FINAL - Week 6: Intro to R Questions with verified
answers
A variable in a data frame is like a vector, and we can apply functions to it
a.) calculating mean call back rate
b.) calculate the # of stereotypically black and white names Ans✓✓✓-a.)
mean(resume$call)
We can use the table function to describe character vectors
b.) table(resume$race)
Arithmetic Operations
Order of Operations Ans✓✓✓-R can be used as a calculator
parentheses can be used to specify order of operations
Bertrand and Mullainathan (2004) Ans✓✓✓-To isolate discrimination, they sent
fictitious resumes to help-wanted ads in Boston and Chicago
They manipulate the perception of race by altering the name on the resume
EX: white-sounding names from the paper: Emily Walsh and Greg Baker
EX: black-sounding names from the paper: Lakisha Washington and Jamal Jones
All other characteristics of the applicant (experience, education, job history) are
randomized
, They then compare callback rates (for an interview) between Black and white
applicants
Calculate call back rate for each data frame
- and its interpretation Ans✓✓✓-Interpretation: Black applicants about 3.2
percentage points less likely to receive a callback, despite these applications being
similar on all dimensions
Conclusion
Recent studies have expanded on this design Ans✓✓✓-Kline, Rose and Walters
(2021) sent around 83,000 applications to the top 100 US employers
On average find lower callback rates for Black applicants (about 2.1 percentage
points)
Around 9.7 percent of white applicants receive a callback, while only 6.4 percent
of Black applicants receive a callback
- therefore, there is a 3.3 percentage point disparity in callback rates. This is a
substantial disparity
- the baseline callback rates are quite low in this setting. A 3.3 percentage point
disparity implies white applicants are more than 50 percent more likely to receive
a callback
highlights a stark disparity in labor-market outcomes
- And given the empirical design: randomizing race across applications, these
disparities cannot be driven by other factors that employers take into account
when making hiring decisions
answers
A variable in a data frame is like a vector, and we can apply functions to it
a.) calculating mean call back rate
b.) calculate the # of stereotypically black and white names Ans✓✓✓-a.)
mean(resume$call)
We can use the table function to describe character vectors
b.) table(resume$race)
Arithmetic Operations
Order of Operations Ans✓✓✓-R can be used as a calculator
parentheses can be used to specify order of operations
Bertrand and Mullainathan (2004) Ans✓✓✓-To isolate discrimination, they sent
fictitious resumes to help-wanted ads in Boston and Chicago
They manipulate the perception of race by altering the name on the resume
EX: white-sounding names from the paper: Emily Walsh and Greg Baker
EX: black-sounding names from the paper: Lakisha Washington and Jamal Jones
All other characteristics of the applicant (experience, education, job history) are
randomized
, They then compare callback rates (for an interview) between Black and white
applicants
Calculate call back rate for each data frame
- and its interpretation Ans✓✓✓-Interpretation: Black applicants about 3.2
percentage points less likely to receive a callback, despite these applications being
similar on all dimensions
Conclusion
Recent studies have expanded on this design Ans✓✓✓-Kline, Rose and Walters
(2021) sent around 83,000 applications to the top 100 US employers
On average find lower callback rates for Black applicants (about 2.1 percentage
points)
Around 9.7 percent of white applicants receive a callback, while only 6.4 percent
of Black applicants receive a callback
- therefore, there is a 3.3 percentage point disparity in callback rates. This is a
substantial disparity
- the baseline callback rates are quite low in this setting. A 3.3 percentage point
disparity implies white applicants are more than 50 percent more likely to receive
a callback
highlights a stark disparity in labor-market outcomes
- And given the empirical design: randomizing race across applications, these
disparities cannot be driven by other factors that employers take into account
when making hiring decisions
