Black Names Aren’t That Simple

In correspondence audit studies, nearly identical applications or resumes, altered only in the name used, are submitted to the same listing. Researchers use distinctively racialized names to communicate information about the applicant, and differences in callback rates are taken as evidence of discrimination. Such studies, however, have produced inconsistent estimates of racial discrimination in employment.

In a recent article published on SocArXiv, Michael Gaddis argues that this variability is largely the result of a lack of scientific rigor in how researchers select “Black” names. Gaddis used New York state birth records from 1994-2012, which contain information about individuals’ names, race, and mother’s education, to identify racialized names across the class spectrum. He then surveyed a large number of people to determine whether individuals could correctly perceive the race of names as they align with population-level naming practices and use in past correspondence studies. Finally, Gaddis used these results to evaluate whether individual racial perceptions of names explain variations in previous studies’ estimates of discrimination.

Gaddis’s results suggest that names given by highly educated Black parents are less likely to be perceived as “Black” than names given by less educated Black parents—that is, the “race” discrimination was actually affected by class differences in naming conventions. Kugelmass’s study, noted in this section (“screening therapy clients by race and class”), went beyond names to manipulate race using voice actors. A large body of social science evidence, however, has incorrectly operated under the assumption that all Black names are alike.