Harvard Race Test: Conditioning people to fail
Not too long ago I decided to take the Harvard Race Test (if you’ve never taken it I strongly DO NOT recommend), which requires you to distinguish faces of European and African origin.
I am always quite skeptical of these tests since they are often made to further prove this hypothesis that most Americans have an automatic preference for white over black (the test states that even before you take it and you can even see it from the results pictured), but I still decided to try it out.
At first you are asked to click on a certain key when a white face appears and another key when a black face appears. Okay- I tell myself- ill give this a try. Who knows, maybe it will pick up something in my reaction time and I will learn a little bit more about myself? WRONG.
After this first task you are asked to click on the same keys when you see a ‘good’ word and a ‘bad’ word. I go through the task and evidently both faces and words categories are merged together for the third task. I have a question for you now: Was the bad word linked to the black or white face at first?
a) White face
b) Black face
If you answered b, congratulations! By merging the categories together your brain is conditioned to click faster on bad/black and good/white. This type of test is only designed so that they can self prove their hypothesis! In the fourth task, when these categories change and you have to click on good/black and bad/white you are most likely to make mistakes due to the prior conditioning! As you can see in the illustration, you are first conditioned by putting two categories together and then tested on your ability to dissociate this conditioning- but does this really have to do with your preference for a certain race? Or could it be how difficult it is for you to dissociate categories?
Although I may be completely off (of course I am no ‘expert’ and I am currently criticizing a Harvard test) I do believe that you could get opposite results if you started off the test by conditioning individuals by clicking the same key for good/black and bad/white and then asking them to change.
When I did the test, I was so conscious of what was happening and shocked that this was a way to measure preferences between black and white individuals that I tried my hardest not to become ‘conditioned’. And it worked! My results were the opposite of their hypothesis; I had a slight preference for African origin individuals as compared to European origin individuals.
Having had some exposure to different tests and the construction of tests I find this one quite ‘ridiculous’ if I can use that word- or even quite childish. It is trying to prove a stereotype by conditioning individuals to fail- and I am sure many people easily fall in this trap!
If you want to a look at this test for yourself, be my guest: https://implicit.harvard.edu/implicit/selectatest.html (it is the race IAT)
