The Implicit Association Test and the Catch-22 of Developing Striking Tests
The Implicit Association Test has gained much attention since the first time it has been introduced by Anthony Greenwald in 1998 in the Journal of Personality and Social Psychology. The IAT measures the implicit attitudes and stereotypes people may hold about certain people or things. This is a complex measure to make, as “implicit” means that even the person who holds the attitude is unaware of this attitude, as it is outside of that person’s consciousness. Therefore it is hard to dispute the results of the test, since the subjects themselves are not aware of the truth.
An example of what the IAT would measure is the conviction a person may have towards homosexual people. The concept is that a response is easier to make when two subjectively related items share the same response key. A subject is shown a photograph of either a homosexual couple or a heterosexual couple and subsequently has to categorize the photo into one of two categories (as seen in the photo below).
The “good” and “bad” label is switched to the opposing side throughout the test. A subject is thought to have an implicit preference for heterosexual people if they respond faster when the phrase “gay people” and the word “bad” share a response key than when “gay people” and the word “good” share the same response key. This would be reversed when categorizing “straight people”. This feedback may then be used as a diagnosis, which I personally regard as problematic. There are many things that can affect results such as handedness, familiarity with the words and pictures, and the opinions of the culture that surrounds us, and therefore the feedback may not be diagnostic of actual behaviour towards homosexual people. An IAT developer, Brian Nosek, has even acknowledged this.
The good news is that professionals that do their research and stay informed are aware of this, and therefore will not take the results of the test as diagnostic. They may view this feedback as an opportunity to inform people about these implicit stereotypes and attitudes that we are unaware of, and how they may affect our actions, but will not treat it as a be-all and end-all of the subjects’ actual behaviour. The bad news is tests such as the IAT tend to get a lot of media and public attention, with tons of features in newspapers and on TV, radio, and the Internet. Once the general public learns of this, the test can easily fall into the wrong hands and may be applied and used as diagnostic by people who do not have the professional experience and knowledge that is required to properly carry out the test and understand its results. This is the catch-22 of psychological testing.
Professional test developers want their tests to gain popularity and attention, but once they do, the tests are more likely to be carried out by the inexperienced, which not only damages the test’s credibility but also the developer’s. So, while the IAT may be an amazing development, in the wrong hands, it may do more harm than good. Subjects may get the “diagnosis” that they are extremely biased against African Americans, and may begin to question themselves morally, when the score may not actually reflect their exact behaviour in real life. This also highlights the importance of validity. In conclusion, it is important to be careful with controversial tests, especially those that have the potential of gaining ample public recognition.