Back to [Archive] Other discussions

1
Is "the hereditarian hypothesis almost certainly true"?
In a recent blog post, Emil argued that a racial-IQ hereditarian hypothesis was almost certainly true because:

(a) a race-IQ HH could readily be falsified by showing an absence of a correlation between genomic race and IQ
(b) some research teams must have looked at admixture-IQ data
(c) a non-association would have been published if it was found
(d) yet no results have been published

But has anyone bothered to email around for results? I just came across the following, for example:

Akshoomoff, et al. (2014). The NIH Toolbox Cognition Battery: results from a large normative developmental sample (PING). Neuropsychology, 28(1).

Estimating the effects of sociocultural factors within a highly diverse sample such as PING is challenging. The multi-site design produced a sample in which participants came from many different ethnic communities and many had mixed backgrounds. Because genotype information was available for the PING participants, we chose to use a set of genetically derived estimates of racial ancestry to estimate effects that could reflect differences in sociocultural background...To examine and control for the influences of race/ethnicity on test performance, genetic ancestry factors (GAFs) were calculated to estimate the proportion of European, African, American Indian, East Asia, Central Asia and Oceania ancestry for each participant, based on genotype analysis (methods detailed below)...The second model included the two separate SES variables (i.e., parental education and household income) in addition to the base model; SES variables considered were highest education of the parents and annual family income, both added as linear terms. The third model added genetic ancestry factors (GAFs) as linear terms. For each pair of nested models we computed R-squared statistics and performed chi-square tests to determine whether the added terms represented a significant contribution over and above the terms already in the model...The addition of genetic ancestry factors (GAFs) to the base+SES accounts for an additional .5% to 1% of the variance over and above SES for most NTCB scores, except for Vocabulary, for which it contributes an additional 2.3%.


The authors found the genomic ancestry explained some cognitive variance (about as much as SES); however, they didn't mention if it did so independent of social racial identity. Undoubtedly, they checked -- they didn't look at the association between genomic and social race, let alone use an index of genomic race in the first place, for nothing.

My question though is: Has anyone bothered to just ask? (In this case: Was genomic ancestry an independent predictor of cognitive ability when taking into account social identity?)
Admin
Good idea. Let's write a standardized email and send it to all corresponding authors who did a study that involved US blacks, asking them to check the data for a correlation between cognitive ability or SES and amount of African ancestry. We can point to the previous diabetes study as a reference for the method.
Good idea. Let's write a standardized email and send it to all corresponding authors who


We should start by emailing the authors of the mentioned NH toolbox paper. One would only need to know the correlation between e.g., PPVT and each ancestry component for each ethnic classification (e.g., "African American", "Mixed", "White" ).
Admin
It's easy to control for skin color, if they would just gather the information.

And yes, that is how they usually explained it. If you search for IQ x skin color correlations.

Apparently Lynn did a study: http://link.springer.com/article/10.1023/A:1014572602343

The relation between skin color and intelligence was examined in a representative sample of 430 adult African Americans. A statistically significant positive correlation of 0.17 was obtained between light skin color and intelligence. It is proposed that the result supports the hypothesis that the level of intelligence in African Americans is significantly determined by the proportion of Caucasian genes.


Higher wages
http://jhr.uwpress.org/content/XLII/4/701.short

This paper develops and tests a theory, referred to as “preference for whiteness,” which predicts that the interracial (white-black) and intraracial wage gap widens as the skin shade of the black worker darkens. Using data drawn from the Multi City Study of Urban Inequality and the National Survey of Black Americans, we report evidence largely consistent with the theory. Moreover, we decompose the estimated interracial and intraracial wage gaps, and find that favorable treatment of lighter-skinned workers is a major source of interracial and intraracial wage differences as predicted by the theory.


Nisbett discussed it in The Black-white Test Score Gap.

Color Differences in the Socioeconomic Status of African American Men: Results of a Longitudinal Study

Using a longitudinal design that links a sample of southern-reared African American men to their childhood census records (collected in 1920), this study attempts to replicate recent findings documenting the influence of skin color on the socioeconomic attainment of African Americans. The childhood census records used in this study classify African Americans as either black or mulatto, allowing for a unique investigation of color stratification in adult life. Consistent with previous research, findings point to the importance of phenotypic characteristics in influencing the life chances of African Americans. Subjects identified as mulatto enjoyed modestly higher adult socioeconomic status compared with subjects identified as black. While the mulatto advantage attenuates slightly once origin characteristics are considered, multivariate results indicate that differences in social origins are responsible for only 10 to 20 percent of the color gap in adult attainment. Findings suggest that color bias (colorism) rather than family background may be responsible for the bulk of color differences in the socioeconomic status of African American men.


Jensen discussed it in The g Factor, p. 481:

Skin Color and IQ. Earlier researchers relied on objective measures of skin color as an index of the amount of African/European admixture. In sixteen out of the eighteen studies of the IQ of American blacks in which skin color was measured, the correlations between lightness of skin color and test scores were positive (ranging from +.12 to +.30).[67]

Although these positive correlations theoretically might well reflect the pro­ portion of Caucasian genes affecting IQ in the hybrid blacks, they are weak evidence, because skin color is confounded with social attitudes that may influ­ ence IQ or its educational and occupational correlates. It is more likely that the correlations are the result of cross-assortative mating for skin color and IQ, which would cause these variables to be correlated in the black population. (There is no doubt that assortative mating for skin color has taken place in the black population.) The same is of course true for the other visible racial char­ acteristics that may be correlated with IQ. If, in the black population, lighter skin color (or a generally more Caucasoid appearance) and higher IQ (or its correlates: education, occupation, SES) are both considered desirable in a mate, they will be subject to assortative mating and to cross-assortative mating for the two characteristics, and the offspring would therefore tend to possess both char­ acteristics. But any IQ-enhancing genes are as likely to have come from the African as from the European ancestors of the hybrid descendants.

In general, skin color and the other visible physical aspects of racial differ­ ences are unpromising variables for research aimed at reducing the heredity- environment uncertainty of the causal basis of the average W-B difference in g.


The [67] note is:

67. A review of these studies and complete references to them are provided in Jensen, 1973, pp. 219-230.


Jensen 1973 is Educability and Group Differences.

---

I'd like to do MCV on the skin color x IQ correlation. The genetic models clearly predicts g-loaded Jensen coefficient, while it is unclear what environmental models predict. If it is confirmed to be positive, this constitutes evidence for genetic models.
I'd like to do MCV on the skin color x IQ correlation. The genetic models clearly predicts g-loaded Jensen coefficient, while it is unclear what environmental models predict. If it is confirmed to be positive, this constitutes evidence for genetic models.


Where have you been? In a series of HV posts, I tested and largely disconfirmed IQ-colorism. I've just been really ill, so I haven't had a chance to formally write up the results: there's a strong Jensen effect on the color-IQ correlation, differences are mostly between families, differences are longitudinally stable, SES outcome-color associations can largely be explained by adolescent IQ differences, color covaries with self and parental reported racial ancestry and both covary with IQ, etc.

http://humanvarieties.org/2013/04/07/color-differences-corrections-and-further-analysis-part-2/
http://humanvarieties.org/2013/03/25/pigmentocracy-a-longitudinal-approach/
http://humanvarieties.org/2013/02/02/more-than-just-colorism-part-1/
http://humanvarieties.org/2013/03/06/color-differences-ubiquitous-yet-understudied/
http://humanvarieties.org/2013/02/22/colorism-in-america-1/

Unfortunately, as I said, I never got around to publishing the results. (Though I made sure to send them off to notable colorists.) I should note that while a colorism-IQ hypothesis can readily be ruled out, it's more difficult to rule out a shared environmental + exogamy one and even more so to rule out an assortative exogmany one.

Whatever the case, one step at a time.

(A couple of samples do have color, cognitive, and genotypic data, such as the CARDIA study, so one could in principle do more sophisticated analyses. There's no shortage of data, just access to it.)
What would be most significant, if this data exists, is if genomic ancestry correlates with IQ even when taking into account all visible measures of ancestry.


Since, in the AA population, the correlation between skin color and ancestry alone is around 0.45, I imagine that the correlation between a principle factor of all anthropometric indexes of continental ancestry and genotypic ancestry would be high. In short, you would run into a colinearity problem.

You can rule out colorism (broadly constructed) other ways, though -- for example by looking at sib data. Anyways, one thing at a time. Can someone here draft an email requesting clarification about methodology and supporting data. My mind has been rather fuzzy of late, so everything I write comes out extra dopey; thus, it would be best if I didn't author the request.

Post the draft here.

Thanks.
Admin
I'd like to do MCV on the skin color x IQ correlation. The genetic models clearly predicts g-loaded Jensen coefficient, while it is unclear what environmental models predict. If it is confirmed to be positive, this constitutes evidence for genetic models.


Where have you been? In a series of HV posts, I tested and largely disconfirmed IQ-colorism. I've just been really ill, so I haven't had a chance to formally write up the results: there's a strong Jensen effect on the color-IQ correlation, differences are mostly between families, differences are longitudinally stable, SES outcome-color associations can largely be explained by adolescent IQ differences, color covaries with self and parental reported racial ancestry and both covary with IQ, etc.

http://humanvarieties.org/2013/04/07/color-differences-corrections-and-further-analysis-part-2/
http://humanvarieties.org/2013/03/25/pigmentocracy-a-longitudinal-approach/
http://humanvarieties.org/2013/02/02/more-than-just-colorism-part-1/
http://humanvarieties.org/2013/03/06/color-differences-ubiquitous-yet-understudied/
http://humanvarieties.org/2013/02/22/colorism-in-america-1/

Unfortunately, as I said, I never got around to publishing the results. (Though I made sure to send them off to notable colorists.) I should note that while a colorism-IQ hypothesis can readily be ruled out, it's more difficult to rule out a shared environmental + exogamy one and even more so to rule out an assortative exogmany one.

Whatever the case, one step at a time.

(A couple of samples do have color, cognitive, and genotypic data, such as the CARDIA study, so one could in principle do more sophisticated analyses. There's no shortage of data, just access to it.)


I didn't read all of the skin color posts. So if I read that, I most have forgotten. But those results should be written up for a journal. They are worth publishing.
But those results should be written up for a journal. They are worth publishing.


We need to prioritize.

Right now I am reviewing genomic ancestry x education studies. There are dozens of them and they don't always show the predicted correlations, at least when it comes to Amerindian ancestry. As for genomic ancestry and cognitive ability, here was another study in which the authors could have checked for an association (assuming that they didn't):

Raj, T., Chibnik, L., McCabe, C., Stranger, B., Barnes, L. L., Weir, D., ... & De Jager, P. (2013). Genome-wide association study and admixture mapping of age-related cognitive decline in African-Americans. Alzheimer's & Dementia: The Journal of the Alzheimer's Association, 9(4), P678-P679.
Someone here talked about pigmentocracy and the fact that lighter skin color advantage can be explained by social theories. There is one way to debunk this; by correlating sibling difference in IQ/income/education with skin color. Such result has been reported for the NLSY97. The correlation was clearly stronger for between family differences, but modest in within-family. If color itself was the leading cause of SES differences, I would expect meaningful correlation within families as well.
http://humanvarieties.org/2013/02/22/colorism-in-america-1/

If someone wants it, I can publish that and improve the result reported in the link above. For example, if there is enough data on income, I can use latent variable on incomes measured at 3 different points in time. I can also apply error-in-variables regression, available in Stata but not SPSS and R. This allows you to correct for measurement errors by assuming a certain level of reliability (e.g., 0.7, 0.8, 0.9).
If someone wants it, I can publish that and improve the result reported in the link above. For example, if there is enough data on income, I can use latent variable on incomes measured at 3 different points in time. I can also apply error-in-variables regression, available in Stata but not SPSS and R. This allows you to correct for measurement errors by assuming a certain level of reliability (e.g., 0.7, 0.8, 0.9).


If you could, that would be great; if you recall, it did take a long time to write the syntax.
Admin
I didn't read all of the skin color posts. So if I read that, I most have forgotten. But those results should be written up for a journal. They are worth publishing.


Is this another example of something that you wouldn't want to publish in a legacy journal? The colorism hypothesis is so prevalent, it really deserves to be debunked in a place where advocates of that hypothesis will see it. I think you ought to weigh your dislike of Elsevier journals against how much more of an impact this paper will have if it's published there.


It is better not to publish in Elsevier. Elsevier is really, really bad for science. By publishing results in OP (or other proper open science journals), one helps change the publishing climate. A change is needed. One will have to get the attention of other researchers through other means that I listed in another thread:

To improve the impact factor of the metajournal, one will have to do some marketing. One idea is to send an email to all possibly interested colleagues at the publication of one's paper. I have done this once and will continue with the practice. I also tweet every time a paper is published (including others' papers). Additionally, I blog my own papers. All authors should do this.

Further steps that authors can take is to put their papers on all social networking sites for academics. I signed up for most of them: ResearchGate, Google Scholar, ORCID. One can also post one's papers on discussion groups e.g. on Reddit or Facebook/Google+. Finally, always have a personal website with a publication list with free PDFs.
Admin
I don't there is any chance to educate politicians about differential psychology anytime soon. That is more of a long-term goal, mostly unrealistic. It is even more unrealistic in the USA. Luckily, I don't live there. :) (Neither does Piffer.) To attempt to sway policy-makers opinion by publishing in legacy journals (i.e. Elsevier) is a very poor reason. They don't care about journal articles published in Elsevier journals either.

There are also selfish reasons for publishing in open journals, e.g. there is evidence that papers in open journals get cited more.

Many people like to be pragmatic, but I'm more idealistic. Someone has to push the issue for something to happen.
Admin
The views of policy-makers are reflective of the consensus among academics, although they sometimes lag behind it. Global warming is a good example. Politicians in the United States used to be reluctant to accept that humans were contributing to global warming, but as the academic consensus in favor of that idea has become more firmly-established, politicians have gradually been coming to accept it as well.


Climate science has the benefit (in this case) of being associated with leftist politics, so it gets an automatic boost from academia (since academia is largely leftist). DP and BG (diff psych and behav gen), on the other hand, are associated with conservative, nationalist and economically liberal political views, which are held by a minority of scientists. In other words, the bias is against it. I don't see the results of DP and BG becoming mainstream anytime soon. I have plenty of experience with Danish students from many faculties, and almost none of them are open to these ideas.

Nature is run by the Nature Publishing Group, who also runs Frontiers. Basically, they are making money both off legacy journals, as well as the Gold OA journals. Brilliant business move, bad for science. PLoS has insane publication fees. The mainstream multidisciplinary one is PLoS ONE IIRC, which charges 1350 USD.

There is not so much to write about. The issue has been settled for 3.4 decades (since 1980). The paper you link to cites no references. It is just humanities nonsense published in nature. Whenever I see stuff like that I always think of the attached picture.

I am afraid I don't see much to gain from publishing in mainstream journals, aside from a huge dent in my bank account and some academic prestige to get me a PHD. I'll try to get a PHD without.

Incidentally, I just finished reading this book, The Digital Scholar: How Technology is Transforming Academic Practice. Perhaps it can explain where I am coming from.
The fact that Nature commissions inanity such as this: http://www.nature.com/nature/journal/v513/n7518/full/513306a.html is evidence that they could not care less what is true or not (when it comes to certain topics). I think the main reason for Nature's very dumb views on race differences are 99% due to ideology + 1% business sense. Showing them a few papers by Jensen is no cure for this.

A better way to change people's minds would rather be to convince them that race realism isn't going to have all the negative consequences they fear. For examples of what I am talking about see Jonathan Haidt* and his book on morals and politics.

*You should read him anyways. After Mller and Pinker, he is about the only racial heriditarian who is "respectable". See his Frostian article here for an example of HBD in public:
http://edge.org/response-detail/10376
It is not true that the only way to get noticed is to publish in top journals, although it surely helps. Sometimes, I email simultaneously several people. They rarely respond, but those who did usually say I ask good questions. Some people even asked me what's my diplomas on this field even if I have none of these (sometime, psychology, sometime economics, depending on the people with which I am talking to). If they are ready to take you seriously, and read your stuff, that does not mean they are ready to publish at OP. I guess writing articles take time and energy. For this reason, they certainly prefer to publish in renowned journals, for glory and all. Even if the journal in question asks them to pay a little bit for publication. (speaking about paying, I have heard that the Duyme et al. 1999 "How can we boost IQs of "dull children"?: A late adoption study" had been refused first, but they paid extra money so they get published -- I don't know if that story is true however. It's just what I have read somewhere.)

---

Emil, it seems to me that "liberal" and "leftist" have the same meaning in the USA.
Admin
For those that didn't click Gilfoyle's link from before:

A full-throated, intellectually rigorous anti-racism must critically assess both biological and cultural evidence about race. It must acknowledge that no work on race science can be free of ideology — and, precisely for that reason, it must not place historical actors before a moral green screen showing an image of contemporary values. Rather, it must set the stage for each scene with meticulous, empathetic historical detail. Such work would allow the scientific study of 'racial superiority' — inherently grounded in subjectivity and bias — to fall on its own sword.


Antiscientific sentiments in Nature? Okay then..
Admin
Perhaps you can convince Pinker to get others to examine the method and results. After all, they are public. Since he has a lot of influence, perhaps he can set up an academic group of anonymous scholars (so they can be honest) that can examine it.
1