Biased? Probably.

January 22, 2021

Ever left an Airbnb without waving a friendly good-bye to the hosts?  If you’re white, probably no big deal.  If you’re Black, well the hosts may have thought you were trying to hide something.

New research in Policy Insights from the Behavioral and Brain Sciences uses this scenario and others as examples of implicit bias based on a person’s “category,” such as race or gender.

Often people are not aware their impressions of someone are rooted in bias.  And it’s this implicit bias that must be combatted if we are to make progress in addressing discriminatory behaviors, authors Bertram Gawronski, Alison Ledgerwood and Paul W. Eastwick argue in “Implicit Bias and Antidiscrimination Policy.”

Consider two job candidates; one is male and the other female.  In one scenario, the male candidate had more education but less work experience and the female has less education but more work experience.  In another, the qualifications were reversed for the male and female candidates.

In both scenarios the male candidate was preferred, and the interviewers justified their decisions “with whatever qualification made [the male candidate] superior to the female candidate.”

Awareness is essential in judging people for who they are, not based on social group. Once a decision maker knows he or she is exhibiting bias, strategies can be used to combat those discriminatory practices.  In “consider-the-opposite,” an interviewer might deliberate on whether he or she might have made a different choice if the qualifications of the two candidates were swapped. 

It’s hard for people to combat implicit bias on their own, and that is among reasons the authors advocate change at the organizational level — “creat[ing] contexts in which discrimination is less likely to occur.”

“Blinding” decision makers, so that they have no idea of a candidate’s gender, race or other defining trait, have been successful.  But if blinding is not possible, decision makers might agree on “clear specifications” before interviewing job candidates, such as educational attainment. 

Still, bias can figure in.  Suppose qualities selected for identifying candidates include “assertive,” “confident,” and “leadership potential.” Well, those might tend toward a male candidate.  How about adding: “excellent communicator” and “inspires teamwork”– as the authors explain, “so that the resulting list of desired criteria became more balanced.”