Wikipedia talk:Requests for mediation/Race and intelligence

Note

I have requested (several days ago) for assistance with this mediation, as my health is still uneven and I am unable to give this the time I would strongly prefer. Unfortunately, the other Mediation Committee members are busy, and no one has yet been able to assist. I appreciate your continued patience and assure you that although this is proceeding slowly,I have not abandoned you. Thanks much to all of you for understanding. KillerChihuahua^?!? 17:31, 26 March 2007 (UTC)

Some ground "rules"

I have removed the previous non-mediation discussion which was here[1], which I see as indicative of a problem we might face in this mediation: This is a complex mediation, with a number of distinct issues and a fairly large pool of editors involved. You'll have to let me drive, or this will turn into the same kind of chaotic unorganized dispute as has been seen so far. In order for mediation to succeed, it is important for the participants to be able to focus on one issue at a time, and work within the framework not disrupt it. I have every confidence that the editors involved all want to resolve this, or you would not have all agreed to mediation. I plan to work through the list of Issues to be mediated one at a time, and hope to post a corresponding list of Agreed items as each item is resolved. That said, the first item is What should the organization of the main article be? and I invite each of you to post a preferred structure. If you support an already posted structure, simply say Support EditorFoo's version. Ok?

The second "ground rule" is one you all know by heart, so it will come as no surprise: comment on content, not contributor. I will cheerfully and promptly remove any posts on this page which involve speculation on any editor's bias, ancestry, or any other personal aspect of that editor. Please feel free to complain on my talk page or email me, I would prefer this page be limited to what promises to be a lengthly and complex mediation, and not individual complaints about the mediator or other parties. KillerChihuahua^?!? 17:50, 21 February 2007 (UTC)

Per request, What should the organization structure/relationships of the many articles be? has also been opened for views to be posted. Add to both or either, but be aware that all of you need to voice a position on both. KillerChihuahua^?!? 23:24, 23 February 2007 (UTC)

15:34, 9 March 2007 (UTC): I am semi-back, thanks for your patience while I was "out sick". I note we have heard from the following parties:

On the MAIN article structure:

• JereKrischel
• Kevin Murray

On the family of articles structure/relationship:

• WRN
• Futurebird
• Kevin Murray
• JJJamal

Editors we have not heard from at all:

• Ultramarine
• Ramdrake

I will request input from the two editors not heard from, and those of you who have entered a position in only MAIN or ARTICLE please add your position to the other section. KillerChihuahua^?!? 15:34, 9 March 2007 (UTC)

And now Ramdrake is the only editor not participating. The last edit by Ramdrake was 10 March 2007. Hopefully it is just a time concern and we will hear from that editor shortly. KillerChihuahua^?!? 22:26, 13 March 2007 (UTC)

What should the organization of the main article be?

JereKrischel's suggestion

0. lead - 3-6 paragraphs
1. background/history (subarticle - Race and intelligence (history))
     - 2-3 paragraphs
1.1. intelligence (main: IQ, intelligence)
     - 1-2 paragraphs
1.2. race (main: race)
     - 1-2 paragraphs
2. research (subarticle - Race and intelligence (Research)
     - 2-3 paragraphs
3. media portrayal (subarticle - Race and intelligence (Media portrayal)
     - 2-3 paragraphs
4. controversies (subarticle - Race and intelligence (Controversies)
     - 2-3 paragraphs
5. end material (subarticle - R&I (references) or switch to footnote ref system)

In my view, research contains three major sub-articles:

• Race and intelligence (explanations) - an outline of the various hypotheses
• Race and intelligence (test data) - an outline of research results, hopefully sans any assertions of meaning (save that for the next section)
• Race and intelligence (interpretations) - specific interpretations of data. This would include a large number of meta-analysis wars had by several researchers of R&I

My concerns generally circle around whether or not this article is primarily about "Race and intelligence research", or "Race and intelligence". One seems much more expansive than the other, and without organizing the research portion into distinct areas, I think that excessively detailed arguments wind haphazardly throughout the structure.

For example, without segregating "interpretations" from "test data", we have problems where one editor will make a citation to a certain study, and a claim about what that study means. Then another editor will make a citation refuting that claim, and possibly citing another study and making additional claims of their own. This can go on endlessly (and does in many places right now). I think this "ad hoc, argue every point, counter point, and counter counter point" methodology leads to a disjointed series of articles.

On the other hand, if under "test data", we simply stated the GRE test results (without claiming they proved a specific hypothesis), or IQ test data, we can avoid this furball. Under "interpretations", we can more easily summarize conflicting interpretations about various categories of data, rather than conflicting interpretations about each specific point made in every journal article over the past 20 years.

I also think the basis of this specific dispute is because of a fear that my proposed structure diminishes the importance of research on R&I. Although inevitably placing research at something besides the root level will diminish its dominance of the article, I don't believe that this is being done in bad faith (specifically to denigrate research that asserts genetic causes to racial gaps), or that a loss of dominance is a loss of any importance.

Disagreements with futurebird

I think that "interpretations" is a research topic, and shouldn't necessarily be under "controversies". I guess I see "interpretations" as being specific to study citations, and controversies being more broad. For example, one researcher may interpret some set of test data as showing 60% genetic contribution to the B-W gap. Another may interpret that same set of test data as showing 0% genetic contribution to the B-W gap. The "controversy" section would probably deal with one researcher insisting that blacks are inferior to whites genetically, without specifically speaking of a given study...a lot of the "aggregationalists" such as Rushton would fall into this category. This kind of division I think allows us to avoid a spiral of death, where a controversy is named, a few studies cited, then a few counter studies cited, then counter counter studies, etc, etc, etc. If we keep the specific details of "interpretations" to specific details around single studies, and keep the controversies more general, we may be able to avoid this anti-pattern currently in the article.

The sections I see under "controversies" are:

• Race and intelligence (Utility of research)
• Race and intelligence (Potential for bias) (poorly named, alternative suggestions welcome)
• "Policy Implications" (currently no sub-article)
• other sections without sub-articles

Utility of research is the general category of disputes about the fundamental questions being asked - the meta-questions of "what is race?", "is race real?", "do the answers to these questions really have any meaning?", etc.

Potential for bias is a terrible name, but essentially seems to be a catch-all for ad-hominem attacks on the other side - I think maybe WRN is right that these sections need to be organized in some different manner.

Disagreements with WRN

Regarding "consequences" or "significance" or "practical importance" of group differences., I have some severe difficulties with those characterizations. The pattern ostensibly cited is the "practical validity" pattern - where "validity" is the correlation between score and outcome. The model for this is "the practical validity of IQ", where we measure IQ scores and outcomes such as income, health, marriage, number of children, etc.

This idea of "practical validity" doesn't apply very well to a category with a dual nature, since the direction of what is considered a "score" is a matter of POV. For example, in a number of hereditarian books and papers, arguments about how IQ differences between groups correlate with non-IQ differences between groups are made. In short form, the illustration they seek to show is "when IQ is controlled for, these gaps between races in Y diminish by X". This prejudges the question by both implying that IQ is a natural function of race, and that it isn't racism which causes differences.

You can turn it around, and seek to show "when race is controlled for, these gaps between high and low IQ groups in Y diminish by X". (I.e., high IQ people may be richer than low IQ people, but if control for how "black" someone is, you find that the gap diminishes, implying that people are richer or poorer based on race, not IQ.)

The assertion of "consequences" or "significance" or "practical importance" is a specific interpretation of the data.

Comments - Kevin Murray

--Kevin Murray 23:45, 23 February 2007 (UTC)

• I only have few issues with Jere's proposed layout, although it references sub-articles which I don't condone. I propose below (section 3.3) that this project be contained within only two articles: R&I and R&I research. And I advocate that history be included as a major section here at R&I, except that a specific history of research be included at R&I research. I use the existing names abbreviated to "R&I" here without prejudice to my preferred "Populations and intelligence" (discussed below in 3.3.2).

0. lead - 3-6 paragraphs

1. intelligence (main: IQ, intelligence)
     - 1-2 paragraphs
2. race (main: race)
     - 1-2 paragraphs
3. history (excluding specifics on history of research)
     - 4 - 10 paragraphs
  3.1 summary of history of research (reference to reserach history section at R&I research)
     - 1-2 paragraphs
4. research summary (subarticle - Race and intelligence (Research)
     - 3-4 paragraphs
5. media portrayal 
     - 4-10 paragraphs
6. controversies 
     - 4-10 paragraphs
7. end material (switch to footnote ref system, with separate bibliography common to the project)

WRN

Research should not be a single section in the main article. There should be three individual section for the three sections currently described as research: (not necessarily these titles)

1. Average racial differences on intelligence related measures)
2. Explanations for average racial differences on intelligence related measures)
3. Practical validity (or whatever you call it)

What should the organization structure/relationships of the many articles be?

WRN's version

general

1. The article uses and should use summary style, which "focuses on topics". The main article (Race and intelligence) should be "a primer on a whole topic". It should have many daughter subarticles, with summary sections left in the main article headed by a link to the daughter articles.
2. It would be preferable for every significant topic to have a summary section in the main article and its own subarticle. My idea of significant is expansive -- I would prefer to see that every topic about which a significant amount of material can we written be given its own main-article facing summary section.

specifics

• both race and intelligence/IQ should have summary sections in the main article to provide salient background, akin to summary style
• a subarticle and main-article summary section are definitely needed for each of the following:
  • history
  • "test data"
  • explanations
  • "interpretations"
  • media portrayal
• one or more subarticles and main-article summary sections are needed for the material in
  • utility of research
  • potential for bias
  • other "controversies"
• if the current reference system is maintained, Race and intelligence (References) is also needed

possible issues of contention, and areas of doubt

• no research article --
  • the purpose of a top level research article is ostensibly to provide a summary of all "research", but the research itself includes 2 (or 3) largely independent topics. the clearest case of this is "test data" and explanations. all major sources that summarize the literature appear to separate these issues. The APA report has two sections ("Mean Scores of Different Ethnic Groups" and "Interpreting Group Differences") which correspond to these two topics. The WSJ statement has two sections ("Group Differences" and "Source and Stability of Between-Group Differences"). Jensen (1998) has two chapters ("Population differences in g" and "Population differences in g: causal hypotheses"). And so on. The dividing line we chose between the two versus the lines chosen by each of these sources is a little fuzzy, but the importance of separating discussion of them should be clear.
  • however, i think a case can be made for having a "research" super-article which provides a more technical summary of all of the same material that appears in a general-audience summary on the main page. i have no objections to this use of a technically-oriented research summary-article, if it doesn't otherwise alter the organization I've outlined.
• "interpretations" as a top level item -- I believe the content of this article fails to fit appropriately in either of the other two "research"-related topics and it is not directly subsumed in any other topic. Like media portrayal, it should be a separate top-level topic. in current and previous discussion of this article, i believe there is a consistent misunderstanding or confusion of what the article's contents are/are not. it is not the same as a "public policy" discussion. the title "interpretations" is a mislabeling. it is actually better called "consequences" or "significance" or "practical importance" of group differences.
• controversies topic -- the distinction between "utility of research" and "potential for bias" has become blurred, and the other content of "controversies" is growing. The controversies topic is clearly expansive, but those particular sub-articles may no longer be the appropriate way to divide that topic. a reassessment should be made and the material may need to be reapportioned into some number of differently divided subarticles. any major subarticles should have main-article facing summaries and the remaining "other" controversies should be summarized briefly in the main article as well.

comments

• Kevin's comment re: Populations and Intelligence -- there are advantages and disadvantages of that, which are probably better answered at a later time. if going down that route, Population Differences in Intelligence would be more specific.
• FB's comment re: But, Race and intelligence research still needs it's own subarticle. -- there are now a number of permutations on that idea. Can you give a specific suggestion?
• I don't know anyone who claims that "Race and intelligence" is a topic limited to only recent research results, but research results are clearly the largest single aspect of this topic, receiving the largest fraction of the attention in scholarly sources.
• I believe Ultramarine may be agreeing with my assessment. Or perhaps I am agreeing with Ultramarine's assessment. The exception being the "interpretations" article doesn't appear in Ultramarine's outline. The issue of a "research" subarticle appears to still be an issue of contention. The only use I can see for it is as a nontechnical summary of the entire research topic, but this isn't necessarily a good idea.

Futurebird's thoughts

1. Race and intelligence research needs it's own subarticle. If there is no section for research the main article will seem like an article that is only about research. Some sources question the very fact that research is being done at all. Other sources see the topic differently, as one related to the history of racial prejudice as much as a question of science. Still other sources are interested in the way that the media portray the intelligence of people of different races, rather than the findings of studies. Research is, of course, one of the most important aspects of this topic, but it is not the only area. Research should have a big section in the main article, but it still needs it's own subsection. Subsections of the research article should be linked to the main article. It should be expansive in some of the same ways WRN has described, but it should avoid inadvertently making a statement about what aspects of the topic are the most important through the structure of the article.
2. Race and intelligence (utility of research) and Race and intelligence (potential for bias) are confused topics at this point. (I do agree with WRN on this, in some ways) I'm open to suggestions about what we should do with these topics, so long as the material in them remains in the controversies section, or in a sub-section of the controversies section. Naming these articles will be a sensitive matter.
3. Race and intelligence (interpretations) belongs in the Race and intelligence (Controversies) section, not in the section for research. This section explains the interpretations that various scholars have of the gap. All of the explanations (both liberal and conservative) are slightly political and controversial in nature. There is no agreement about how gaps ought to be understood among academics. The work in this section isn't about experiments, but rather it about comparing data from experiments that others have done and then drawing conclusions about what should or should not be done.futurebird 03:33, 22 February 2007 (UTC)

Compromise and comments

• I'm willing to compromise given the points raised by JK and have Race and intelligence (interpretations) be a sub-section of research. But, Race and intelligence research still needs it's own subarticle.futurebird 13:12, 22 February 2007 (UTC)

• The history section should be integrated as it pretty much is now including research and things that are now considered "scientific racism" we need to do this because the line is blurry and debated.

• The research sub-article should not contian it's own history section. It should focus on modern work. futurebird 15:15, 3 March 2007 (UTC)

• I think Ultramarine's idea also makes a lot of sense. I'm willing to give up on the "race and intelligence research sub-article.futurebird 22:16, 10 March 2007 (UTC)

Stalled?

I hate to be impatient, but a few of us are wondering if this process has stalled? futurebird 22:00, 1 March 2007 (UTC)

Comments - Kevin Murray

--Kevin Murray 14:08, 22 February 2007 (UTC)--Kevin Murray 13:50, 22 February 2007 (UTC)--Kevin Murray 13:43, 22 February 2007 (UTC)--Kevin Murray 01:58, 23 February 2007 (UTC)

Overview

• I am modifying my position to accept Race and intelligence research as the proper place for much of what was in the article prior to 2/11/07, including the actual research and the history of the research. Seeing this information objectively kept together would reduce my objection to the recent and rapid evolution of the "main" article.
  • I can see the history of Race and intelligence as separate from the history of Race and intelligence research. I would prefer that the history be distributed between these articles rather than as a stand-alone article.
• I see a risk for developing too many sub-articles; with a contentious topic there is too much risk of vandalism and POV pushing within the permutations. I see evidence of this already.
  • Eliminating off-point discussions of political and social issues could reduce the need for extensive permutations from this topic. Much of what is creeping into this series of articles is better relocated to articles dealing with US race relations and civil rights history.
  • I would prefer these R&I articles to focus on a global view if possible. Clearly much research is US based which presents an inherent bias to the article which will be hard to overcome.
  • However, there is no question that the topic has expanded beyond what is manageable in one article; two or three seem an appropriate balance.
• Should we be labeling these articles "Race and Intelligence" or "Populations and Intelligence" as the research seems to differentiate among populations which are not specifically races.
  • "Hispanics" in a US context are substantially speakers of a common language, and the general south or central American populations are predominantly European, and Native American, with substantial Sub Saharan populations.
  • Afro American/ Black seems blurred to me as the "one-drop rule" and affirmative action have biased the perceptions and self-perceptions of racial identification.
  • Asians are also unclearly defined

Specifics

• Race and intelligence should include substantial discussion of: media portrayal, other "controversies", and a general history of the topic.
• Race and intelligence research should include "test data", "interpretations", and a specific history of the research
• Utility of research and potential for bias (systemic & cultural) sections are important at Race and intelligence research, but should be also be included at Race and intelligence, in a brief summary form, as should an overview of the research.

Comments - JJJamal

• The main page should not give the impression that the topic "Race and intelligence" is only a topic for scientific research. I don't care about the names of the subpages, however it is organised this should be the final result.

• The history section should tell the history history of the entire topic. Not just the research. I think if we did what Kevin is suggesting we'd end up with two history sections. Let's keep it all in one place.

• The test data section is not working. I think it should be split and merged.

JJJamal 11:39, 26 February 2007 (UTC) Bold text

Comments - Ultramarine

• The most important point is that "Race and intelligence" is not a topic limited to only recent research results, as sometimes have been argued. It also inlcludes, for example, racial stereotypes throughout history, use in media and by racists, and motivations and funding of research.

• My proposed structure would be:
  • Race and Intelligence
    • Race and Intelligence (Historical views and their influence)
    • Race and Intelligence (Current media portrayal)
    • Race and Intelligence (Average racial differences on intelligence related measures)
    • Race and Intelligence (Explanations for average racial differences on intelligence related measures)
    • Race and Intelligence (Controversies surrounding research)

I do not think there is a need for a "Race and Intelligence (Research)" article since it will only repeat material already in other articles and "Race and Intelligence" is already a summary article.Ultramarine 21:20, 9 March 2007 (UTC)