Jane Jones vs Latoya Jones.
Latoya gets fewer interview offers. That made the news. But that part of the research didn't make it past peer review. The tentative conclusion was race was what mattered. But it was kicked back for lack of a control: Was it race or naming conventions that mattered? They were simultaneously checking both, but claiming it was just one. Seems silly, but it's an important point.
So there was a bit of a re-do and they expanded the study:
Name: Jane Jones
Race: White
vs
Name: Jane Jones
Race: Black
vs
Name: Latoya Jones
Race: Black
If it's race that's at issue, you'd expect the two black applicants to get fewer interview offers. Perhaps the numbers wouldn't be radically different because then a lawsuit might discover something that is clearly racism, but you'd expect to see some sort of difference. Instead, the two Janes got about the same number, while the Latoyas got essentially the same results as in the first part of the study. That caused the researchers to reject their earlier conclusion, even though the MSM drew precisely that conclusion.
Instead the researchers had to conclude "racism," but pushed it back to a race/culture correlation, with some interviews that suggested as much. "Jane Jones" was likely to fit in with office culture; "Latoya" was more likely to not fit in with office culture.
Going back over the original data they realized that this was what they'd observed in their first data set. And, no, those weren't the names used; a variety of names were used, but pretty much everybody had a fairly strong intuition as to whether a name was probably almost only white, common, or pretty much only African-American. "Buffy" vs "Jaela", "Phoenix" vs "Jaquon", hence the (flawed) design study.
I'd liked to have seen a "Malika Jones (race: white)" as an option, but I don't recall that being in the research protocols.
Other studies show a lot of implicit bias as well. A study in Colorado that I read years ago looked at police and how quickly they pulled the trigger. Some were big city police, some small town or rural police; the images they had to evaluate were black or white, young or old, male or female. They found urban/rural didn't much matter; training did. They also found that civilians, black, white, male, female, young or old were more likely to pull the trigger than any group of police. As part of the follow-up, police groups that were outliers underwent training for several weeks and "lost" part of their "racism" as far as willingness to shoot young black males. While the researchers still said "racism," they had to acknowledge that the statistics also said that the police attitudes weren't all racist but also grounded in crime statistics. Like it or not, people are often still considered to be part of a group and to share group characteristics. In fact, some people insist on being seen as part of a group.
Another study looked at fear response using fNMR. You'd have your brain imaged as pictures were flashed before you. If you were white, you showed increase fear reactions when shown a picture of a young black male. That could be mitigated by putting him in a suit and tie, but not eliminated. Young white males in "gansta" clothing didn't trigger the same levels of fear. This study was also bounced in peer review because there was no control--all the subjects were white. The control group chosen was a heterogeneous group of African-Americans matched to the white subject group. The conclusion was thrown out because the black controls patterned as did the white subjects. This was in the late '90s or early '00s near USC, and they had a black gang problem at the time. The subjects and controls were mostly local and all had generalized the same fears. So the researchers, wedded to the idea that it was racism and not just generalization based on how humans interpret statistics (which is to say, badly) pushed the locus of the racism to the news and media, so that the white-dominated media was racist and the AA community had internalized white racist norms. That made it past peer review because that's what people wanted and still want to think. The alternative response, however, was quickly suggested and there was no defense against it except, "Um ... Let's not say that."
in future: discourage racism, and make it harder to get access to guns.
= new reply since forum marked as read
