SBP: A Perfect Match? How Job Demands Shape Gender and Minority Differences in Hiring

University Of North Carolina At Chapel Hill, Chapel Hill NC

Investigators

Abstract

Continued status differences in occupational attainment prompt concerns regarding whether and to what extent such differences are a function of bias in the hiring process. A critical point in this process is when job applicants either do or do not receive call backs—request for interviews or more information—after applications have been submitted, and whether such differences vary by social status. This project will examine whether and how job demands – the required skills and expectations associated with a job opening – influence gender and racial differences in offering callbacks. The project will examine whether employers are more or less likely to decline callbacks for women and minority job applicants when the job demands are vague vs. quantified, feminine-typed vs. masculine-typed, and when applicants meet or do not meet the required job demands. The project will additionally examine how such patterns, and the association with job demands, depend on job level and upward vs. downward job mobility. Biases faced by women and minorities are costly, not only to the applicants but to companies and the broader economy that may be missing out on matching talented individuals to appropriate jobs. By elucidating the specific mechanisms and conditions under which biases occurs, this research will demonstrate how interventions by employers and government leaders can reduce inequality in the labor market. To study how job demands affect gender and racial inequality levels, and how demands relate to bias across job level and upward or downward job mobility, the project will field a large-scale audit study. This field experiment study will involve varying applicants by perceived race and gender, signaled by name. The project will submit approximately 20,000 job applications to online job openings, recording callbacks – requests for interviews or more information – across experimental condition. The scope conditions will be limited to four occupations: accountants, software engineers, sales professionals, and human resource managers. Approximately 5,000 job applications will be submitted to each occupation, dividing applications between entry-level and mid-level positions, and between applicants who are in early- or mid-career stages to represent attempts at upward vs. downward job mobility. The job openings in the audit study will then be matched to proprietary skill data on online job postings. By merging these data, the project will identify and characterize the bundle of job demands for each unique job opening. Computational text analysis techniques such as topic models, together with information in the dataset, will be used to create variables to categorize demands by level of vagueness, associations of masculinity or femininity, and the extent to which applicants meet the demands. Logistic regression models predicting callbacks will be used, modeling interaction effects with the job demand variables and applicants’ race and gender. The project will assess whether and how job demands affect bias within occupations and job level, and whether demands mediate the relationship between bias and job mobility. Findings will inform sociological theories regarding status differences in the operation of labor markets, particularly for professional occupations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.