Discrimination in the labour market is illegal in many countries and leads to unnecessary costs for society, organizations, and employees (OECD, 2020). Despite this, discrimination only decreases slightly over time (Galos & Coppock, 2023; Lippens et al., 2023; Quillian et al., 2017). Biased outcomes in hiring lead to the devaluation of certain groups of employees (Jones et al., 2016), and discrimination occurs when individuals are treated differently based on their group membership, such as gender or nationality (Allport, 1954). Research has yet to provide answers on how to reduce bias and discrimination in personnel selection (Dhanani et al., 2018). This is of particular concern in the context of public employers for three reasons: First, the selection of public employees is key to the effective delivery of public services (Ashraf et al., 2020). Second, laws in many countries require the selection of the best when granting access to public service jobs (Jakobsen et al., 2023). Third, poor selection practices threaten the goals of representative bureaucracy (Jankowski et al., 2020).

This manuscript aims to contribute to the literature in two ways: It tests both the effectiveness and feasibility of the popular approach of blinding, i.e. not revealing demographics of applicants in the selection process prior to a potential job interview (Fath et al., 2023; Neumark, 2021). This manuscript presents a pre-registered field experiment (n>4200 real-world applications, nested in 293 hiring processes; data collection completed in 07/2024), co-designed with a people management software company that wants to improve selection processes among its public and private users (Jakobsen et al., 2024).

It is tested whether blinding can increase the hireability (i.e. job interview invitations and hiring) of two groups (female, non-Danish applicants) by randomizing the selection processes into either into a blinding or an active control group. In the latter, the gender and nationality are not blinded, but instead selecting managers are asked to evaluate applicants on the basis of three key competencies that they had to pre-register before the application period.

To assess the feasibility and practical relevance of the blinding approach, we present an additional expert study (n=76 recruiters) which shows that practitioners predict limited effects of a blinding intervention on hireability. They predict that blinding can only increase the proportion of non-Danish applicants invited to a job interview. They do not expect blinding to have a significant effect on the final hiring of non-Danish applicants or on the hireability of women.

This research examines the perspectives of DEL employees and job seekers on the usage of the Employment Services South Africa (ESSA) system. ESSA, also known as Public Employment Services in other parts of the world, is a public service job recruitment system that connects job seekers with available job opportunities. The system's primary aim is to place unemployed South Africans in available job opportunities by facilitating information sharing between employers and job seekers based on the job seekers’ qualifications and experience.

Using a qualitative approach, data was collected from 34 respondents, including employees in the Department of Labour and job seekers through interviews. The findings revealed that many unemployed job seekers do not effectively use the ESSA system. While the ESSA system is valued by DEL employees, its adoption, perceived usefulness, accuracy, and effectiveness among job seekers are lacking. The system still struggles with low visibility, a lack of awareness, and trust among job seekers, who often look elsewhere for effective recruiting. To address these challenges, the study recommends that the Department of Labor focus on enhancing outreach and education to increase awareness among job seekers. This can be achieved through targeted campaigns, leveraging social media and online platforms where job seekers are likely to spend their time. The study also suggests redesigning the user interface to be more intuitive and user-friendly, implementing robust feedback mechanisms that allow users to report issues and suggest improvements, conducting regular audits of the system data to ensure accuracy and timeliness, and clearly communicating the unique benefits of the ESSA system over other job search platforms.

An expanding body of research on representative bureaucracy is dedicated to investigating the individual level effects of gender representation on student academic performance (e.g., Xu & Meier, 2021; Zhang, 2019; Doornkamp et al., 2021). The dominant expectation in these studies is that student-teacher gender congruence positively affects academic performance. At the same time, bureaucratic representation may also function at the organizational level, where bureaucrats may influence clients they do not directly interact with (Keiser et al. 2002; Meier & Nicholson, 2006; Favero & Molina, 2018). This study contributes to the literature on representative bureaucracy by examining how individual level student performance is associated with representation in portions of the school with which students do not (always) directly interact, both in terms of management composition (vertical) and staff composition (horizontal).

We hypothesize that both management composition and staff composition moderate the relationship between student-teacher gender congruence and student performance. As the proportion of women in management positions increases, the impact of student-teacher gender congruence on the performance of female students increases. This is because increased female representation in management is likely to foster an internal environment supportive of representative advocacy (Keiser et al., 2002). Additionally, as the number of female teachers increases, critical mass effects come into play (Keiser et al., 2002; Meier & Xu, 2023), and the influence of student-teacher gender congruence on student performance grows. Furthermore, we explore contagion effects (Atkins and Wilkins, 2013; Li, 2021; Meier & McCreab, 2022) by investigating whether female students under male teachers (i.e., lacking congruence) exhibit improved performance when those male teachers work in an environment with a higher proportion of female colleagues.

To empirically test our hypotheses, we use a unique longitudinal dataset consisting of 400,685 students nested within 193 Dutch secondary schools, covering the entirety of students’ academic journeys. This dataset includes student-level information such as gender, subject-specific academic performance, and teacher gender by subject and year, as well as school-level data like gender distribution among school management and teaching staff, and total student enrollment. This comprehensive dataset also permits us to explore the hypotheses pertaining to both female and male gender congruence, ensuring a comprehensive analysis.

References:

Atkins, D., & Wilkins, V.M. (2013). Going beyond Reading, Writing, and Arithmetic. Journal of Public Administration Research and Theory 23(4):771–90

Doornkamp, L., Groeneveld, S., Groeneveld, M. G., Van der Pol, L. D., & Mesman, J. (2021). Understanding the Symbolic Effects of Gender Representation: A Multi-Source Study in Education. International Public Management Journal, 1-23.

Favero, N., & Molina Jr., A.L. (2018) Is Active Bureaucratic Representation an Organizational-Level Process? The Indirect Effect of Bureaucrats on Clients They Don’t Directly Serve. American Review of Public Administration 48 (1): 3–17.

Li, D. (2021) Spillover Effects of Minority Representation on Majority Bureaucrats’ Behavioral Change. Public Administration Review 81(6):1071–91.

Keiser, L. R., Wilkins, V. M., Meier, K. J., & Holland, C. A. (2002). Lipstick and logarithms: Gender, institutional context, and representative bureaucracy. American political science review, 96(3), 553-564.

Meier, K. J., & Xu, X. (2023). Critical thoughts about critical mass in representative bureaucracy: A theoretical exploration and empirical illustration. Governance, 36(3), 887-907.

Meier, K. J., & McCrea, A. M. (2022). Sit it out or dance: representative bureaucracy contagion effects in health care. International Public Management Journal, 1-21.

Meier, K. J., & Nicholson‐Crotty, J. (2006). Gender, representative bureaucracy, and law enforcement: The case of sexual assault. Public Administration Review, 66(6), 850-860.

Xu, X., & Meier, K. J. (2022). Separating symbolic and active representation: A mixed methods study of gender and education in China. Public Management Review, 24(9), 1429-1451.

Zhang, Y. (2019). Representative bureaucracy, gender congruence, and student performance in China. International Public Management Journal, 22(2), 321-342.