One of the core authors of the IAT paper administered here to measure implicit bias 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.
Many of us wish that there was a simple link between an unconscious bias, as measured by the popular IAT test administered by Hardvard or similar tests, and biased behavior. This would give a clear path to fix any bias problem. However, 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 behavior. You can find a summary of the article and its findings here [2].
The paper examining 499 studies over 20 years involving 80,859 participants that used the IAT and other, similar measures. They discovered two things: 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. 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."
"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?’"
This is my biggest confusion reading the summary. Since all kinds of inputs can lead to the same measured output (such as being more scared about the entire theme and self-censoring), how would they single out 'implicit bias' and decide to label it as the cause?
"Implicit biases are malleable. Our brains are incredibly complex, and the implicit associations that we have formed can be gradually unlearned through a variety of debiasing techniques."
I don't know... isn't it almost by definition not 'implicit bias' then?
Yes, it is shoddy science or possibly not science at all. Testing an assertion using known measures that have been shown to not test what you want to test does not add to the scientific knowledge and instead subtract by adding noise.
Building upon tests that has not held up to scrutiny without having a good argument for why it is different this time violates how science explores and builds knowledge. Science is a technique that produce a network of knowledge [1], where the core nodes are viewed as facts because they have successfully held up to scrutiny by other pieces of knowledge and has worked as a foundation for other pieces of knowledge.
In science pieces of knowledge that builds upon claims that are shown to be invalid, such as the IAT test, are viewed as inheriting the problems of the knowledge it builds upon.
Comments
One of the core authors of the IAT paper administered here to measure implicit bias 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.
Many of us wish that there was a simple link between an unconscious bias, as measured by the popular IAT test administered by Hardvard or similar tests, and biased behavior. This would give a clear path to fix any bias problem. However, 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 behavior. You can find a summary of the article and its findings here [2].
The paper examining 499 studies over 20 years involving 80,859 participants that used the IAT and other, similar measures. They discovered two things: 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. 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."
"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?’"
[1] https://www.researchgate.net/publication/308926636_A_Meta-An...
[2] https://www.chronicle.com/article/Can-We-Really-Measure-Impl...
This is my biggest confusion reading the summary. Since all kinds of inputs can lead to the same measured output (such as being more scared about the entire theme and self-censoring), how would they single out 'implicit bias' and decide to label it as the cause?
Also http://kirwaninstitute.osu.edu/research/understanding-implic...
"Implicit biases are malleable. Our brains are incredibly complex, and the implicit associations that we have formed can be gradually unlearned through a variety of debiasing techniques."
I don't know... isn't it almost by definition not 'implicit bias' then?
Yes, it is shoddy science or possibly not science at all. Testing an assertion using known measures that have been shown to not test what you want to test does not add to the scientific knowledge and instead subtract by adding noise.
Building upon tests that has not held up to scrutiny without having a good argument for why it is different this time violates how science explores and builds knowledge. Science is a technique that produce a network of knowledge [1], where the core nodes are viewed as facts because they have successfully held up to scrutiny by other pieces of knowledge and has worked as a foundation for other pieces of knowledge.
In science pieces of knowledge that builds upon claims that are shown to be invalid, such as the IAT test, are viewed as inheriting the problems of the knowledge it builds upon.
[1] https://www.youtube.com/watch?v=bxdBRKmPhe4&feature=youtu.be...