In their 2004 study of racial discrimination in employment markets, Bertrand and Mullainathan sent resumes with varying characteristics to firms advertising job openings. Some firms were sent resumes with putative African American names, while other firms received resumes with putatively Caucasian names. The researchers also varied other attributes of the resume, such as whether the resume was judged to be of high or low quality (based on labor market experience, career profile, gaps in employment, and skills listed). The table below shows the rate at which applicants were called back by employers, by the city in which the experiment took place, and by the randomly assigned attributes of their applications.
City | Boston | Chicago | ||||||
---|---|---|---|---|---|---|---|---|
Quality of resume | Low | High | Low | High | ||||
Applicant name | Black | White | Black | White | Black | White | Black | White |
% received call from employer | 7.01 | 10.15 | 8.50 | 13.12 | 5.52 | 7.16 | 5.28 | 8.94 |
N | 542 | 542 | 541 | 541 | 670 | 670 | 682 | 682 |
Q: Propose a regression model that assesses the effects of the treatments, interaction between them, and interactions between the treatments and the covariate, city, where city is coded 1 for subjects in Boston and 0 for Chicago. Estimate the parameters in your regression model. Interpret the results.
Reference: 2003, Marianne Bertrand Sendhil Mullainathan, ARE EMILY AND GREG MORE EMPLOYABLE THAN LAKISHA AND JAMAL? A FIELD EXPERIMENT ON LABOR MARKET DISCRIMINATION,