Working Paper: NBER ID: w28319
Authors: Hadar Avivi; Patrick M. Kline; Evan Rose; Christopher R. Walters
Abstract: Correspondence experiments probe for discrimination by manipulating employer perceptions of applicant characteristics. We consider the gains from dynamically adapting the number and characteristics of fictitious applications to the sequence of employer responses received so far. Calibrating the employer callback process to data from a recent correspondence experiment by Nunley et al. (2015), we find it is possible to cut the number of applications required to detect a fixed number of discriminatory jobs roughly in half relative to the static benchmark design that sends the same number and mix of applications to all jobs. These gains are achieved primarily from abandoning jobs with very low callback probabilities and those that demonstrate a willingness to call black applicants.
Keywords: No keywords provided
JEL Codes: C9; J18; J71
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| adaptive correspondence experiments (C90) | reduce the number of applications needed to detect discriminatory jobs (J79) |
| abandoning jobs with very low callback probabilities (J63) | reduce the number of applications needed to detect discriminatory jobs (J79) |
| abandoning jobs that show a willingness to call black applicants (J63) | reduce the number of applications needed to detect discriminatory jobs (J79) |
| adaptive strategy (O33) | more efficient identification of discriminatory employers (J79) |
| correspondence experiments (C93) | target investigations more effectively (C99) |
| discriminating job (J70) | odds of being called back are approximately 36 times higher for white applicants than for black applicants (J79) |