Confidence Intervals for the Difference in Population Proportions: Learn It 1

  • Calculate a confidence interval for the difference in proportions of two groups.
  • Make conclusions based on a confidence interval.

Who gets more job callbacks?

Do job callbacks differ based on the perceived gender of the applicant?

To examine this question, we will analyze data from an experiment[1] that assessed the impact of gender and race on a job applicant receiving a callback, the opportunity to progress to the next part of the applicant process based on their application.

Note: The researchers randomly assigned names that are commonly associated with particular races and genders.

Though we have an estimate, or a “best guess,” for the difference in proportions of applicants who received callbacks between the two groups, we expect there is some variability. In other words, if we calculated the difference in the proportions of applicants who received callbacks from two other random samples of [latex]3,746[/latex] applicants who were perceived as female and [latex]1,124[/latex] applicants who were perceived as male, we would expect to get a different (yet probably close) value of [latex]\hat{p}_F - \hat{p}_M[/latex] compared to what we did in the previous exercise.


  1. Bertrand, M. & Mullainathan, S. (2004). Are Emily and Greg more employable than Lakisha and Jamal? A field experiment on labor market discrimination. American Economic Review 94(4), 991–1013. DOI: 10.1257/0002828042002561. http://www.nber.org/papers/w9873