Interracial Interaction

Issues>Outcomes>Interracial Interaction

How does implicit bias shape the way we interact with people of a different race?  And can it predict how we’ll behave?  John F. Dovidio, Kerry Kawakami and Samuel L. Gaertner sought the answers to these questions in their 2002 study Implicit and Explicit Prejudice and Interracial Interaction. In it, they examined the ways that explicit and implicit attitudes shape the ways we interact.  They hypothesized that the Implicit Association Test would be a good way of predicting the non-verbal ways that white people interact with African Americans (such as body language, posture, facial expressions etc.) while explicit tests would accurately predict how people interact verbally and- as described in detail below- the study confirmed the hypothesis.

The researchers directed  forty students to fill out an explicit bias test—Brigham’s (1993) 20-item Attitudes Toward Blacks Scale—at the beginning of their academic semester and several weeks later the same students all took an IAT focusing on images of African American or White faces and positive and negative words.

During the next phase, the students were  told they would participate in a set of conversations in an “acquaintance process study”.  The students were informed that they’d be talking with other participants in the study (though the other people were confederates of the testers). The subjects had two conversations, both about dating, one with a White confederate and one with a Black confederate. The conversations were videotaped with the cameras angled in such a way that the race of the confederate was not visible. The testers observed the interactions without seeing the confederate’s race and rated the body language of the student for friendliness.

After the conversations, the confederates and the subjects filled out an “impression questionnaire”, rating the verbal friendliness of the encounter on a scale of 1 to 7. behaved”.

By crunching the numbers, Dovidio found a strong correlation between explicit measures of bias and verbal friendliness.  At the same time, the IAT was shown to correlate strongly with the level of body-language friendliness the testers observed on the race-blind video tapes.  In other words, implicit measures of bias accurately predicted the physical cues a student would exhibit, while white students’ self reports about their attidues toward blacks correlated with the friendliness of their language only.