Name:

Description:

This experiment data comes from a study that sought to understand the influence of race and gender on job application callback rates. The study monitored job postings in Boston and Chicago for several months during 2001 and 2002 and used this to build up a set of test cases. Over this time period, the researchers randomly generating resumes to go out to a job posting, such as years of experience and education details, to create a realistic-looking resume. They then randomly assigned a name to the resume that would communicate the applicant's gender and race. The first names chosen for the study were selected so that the names would predominantly be recognized as belonging to black or white individuals. For example, Lakisha was a name that their survey indicated would be interpretted as a black woman, while Greg was a name that would generally be interpretted to be associated with a white male.

Variables:

• job_ad_id - Unique ID associated with the advertisement.

• job_city - City where the job was located.

• job_industry - Industry of the job.

• job_type - Type of role.

• job_fed_contractor - Indicator for if the employer is a federal contractor.

• job_equal_opp_employer - Indicator for if the employer is an Equal Opportunity Employer.

• job_ownership - The type of company, e.g. a nonprofit or a private company.

• job_req_any - Indicator for if any job requirements are listed. If so, the other job_req_* fields give more detail.

• job_req_communication - Indicator for if communication skills are required.

• job_req_education - Indicator for if some level of education is required.

• job_req_min_experience - Amount of experience required.

• job_req_computer - Indicator for if computer skills are required.

• job_req_organization - Indicator for if organization skills are required.

• job_req_school - Level of education required.

• received_callback - Indicator for if there was a callback from the job posting for the person listed on this resume.

• firstname - The first name used on the resume.

• race - Inferred race associated with the first name on the resume.

• gender - Inferred gender associated with the first name on the resume.

• years_college - Years of college education listed on the resume.

• college_degree - Indicator for if the resume listed a college degree.

• honors - Indicator for if the resume listed that the candidate has been awarded some honors.

• worked_during_school - Indicator for if the resume listed working while in school.

• years_experience - Years of experience listed on the resume.

• computer_skills - Indicator for if computer skills were listed on the resume. These skills were adapted for listings, though the skills were assigned independently of other details on the resume.

• special_skills - Indicator for if any special skills were listed on the resume.

• volunteer - Indicator for if volunteering was listed on the resume.

• military - Indicator for if military experience was listed on the resume.

• employment_holes - Indicator for if there were holes in the person's employment history.

• has_email_address - Indicator for if the resume lists an email address.

• resume_quality - Each resume was generally classified as either lower or higher quality.

Details

Because this is an experiment, where the race and gender attributes are being randomly assigned to the resumes, we can conclude that any statistically significant difference in callback rates is causally linked to these attributes.

Do you think it's reasonable to make a causal conclusion? You may have some health skepticism. However, do take care to appreciate that this was an experiment: the first name (and so the inferred race and gender) were randomly assigned to the resumes, and the quality and attributes of a resume were assigned independent of the race and gender. This means that any effects we observe are in fact causal, and the effects related to race are both statistically significant and very large: white applicants had about a 50% better chance of getting a callback than black candidates.

Do you still have doubts lingering in the back of your mind about the validity of this study? Maybe a counterargument about why the standard conclusions from this study may not apply? The article summarizing the results was exceptionally well-written, and it addresses many potential concerns about the study's approach. So if you're feeling skeptical about the conclusions, please find the link above and explore!

Link To Google Sheets:

Rows:

Columns:

Files:

License Type:

References/Notes/Attributions:

Source

Bertrand M, Mullainathan S. 2004. "Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination". The American Economic Review 94:4 (991-1013). http://www.nber.org/papers/w9873

R Dataset Upload:

Use the following R code to directly access this dataset in R.

R Coding Interface: