the cover image of LLCS's book about population genetics in the race article, only in the article about the book or the author. I don't have a clue about the publisher, but WP has been on lock-down about fair use. There is a group of editors deleting all promotional head shots and replacing them with free images whenever they are available, even if they are of lower quality. --W. D. Hamilton 03:30, 19 January 2007 (UTC)
 * The same applies to the SES graph, which is from The Bell Curve. The publisher can't give you fair use permission, and I doubt Free Press is going to release the image into the GFDL. However, there would be no need to do so as we can circumvent the copyright by making our own graphs.

Other problematic OR images
The scatterplot and East asian graphs do not contain sufficient information regarding the dates or sizes of the studies mentioned. Simply stating "dozens of studies" strongly implies OR was performed to choose those studies. If Lynn has presented graphs of this sort, please simply use the graphs he presents, and cite them, rather than presenting his particular choice of studies as undisputed or undoubtedly factual. --JereKrischel 21:44, 18 January 2007 (UTC)


 * No single graph can contain all of the information in a massive data table. The data presented is based on what Lynn highlights as important and what the context of the article indicates to be relevant. Your claim "presenting his particular choice of studies as undisputed or undoubtedly factual" -- is groundless -- a creation of your own intellect, and thus not suitable for making such decisions. I know of no claims that Lynn's collection of studies are unrepresentative -- I know of other criticisms, such as the averages for the S.S. African scores -- but no criticisms that Lynn is selectively reporting. --W. D. Hamilton 22:45, 18 January 2007 (UTC)


 * But, even for most of the others, 'direct evidence' is putting it strongly, as even a cursory glance at the motley tests, dates, ages, unrepresentative samples, estimates, and corrections show. . This article clearly claims that Lynn's collection of studies are unrepresentative. --JereKrischel 02:46, 19 January 2007 (UTC)


 * Are you talking about the scatter plots or the map? That article is criticizing IQatWoN. The problem (per the article) there is the estimation of an average IQ for a whole country based on 1 or 2 tests at best, usually tests of children. The tests done in the developing world are often the subject of this criticism for other reasons. The dozens of data points in each the two scatter graphs are not the subject of criticism overall (developed world, no claim to represent whole countries, just subpopulations), and the values shown in the RDI map consist merely of estimates of the average IQ of the indigenous subpopulations of each region. The 100s of estimates in IQatWoNs is obviously open to criticisms that the 10 averages from RDI are not. Thus, we need to be specific on this point. --W. D. Hamilton 02:53, 19 January 2007 (UTC)
 * p.s. Calling the images OR makes it a strain to stay on the topic that you really seem to be discussing. --W. D. Hamilton 02:56, 19 January 2007 (UTC)


 * Both. What "10 averages" are you talking about? --JereKrischel 02:58, 19 January 2007 (UTC)


 * The ten racial group averages. The citation you offered is about linking reaction time measures to global variation in IQ scores. IQatWoN may be criticized for this, but it's not the use that Rushton and Jensen (2005) find (the problem with Lieberman's mischaracterization of their views aside). Here is what Deary said about the limitations of reaction time:
 * This leaves the question of what these variables mean in terms of brain processing. Linking mental test scores to cognitive variables is only really productive when the cognitive variables are themselves theoretically tractable. Otherwise one has merely linked an unknown to another unknown.
 * This clearly does not mean what the is suggested in the current footnote. It is about the ability of mental chronometry to advance the understanding of intelligence. --W. D. Hamilton 03:07, 19 January 2007 (UTC)


 * Ten racial group averages? I'm sorry, I just don't see any image with more than 8 races listed.  Can you be more specific? --JereKrischel 03:13, 19 January 2007 (UTC)

They're not all in the images. The ten are in the book. --03:21, 19 January 2007 (UTC)

To get back on track, the criticism of IQatWoN says "unrepresentative", but about *current* national IQ not that the studies are unrepresentative of the groups they actually test. --W. D. Hamilton 03:22, 19 January 2007 (UTC)


 * The criticism of IQatWoN data does include the assertion that the studies are unrepresentative. Their scheme is to take the British Ravens IQ in 1979 as 100, and simply add or subtract 2 or 3 to the scores from other countries for each decade that the relevant date of test departs from that year. The assumptions of size, linearity and universal applicability of this correction across all countries are, of course, hugely questionable if not breathtaking.  I think you're misreading the Heredity article. --JereKrischel 03:28, 19 January 2007 (UTC)


 * That's a criticism about the Flynn effect correction, not the representativeness. That is a separate criticism, which could be included. It basically comes down to a criticism about a each study varing a few points one way or the other. It applies to the world-wide scores, where the Flynn effect may not be as well measured (I don't really know if that's true) as in developed countries. It clearly does not apply to the scores from developed countries, where the Flynn effect is well charted. --W. D. Hamilton 03:33, 19 January 2007 (UTC)


 * A criticism of whether or not the data is accurate is equivalent to a criticism of the representative nature of the data. Does the data "represent" the truth?  The review clearly asserts that it does not. --JereKrischel 03:52, 19 January 2007 (UTC)


 * "representative" in the context of this critique means that the sample is "representative" of the overall (national) population. all of the details point that as being the meaning of "representative" here. reliability or accuracy are completely different kinds of criticisms than being "representative". the flynn effect criticism is about the values being directly comparable, which is close to "reliable" but in a very specific way. --W. D. Hamilton 06:35, 19 January 2007 (UTC)


 * I'm not sure where you're getting "reliable" from. The article said, But, even for most of the others, 'direct evidence' is putting it strongly, as even a cursory glance at the motley tests, dates, ages, unrepresentative samples, estimates, and corrections show. They only use the word "reliable" when talking about GDP: With the measures of GDP, L&V admit that estimates may be 'highly unreliable for developing countries' (p 83).  --JereKrischel 21:14, 19 January 2007 (UTC)


 * the point is that the representativeness question is specific to iqatwon, per below. --W. D. Hamilton 07:07, 20 January 2007 (UTC)

Map and data from a non acadmic trade book from a very dubious publisher


The section "world wide scores" contains material from a non acadmic trade book from a very dubious publisher, criticzed for publishing anti-semitic literature. The map should certainly be removed and a disclaimer regarding the publisher and the non academic status must be included.Ultramarine 22:21, 11 January 2007 (UTC)


 * The NY Times, WSJ, Penguin books, and others are not an academic press, nor are most publishers of book. The publisher itself is of absolutely no importance to the content of the book; the academic press / trade book distinction is irrelevant. The author is of course noteworthy, and the author is noted. In the case of this particular book, it is clearly a follow-on to Lynn's earlier paper by the same title and his previous book. No argument has been given about removal of the map other than the publisher. --W. D. Hamilton 22:41, 11 January 2007 (UTC)
 * No, it is not. University Press is quality mark which is why researchers publish their serious works there. Anyone can publish a private trade and make any claims.Ultramarine 22:47, 11 January 2007 (UTC)


 * That's simply beyond the scope of what WP demands from sources, and moreover is defeated by the fact that the book has been reviewed twice in peer-reviewed journals. This indicates that the authors of these journals take book as a serious scholarly work. Your opinions and those of search light magazine regarding the publisher of the book are not important to WP policy in light of that fact. --W. D. Hamilton 22:55, 11 January 2007 (UTC)
 * Sorry, self-published sources are not allowed in Wikipedia. Many strange books like Mein Kampf has been reviewed in academic journals. Why did the author not use an academic publisher? The argument that he is persectued by everyone does not hold since the previous book was published by an academic publisher.Ultramarine 23:27, 11 January 2007 (UTC)
 * The book is treated as a scholarly work by other scholars writing in the peer reviewed literature. This is the crucial and *sufficient* matter of fact on the issue. Arguments about the reputation of the book's publisher are thus moot. The suggestion that this isn't a scholarly work because it isn't published by an academic press are not supported by the treatment this book has received by other scholars. Publication by an academic press is not a necessary condition on scholarly work, as evidenced by the many scholarly works that are "trade books" (e.g. Free Press (publisher)). --W. D. Hamilton 23:44, 11 January 2007 (UTC)
 * If you want to add something to the article to add support for the book and its claims, do so. However, again, we should certainly state the opposing views also. Wikipedia does differentiate sources on reliability, as do scholars in general. Anyone can self-publish something, which is not possible with University Press. What scholarly (not popular, having references and using statistical research) significant works are you thinking of that are self-published?Ultramarine 08:20, 12 January 2007 (UTC)
 * I don't understand the label self-published, but I assume you mean "trade book". Just look at the various works by scholars that are published by commercial (trade) publishers rather than academic imprints. Large reviews of subjects, because they are of broader interest, often get published by non-academic imprints because they are money-makers. The Bell Curve and The Blank Slate come to mind. I can't say whether Lynn was hoping to make more money or whether he's so controversial that academic imprints couldn't afford to have a "g Factor" incident where they print a book and then pull it off the shelves. In any case, you can't exclude Lynn's book and his findings from discussion on the basis of attacking the publisher. Outside of that criticism, I don't see any reason why the map is inappropriate. --W. D. Hamilton 18:08, 12 January 2007 (UTC)
 * But you are proving my point, the Bell Curve is just a summary of prior research for the popular reader. Obviously it is possible to exclude and qualify sources due to its nature, Wikipedis does this all the time when considering reliability. Unreliable sources are simply excluded from Wikipedia without any demand that every claim in them should be contradicted in detail. However, we can cerainly mention the book if we include a disclaimer regarding its reliability.Ultramarine 18:17, 12 January 2007 (UTC)
 * Regarding the map, it could be included if we have the disclaimer in the text.Ultramarine 18:22, 12 January 2007 (UTC)

[edit conflict]
 * Point out the opposing view about the rank order of IQ scores from different regions (all that's visible in the map). What comparable data source exists that can make claims about world-wide IQ that we could contrast with Lynn's rankings? Is there a good reason not to point out that there are four major clusterings of IQ scores worldwide and show how they are arranged on a map? I suspect what you mean is concern about the lowest scores. However, the map doesn't make any scores directly visible. Whether average African IQ is 70 or 80, it's lower than N. African IQ, which is what we're trying to convey. --W. D. Hamilton 18:24, 12 January 2007 (UTC)

[edit conflict] you could write the appropriate "disclaimer" regarding the dispute about the reliability of low test scores, but keep in mind what is and isn't visible in the map. --W. D. Hamilton 18:24, 12 January 2007 (UTC)
 * How about the current version? Ultramarine 18:36, 12 January 2007 (UTC)
 * Basically good. I see some other problems not related to your edits that I'm going to flag. --W. D. Hamilton 19:27, 12 January 2007 (UTC)

Map is totally violating WP:NOR. If such an image exists in Lynn's book, scan it and use it as fair use. Independent generation and compilation "based on Lynn's values" is clearly an inappropriate synthesis. --JereKrischel 08:50, 14 January 2007 (UTC)


 * Also removed OR image of bell curves and IQ by race - without citation of data, definitely violates WP:NOR. Suggest someone find an existing image presented in some research, scan it, and use it as fair use. --JereKrischel 08:56, 14 January 2007 (UTC)


 * I agree, remove it. futurebird 00:10, 17 January 2007 (UTC)

Is Gottfredson 2004 an unpublished source?
futurebird 15:36, 21 January 2007 (UTC)
 * This is the source for the chart: Gottfredson 2004 Social Consequences of Group Differences in Cognitive Ability . The charts seem to be from page 43. Please take a look at this PDF.
 * I think that the caption of the image should be expanded, as shown to the right, or else the image needs to go.
 * I don't know where the curves for 'Asian' and 'Hispanic' came from, they are not in the charts in Gottfredson 2004. Are they original research?
 * And, please tell me in which academic journal was Gottfredson 2004 published?
 * Why in the archives does it say these data are from 1981? That makes it 25 years old!

Futurebird, you are quite confused. Gottfredson's data is not from IQatWON - I can only assume you're filling this in by inference from the fact that IQatWoN has been a popular topic of conversation. The Black and White curves Gottfredson uses in similar graphs are from a 1981 standardization of the adult IQ test called the WAIS. As the note shows, the gap they show are about right for college-age population. She has published this kind of graph half a dozen times or so in various journals. Gottfredson 2004 is just the most easily accessible source for the four groups. The graph is a probability distribution function. The y-axis is not meaningful to most people, but it could be said to be % of the population, however the scale is vary small as you have to integrate over the entire range to get a value of 1 for such a graph. --W. D. Hamilton 18:48, 21 January 2007 (UTC)


 * I'm looking at the PDF (It has no footnotes!) and I can't tell where she got the data. Saying it was WAIS isn't an answer to my question, that just tells us what kind of test was used.

"The racial IQ gaps in the United States, portrayed in Figure 3, will be used to illustrate the social processes and controversies that such differences create within a multiracial nation. The two bell curves in the upper part of the figure, which represent blacks and whites, make  several important points. First, both groups produce a bell curve that covers the full range of  what is called normal intelligence (IQ 70-130). Second, the major difference between the two  bell curves is that the black distribution is centered about 15 points lower on the IQ continuum  than the white distribution when measured by the most widely-used test of adult intelligence, the  Weschler Adult Intelligence Scale (WAIS). Third, the two bell curves overlap a lot, so it is not  possible to know anyone’s IQ by knowing their race.     --   Insert Figure 3   --"


 * Figure 3 is the graph this graphic to the right is meant to match. It only shows two bell curves, not 4, though the data are given at the bottom of the page. There is no mention in this link of the source for this data. Who took these IQ tests, who picked the people to take the test... etc. How did they identify a person's race?


 * Also, could you please tell me the name of the journal where this was printed?

futurebird 19:33, 21 January 2007 (UTC)


 * The 1981 WAIS standardization is a single testing event that was published ca 1987. This article is in a book whose title is incomprehensible to me. But all of that is irrelevant. --19:35, 21 January 2007 (UTC)


 * So, we don't know if this paper has been published? We'll have to assume it has not been published making it an invalid source. Thanks for the informtion on the test. I'm looking in to it now. futurebird 20:45, 21 January 2007 (UTC)


 * It's listed as published in 2006 on Gottfredson's CV. It was first posted in 2004. --W. D. Hamilton 20:51, 21 January 2007 (UTC)


 * But, without the name of the journal, there is no way to verify this. futurebird 21:02, 21 January 2007 (UTC)


 * Click the links. I've added several other instances of Gottfredson publishing the same figure to the legend text. --W. D. Hamilton 21:06, 21 January 2007 (UTC)

Thank you for the other sources, I'm looking in to them now.

Sources cited for this graphic: 

I pasted the sources here because they vanished from the caption as I was reading them. None of them seem to be from peer-reviewed main stream journals, they seem to be from books, I'll need to look in to the books-- I wish there was a way to see the footnotes. And there is still no journal for gottfredson 2004. futurebird 21:56, 21 January 2007 (UTC)

Data in graph are 26 years old
Why has this simple fact which relates to the credibility of the data been removed again? futurebird 20:45, 21 January 2007 (UTC)


 * I don't know why it was removed, but it should be phrased as a date, because what's 26 years now is 27 years next year. One problem is that the size of the gap shown is merely chosen by Gottfredson to correspond to the most common view (for adults). --W. D. Hamilton 20:49, 21 January 2007 (UTC)

The bell curves are inaccurate in this graphic.
Thanks for the sources. It's a big help.


 * According to the caption on page 26 of the "Comprehensive Handbook of Multicultural School Psychology" one if the three sources listed for this graphic the standard deviation for the various bell curves is not the same. It is 13.0 for blacks and 14.7 for whites. O opened this image in photoshop and over-layed the curves. They are all identical. The curve for whites should be more spread out.


 * Adding bell curves for other "Asian" and "Hispanic" seems to be original research since the graph in these sources only has curves for black and white. The numbers are given, but doing a normal regression on these numbers amounts to research. However, since the bell curves are all identical, I doubt a normal regression was even done when this graph was created.


 * This style of graph was used in conjunction with labels such as "Clerk, teller, police officer." or "Attorney, chemist, executive." and other professions. It is meant to be used to show the relation between these labels, IQ, and social pressures Without them the graph becomes even more meaningless since there is no scale for the y-axis. One might assume it was based on population or some such thing. They ought to be replaced or the caption should reflect the original intent of the graphic.futurebird 21:44, 21 January 2007 (UTC)

The most informative figure legend is found in -- "Relative representation of American Whites, Blacks, Hispanics, and Asians at different points along the IQ continuum". The figures are built on the underlying data table shown in the same graph. Gottfredson's intention is clear, as you cited above -- ''The two bell curves in the upper part of the figure, which represent blacks and whites, make several important points. First, both groups produce a bell curve that covers the full range of what is called normal intelligence (IQ 70-130). Second, the major difference between the two bell curves is that the black distribution is centered about 15 points lower on the IQ continuum than the white distribution when measured by the most widely-used test of adult intelligence, the Weschler Adult Intelligence Scale (WAIS). Third, the two bell curves overlap a lot, so it is not possible to know anyone’s IQ by knowing their race.'' These are the same points being made in our graph, only we show the curves that Gottfredson omits using her data table to build the missing curves. Your concerns about height/SD are misplaced. The best way to represent the points Gottfredson is making is to use the style chosen by Gottfredson. --W. D. Hamilton 22:13, 21 January 2007 (UTC)


 * Whoa, trying to assert Gottredson's "intention" smacks of OR. Removed image, now that it is clear that you've completely constructed a novel synthesis based on multiple sources. --JereKrischel 10:26, 22 January 2007 (UTC)


 * You couldn't have actually looked at the papers. The graphic is the same in each of them. Gottfredson's "intention" (i.e. what she's arguing in the graphic) is what she says the graphic means, copied from one of the three papers. There's nothing in NOR that supports the removal of material that is so strongly sourced in this manner. I will escalate this if need be. --W. D. Hamilton 18:34, 22 January 2007 (UTC)


 * I did look at the paper you cited. Whether or not her "intention" was to put four curves side by side or not is purely speculative.  The graph removed is NOT even remotely the same as the graph Gottfredson has in her paper.  Please escalate at your convenience, I believe we need input from other editors at this point. --JereKrischel 19:53, 22 January 2007 (UTC)


 * (1) You wrote previously clear that you've completely constructed a novel synthesis based on multiple sources. So you are no longer making this claim? (2) Whether or not her "intention" was to put four curves side by side or not is purely speculative. - Not at all what I listed as her intention. Her intention was to demonstrate the three points she lists in the quoted text. Thus our figure legend makes the same point. The addition of a criticism sentence is the only thing that doesn't come from Gottfredson. (3) The graph removed is NOT even remotely the same as the graph Gottfredson has in her paper. - It's nearly identical. The black and white curves are copied verbatim and the asian and hispanic curves are added to Gottfredson's specifications. I believe we need input from other editors at this point. - If the only claim is that the graph isn't supported by the source, then we'll soon have that answer from the deletion debate. --W. D. Hamilton


 * That's not the only claim.futurebird 20:14, 22 January 2007 (UTC)


 * 1) You've clearly cited the multiple sources you've used. Can you give us a single source where we can find a graph with all four curves drawn as you've done?  2) You cannot reliably assert you represent Gottfredson's intentions.  3) You have taken Gottfredson's curves out of the context of the graph she presented, and then added two additional curves - a clear example of OR.


 * To reiterate, the claims are -> you've altered the graph presented by Gottfredson in a novel and original manner, and you've used multiple sources to synthesize your graph. If you can provide a citation to a graph which has the four curves, laid out in the same manner as you have, you can mollify my concerns. --JereKrischel 20:37, 22 January 2007 (UTC)


 * Gottfredson's graph, which is published in identical form in at least three different papers, presents two curves (Black and White) but also a data table which gives the distributions for Asians and Hispanics -- all in the same "figure". The image in WP has Gottfredson's black and white curves with identically rendered curves for Asians and Hispanics based on the data Gottfredson reports in her figure -- they are laid out the same as the source. The discussion of "intention" appears to be distracting you, and it is not that important. --W. D. Hamilton 21:34, 22 January 2007 (UTC)


 * Again, deciding to combine data from a figure that is being presented in two very separate ways is an inappropriate synthesis. We cannot know what Gottfredson's intention was, or why she decided to graph two of the curves, and not graph the others.  Synthesizing a new graph from her graph and subsequent data tables simply isn't appropriate.  Especially if Gottfredson's graph is published in identical form in three different papers, it behooves us not to change it in some novel way. --JereKrischel 23:47, 22 January 2007 (UTC)

Could this resolve the dilema?
-- —Preceding unsigned comment added by Kevin Murray (talk • contribs) 21:41, 22 January 2007


 * I would support this as an interim solution until consensus can be determined on the old image. If there are no objections, I will replace the image in the article with this one, but leave the page protected. --causa sui talk 21:47, 22 January 2007 (UTC)
 * I don't object. The other image (4 curves) was an attempt to circumvent certain objections to a graph like this one (black and white). --W. D. Hamilton 21:56, 22 January 2007 (UTC)


 * Agree This image should be noted as derived directly from Gottfredson, and detailed argument should be avoided in the caption. --JereKrischel 23:43, 22 January 2007 (UTC)

This is better because
 * At least the curves match the essay.
 * At least you're not mixing data sets.

However futurebird 23:54, 22 January 2007 (UTC)
 * It still has no y-axis, could we please add one?
 * Caption must note the date for the data.
 * Should not be given undue prominence in the article or used as a logo.

I think we have access to more recent data than this. Since we're drawing the curves anyway, can't we find more recent means and SDs to use for this graph? Or would that be original research? futurebird 00:30, 23 January 2007 (UTC)

If someone wants to give me the data for the y-axis, I can include it. If there are new data I can adapt to those as well. --Kevin Murray 00:38, 23 January 2007 (UTC)
 * See the deletion discussion for the old image. There are more recent data mentioned here: .Ultramarine 01:17, 23 January 2007 (UTC)
 * All of that is summarized here: Race_and_intelligence. Flynn and Dickens finds an adult BW gap of ~1.1 sd, but a smaller gap for children. This is apparently not new, but no one really thought much of it before Flynn -- but as Flynn points out Jensen recognized it previously. What is new in Flynn's report is that the gap found among children is smaller than previous (i.e. 60s/70s) reports. --W. D. Hamilton 01:31, 23 January 2007 (UTC)
 * IIRC, the newest adult IQ data in Dickens and Flynn's analysis is the 1997 AFQT, which I believe shows exactly the same gap as the 1981 WISC-R as currently drawn. --W. D. Hamilton 02:16, 23 January 2007 (UTC)
 * Actually, the WJ-III, which they discard for some reason was 2001, but it showed a 1.05 sd gap for adults and a 0.99 SD gap for children. --W. D. Hamilton 02:19, 23 January 2007 (UTC)

Problem with just a BW graph
(1) Previously argued to be more 'jarring' than a graph that shows a diversity of groups. (2) Although there are surely fewer studies over less time for Hispanic and Asian groups, there's plenty of contemporary data. Roth et al 2001 do a meta-analysis on Hispanics. The number for East Asians appears slippery because they seem to be getting relatively higher all the time, but they're somewhere above 100. --W. D. Hamilton 01:31, 23 January 2007 (UTC)


 * Most of the debate has been about blacks and whites in the US. Having the "jarring" graphic is a part of that history. If you add historical context it will not be such a big problem.futurebird 01:36, 23 January 2007 (UTC)
 * In the 70s and 80s it was. In the 90s and 00s, the scope has expanded, coincident with the population expansion of Asian and Hispanic minorities in the U.S. --W. D. Hamilton 02:16, 23 January 2007 (UTC)

From a recent NYTimes article: --W. D. Hamilton 03:03, 24 January 2007 (UTC)
 * In California, the rise of the Asian campus, of the strict meritocracy, has come at the expense of historically underrepresented blacks and Hispanics. This year, in a class of 4809, there are only 100 black freshmen at the University of California at Los Angeles — the lowest number in 33 years. At Berkeley, 3.6 percent of freshmen are black, barely half the statewide proportion. (In 1997, just before the full force of Proposition 209 went into effect, the proportion of black freshmen matched the state population, 7 percent.) The percentage of Hispanic freshmen at Berkeley (11 percent) is not even a third of the state proportion (35 percent). White freshmen (29 percent) are also below the state average (44 percent).

Caption
Since saying "the data are 26 years old" requires constant updating, how about something like "18 years after The March on Washington" Since the data are all from the US, this should be OK. It would also help to add some much needed historical context and sensitivity to this article. The caption should, of course, note that this isn't international.futurebird 01:04, 23 January 2007 (UTC)

Graph is US-centric
Two-curve graph from Gottfredson should not lead the article, unless the article is only about the United States. Better in a US section. Jokestress 17:54, 23 January 2007 (UTC)

"Countering systemic bias" is a good idea when editors are the source of that bias, but in this case the "systemic bias" exists within the literature on this topic. The U.S. isn't just a section, it's the whole article with only one subsection expanding beyond the U.S./developed world. That's on par with what's written in the major works on this subject. --W. D. Hamilton 18:01, 23 January 2007 (UTC)


 * Perhaps then, it should be renamed to Race and intelligence (United States). --JereKrischel 05:02, 24 January 2007 (UTC)


 * I like this idea better than calling it 'groups and intelligence' though, a fair ammount of this relates to the UK too. So I don't think it'd work. futurebird 00:23, 25 January 2007 (UTC)

IQ by race table
The "IQ by race" table should be removed from this article. It is redundant to the "Average gaps among races" subsection/subarticle (where it would fit in the scheme of the article). I might also argue that building such a table from multiple sources is a task beyond the scope of a WP article -- it has already been done in a fashion by Lynn's Race Differences in Intelligence, which is a better substitute. --W. D. Hamilton 20:47, 21 January 2007 (UTC)

Your opinion Hamilton! Please do not remove this material again unless you can build a consensus. The table has been tagged to stimulate further discussion and research --Kevin Murray 21:41, 21 January 2007 (UTC)

I have not been happy with the graphic at the beginning of the article. I belive that the "race" categories are overly aggregated, and that hispanic is a false race classification. I would prefer to see a table with even more precision than offered by what I brought in today from another WP article. --Kevin Murray 21:52, 21 January 2007 (UTC)

Perhaps the chart should be reorganized to be sorted by source rather than by IQ statistic.

Considering the validity of the data in the chart
 * Is this source credible? http://iqcomparisonsite.com/NationalIQs.aspx  If so, should we braoden the scope of the table to include all of the data at the site.  If not, can the information be verified at the source cited by the site?  --Kevin Murray 22:47, 21 January 2007 (UTC)

The table at Race Differences in Intelligence, suggested by Wdhamilton is repeated below. The statistics are quite different in some categories. Perhaps the data should be included in the table here, and the chart reordered (as I suggested above) by study rather than by IQ.

The IQ by race table in this article is identical to that contained in another article, IQ by ancestry, which has been nominated for deletion. The overwhelming consensus at the AfD discussion has been for Deletion, not merger. This table is 100% original research and unverified claims, and if you want further justification for that look at the AfD discussion for the article linked above. I would suggest you take it and completely revamp it if you want it kept, but as it stands this section is a complete violation of WP:OR. As such I am removing it. Don't bother claiming lack of consensus again, all the consensus necessary is in the AfD.-- Dycedarg &#x0436; 23:15, 21 January 2007 (UTC)
 * Nonsense, the discussion at the other article is not about a merger or relocation of the information. That article is being deleted as not notable on its own strength.  The discussion of merger is not relevant to the AfD.  This will remain here until a consensus otherwise is reached.--Kevin Murray 01:32, 22 January 2007 (UTC)
 * How can you claim that this is original research when the table cites the sources of the infromation? --Kevin Murray 01:33, 22 January 2007 (UTC)

(1) The RDI table given above is only a subset of the studies in the book Race Differences in Intelligence. Scores for minority groups living in the developed world are also part of the book. See the various book reviews cited in the RDI article, especially this one.

(2) If we preserved the existing structure of the article, the table's content belongs in "Average gaps among races" subsection/subarticle. As a WP:Summary style article, there is no room for such an extensive review in the main article (see 3)

(3) To the extent that this exercise has already been performed by Lynn in Race Differences in Intelligence, we should prefer his work to our own substitute unless there is a particular reason to do otherwise. Lynn's work takes a 334 page book, and so it's unlikely that we could substitute. --W. D. Hamilton 23:18, 21 January 2007 (UTC)
 * Clearly we should not be replicating the work of scientists; we should only be presenting an encyclopedic summary. In this case since there appears to be no consensus among the "experts" it seems that we need to report the prevailing divergent views.  I think that this table could be modified to fit that need. --Kevin Murray 01:44, 22 January 2007 (UTC)
 * Did you even look at the AfD discussion? You should, because it is most certainly relevant. It is not being considered for deletion because of a lack of notability, it is being considered because it is original research. Something that is original research when it is on its own in an article does not magically cease to be original research when you put it into another article. It is original research because for one thing, it is not properly sourced, but for another thing, taking the results of different studies on this topic and comparing them is original research. As I said in the AfD: "It draws the IQ's from a number of different sources. Each source almost certainly varies widely with regards to what the sampling methods and IQ tests it uses were, which means that the results of each source can not be compared in this manner without inherent inaccuracy. A list like this does nothing but result in completely erroneous and misleading conclusions and is in fact utterly useless." Comparing the results of these studies on your own is inherently original research, and will never cease to be unless you either draw all the IQ data from one source or rewrite the format of the table completely. THAT is the consensus on the AfD, and YOU are the one ignoring it. How can you claim that a consensus that has been reached is suddenly not applicable just because it happens to be on another page? Either rewrite the table or leave it gone. In its present form it is absolutely not valid for inclusion. -- Dycedarg &#x0436; 02:14, 22 January 2007 (UTC)

'''Stop removing the table or I will report it as VANDALISM" --Kevin Murray 02:37, 22 January 2007 (UTC)
 * I will agree with you that the table should be rewritten, but I'd like to see some consensus first on what it should contain --Kevin Murray 02:37, 22 January 2007 (UTC)
 * I support the deletion of the other article, but there is broad support for redirect to this article. Since redirect is not specifically the topic for discussion at the AfD, it is not a valid "vote" forum on the proper inclusion here.  This is the forum at which that discussion should be held.  --Kevin Murray 02:37, 22 January 2007 (UTC)

I don't think there was any consensus on if the table should be kept. I don't support it being in this article. It is a patchwork of original research.futurebird 02:46, 22 January 2007 (UTC)
 * As I said above (1) I'm not happy with the graphic which is now at the beginning of this article which is overly aggregated. (2) I'm not happy with the table as is; it needs better referencing to the sources (3) this hype about original research is nonsense -- this is merely a table presenting the findings from the research of others AKA: secondary research. Primary research would be administering the IQ tests. --Kevin Murray 03:16, 22 January 2007 (UTC)

With the permission of the contributors here, I will follow Dycedarg's suggestion and modify the table. I would feel hypocritical if I started changing it dramatically without some consensus. I wouldn't object to removing it while I work on it. My suggested revisions would be:
 * (1) Include the results of more studies
 * (2) Change the order of the data to be alphbetical first by study/source and then by group to demphasize the IQ order
 * (3) Provide standard footnotes to the source material
 * --Kevin Murray 03:16, 22 January 2007 (UTC)

I removed it again. As you just stated that you wouldn't mind removing it while you work on it, I trust you won't object this time. By the way, please don't make stupid threats in a debate like this. What I was doing wasn't vandalism in any sense of the word, and any admin worth his salt would have seen that immediately. I don't think I would object to the table if it is arranged in the manner you just suggested, but make sure all the sources are proper and valid. Some of the sources for it at the moment are neither.-- Dycedarg &#x0436; 03:26, 22 January 2007 (UTC)
 * Under the circumstances I agree. I'll pass on any further debate about my stupidity and your vandalism --call it a truce for a better purpose. Cheers! --Kevin Murray 03:54, 22 January 2007 (UTC)

I'll rephrase my criticisms. (1) The table is not appropriate for the main article, but if done in a fashion conforming to policy, it could complement the discussion of world wide scores in the "average gaps" section. (2) The topic of world-wide scores is a very small part of the overall topic (the scholarly literature is the standard against which we make that judgment). (3) The table would not be an appropriate substitute for a that shows the gaps, overlaps, and ranges of the major U.S. ethnic groups (where the vast majority of the literature focuses). --W. D. Hamilton 03:39, 22 January 2007 (UTC)
 * The lead image carries a bold statement, but does it offend more than educate? Let's see whether I can put together something that will be meaningful.
 * Perhaps the main article is getting too long and should be divided.

--Kevin Murray 03:54, 22 January 2007 (UTC)

The table under discussion has been modified to (1) segregate the data into the various studies, (2) remove a ranking system, and (3) provides specific sources for the data. It has been inserted here. It is disputed whether this should be in this article or as a complementary article. Right now the concept of a standalone artice and the "lead image" for this article are both being discussed in the deletion process. --Kevin Murray 10:29, 22 January 2007 (UTC)
 * Regardless, this an original research synthesis of various studies and certainly not complete. Also, not needed since there are separate are already separate articles for Lynn's books.Ultramarine 15:47, 22 January 2007 (UTC)
 * Nonsense, this has been carefully discussed and modified to remove the hint of original research. Please leave this in place so that more editors can discuss the material.  The original research claim came from an objection to having the data from multiple sources ranked together from highest to lowest IQ, and without specific references.  Many items were removed and the studies have been kept together with separate headers.  I would not be opposed to other modifications; however, the data is pertinent and should be displayed in one form or another as an overview/summary of this subject.  --Kevin Murray 16:41, 22 January 2007 (UTC)
 * If we should present them here instead of in the appropriate subarticles, then NPOV also requries inclusion of concerns regarding the reliabiility of these sources.Ultramarine 18:10, 22 January 2007 (UTC)

The article where the table would be appropriate, if it is appropriate at all, would be Race and intelligence (Average gaps among races). --W. D. Hamilton 03:06, 24 January 2007 (UTC)

Large scale delettions
An anonymous user just deleted parts of the article, also making an incorrect statement, since the additions had been carefully explained. Obviously the article is not NPOV.Ultramarine 18:36, 22 January 2007 (UTC)
 * Did you warn said user? I thought it seemed like vandalism. futurebird 20:10, 22 January 2007 (UTC)

This page is protected
I've protected this page from all edits and moves by non-admin users per the Protection policy due to a protracted edit war. Please resolve disagreements about the content of the article on the talk page before reverting. --causa sui talk 21:39, 22 January 2007 (UTC)

Why?
If it is in fact proven that people of African descent are the dumbest people on the planet and in history, what purpose will it serve? Will the elitists that control the globe then switch focus on what role these people should serve to humanity? This stuff will be used by extremist to subjugate the lower group all over again. This Bell Curve does have one noticeable effect on the people who are in the higher group - a dramatic increase in CQ or Conceit Quotient. This will boost the confidence and self esteem of those on the highest rung and push them to achieve even more while smashing the self-worth of those on the lowest since it is something that they supposedly have no control over. Apathy 05:34, 23 January 2007 (UTC)


 * If you have some sources for these ideas you will be my hero... FOREVER. Seriously, I think we all know this. That is one of the reasons it is important work on making these articles as objective and balanced as possible. I hope you plan to stay around and help out. Up until very recently "respectable scientific sources" endorsed racist ideas as scientific fact. These views are rapidly growing out of favor, but they die hard, as you can see. It's our job to place the scientific ideas in historical context and explain them as neutrally as possible. Are you up to that task? I hope so, it's wearing me out!
 * And to answer your question? If it is proven than black people are dumb than we don't need to care about what happens to them. We don't need to acknowledge the impacts of colonialism or slavery or racism. In fact we can continue to segregate and degrade people without any guilt. Won't that be great?futurebird 05:57, 23 January 2007 (UTC)
 * I happen not to believe that black people are dumb (or any dumber than I am, and I am not black). That said, let's for the sake of argument accept the premise, "If it is proven than black people are dumb ..."  This historically may very well have indeed been used by people in power to justify the belief that "we don't need to care about what happens to them. We don't need to acknowledge the impacts of colonialism or slavery or racism."  But, I want to point out, that this is not the logical consequence of the claim.  Let's look at analogous cases.  Say that a certain population suffers from a higher incidence of diabetes or sickel-cell anemia.  We do not just let those people die.  Instead, we ask doctors to monitor members of the population carefully to diagnose the disease as soon as possible, and we set scientists to work on treatments to manage or even cure (we can hope) the diseases.  Another analogous case: I cannot see well, for genetic reasons.  But I was not taken out of school when I was eight, and left to care for myself or put to work at some task that does not require reading or careful vision.  No.  Humanity long ago had invented eye-glasses, and I was prescribed and provided with eye-glasses.  These glasses do not change my genes nor my biological limitation (still can't read without them, or recognize people's faces), but they fully compensate for the defect.  If people of a certain population are genetically predisposed to certain cognitive or learning deficits, we should not automatically assume that these cannot be compensated for, perhaps through different pedagogies or educational environments, or just more intensive instruction.  Look, even a dyslexic can become president of the United States.  My point is this: if some people are treated as inferiors and not given the opportunities others are, it is generally not because they were born with some genetic disadvantage.  It is on the contrary because they are thought of as inferiors that they are not given the same education, training, or technological assistance necessary and available that helps people born nto socially advantaged groups to manage or overcome their genetic disadvantages. My pioint is we should not accept the following reasoning as necessary or logical: "If it is proven than black people are dumb than we don't need to care about what happens to them. We don't need to acknowledge the impacts of colonialism or slavery or racism."  The very logic of this statement - the relationship between the premise and the conclusion - is the logic of racism. We need to be clear that the premise does not naturally lead to the conclusion.  Slrubenstein   |  Talk 12:59, 23 January 2007 (UTC)


 * I'm sorry for the sarcasm. (It never reads well online.) I'm simply find it hard to ignore the fact that some of these "objective" sources of sciences have ties to white supremacist groups. It makes it hard to view their research as neutral. I think that we should not avert our eyes from potentially harsh truth, be we also should pay careful attention to the source of the truth and ideological aims of their sponsors. This article needs more mainstream sources.


 * We need to be able to explicitly define what it is that we are talking about. What do we mean by "race" "black" "white" is the definition clear enough for scientists to use? And what do we mean by intelligence? Unfortunately it seems that there is no consensus on the first idea and a heated debate raging about the second.


 * The intro to this page is very good in laying out these questions, but after that it breaks down. We also need to place ideas about race and intelligence in historical context. It's not like this is pure science with pure motives. Sometimes, the very questions you choose to ask have the effect of setting an agenda. Ignoring that is diverting our eyes from 'the potentially harsh truth' as well. futurebird 13:14, 23 January 2007 (UTC)


 * Look, with all due sympathy to the emotions expressed here, can I ask you both to get on topic? There are places to argue consequences - this is not one of them. The article should simply report the data. Arker 13:04, 23 January 2007 (UTC)


 * We need to talk about the way to revise this article so it is more sensitive and less one-sided. Part of that is exploring these issues. futurebird 13:14, 23 January 2007 (UTC)


 * There is no reason for this article to "simply report the data" when the assumptions behind the collection of the data and the interpretation of the data is controversial. In such cases our NPOV policy demands not only that we provide multiple points of view but also provide the context for each point of view, and provide an account of the controversy. Slrubenstein   |  Talk 13:26, 23 January 2007 (UTC)


 * I agree. Historical context is essential. futurebird 19:17, 23 January 2007 (UTC)

restore the racism subsection
it was removed by JK and had discussions about what various scholars thought about racism in relation to this subject. Slrubenstein's POV that the assumptions behind the research are racist was described and debated in that section. (yes, i know the whole last section needs work.) --W. D. Hamilton 18:04, 23 January 2007 (UTC)


 * Can you point me to a revision where I can look at the subsection? futurebird 19:09, 23 January 2007 (UTC)


 * Oh and thank you for making this suggestion. I feel like we've been butting heads for far too long over this topic. futurebird 19:17, 23 January 2007 (UTC)

All of the material should be spread amongst: A reminder. NOR and NPOV place great constraints on what is appropriate and not appropriate for inclusion in WP articles. The safest way to write for controversial topics is to find 2ndary sources that describe the situation, and to relate what they describe as the salient points. In the thread above, you and several other editors say that this article requires historical background and discussion. However, this would be very difficult to accomplished in accord with NOR by consulting primary sources or using 2ndary sources that are several steps removed. You would need a book or article that says what the important historical background for race and intelligence is, and report what they say is the important background. --W. D. Hamilton 03:14, 24 January 2007 (UTC)
 * Race and intelligence (Public controversy)
 * Race and intelligence (Accusations of bias)
 * Race and intelligence (Utility of research)
 * Race and intelligence (Media portrayal)


 * WD, forgive me, but I think you still need to address the issue of balance. You may in fact do a NPOV job of citing literature, but when all you do is cite one POV, the article becomes POV pushing.  If you would do us the favor of finding just as much material contrary to your POV, and adding it in a neutral manner, as you do with the material that supports your POV, this article would be much better off.  You seem to be very well connected to the pro-hereditarian movement, and continually cite some of the latest research from prominent Pioneer Fund folk to bolster your case.  If you could spend even a fraction of that energy reporting on the latest research that repudiates the pro-hereditarian POV, you'd do us all a great service. --JereKrischel 05:01, 24 January 2007 (UTC)

the ruin of this article is not following policy
if we don't treat the multi-author collective statements from the APA and WSJ and the S&R survey data in relation to this article they way the IPCC is treated in relation to the article on global warming, then there's little if any hope for this article. if straightforward claims like 'the BW gap is ~1SD' are removed despite citation to these sources (and despite quotations that directly support the claims), and if this principle is applied further, then there will be no content left. If the major views of the cause of group differences are being rewritten from "partly genetic" and "environment only" to "primarily genetic" and "primarily environment" then the factual accuracy of this article will likewise degrade to a worthless state. --W. D. Hamilton 17:18, 24 January 2007 (UTC)


 * As per the APA statement, this article should be titled Groups and intelligence (United States). I find the reliance on only three "multi-author collective statements" to be misguided in general though. I think you are being hypocritical in your selective view of these consensus statements. The Genetic Hypothesis. It is sometimes suggested that the Black/ White differential in psychometric intelligence is partly due to genetic differences (Jensen, 1972). There is not much direct evidence on this point, but what little there is fails to support the genetic hypothesis. and These groups (we avoid the term "race") are defined and self-defined by social conventions based on ethnic origin as well as on observable physical characteristics such as skin color. None of them are internally homogeneous. , both illustrate a view you seem to be determined to eliminate and avoid in the article.  --JereKrischel 20:01, 24 January 2007 (UTC)


 * Two issues (1) the importance off these souces and (2) your charges aimed at me.
 * First, (2) -- this is false charge. The views you are describing are found in the article along with their counter-views. I am not trying to suppress them at all. Indeed, I was the first to add these arguments based on reading the APA report. However, this is a distraction from (1), and fails to explain the examples of removal and mischaracterization of material from these sources as I've described above. Ironically, your removal of massive amounts of material from the explanations section does the most to eliminate the detailed description of these arguments.
 * Second, (1) -- you have offered no reasoning as to why we should not treat these multi-author sources as the equivalent of the IPCC report. They clearly differ in the way they present the relative popularity of the partly-genetic and environment-only explanations, but this is a trivial problem for us as that is recognizably a point where expert opinion is most divided (the central debate as S&R put it). It is the common consensus among the three sources that constitutes our own IPCC. --W. D. Hamilton 22:55, 24 January 2007 (UTC)


 * See Mainstream Science on Intelligence regarding one of your preferred sources. 1/3rd of the signatories were Pioneer Fund grantees, and only 52 of the original 100 distributed to seemed to be willing to sign at all.  The APA statement seems much clearer, and calls it the "Genetic Hypothesis", not the "Partly-Genetic Hypothesis".  The other categories of explanation they use are, "Socioeconomic factors", "Caste-like minorities", and "African-American Culture".  Again, I assert we shouldn't use the weasel words used by pro-hereditarians to characterize the debate, even if they found 52 people to sign an article printed in the WSJ. --JereKrischel 03:32, 25 January 2007 (UTC)


 * WSJ - Actually, 131 were distributed to, but only 100 responded by the deadline. Of those 100, 11 said they were not expert enough to say. 28 of the 37 who didn't sign had no disagreement with the content, leaving 9. Of those not signing, disagreement with the definition of intelligence given was the modal reason for not signing. How often will you find that much agreement among experts under the given conditions? As noted elsewhere, one third of the APA authors were also WSJ signatories. Each of the 3 sources has good and bad features (trade offs between numbers of authors and depth of detail, etc), but the consensus of the 3 is a very powerful source. Roughly, the S&R survey establishes the prevalence of each view, the WSJ gives the most concise statement of the majority expert opinion even if it elides over detail, and the APA gives the most thorough review with the exception of their coverage of the partly-genetic hypothesis, which they don't detail at all (and were later criticized for in letters accompanying the publication.) --W. D. Hamilton 03:42, 25 January 2007 (UTC)


 * Do you have a citation for your assertions regarding who didn't sign and why? It may be good to include in the article on the topic.  Also, 1/3rd hardly qualifies as a majority, and the discrepancy between the APA article and the WSJ article seems to prove that they were unable to reach a consensus when put in a group of 9 on those issues.  Your characterization of three articles as somehow an undeniable "consensus" is really quite a stretch.  The APA source seems fairly good, but the S&R survey does not nearly say what you think it does, and the WSJ article by Gottfredson is hardly non-controversial.  Again, you are pushing for the usage of pro-hereditarian and pro-biological determinism language and rhetoric to frame the debate, which is clearly POV pushing. --JereKrischel 03:54, 25 January 2007 (UTC)


 * See the article you linked . 1/3rd hardly qualifies as a majority -- we don't know if the others were asked or if they responded; the "majority" is in the fact that 52 signed saying that they think that's the mainstream view (not just their view), and in the intersection with the S&R survey. Your characterization of three articles as somehow an undeniable "consensus" is really quite a stretch. - I don't claim they are a consensus, I claim that there are many (most) issues of consensus in the intersection of the three. the S&R survey does not nearly say what you think it does - could tell us how i misunderstood it and why/how everyone else who cites it misunderstands it as well (recall the sternberg interview). the WSJ article by Gottfredson is hardly non-controversial -- show me the citations that support your claim. pro-biological determinism -- huh??? clearly POV pushing -- if you mean the POV that some things are clearly spelled out in these sources. --W. D. Hamilton 04:02, 25 January 2007 (UTC)


 * You're right, we don't know if the 3 of 9 were even asked, just as we don't know if the characterization of 52 individuals properly reflects what is "mainstream". My use of pro-biologicial determinism is simply a reflection of the POV layout given in the Britannica article.  Perhaps you'd like to adopt it.  Again, to be very clear, your arbitrary characterization of the two sides being "partly-genetic" and "environment only" is POV pushing.  You frame the entire debate along those lines, and use the powerful rhetoric of the pro-hereditarian side to build an imbalanced article.  Furthermore, your assertion that you have justification for such an arbitrary drawing of lines, by citing the APA, S&R, and WSJ article are just not tenable.  If the sides are so clearly defined, why does the APA statement not include those terms "partly genetic" and "environment only"?  Like a rorschach test, you are seeing what you want to see. --JereKrischel 04:24, 25 January 2007 (UTC)


 * If what is clear to me is actually an illusion, then simply claiming that I'm wrong will do nothing to convince me of it. I've offered citations and quotations. You've offered nothing but attempts to label me. --W. D. Hamilton 04:36, 25 January 2007 (UTC)


 * I'm afraid that I cannot convince you that you're wrong, if you're wedded to a specific illusion by faith. I can make the case to other neutral third parties that your illusion is not the truth, however, which I think I've done fairly well.  That being said, please help me convince you - tell me, what kind of evidence would you accept that would make you believe that "partly-genetic" and "environment only" are not appropriate categories for framing the debate?  Would a reference from another encyclopedia help?  Or articles in the press or scientific journals framing the debate in a different manner?  To put the shoe on the other foot, I would be convinced that "partly-genetic" and "environment only" are proper frames of the debate if you could produce say, 10 papers over the past 5 years published in scientific journals unconnected to the Pioneer Fund, arguing the anti-racialist POV, clearly using those exact terms.  --JereKrischel 05:14, 25 January 2007 (UTC)


 * that "partly-genetic" and "environment only" are not appropriate categories for framing the debate?  you could begin by trying to cite sources that frame the debate in some other fashion. the multiauthor papers all appear to recognize this as the major point of dispute. I would be convinced that - I think this is a damning admission. what kind of burden of proof is that? i'm not sure there even are 10 review papers that would cover the subject in the last 5 years from any authors. --W. D. Hamilton 05:41, 25 January 2007 (UTC)


 * Race--Social, Biological, or Lemonade? By Carey, Gregory American Psychologist. 61(2), Feb-Mar 2006, 176. - Placing a complicated construct like race into a discrete “social” or “biological” box makes as much sense as asking whether lemonade is (a) lemon juice, (b) water, or (c) sugar.


 * Sternberg, R. J., Grigorenko, E. L., & Kidd, K. K. (2005). Intelligence, race, and genetics. American Psychologist, 60, 46-59. - Where does race fit into the genetic pattern? Actually, it fits nowhere. Race is a socially constructed concept, not a biological one. It derives from people's desire to classify.


 * Helms, J. E., Jernigan, M., & Mascher, J. (2005). The meaning of race in psychology and how to change it: A methodological perspective. American Psychologist, 60, 27-36. - Equating race with racial categories gives scientific legitimacy to the conceptually meaningless construct of race, thereby perpetuating racial stereotypes and associated problems in society. Moreover, it permits the discipline of psychology to function as an “objective” science even though it has granted a conceptually meaningless concept (i.e., race) so central a role in its theory, research, and practice (Fairchild, 1991; Zuberi, 2001).


 * Race, genetics, and IQ. By Nisbett, Richard E. Jencks, Christopher (Ed); Phillips, Meredith (Ed). (1998). The Black-White test score gap. (pp. 86-102). Brookings Institution. x, 523 pp. - Abstract: In this chapter, the author reviews the evidence on whether the Black-White IQ difference is wholly or substantially due to genetic factors...the most relevant studies provide no evidence for the genetic superiority of either race, but strong evidence for a substantial environmental contribution to the IQ gap between Blacks and Whites. (We might, according to this article, label the two POV "substantially genetic" and "substantially environmental".)


 * Explaining race differences in IQ: The logic, the methodology, and the evidence. By Mackenzie, Brian American Psychologist. 39(11), Nov 1984, 1214-1233. - He uses the terms "Environmental Models" (The Sociologist's Fallacy), "Genetic Models" (The Hereditarian Fallacy), and "Jointly Genetic/Environmental Models". He puts folk like Jensen in the "Genetic Model" category, not "partly-genetic".  In summary, the first part of the hereditarian fallacy is to assume that the failure to identify specific environmental causes of race differences in IQ is sufficient to refute an environmental hypothesis generally. It is a fallacy because our research methods do not require us to pinpoint the environmental influences, any more than the genetic ones, in order to estimate the relative contribution of each. The second part of the hereditarian fallacy is to assume that the refutation of an environmental hypothesis is sufficient to confirm a genetic one. It is a fallacy because the failure of one hypothesis can strengthen the case for a competing one only if that competing one has some independent empirical or theoretical backing of its own.


 * I'm sure I could find more sources for you. The fact that you cannot come up with a "I would be convinced that", seems to clearly identify your POV as a matter of faith, not reason.  I would suggest that until you are intellectually able to specify some form of proof that would change your mind, you aren't a very good person to determine what is NPOV. --JereKrischel 05:50, 25 January 2007 (UTC)

What are these sources saying and what are they not saying? I understand their POV very well, and it's not that the central question of group differences isn't the extent to which genetics contributes. It's that in answer to the question genetics can't possibly contribute. --W. D. Hamilton 05:55, 25 January 2007 (UTC)


 * I think you overestimate your understanding. They are clearly saying that race is an invalid category.  They are not saying that there are only environmental factors involved in intelligence.  For example, take a group of "white" people and split them into two arbitrary groups based on the second letter of their astrological sign, you may find a difference in intelligence that can be traced to a genetic cause, just at random.  The difference would be very small, but it would exist.  Now, in this case, it is perfectly reasonable to say that there is a genetic component to the difference in IQ between these two groups, but also perfectly reasonable to say that this genetic component is simply an artifact of random chance.  This is not an "environment only" explanation.  This is a critique of the arbitrary nature of the categorization. --JereKrischel 06:09, 25 January 2007 (UTC)


 * They are not saying that there are only environmental factors involved in intelligence. -- of course they are not, because they all agree that differences within races are due to a mix of genetics and environment. to the extent that they are getting at the question of the cause of observed group differences, they are siding with no genetics. nothing in these papers argues that the debate isn't about genetics/biology versus no biology. Placing a complicated construct like race into a discrete “social” or “biological” box makes as much sense as asking whether lemonade is (a) lemon juice, (b) water, or (c) sugar. -- the author is arguing that race is clearly a mix of both biological and social elements and cannot be reduced to either. Where does race fit into the genetic pattern? Actually, it fits nowhere. Race is a socially constructed concept, not a biological one. It derives from people's desire to classify. - in contrast to the last author, they are saying that there is no underlying biology to race. Equating race with racial categories gives scientific legitimacy to the conceptually meaningless construct of race, thereby perpetuating racial stereotypes and associated problems in society. Moreover, it permits the discipline of psychology to function as an “objective” science even though it has granted a conceptually meaningless concept (i.e., race) so central a role in its theory, research, and practice (Fairchild, 1991; Zuberi, 2001). - they are saying that the biological notions of race are not valid, and have no connections to the social characteristics of the term. none of these papers support your claims. --W. D. Hamilton 06:22, 25 January 2007 (UTC)


 * All of these papers support my claim -> they are all attacking the scientific utility of the simple categorization of "race". The first claim, that the concept of race cannot be reduced, is clearly a critique of using race as category - after all, if race is significantly a social category (i.e., significantly environmental), then measurements of race differences only really reflect biological differences if every environmental factor is controlled for.  Similarly, asserting that race is completely a social construct is even a stronger attack on using it as a category for asserting genetic based differences - any genetic based differences would simply be a factor of random chance.  Lastly, calling race a "conceptually meaningless concept" is certainly the most devastating attack to be made, calling into question the very foundation of past and present research in the area.  You seem to have a phenomenal ability to see only what you want to, and I think that makes your contributions to this article, in the spirit of NPOV, particularly difficult for you.  Please, if you can identify some manner of evidence that would change your mind, please do.  Otherwise, it seems that you are simply arguing a matter of faith, and it probably behooves us to have other editors advocate for your POV. --JereKrischel 06:31, 25 January 2007 (UTC)


 * Per below, I don't see how you get from your premises to your conclusions. Nothing in these articles suggests that the S&R are wrong that the central question of the debate is about the contribution of genetics to phenotypic race differences. They're simply arguments in the debate -- your own characterization of them puts them in terms of genetic and environmental causes.
 * You seem to have a phenomenal ability to see only what you want to, and I think that makes your contributions to this article, in the spirit of NPOV, particularly difficult for you. Please, if you can identify some manner of evidence that would change your mind, please do.  Otherwise, it seems that you are simply arguing a matter of faith, and it probably behooves us to have other editors advocate for your POV. -- JK, I'm a geneticist. I don't know how many thousands of scholarly articles I've read as a student and a scientist, but I'm pretty sure I know how to read a paper. There's nothing difficult about NPOV for me. Understanding that there can be a diversity of well-founded ideas on a topic is integral to my daily work. But if you're going to get personal -- I have to express my doubt about your ability to understand what you're reading with the precision to apply what you've read accurately to this topic. --W. D. Hamilton 07:01, 25 January 2007 (UTC)


 * Again, I think you misunderstand. S&R's survey did not determine "the central question of the debate", they merely framed it in one way.  There are other frames which you don't seem to be able to comprehend.


 * It seems that as a geneticist, you've not only found a profession, but a faith. You are unable to give any examples of evidence that would change your mind on this topic, and frankly, no personal offense intended, if you are unable to even conceive of evidence that would alter your POV you are not in possession of a rational POV, much less the ability to contribute in an NPOV manner.


 * You are absolutely right to doubt my ability to understand, but in that sense I am your perfect advisor - I am everyman, John Q. Public, and if you cannot express things in ways that are easily understandable to me, you probably aren't being understood by a great many others.


 * Please, you seem to be a dedicated and intelligent person who is very strongly attached to one POV. I'm sure you could copy all of the R&I articles you've worked on to your gnxpwiki.com, and create a wonderful professional journal level wiki based on the promotion of the pro-racialist POV.  It would certainly be a worthy task.  But wikipedia just isn't the place for your extensive, and by your admission incomprehensible to the average man, meta-analysis and research.  Examine how far you've taken this article away from what something like Encarta or Britannica have produced.  And please, consider moving your original research and energy to your own mediawiki installation. --JereKrischel 07:24, 25 January 2007 (UTC)


 * JK, John Q. Public isn't going to try to read Sternberg's article to determine whether he does or does not agree that the causation debate is divided into two major POVs. If they did, they might be quite confused. However, a person familiar with these topics would recognize Sternberg's argument (where it touches on intelligence) is a in the class of arguments in favor of an environmental explanations for group differences. They write Of seven published studies, six supported primarily environmental interpretations of group differences, and only one study did not; the results of the nonsupporting study (Scarr & Weinberg, 1976, 1983) were equivocal. The article comes from a series on the use of race in psychology, and thus the lengthy exposition on their view of the non-utility of racial classifications. --- Arguing that race should be used in research isn't an argument that race doesn't exist (note the extensive discussion of racial groups social underpinnings and denial of any biological connection). It's an argument that race is a social phenomena (and sometimes that some other grouping would be appropriate, but I don't see that in the Sternberg piece). Nothing in that indicates that the causation debate isn't divided into two camps, as your admission that in the section below demonstrated. --W. D. Hamilton 17:19, 25 January 2007 (UTC)


 * Again, wikipedia is for John Q. Public. It is not a professional journal amenable to detailed meta-analyses according to one POV.  Note even in your quote the term used is primarily environmental, not "environment only".  The causation debate is not the only debate being had, as exemplified by the extensive discussion of a denial of any biological meaning of race.  And the causation debate does not line itself up in the manner you suggest, with "partly-genetic" being somehow exclusive from "primarily environmental".  I still don't understand why you insist on using the weasel-words utilized by only one POV, and want to censor the existence of significant questions regarding the very foundation of the question being asked. --JereKrischel 18:57, 25 January 2007 (UTC)

re:long debates
A talk pages isn't for arguing a point to "prove" someone, or someones position is wrong. It is for discussing the content of an article. As for the content, I think the first 2/3d's of the article should describe the position that racial groups can be categorised and that certain races show higher or lower IQ's. The latter 1/3 of the page should portray the dissenting view (in many cases a direct response to the hereditarian view) that "race" is a artificial construct and that the definition "intelligence" varies depending on the researcher. We can let the the reader decide which is the "right" position. Mytwocents 07:17, 25 January 2007 (UTC)


 * That makes not sense. Race is increasingly viewed as primarily a social construct, one that may be of little use to science. Why should we devote 2/3 of this article to the idea that there are scientifically meaningful human races with significantly different innate intelligence which, although it is shaped by environment, for those inferior races, it will never be ameliorated. futurebird 13:40, 25 January 2007 (UTC)


 * Point well taken, Mytwocents. I believe both WD and I are trying to determine the appropriate content for the article in our argument however.  In regards to the layout, it seems to me that it should be at least 50/50...but more importantly, not a tit-for-tat collection of point and counter point ad infinitum.  We need to be much more compact, and eliminate the inside-ref editorializing and debating.  If anything, I would suggest we try and build a completely new page based on consensus, starting off with a proposed section layout.  Perhaps we can get a start on /Proposal 1 together? --JereKrischel 07:24, 25 January 2007 (UTC)


 * There's no WP paper shortage. In a summary style article, any important level of detail should be filled into subarticles. 50/50 isn't NPOV -- NPOV doesn't mean equal time -- it means proportional representation to the literature. --W. D. Hamilton 17:30, 25 January 2007 (UTC)


 * Just because you have a prolific pro-hereditarian like Rushton who floods with paper after paper, rehashing his old meta-analyses, does not mean we should simply measure the proportion of the literature to decide a proper ratio. In fact, it would be difficult for either of us to determine a metric to determine the actual proportion of positions in the literature (especially given the plethora of actual positions held not conforming to "pro" and "con").  That being said, the most fair and reasonable idea is to go 50/50.  I may want 90/10, you may want 10/90, and we both may have some metric that supports that - but we'll never agree on those metrics, and it's probably not worth arguing. --JereKrischel 18:49, 25 January 2007 (UTC)

These threads are getting confused. The point of this thread is to establish the relevance of these sources for the whole article/series, not just the topic of the other thread. --W. D. Hamilton 04:41, 25 January 2007 (UTC)

''Again, to be very clear, your arbitrary characterization of the two sides being "partly-genetic" and "environment only" is POV pushing. You frame the entire debate along those lines, and use the powerful rhetoric of the pro-hereditarian side to build an imbalanced article.''

I have noticed this two across multiple articles in this series. This in addition to a lack of historical context leaves much of the series on race and intelligence highly unbalanced. And unencyclopedic. I plan to work on one section at a time. Restoring balance where I can. JereKrischel, I'd love to know your thoughts on my changes to Race and intelligence (Utility of research). There is still a lot of work to be done on that one.futurebird 13:28, 25 January 2007 (UTC)


 * I have no problems with long debates especially when people at least appear to be trying to communicate and draw on relevant material. But I must admit to feeling a bit lost.  I have the feeling that both WDH and JK feel that in addition to there being different positions in a particular debate, there are different positions as to how to characterize the debate.  I would really appreciate it if each of you could as simply and concisely as possible sum up what you think are the different positions in the debate, and what are the different ways of characterizing the debate if indeed you thin there are multiple ways of characterizing the debate.  I ask this because I am getting pretty confused.  I get the sense sometimes that someone thinks someone thinks the debate is 100%genes vs. 100% environment vs. some mix of the two, but are there really any people who advocate either of the first two?  I also understand that for JK it is an important point that race is a social construct.  But it seems to me - and I admit I have read less than either of you - that the question is whether in the US at least (1) are there are statistically significant differences in IQ results among distinct groups (2) are these groups isomorphic with what Americans colloquially call "races," (3) if so, is this isomorphism (or highly positive correlation) because the race corresponds to a genetic population or (4) because the race occupies a specific socio-economic location in American society or (5) is the high correlation a purely statistical artefact, the result of certain assumptions that went into the modeling of the data that assumed the existence of races ... or other things?  If my comments are confused or confusing, please ignore them because I am not looking for a lengthy repetition of points you have already made but rather asking for us to pause a moment and try to summarize what we think the conflicts - between EDH and JK, and among the scientists EDH and JK are talking about - are, fundamentally, about.  Is it possible to sum this up concisely? Slrubenstein   |  Talk 17:50, 25 January 2007 (UTC)

Slrubenstein -- here's your short summary, thank you for weighing in: The debate is: what is "the role of genetic differences in the black-white IQ differential"? The major positions are: Of course, there is diversity within each of those two positions, and there is a substantial portion of experts who are well read enough but still don't have an opinion in favor of (1) or (2). Arguments about the nature of race that suggest there should be no genetic difference between races fall in position (2) and those that that argue for (1) will often argue that there are genetic differences between races (e.g. Jensen 1998). The quotes above come from Snyderman and Rothman (1987). --W. D. Hamilton 17:58, 25 January 2007 (UTC)
 * 1) "the difference to be a product of both genetic and environmental variation" - shorthand "partly-genetic"
 * 2) "entirely due to environmental variation" - shorthand "environment only"

(edit conflict) - ::::This conundrum has existed to my knowledge as long as the article. What I found, if I widened my search to more than one group of researchers, is that the answers will vary depending on who you ask. This is a politically loaded question, especially in the United States. You will get midely differing, if not opposite answers, depending on whether you ask psychologists, psychometricians, anthropologists, geneticists, and even depending on which school of thought they follow (e.g. "The London school"). I know this hardly answers any questions, but that's probably the most honest way to represent the field.--Ramdrake 18:06, 25 January 2007 (UTC)
 * Short reply - I think you could have titled the article 'intelligence (race)'. You can talk about race without intelligence coming up but you can't talk about intelligence w/o race coming up. That's where most of the literature

is and that's where the public attention is. --W. D. Hamilton 18:28, 25 January 2007 (UTC)


 * You can clearly talk about intelligence w/o race coming up. The well-orchestrated public relations campaign of the Pioneer Fund and its affiliates aside, there is a wealth of research on intelligence that has no position on BGH. --JereKrischel 18:45, 25 January 2007 (UTC)


 * Also, the field of neuroscience is replete with articles about intelligence that don't mention anything about race.--Ramdrake 18:53, 25 January 2007 (UTC)


 * The debate is: what is "the role of genetic differences in the black-white IQ differential"?
 * The major positions are:
 * "the difference is substantially genetic" - shorthand "partly-genetic" by pro-hereditarians, also called "genetic hypothesis" or "genetic model" or simply "genetic"
 * "the difference is substantially environmental" - shorthand "environment only" by pro-hereditarians, also called "environmental hypothesis" or "environmental model" or simply "environment"
 * "the groups defined (race) are improper classifications for a genetic model, regardless if the difference is genetic or environmental" - a common position that challenges the very foundation of the question without addressing the question of genetic or environmental differences between groups.


 * The problem WD has is that he cannot accept that the foundation of the question is of significant debate. Given the rich number of pro-hereditarian papers he's read that have characterized the debate with the weasel-word term "partly-genetic", it seems he believes that such a characterization is both normal and neutral.  Ramdrake is correct, it is a politically loaded question - and I believe the only proper NPOV way of characterizing the debate is to make sure that all positions are presented, even though that offends the POV of pro-hereditarians. --JereKrischel 18:43, 25 January 2007 (UTC)


 * I agree with JK that representing the different positions as a spectrum with the "wholly environmental" and the "wholly genetic" as the opposite ends is probably more representative of the literature outside of the circle of the London school.  The interesting part of it is that it clearly desmonstrates how few people there actually are at each end of the spectrum, and begs that the most important question to ask is "how much of it is genetic"? rather than "is it at all genetic in origin or not"?--Ramdrake 18:53, 25 January 2007 (UTC)


 * I've commented on this in the section below. it's the regardless if the difference is genetic or environmental that doesn't hold up. No one who argues the first half of (3) argues that the difference is likely genetic, they are saying it is likely environmental (e.g. the Sternberg paper JK cites) so (2) and (3) are the same. (To do so would be very odd.) This paper lists 102 possible causes for the observation of the BW gap, but I don't think that means there are ~102 positions. --W. D. Hamilton 21:04, 25 January 2007 (UTC)


 * Without coming out and actually saying so, researchers like Rushton (and a few other Pioneer fundees) insist so much on a genetic cause that their position does look like it's a "mostly-genetic" explanation, whereas they officially say it's about 50/50. Others, like Lieberman, say that if such a genetic component exists 1)it's not really race-based as the genetic difference is far more likely to be clinal than cluster-like (based on the works of Beals and al, which is being interpreted very differently by Rushton!) and 2) possibly accounts for just a few percent of the gap. So, yes, go re-read some of Rushton's hypotheses and it does come across as "likely or mostly genetic".--Ramdrake 22:16, 25 January 2007 (UTC)

Okay, here is my immediate take on it. It seems to me that JK's account is slightly better -well, let me just say for the moment, more interesting (see below, maybe it is not better) - than WDH's in that he is identifying two dimensions along which there is debate: one dimension has to do with the relative role of genetics and environment in explaining differences among groups, the second dimension is how to characterize those groups. I think there is a big flaw in the way JK lays out these two dimensions because two dimensions should produce four points, not three. In other words, for the second dimension (what are the groups we are testing) there should be at least two possibilities, each expressed positively (some say x, some say y). Instead, two possibilities are conflated in a negative postion (y is wrong). Instead, it would make more sense to lay out two positive but competing views. And here I think is the source of WDH's concerns, and I think they are valid. If a degree of the variation is genetic, and if we reject race as a way of talking about these genetic populations, how then do we talk about them, and why do they seem to correlate with race? Unless JK can address this in a satisfying way, then I think WHD's point may be right/his approach better: if I understand him, he is saying that even if we are looking at two dimensions they are not orthoganal. Position 3 is a subset of position 2, and there remains the fourth position, which JK did not list but which would be the opposite of position 3, which would be a subset of position 1. Does this make sense? Again, I am just trying to sort out the differences between JK and WDH for now. I believe I had proposed two alternatives: If WDH is correct that there are really just two sides to the debate, they the way forward would be to show how each of those sides subsumes points raised by JK, and we need to think about how we can discuss more clearly the way each of the two position rests on certain assumptions or subsumes these other points. If JK is correct then he has to provide the fourth term and show how the two dimensions are orthogonal. Slrubenstein  |  Talk 11:36, 26 January 2007 (UTC)


 * You're right, Slrubenstein, there are four points, although one of the quadrants is hardly ever proposed in the literature -> racialist/hereditarian, racialist/environmentalist, non-racialist/hereditarian, non-racialist/environmentalist. Here's a table once on Race and intelligence (explanations):

Comparison of explanations
The possible explanations for observed differences in intelligence between racial groups generally occur along two axes - how valid a proxy race is for most genes that control intelligence, and primarily-genetic/primarily-environmental.


 * Insofar as why things "seem" to correlate with race, my personal understanding is that it can be seen as a combination of joint genetic/environment interaction. That is to say, differing environments can affect gene expression, so a genetic difference may exist, but to say that the cause is genetic is misleading - change either the gene or the environment, and the end result is different.  Some pro-racialists claim that such genetic/environment interactions support their hypothesis as a distinction without a difference (since there is some role for genes), and some anti-racialists insist that the distinction is terribly important, and the assertion that the difference is "significantly genetic" is both misleading and misguided. --JereKrischel 18:52, 26 January 2007 (UTC)


 * I would like to know whether WDH thinks that this table proides a reasonable basis for organizing the article, or part of it, an if not, why not (of course others are welcome to chime in, esp. Ramdrake who has been following this debate closely, but I am especially concerned to know how WDH responds to this. Slrubenstein  |  Talk 13:08, 27 January 2007 (UTC)


 * I know you didn't ask me, but I support it. -- futurebird 13:28, 27 January 2007 (UTC)


 * SLR, i missed when this question was asked, but I think this table represents an original and inaccurate presentation of views about the cause of group differences. On one hand, literature divided along the race (columns) axis is mostly not about intelligence and doesn't claim to be about it. On the other hand, to the extent that literature examining the race question does does claim to have implications for the cause of group differences in intelligence, it does not claim to do so in an orthogonal manner -- rather there is a correspondence between no-race and environment as the one POV and race and genetics as the other POV. I don't see how it could organize the article overall as there are many subtopics where neither the cause of group differences nor the nature of race plays an important part. Murray and Flynn (2006) managed to have an important debate race and intelligence without ever directly bringing up either of those topics. This may not have been clear -- I don't think the question about group differences organizes the article. It just happens to be the issue that is most hotly debated. It's the issue where there's the least agreement and the most contention. --WD RIK NEW 19:11, 27 January 2007 (UTC)

I don't see how this is original research. It gives us a framework for organizing this complex issue. It makes it clear that denying that race is a solid genetic indicator for intelligence is not the same thing as:


 * 1) Saying intelligence is not partially genetic.
 * 2) Saying intelligence is not a partially genetic trait that affects genetic groups.
 * 3) Saying that race is not a useful form of genetic grouping.

In the present organization these three ideas are used as straw-men, when, in fact, all three of these things may be true and the hypothesis that says that race is a valid genetic grouping, and that it significantly linked heritable intelligence may still be false. This a subtle, but important, point that is glossed over in the present article structure. Research that supports one of these three hypothesis is used to support the entire "heritable racial intelligence" hypothesis, and that is a misrepresentation of the work. Only research that makes this total connection should be portrayed as such. Moreover, the tone of the article should reflect that fact that these three hypothesis are still all unanswered questions in mainstream science (though, #1 may be more widely accepted.) futurebird 21:33, 27 January 2007 (UTC)


 * I'll also note that there is another axis that I've come across, only once so far, but it may be more prevalent than I expect - there is the question as to the ordering of the "races". There are those that have asserted that there may very well be race-based differences based on genetics, but assert that a large environmental impact on certain races is masking the true hierarchy.  That is to say, where Rushton says the order is B-W-EA, from lowest to highest, others may assert that once proper controls for environment are imposed, the order is actually W-B-EA, or W-EA-B.  I suspect that this ordering, this assertion that the genetic difference has been found in a certain order, is also a big reason why such findings are so controversial.


 * This becomes a sociological issue, but to entertain it for a little while...how would we all feel if we did a perfect study, and found that there is a genetic difference between races, but that the superior intelligence of blacks was being masked by incredibly negative environmental factors? Would I still be arguing that race isn't a valid proxy?  Would WD/RIK still be defending the pro-hereditarian view?  I think the issue that this touches is the one of conclusions - if we are to believe that the B-W gap in the U.S. is the "natural" state of affairs, we can justify the lower socioeconomic position of blacks by blaming genetics.  However, if we were to believe that the "natural" state of affairs would be a W-B gap, we might look very critically at our society and wonder how it so dramatically diminishes the superior intelligence of blacks. --JereKrischel 07:54, 28 January 2007 (UTC)

Groups and Intelligence
As per the major source (APA) pointed out by WD, I suggest that we rename this series of articles to Groups and intelligence (United States), or Groups and intelligence (Developed countries). --JereKrischel 20:09, 24 January 2007 (UTC)
 * That's not such a bad idea. Though, it might be too confusing, and there would be a redirect. futurebird 20:16, 24 January 2007 (UTC)

The "groups" in the APA report were both race and sex. This article is not about sex and intelligence, but rather about race and intelligence as the common sense meaning of those words. the APA report also discusses individual differences in intelligence, but that is also not the topic of this article. as previously discussed at length and many times on the talk page, the formulation of "race and intelligence" is the most common formulation is the title of books, journal articles, and other encyclopedias. the distinction between developed and developing countries appears to be unrelated and unsupported. --W. D. Hamilton 22:44, 24 January 2007 (UTC)

Race is a term which is highly charged and the very concept of races of men is being disputed or at least lacks clarity in definition. For example, what is "black" or "afro-american" in the US? Is it 50% sub-saharran african? 10%? 2%? Are Egyptian imigrants Afro-Americans? The American Anthropological Association has an interesting dicsussion of race at http://www.aaanet.org/stmts/racepp.htm --Kevin Murray 00:15, 25 January 2007 (UTC)

Other alternatives to the current title might be (1) Ancestory and Intelligence (2) Heredity and Intellegence (3) "Genetics and Intellegence (4) Populations and Intellegence (5) Variations of Human Intellegence. Another alternative is to come up with a substitute word for intellegence; since a major critisism is whether IQ testing is valid test of true intellegence, perhaps we subtiture the term "IQ Scores" or "IQ Testing."  For example "Variations of Human IQ Scores."   Or "Variances in IQ Among Populations" with redirects from more likely search titles. --Kevin Murray 00:15, 25 January 2007 (UTC)


 * Kevin, there would need to be good reasons for such a change. Britannica uses race and intelligence. Encarta has a similar formulation. The book by skeptic Fish uses the same title. There exists articles on race and articles on intelligence, and their titles persist despite the imprecisions you mention. (In reality, the distribution of African admixture among African Americans is quite tight, centered around ~80%, and IQ is the de facto measure of intelligence, with the caveat that it doesn't capture the full range of common sense meanings of the term.) Previous discussion of this topic occurred here and here. --W. D. Hamilton 00:24, 25 January 2007 (UTC)


 * Britannica actually uses "Race" and intelligence. Perhaps we should put quotes around Race as well. --JereKrischel 03:35, 25 January 2007 (UTC)


 * Britannica also is much more encyclopedic in its approach to "Race" and Intelligence. It also denotes the "hereditarian belief" and "biological determinism", and does not use the term "partly-genetic".  It may be a good example for us to follow for a more encyclopedic article, rather than an in-depth meta-analysis research paper on the evidence for and against the hereditarian POV. --JereKrischel 03:43, 25 January 2007 (UTC)


 * I think an article on "Race and intellege" makes sense, since, from a historical perspective, that is how this controversy was originally framed. Also, I think it is what people will search for fisrt-- and those other topics are releted but not the same. futurebird 00:28, 25 January 2007 (UTC)


 * I don't have a strong feeling, and if WD and FB concur, that's good enough for me. --Kevin Murray 01:54, 25 January 2007 (UTC)

All right then, the name will not be changed. --causa sui talk 01:55, 25 January 2007 (UTC)