The Implicit Association Test (IAT) is a cool little experimental procedure for testing for unconscious associations people may have between certain concepts. It's been used in particular in research on prejudice. The participant in the experiment sits at a computer and is required to sort various word and/or pictures into one of two categories as fast as possible. For example, one might sort of a series of words (happy, sad, angry, cheerful, fantastic, murder) into the categories of either 'good' or 'bad', or sort various faces into the categories of either 'white' or 'black. The clever part of the test, however, is when the participant is required to sort a combination of things (e.g., words and faces) into various pairings of categories (e.g., 'black OR good', 'black OR bad', 'white OR good', 'white OR bad'). The amazing thing is that even amongst those desperately trying not to be racist, reaction times for the prejudiced pairings (e.g., 'black OR bad' and 'white OR good') are faster than for the other pairings. The theory is that pairing 'black' with 'bad' and 'white' with 'good' is easier, even amongst people who abhors racism, because we've all to some degree learned to associate those two concepts in our culture...it's an 'implicit association'.
There's a great little interview over at Edge with Mahzarin Banaji and Anthony Greenwald, inventors of the IAT. Check it out.
There's also a website over at Harvard that lets you take the IAT yourself. I highly recommend it. (Follow the 'demonstrations' link to start with). There's a number of different versions you can take, including ones designed to test for predjudices regarding religion, sexuality, politics, race, gender, disability etc.
I thought I'd give the disability version a go. I was presented with various symbols related to either disability (wheelchairs, crutches etc.) or ablebodiedness (skiing, running, etc.) as well as positive and negative words. My results:
"Your data suggest a slight automatic preference for Abled Persons compared to Disabled Persons."
This, despite the fact that I myself have quite the cripple credentials, having a physical disability myself. ;-) Wow!
And here's how others did:
This kind of stuff is experimental psychology at its best.
Wednesday, February 13, 2008
Predjudice at the Puddle: results of the Implicit Association Test
Subscribe to:
Post Comments (Atom)
4 comments:
Self loather! ;)
Be the wind beneath your wings, dont wait for Bette Middler to come to your rescue.
Honestly, when I took it (the one about fat vs. skinny) I had two major problems. One, that I couldn't always tell what they were trying to show in the pictures, since they were all distorted; Two, that it seemed like the order of the pictures was biased to influence the outcome. I talked to someone else about this and he said they are supposed to be random, so I went back and took another one (now I don't remember which one) and noticed the same thing.
I think it's a great idea, it's the execution that seemed unconvincing. I kept thinking of my (incredibly prejudiced) advisor, and how he would never have been enlightened by something with such obvious flaws in the experimental design.
Hmmm. I suspect that the implementation for the web might be have been oversimplified for technical reasons. In the lab I'm pretty sure it is supposed to be randomised, and I think participants are supposed to do quite a few more trials (i.e, hundreds and hundreds).
I went to the site and got bored. Does that mean I've got ADD?
Post a Comment