Implicit Association Test

[ahegeman 0]

Here's a link to the "Implicit Association Test" project at Harvard. These are tests designed to highlight unconscious biases you may have in various areas such as race, gender, age, presidents, etc. I've only taken the Race IAT, and found it interesting, though not necessarily accurate. Any statements I make below are based on just that test. I've tried to take some of the others but haven't gotten them to work. I suspect the site is perhaps a bit overloaded with people linking over from wherever I found it posted, though I cannot remember where that was.

If you are interested, you may want to take the tests before reading the rest of my post as it may screw up your testing.

Its a timed test that requires you to respond to flashed images and words by placing them into one of two categories quickly. In the Race IAT, you will be asked to quickly categorize "bad" words and "good" words, and pictures of black faces and white faces by pressing buttons on your computer. If you get the categorization wrong, the test lets you know so you can fix it. There is a control section where they simply test your ability to categorize things by first categorizing just "bad" words vs "good" words, and just black faces vs. white faces. The "e" and "i" keys represent the categories.

The meat of the test, however, is in the last two sections. There you will be asked to categorize the same words and pictures one after the other, randomly mixed together. In one of the last two sections, the "e" key is for "bad" words and black faces, and the "i" key is for good words and white faces. In the other, it is partially switched, with the "e" key for "good" words and black faces, and the "i" key for "bad" words and white faces.

Wanting to answer as quickly as possible, you will find it easiest to do the categorization by saying to yourself, "OK, for this part black equals good and white equals bad", then for the next part, you will proceed under the association that black equals bad and white equals good. I suspect comparing how well you perform through each section is the basis for their comparison. If you perform the categorizations well when you operate under the association black=bad and white=good, then you make that association easier and thus have a preference for white. Supposedly I have a slight preference for black, and I think that result is because I got 3 categorizations wrong under the white=good scheme, and 1 categorization wrong under the white=bad scheme, so I didn't perform as well when I had to say to myself that black is "bad".

I'm not sure its an entirely solid methodology because the last section is where you have to make the black=bad association, and that's where I became aware of how the testing worked, and its probably pretty useless if you know what's going on. Perhaps they randomize the order of the sections - I don't know since I've only taken the test once. Despite what billvon may think, I like to think I have no preference either way, but of course that's the whole point of the test. I found it interesting, though.

edited to add: Dang. Sorry for the long post.

---------------------------------------------------------------
There is a fine line between 'hobby' and 'mental illness'.
--Dave Barry

[Cornholio 0]

I don't find the method for that test to be accurate.
For instance, you can use anything in place of the two choices. It's your brain that get's the two mixed up.

For instance, if they ran the test the same way with the types of cars : compact or truck, and then good and bad feelings. They same results will come out. For instance, if they reversed the order of the testing, the same results will come out. You get used to thinking/associating one way, then they mix it up. It has nothing to do with race. IT has to do with how your brain processes information..


Interesting test though.

Butthead: Whoa! Burritos for breakfast!
Beavis: Yeah! Yeah! Cool!
bellyflier on the dz.com hybrid record jump

[ahegeman 0]

That's a good point. You get used to making one association, then they switch it up and you get screwed up on whichever one comes second.

---------------------------------------------------------------
There is a fine line between 'hobby' and 'mental illness'.
--Dave Barry

[Nightingale 0]

Your data suggest little or no automatic preference for Young relative to Old

Your data suggest little or no automatic preference for GAY PEOPLE relative to STRAIGHT PEOPLE

Your data suggest a strong automatic preference for Kerry relative to Bush


Well, duh. i could've just told them that.

[3ringheathen 0]

The issues you bring up are controlled for in the experiments. All possible pairings and the order in which the tests are administered to the subjects are randomized.

-Josh

If you have time to panic, you have time to do something more productive. -Me*
*Ron has accused me of plagiarizing this quote. He attributes it to Douglas Adams.

[livendive 8]

I signed up and just did the first randomly assigned subject. Apparently I have a slight preference for chaos versus order.

Blues,
Dave

"I AM A PROFESSIONAL EXTREME ATHLETE!"
(drink Mountain Dew)