Are Emily and Greg more employable than Lakisha and Jamal? 4.1 Description Bertr

Are Emily and Greg more employable than Lakisha and Jamal?
4.1 Description
Bertrand and Mullainthan (2004) have set up a field experiment to analyze labor market dis- crimination by race. They have sent resumes to help-wanted ads with randomly assigned African-American or White-sounding names and recorded the callback rates for interviews. The differences in the callback rates indicate that differential treatment by race is prominent in the U.S. labor market.
4.2 General direction
The aim of the project is to analyze the determinants of callbacks for interviews using probit (and logit) regression tools. Use the following points as building blocks of your project:
a) The project is based on experimental data. Conceptually, this is different from using observational data which are most often used in econometrics. Make yourself familiar with the consequences of using experimental data by reading Stock and Watson (2011, Chapter 13).
b) Why does it make sense to perform a probit regression of the callback-dummy variable not only on the race dummy variable but also on other regressors?
c) If the data are really obtained from a randomized experiment, what would you expect (i) for the correlation of the race dummy and the other regressors and (ii) for the point estimates of the race dummy coefficients in a bivariate and multiple probit regression?
d) Set up a probit (logit) model in order to determine the determinants of the callbacks. Which regressors are significant, which not? How can you test for significant differences between white and African-American names with respect to the effects of certain regressors. How good is the model fit?
4.3 Data set
The data are provided in the file names.dta (or names.xls). They contain information on all resumes considered in the paper of Bertrand and Mullainthan (2004). However, some variables, e.g. on the zip-code characteristics, are not contained.
4.4 Literature
a) Stock, J.H. and Watson, M.W. (2011), Introduction to Econometrics, 3rd edition, Pear- son.
b) Bertrand, M. and Mullainthan, S. (2004), Are Emily and Greg more employable than Lak- isha and Jamal? A field experiment on labor market discrimination. American Economic Review, 94:4 991-1013.