Suppose you are designing a facial recognition system for police to use in the field while investigating a recent crime to see if anyone with a criminal history is nearby.
(Data taken from: https://en.wikipedia.org/wiki/Incarceration_in_the_United_St...) Because blacks are over-represented in the US criminal justice system (40% of the prison population vs 13% of the population) and because part of what defines "black" is the outward appearance of certain facial features, a facial-recognition algorithm which is trained to recognize criminals, with a cost function based on prediction accuracy alone, and facial features as input parameters is likely going to have false positives that over-represent blacks.
It's very important to consider this when you develop a training set. The developer error (who mostly failed to understand Baye's theorem here) might work something like this: They take 100 innocent people's faces at random. (On average it will have only 13 blacks) Then take 100 random criminal faces from inmates. (On average it will have 40 blacks.)
Then mix up the groups into your training set and assign a prediction score 1 or 0 depending on whether or not your classifier has correctly predicted whether or not a face was in the criminal group. Then, based on no other feature than race, your neural net can get better performance based solely on guessing more often that black people are criminals. That's not a good thing. In fact, if it's looking at a black face from its training set, the odds are nearly 2 to 1 that it's one of the criminals, even though the odds that are at least 2 to 1 against a random black person having a criminal history.
The likelihood of being falsely identified as having a criminal history is much greater based on the only variable of being black. And this type of thing has happened several times already in production systems!
Conversely, the same system, trained on the same data in the same training set, can get higher performance than random by simply guessing that any non-hispanic white person does NOT have a criminal history.
Thus, it's pretty important to correct your training set to reflect the correct Bayesian prior, and the underlying structures that sometimes go by the label "structural racism" or "institutional racism" are essentially exactly that reality in this case.
To be honest, I don't think our understanding of these systems is mature enough for us to be just throwing them out into our societal systems right now. There needs to be a lot more testing because the possibility of unethical results is pretty damn high.
That said, I still think that the concept of structural racism is a bad way to look at this problem, as it's simply one form of a common error when looking at sample sets.
It's now leaking into other fields. I remember when I first heard stem be changed to steam to include "arts" a laughed at how inclusive and utterly useless it is.
Also, I think it would marvel you knowing how much things you would call "scientific study" are also heavily politicized.
If this bothers you to a big extent I would recommend you try to find comfort in thinking about postmodern sociology as a religion different than yours. They won't be bothered by it and it will probably fit your mindset in a more soothing way than thinking about them as scientists. It's not that they are trying to publish their findings in ACM TOPLAS or something, they have their own community and books and kinda like it.
There are plenty of statistically viable ways to determine differences between groups. Even a simple T-test is usually sufficient.
The difference of means does not have to be more than the range in either group. Standard deviation is what’s important here. You use a test statistic to determine the relative value of the real mean within a confidence interval. It’s whether these intervals overlap that you can determine a significant difference between groups.
Take the “Sokal 2.0” affair, where some profs sent obviously bogus research to various journals. Notably, while they were able to get “rape culture among dogs at the dog park” (or maybe it was racism, easy to look up) published in a gender studies journal, they couldn’t get published in sociology journals. Sociology has standards. The absurdities committed by gender studies as an institution don’t falsify racism/other forms of oppression. They might be wrong, but you’re certainly not right.
Mostly political and much of it unreproducible. Desperate political academics using non science to argue their opinions and right and citing each other into fake legitimacy.
Garbage studies garbage standards garbage journals from people who couldn't hack it elsewhere.
> Structural racism refers to “the totality of ways in which societies foster [racial] discrimination, via mutually reinforcing [inequitable] systems...(e.g., in housing, education, employment, earnings, benefits, credit, media, health care, criminal justice, etc.) that in turn reinforce discriminatory beliefs, values, and distribution of resources,” reflected in history, culture, and interconnected institutions (Bailey and others, 2017).
I think I might be misunderstanding, but given this includes “culture”, is this so sufficiently broad such that hypothetical scenarios such as this (no idea if this is accurate) would be captured “white people are culturally more likely to use crystal meth than other racial groups, ergo they are victims of (a certain kind of) systemic racism”?
It seems like this is just a catchall for any kind of error associated with a racial group, and the article is merely cautioning against such errors. If so, it begs the questions “why not just say so?” and “why use such a loaded term like systemic racism?”.