I asked my Race and Politics students to take the Harvard Project Implicit test on race. For those not familiar with the test, subjects are asked to place photos of people into a black/white category, then asked to place words in a good/bad category and then asked to do both simultaneously, moving words or pictures to the right or left of the screen. The cognitive trick occurs when respondents are asked to put words into the Black/good category or the White/bad category. The longstanding negative associations associated with “blackness” are well documented.
In Black Visions, Michael Dawson cites 1990 General Social Survey that shows 54% of whites still believes that blacks are less intelligent than whites. According to the project implicit site, 75-80% of self-identified Whites and Asians show an implicit preference for racial White relative to Black.
To my great surprise, my own scores revealed an implict bias towards African-Americans Whites. My results read as follows:
Your data suggest a moderate automatic preference for African American compared to European American.
The overall biases read accordingly:
Apparently my biases run in the opposite direction as predicted by my phenotype. Has anyone else taken this test? What were your findings? Impressions?
Comments 4
kathy — September 18, 2008
I came up with little to no preference btn Black and White people. I'm white. It was a nice surprise; I was assuming I would carry some of the country's biases, like it or not.
But then I took the gender version and came up with a preference for men over women. Oh well.
Khoa Nguyen — September 19, 2008
I was born in a different country so I have never had a racial black and white. The difference between the skin color doesn't really affect me. However, I have watched American movies since I was a kid and ironically, most of the bad guys in those movies are whites. I assume that is probably the reason why in the shooting quiz, I have a faster shooting speed for white guys than for black guys.
Kenneth M. Kambara — September 19, 2008
I had similar results, but while I think this stuff is interesting, I'm not sure about the convergent validity of it all. I get the idea of trying to uncover latent judgments, but I think there a ton of noise going on in this design, from the stimuli to the clustering of the questions. The work of social psychologists like Zajonc have shown that preferences can be linked to exposure, i.e., the "mere exposure" effect or what I call the "Starbucks effect," independent of any attitudes towards the "objects."
Here's on for the Harvard team...what do fMRIs say? If you show stimuli based on race, which parts of the brain light up.
neelix — September 19, 2008
I did take the test and found that it was an interesting way of determining whether someone might have an implicit bias. My results were I had no preference one way or the other, which is accurate to my feelings on the subject (i dont relate bad with black or white nor good with either) Although the test in particular that i had taken was with McCain and Obama to which i have a little discrepancy because some who take that test can be altered results by there political opinions in thinking Obama is wrong on his views or vice versa with McCain and therefore relating the two together on false terms according to the tests intentions. Interesting though.