Signaling Class: An Experiment Examining Social Class Perceptions from Names Used in Correspondence Audit Studies

Abstract

Field and survey experiments examining racial discrimination and inequality commonly use names to signal race and ethnicity. However, little work has been done to understand how individuals interpret these signals. Despite strong concerns that racialized names simultaneously signal social class, no work has scientifically examined the social class signaling power of names used in previous research. In this article, I conduct a survey experiment to test individual perceptions of social class from names. Respondents are presented with a series of first and last names and asked to state the social class category they associate with each name. In total, 7,936 respondents provide their social class perceptions on 600 different combinations of first and last names. I find that black and Hispanic names are much more likely to be perceived as lower or working class than white names, which are overwhelmingly perceived as middle or upper class. These perceptions are independent of the effect of population-based socioeconomic naming patterns. Overall, this research suggests that scholars must not assume that population patterns in naming are indicative of individual perceptions of names. Instead, scholars should pretest names ethnicity before conducting a field or survey experiment to increase internal validity.