-Professor Pooja Mishra

Despite extensive research efforts to understand the origins of class-based inequalities in the workplace, there remains to be a greater understanding of the institutional-level mechanisms contributing to reinforcing these disparities.

Previous studies have focused on individual-level factors such as self-selection into specific occupations, lack of proactive behaviour, and low self-efficacy.

In this project, investigators researched the influence of a crucial institutional-level context, namely the hiring process, on the challenges individuals from socio-economically disadvantaged backgrounds face when attempting to enter high-status occupations.

They proposed that using classed wording, characterized by terminology aligning with the independent norms of upper socio-economic strata, represents a new mechanism contributing to existing class-based inequalities in the workplace.

The study employed a mixed-method approach encompassing three studies to examine this hypothesis, incorporating archival and experimental analyses.

Study 1 (n=5,121) and Study 2 (n=4,742) leveraged archival data extracted from numerous job advertisements featured on a popular job search website. Study 1 drew upon data from three sources: the General Social Survey (GSS), LinkedIn.com, and the Occupational Information Network (O*NET). The findings of Study 1 indicated a correlation between the presence of classed wording in job advertisements and the socio-economic background of employed individuals. Specifically, occupations hiring individuals from high socio-economic origins (parental socio-economic status falling within the top 10% of the sample) tend to employ more agentic language in their job descriptions compared to occupations hiring individuals from low socio-economic origins (parental socio-economic status falling within the bottom 10% of the sample).

Building upon these results, Study 2 investigated occupations predominantly occupied by currently high socio-economic people and revealed a strong correlation between the use of agentic language and the percentage of employees from high socio-economic origins in those occupations.

This finding challenges the notion that self-selection alone drives the class divide. Following the establishment of class-consistent language in job advertisements, we experimented (Study 3) to examine the impact of such language on job applicants' behavior. The laboratory experiment, conducted at a large public university in India, demonstrates that individuals who encounter job advertisements featuring more agentic language perceive fewer employees from lower socio-economic backgrounds in those jobs/occupations. This anticipation subsequently reduces the sense of belongingness among individuals from lower socio-economic origins, reducing their inclination to pursue these jobs and affecting their application behavior. Taken together, the results of these studies unveil the presence of seemingly innocuous yet systematic differences in the wording of job advertisements, which significantly impact individuals' job application behavior, thereby perpetuating class-based inequalities in the workplace.