Am. Econ. Rev. https://doi.org/10.1257/aer.20181714

Studies of employer preferences and sources of bias in labour markets have often relied on audit studies, in which researchers send fictitious resumes in response to real job ads and assess whether different resumes receive different responses. This method can uncover subtle biases and hidden preferences that might not be openly disclosed. However, the design can be problematic because it relies on deception and coarse measures of employer preferences and decisions.

Credit: NAN / Alamy Stock Photo

A new study by Judd B. Kessler, Corinne Low and Colin D. Sullivan, of the University of Pennsylvania, introduces an alternative method, Incentivized Resume Rating (IRR), that avoids these pitfalls. In an IRR study, employers agree to evaluate resumes they know are hypothetical. In return, they receive a set of real resumes that are matched to their preferences (but disregarding any discriminatory preferences). Researchers gain insight into how employers evaluate candidates, and employers gain a short list of potential matches for the job. The authors demonstrated their method with 72 employers seeking to hire university graduates and found that employers particularly emphasized academic grades and the quantity and quality of internship experiences in their ratings of candidates.

The study provides researchers with a new tool for evaluating labour markets and uncovering bias while avoiding the ethical and practical risks of deception.