Field Experiment Yields Little Evidence of Bias

Author
Tracey Potts
Page Body

Recent research by Cory Koedel, Associate Professor of the Harry S Truman School of Public Affairs, University of Missouri, found little evidence of employer preference for applicants’ race or gender in the early stages of the hiring process. In a follow-up study to Bertrand and Mullainathan’s 2004 resume audit, where the authors found large gaps in callback rates by race, Koedel et al found that by isolating surnames  generally by race, but not choosing names indicating socioeconomic status, and keeping the overall pool youthful, employers did not remarkably distinguish for race and gender in their response rates. Though white applicants were favored, the results were statistically insignificant.

In the Koedel study, nearly 9,000 fictitious resumes were sent to online job postings in seven major cities, in six occupational categories, based on actual resumes by job seekers, but with randomized names to suggest race (Washington, Jefferson, Hernandez, Garcia, Anderson and Thompson). Authors paired first names Chloe, Ryan, Isabella, Carlos, Megan and Brian with the surnames to suggest gender. The resumes all indicated the applicants were recent high school graduates and 85% had some college coursework at a 2-year institution. Employers responded, on average, to 11.4% of the resumes. Koedel’s study is the first to track labor market outcomes for Hispanic applicants using a resume audit design. While the experiment design does not examine discrimination in other stages of the hiring process, Koedel concludes continued research is needed to understand barriers to employment faced by minorities. “Race and Gender Effects on Employer Interest in Job Applicants: New Evidence from a Resume Field Experiment” was originally published in Applied Economic Letters and was highlighted in an article published by the University of Missouri’s News Bureau.

Read the Study

Related Article