"There is a social scientific sort of Holy Trinity. Stereotype Threat is one piece. The other two are Microaggressions and Implicit Bias. And the scientific foundation on which all three are built is exceedingly shaky. In all three cases, the narratives and real world applications are based on highly dubious and uncertain science. And there has been a slow-moving walking back of the extraordinary and extreme claims in all three areas over the last like 10 or 15 years." verbatim quote from a talk by Lee Jussim, "He Comes to Abolish Social Psychology."
Dr. Jussim leads the Social Perception Laboratory at Rutgers. He is also chair of the Dept of Psychology. The credentials are particularly important because these pillars of Social Psychology are his specific area of research and expertise - and he is saying the field has wildly overstepped its real knowledge. He also notes that he considers himself to be a moderate Democrat, but is increasingly being called a right-winger within his field, and sees the same happening to others. He finds it amusing about himself, but worries the effect this name-calling has on others.
This lack of evidence includes trumpeting very modest effects for purposes of publication even when one's own research shows that other factors have a more robust effect. There is some research that shows that identical resumes get treated differently when they are attributed to someone with a white-sounding vs black-sounding name. But the effects are small and uneven, while the factors that employers did respond to, such as degrees, experience, and personal characteristics were "gigantic, some of the largest effects in the social sciences." Right there in the papers that most heavily lean on the the idea that Implicit Bias is real is the evidence that other factors are far more real. Jussim also noted that accepting studies with very small sample sizes, especially when there are similar studies based on ten or a hundred times more individuals is shoddy science.
Stereotype Bias is the idea that African-Americans or women perform worse in some real life when they are given discouraging messages that blacks or women don't do this or that as well as whites or men. The research behind this is not even that they do worse in academic courses because they have these negative thoughts, because that is very hard to test. They do worse in some areas, and so it is assumed that this must be why. I still haven't gotten to the research parts yet. What they find they can examine is the results of short-duration tests, before which they tell some of the students that their group does not usually do well in this subject.
A small effect was found at the lower extremes, but not among the majority. The blacks and women who had scored as well as the whites and men on initial tests were not affected by the discouraging language. Among those who were already scoring poorly, the discouragement seems to have exaggerated the effect and made them a little worse. But the reporting was done with some sleight-of-hand. In the topics that they targeted, such as women taking tests in math, they found that the subsequent overall scores were lower. Yet they had been lower before the Stereotype Threat test - this was not highlighted. Similarly, African-Americans got lower scores on the tests after receiving the Threat. But they got lower scores before the threats too. When you pull those numbers back to apples-to-apples, the remaining effect is very small indeed.
Lee Jussim believes that if truth-seeking does not come first, then social justice can never result and we are just spinning our wheels. That interestingly echoes CS Lewis's First and Second Things. which itself echoes the Gospel of Matthew "Seek ye first the kingdom of God and his righteousness, and all these things will be added to you."
"If you prioritise truth-seeking, as a scientist ...everything follows from truth-seeking, including justice and social justice. If you try to build social justice on false stories and narratives you end up with the Soviet Union and the Cultural Revolution."(Jussim again, same talk.)