Thursday, March 7, 2013

Implicit Association Test


People often fail to share or reveal their true attitudes and beliefs either because they are unable or unwilling to do so. As a result, the Implicit Association Test, or IAT, was developed to evaluate these hidden attitudes by measuring the degree to which specific concepts are associated (Greenwald, McGhee & Schwartz, 1998). The way that the IAT works is that you are presented with an image or word in the center of a computer screen and instructed to assign that word or image to a particular category (located in the upper corners of the screen). This choice is meant to be done quickly so as to measure the automatic responses and thus attitudes towards the categories being tested. For example, in the Young-Old IAT, I was presented with images depicting faces of older people and those of younger people as well as various positive words (joy, glorious, laughter, etc.) or negative words (hurt, evil, etc.). Using the “e” and “i” keys I quickly assigned the words and images to different categories. The categories are presented in various orders and combinations which tests the participant’s ability to learn new pairings. The idea behind this procedure is that easier pairings (denoted by quicker responses) represent a stronger association between two concepts in memory, which should then reflect that person’s true attitudes (Greenwald, McGhee & Schwartz, 1998).
I was definitely a little apprehensive about taking some of these tests simply because the thought of getting results that are inconsistent with my perceived attitudes and beliefs is bothersome. However, I think the IAT’s that I completed did a fairly decent job of predicting my true attitudes and beliefs toward young and old people as well as the relationship between gender and science. The interpretation of my results for the Young-Old IAT suggested that I have a moderate automatic preference for young compared to old. This wasn’t too terribly surprising to me given that society and the media often portray youth in a more positive light and the elderly in a more negative light. I have also had more negative experiences (experiencing the loss of a family member, going to funerals, dealing with the frustrations that come with teaching some people how to use technology, etc.) with older family members which might have contributed to my performance on this IAT.
My data for the Gender-Science IAT suggested that I have a slight association of male with science and female with liberal arts. This result also does not surprise me all that much given that this bias is fairly prominent in our society which is also reflected in the disproportionate gender representation in careers associated with science (which tend to have more men) and liberal arts (which tend to have more women). Yet, my bias seems to be limited which is likely a result of attending a liberal arts university in combination with other factors such as familiarity with female peers majoring in the sciences. 
Taking the different IAT’s was an interesting process in that it brought attention to how easily our attitudes and beliefs can be detected in simple tasks and behaviors. I would say that I was already pretty aware of my perceptions and any biases I may have, but this has definitely provided a new perspective in which to evaluate how I perceive different groups and concepts and has illuminated areas in which I can now work on to reduce or eliminate any prejudices I may possess.


(n= 575)

Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual differences in implicit cognition: The Implicit Association Test. Journal of Personality and Social Psychology, 74, 1464-1480.

1 comment:

  1. So I found your post to be very interesting, considering that it actually prompted me to think about an issue rather than just read through in sustained agreement. In particular, I was caught by your paragraph on Gender-Science IAT, in which you state that a bias for associating men with science more so than with women must be prominent when considering that more academic science positions are held by men (Which I read you to be saying is the result of such a bias being prevalent).
    While this conjecture may be right on, I feel the need to point out that the connection is not necessary. There is actually a pretty interesting debate between Steven Pinker and Elizabeth Spelke on this issue (http://www.edge.org/3rd_culture/debate05/debate05_index.html), with Pinker arguing that an important factor being the individual difference between men and women that are genetic. Let me clarify that Pinker does not state that men are smarter, but rather that men are more likely to have traits like being object-oriented while women are people-oriented, which can explain explain the differences we are talking about. Thus, it is not strange that in psychology, a science focused on people, awards 70% of its PhD's to women (http://www.apa.org/gradpsych/2011/01/cover-men.aspx).
    Now, this whole thing about men being object-oriented and women being people-oriented is also a stereotype and I agree that we should not expect every woman to be people-oriented. However, stereotypes can accurately describe overall groups (just not every member of the group) if they have truth and therefore explain group findings, like women as a whole making up only a minority of academic science positions. However, it is also possible that bias is to blame for the disproportion, and it is also possible that both have influence.

    ReplyDelete