I don't understand why people continue to do IAT tests. One of the core authors of the original IAT paper released a paper last year showing how the test doesn't actually measure bias, and how it is not clear what this kind of test actually measures.<p>Brian Nosek who is one of the co-authors of the IAT test just released a paper [1] that show that changes in implicit bias don’t lead to changes in behaviour. You can find a summary of the article and its findings here [2].<p>The paper examining 499 studies over 20 years involving 80,859 participants that used the IAT and other, similar measures. They discovered two things:<p>* One is that the correlation between implicit bias and discriminatory behavior appears weaker than previously thought.
* They also conclude that there is very little evidence that changes in implicit bias have anything to do with changes in a person’s behavior.<p>These findings, they write, "produce a challenge for this area of research." The finding that changes in implicit bias don’t lead to changes in behavior, Forscher says, "should be stunning."<p>"I see implicit bias as a potential means to an end, something that tells us what to do and some possible remedies for what we see in the world," Forscher says. "So if there’s little evidence to show that changing implicit bias is a useful way of changing those behaviors, my next question is ‘What should we do?’"<p>[1] <a href="https://www.researchgate.net/publication/308926636_A_Meta-Analysis_of_Change_in_Implicit_Bias" rel="nofollow">https://www.researchgate.net/publication/308926636_A_Meta-An...</a><p>[2] <a href="https://www.chronicle.com/article/Can-We-Really-Measure-Implicit/238807" rel="nofollow">https://www.chronicle.com/article/Can-We-Really-Measure-Impl...</a>
Is it like dark matter in cosmology? Except that here, "scientists" observe a social galaxy and note that it's round. According to their beliefs, the galaxy must be rectangular, so they had to postulate an invisible ubiquitous force that makes the galaxy round. They called this dark force many names: inequity, implicit bias, structural racism.
Modern day Lysenkoism. Compromising science for ideology, any ideology, is one of the more destructive things a society can do to itself. Not only does it halt progress, it destroys the credibility of the scientific community.
I’m really not impressed by hacker new’s scholarship on this. Here’s a paragraph from one of the papers cited here. Basically one of the key arguments in this paper is that basis might be justified. Ignoring for a second that biase in real humans is much strong than differences between races, it also is beside the point. Race isn’t something people can control. If people at large are classifying people based on it that will legitimately act to keep the situation locked in place. If I think everyone with red hair is a criminal and act like it people with red hair will have pretty bad lives and a lot more of them will end up turning to crime<p>> Cultural knowledge and accuracy as sources of implicit associations.This analysis has akinship to that of one of the earliest critiques of the IAT (Arkes & Tetlock, 2004). They argued that,rather than reflecting prejudice, the race IAT might reflect knowledge of cultural stereotypes. Our argument, however, goes further. The associations tapped by the IAT may reflect not just cultural stereotypes, but implicit cognitive registration of regularities and realities of the social environment.Consistent with this analysis, IAT response times were faster when stimuli corresponded to ecologically valid base rates (Bluemke & Fiedler, 2009). Even evaluations of groups may come from knowledge of those groups’ social conditions (Payne et al., 2017). For example, groups disproportionately living in unpleasant conditions (such as poverty, crime, rundown housing) may be associated with “unpleasant” more than other groups
The mentioned article appeared in the "News" section of Science, not the "Research" section. There are very different standards for those, this is not a peer-reviewed paper. The snarky remark in the first sentence is a bit misplaced here as this is actually the section of Science that is more like a magazine than a scientific journal.
> those who study implicit bias already assume it exists, and thus will continue indefinitely to find it.<p><a href="https://en.wikipedia.org/wiki/Reality_tunnel" rel="nofollow">https://en.wikipedia.org/wiki/Reality_tunnel</a>
A few questions that are never really addressed in these discussions about "bias", "discrimination", "race", etc.:<p>What <i>is</i> discrimination at a fundamental level? The term originally means to perceive something as validly distinct - e.g. to discriminate between truthhood and falsity, between up and down, between black and white. The term is most often now used with the implicit qualification of "unjust", "unwarranted", "unreal" or "quasi". So-called "racial discrimination" is a charge made when there appears to be some kind of injustice in the evaluation. So does this mean the mark of racial/ethnic/(even religious)/whatever stock is to be totally unrecognized as a heuristic in all judgments? Or just some? If just some, which ones? Why? Where is the line drawn? How many cycles of flawed attempts at psychological programming (the unintended consequences of which are perhaps not even entirely observed or taken into consideration) do we need to go through before ditching this endeavour? As it pertains to law enforcement, is it easier to simply admit that our current system can't be reformed through psy-ops, and that maybe we need to revert to a more subsidiarian/local principle of security in general?