Friday, October 23, 2020

I guess I picked the wrong day to quit reading cognitive trash

It is really rather unfortunate to read Shades of Privilege: The Relationship Between Skin Color and Political Attitudes Among White Americans by Nicole Yadon & Mara C. Ostfeld so soon after having read Science Fictions: How Fraud, Bias, Negligence, and Hype Undermine the Search for Truth by Stuart Ritchie.   

From Yadon and Ostfeld's Abstract:

Shifting racial dynamics in the U.S. have heightened the salience of White racial identity, and a sense that Whites’ social status and resources are no longer secure. At the same time, the growing size of non-White populations has also renewed attention to skin color-based stratification and the potential blurring of racial boundaries. We theorize that Whites with darker skin will be motivated to protect the boundaries of Whiteness due to the loss of status they would face from blurring racial boundaries. Consistent with growing evidence of skin color’s importance for Whites, we demonstrate that darker-skinned Whites—measured via a light-reflectance spectrophotometer—identify more strongly with their White racial identity and are more likely to hold conservative political views on racialized issues than lighter-skinned Whites. Together, these findings offer new insights into the evolving meaning of race and color in American politics.

Clearly a couple of Critical Race Theory (CRT) adherents out on the academic town, tying it on.

Everything Ritchie calls attention to regarding academic impropriety is in evidence here.  Surprisingly, the argument is coherent, believing from CRT, if you do, in the importance of racial thinking, their theory that "Whites with darker skin will be motivated to protect the boundaries of Whiteness due to the loss of status they would face from blurring racial boundaries" makes logical sense.  But everything else is downhill from there.  

I don't have access to the paper itself but from some of their remarks and some footnotes, is seems pretty clear that this is an underpowered, too small and non-random sample, p-hacked, paper using questionable and probably little tested technology ("light-reflectance spectrophotometer"), to find what they want to find.  Almost certainly no pre-registration or any other research leading practice.

Critically, no sample size and no effect size.  Without those, this is meaningless garbage.  Indeed, we are taking on faith that this is even p=0.05 since not even that is affirmed.  

They find that "Whites with darker skin will be motivated to protect the boundaries of Whiteness due to the loss of status they would face from blurring racial boundaries" "identify more strongly with their White racial identity and are more likely to hold conservative political views on racialized issues than lighter-skinned Whites."  Would be useful to know by how much, and what the measure of conservative views might be.  

Importantly, based on the design of the study, this is, as usual for critical theory, a win-win.  If darker skinned whites express more conservative views, then that proves that America is skin color alert.  If White Americans express more conservative views then it at least proves that White Americans are racist as CRT has all along claimed.  Head I win, tails you lose.

The absence of normal critical measures (sample size, sample randomization, effect size, p measure) is enough to dismiss out of hand.  But there is more.  In footnote 11, there is a tell.

This lack of association between skin color and income is surprising. We believe this may be due in part to the student populations in some areas in which we collected data, as well as question wording which asked for one’s annual (as opposed to monthly) income. We observed uncertainty on-site when individuals sought to convert their weekly or monthly income into yearly income.

That they are surprised by a lack of association between skin color and income is an indication of their adherence to CRT.  Under CRT, the system is racist and darker people are unfairly discriminated against based on income.  Plenty of evidence that income is determined overwhelmingly based on behavior, IQ, work practices, education choices, etc.  Race is a minuscule factor.  

Additionally, the fact that participants struggled with their questionnaire wording indicates a lack of study design and testing prior to execution.  

Someone, somewhere, and ultimately it is almost certainly the taxpayer, is paying for low quality slapdash research to support an anti-Enlightenment racist totalitarian ideology.  No wonder academia is struggling.


No comments:

Post a Comment