Implicit Association Test

[ScienceRoar]

Have you taken the Implicit Association Test? It's super cool way of finding out whether you have unconscious biases, eg. do you associate science more with women or men? Do you associate bodyweight/race etc. with positive or negative words?
Perhaps I'm only excited about it because the first two tests I've taken have yielded proud feminist results (slight association with females and science, and males with liberal arts, and no preference for fat or thin people).

Is this a real thing or did you find it on Everyday Feminism?

ah I have seen these unconscious bias tests. They are fascinating if you're into a bit of self development.

@upstartcrow It's a real thing, hosted by Harvard University and emerging from a collaboration between several US universities. I found it on a philosophy blog. I've heard of the theory before, but never seen one you can take at home.

Is it online?

Yes it's online, we had it round at work.

I didn't do all of them but unfotunately I came out with a slight bias against women who work, even though I am a feminist with kids who works full time

So, bias really is subconscious...

could be this? can't remember was a while ago now

My understanding is that while you are helping aggregate results by taking the test, it doesn't really accurately measure your personal subconscious views on a single taking. There are articles about this. Hang on...

here you go:

www.vox.com/identities/2017/3/7/14637626/implicit-association-test-racism

Oh well that is interesting and a bit of a relief Lang!

I thought fuck me if I have unconscious bias against working women ie bias against myself, then how bad could it be amongst other groups!

Yes it's that! @NothingOnTellyAgain. I can't believe I didn't share the link!

Interesting analysis @LangCleg. It seems that you need to take the test a few times, or as a shortcut, be mindful that you might be one category in either direction compared to your reported result.

The implicit bias test is neither reliable nor valid. Even it's designers don't make that claim anynore.

Social psychology is going through a crisis right now because so much of its foundational research can't be replicated. It's not like the discovery of the Higgs bosun where the chances that the observation are down to chance are billions to one: a typical social psychology experiment considers something proven if there's a 95% chance that the results aren't down to chance - which means one in twenty experiments is horseshit.

Unfortunately journals publish positive results but rarely publish negative ones - if a follow-up experiment 'fails' to get the expected results it is just assumed the researchers did the experiment wrong.

The reason this test is still used is that it tells people what they think they know already: that there a lot of prejudice. If people point out the test is flawed they are accused of denying that prejudice exists. But that's nonsense: you can say horoscopes are nonsense without being accused of denying the existence of stars and planets.

I studied social psychology. It has been alarming watching the field implode.

you can say horoscopes are nonsense without being accused of denying the existence of stars and planets.

Off topic but that’s a fab turn of phrase I shall be appropriating, thanks!