What's in a name? Study shows that workplace discrimination begins long before the job seeker shows up for an interview

Thinking of naming your child Keisha or Aisha? How about Rasheed or Tremayne? African American pars across the nation may have to think again, as a recent study has shown that workplace discrimination begins long before the job seeker shows up for an interview.

Indeed, it seems to be in play from the moment the resume hits the human resource manager's desk.

Dr. Marianne Bertrand, a professor of economics at the University of Chicago, and Dr. Sendhil Mullainathan, MacArthur-winning associate professor of economics at the Massachusetts Institute of Technology, have made a significant contribution to the research literature with their new study, "Are Emily and Brendan More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination."

With names chosen from birth records in Chicago and Boston, the researchers crafted sets of resumes--some of higher quality, some of lower--labeled them with either "White-sounding" or "Black-sounding" names and sent nearly 5,000 of them out in response to 1,300 jobs advertised in the Chicago and Boston papers.

The response from colleagues as they designed their deceptively simple study was, "'Oh, yes, you'll find a discrimination effect, a reverse discrimination effect,'" Bertrand says.

Instead, they found that resumes with "White-sounding" names--like Jay, Brad, Carrie and Kristen--were 50 percent more likely than those with "Black-sounding" names to receive a callback. The results were striking, holding both for jobs at the lower end of the spectrum--cashier and mailroom clerk positions--and for those at the executive level. Put another way, a White job seeker would have to send out at least 10 resumes to receive a single contact from a potential employer. A Black candidate, meanwhile, would have to send out 15--and this in a "soft" economy with a relatively low rate of new job creation.

The most intriguing--and troubling--aspect of the study was that the discrimination effect held even for candidates with stronger credentials: those who had gone to better schools, or won awards, or had fewer resume "gaps," periods of at least six months without employment.

"We really thought a higher quality resume would help the African American candidate--that the employer would put less weight on the names," Bertrand says.

And indeed, improving the resume quality helped candidates with White-sounding names significantly--their chances of receiving a callback rose 30 percent. But for candidates with Black-sounding names, "we found none of that. If anything, we found the opposite," Bertrand says.

"It was very counterintuitive," she adds. "One imagined employers looking at the names and kind of screening at that stage, not going any further, not even reading the resume. People in HR (human resources) call that a 'deselection process,' where you see a pile of resumes that you have to get through and do a kind of rapid screen" in order to separate the wheat from the chaff.

"That's exactly how we used to do it," notes Kimberly Wilson, who held a human resources position in a mid-sized social policy research firm in the Washington, D.C., area. It was Wilson's job to cull the stack of resumes--"perhaps around 300" for every position, she says--down to about 20 names that would then be brought before a committee.

"And every time, the committee would be more critical of resumes with Black-sounding or foreign-sounding names. 'Oh, yes, she did a research internship, but it wasn't in health and that's what we're really looking for,'" Wilson says. When the names didn't provide a clue to race, the committee members would often zero in on other data--professional organizations, for example.

"'Oh, he belonged to the Hispanic fellowship group in college," she adds, explaining, "I don't even think it was conscious, certainly not in most cases."

Conscious or not, the MIT-UChicago study demonstrates that employers actively discriminate among job candidates on the basis of race. And attempts by African Americans to improve their chances with more education and more skills don't appear to help at all.

In addition, the study showed that:

Adjusting for gender greatly increased the discrimination effect. There was a difference of 3.35 percentage points--or 50 percent--between the callback rates for all Whites (10.1 percent) and all Blacks (6.7 percent). But the callback rate for the lowest scoring Black female name, Aisha (2.2 percent) was 6.1 percentage points below that of the lowest scoring White female name, Emily (8.3 percent) and 11.4 percentage points below that of the highest scoring White female name. Indeed, five of the nine Black female names--Aisha, Keisha, Tamika, Lakisha and Tanisha--scored lower than the lowest scoring White female name. By contrast, the racial gap between male names was not nearly so pronounced. The lowest scoring White and Black male names--Neil and Rasheed--were only 3.6 percentage points apart.

Applicants who lived in "better" neighborhoods--"Whiter," more educated, higher income--received more callbacks than those who did not. But again, Whites benefited so much more than Blacks, it was not clear that Blacks benefited at all, suggesting that, despite a widespread societal belief in the stigmatizing effect of a "ghetto" address, a good address doesn't help African Americans overcome discrimination either.