Names may trigger hiring bias

Get all the Monitor's headlines by e-mail. Subscribe for free.

E-mail this story Write a letter to the Editor Printer-friendly version del.icio.us [ What is this? ] digg [ What is this? ]

Researchers Marianne Bertrand of the University of Chicago and Sendhil Mullainathan of Massachusetts Institute of Technology sent 5,000 résumés in response to 1,300 job postings in The Boston Globe and Chicago Tribune.

The researchers started by examining birth certificates from the mid-1970s to find common names for blacks and whites.

Common white names included Neil, Emily, Jill, and Anne, while black parents tended to choose names such as Tamika, Ebony, Rasheed, and Kareem.

Using those names to create fictitious résumés, the researchers mailed four résumés to each job vacancy, two low-caliber résumés and two high-caliber ones.

The black- and white-sounding names were evenly divided, so that the only differing factor was the name. The researchers then waited for responses either in the form of a call, letter, or e-mail.

"White" applicants received one response for every 10 résumés mailed, while "black" applicants with equal credentials received one response for every 15 résumés.

"This represents a difference ... that solely can be attributed to name manipulation," says Ms. Bertrand.

Some first names - such as Kareem - could have been assumed to be Muslim, which might have invited another brand of discrimination. But Bertrand says that the last names used - Williams, Jones, and Jackson - were clearly non-Muslim.

"We couldn't find that the Muslim names did worse," she says.

Bertrand says the research doesn't show whether discrimination in job recruitment has waned or grown over the years. "As far as we know," she says, "this could be an improvement."

She says whites tended to be surprised at the results while blacks found that the data "resonated with their personal experience."

As it did with writer and cultural critic Yvonne Bynoe of the Urban Think Tank in Brooklyn, N.Y. Ms. Bynoe, an African-American who has called for federal regulators to perform industry audits similar to the MIT and University of Chicago study, says the most straightforward explanation of the data is "taste-based" discrimination.

"African-Americans, even if they are capable of performing the advertised job and maximizing profits, would be disqualified simply because of their color," she says.

But other, nonracial factors could be involved, says Peter Van Doren of The Cato Institute, a libertarian think tank in Washington. He suggests that government regulation might complicate the picture. "Often the unintended result of regulatory intervention is that employers are less willing to talk to people at all."