The Individual‐Level Effect of Gender Matching in Representative Bureaucracy

• Published date: 01 May 2018
• DOI: http://doi.org/10.1111/puar.12907
• Date: 01 May 2018
• Author: Thorbjørn Sejr Guul

398 Public Administration Review • May | June 2018

Public Administration Review,

Vol. 78, Iss. 3, pp. 398–408. © 2018 by

The American Society for Public Administration.

DOI: 10.1111/puar.12907.

Research Article The Individual-Level Effect of Gender Matching in

Representative Bureaucracy

Abstract: A substantial literature links passive bureaucratic gender representation to better outcomes for represented

citizens. However, because existing studies have been carried out primarily at the organizational level, it is not known

whether these effects are caused by individual-level interaction or by organizational processes. This article investigates

whether gender matching in individual relationships affects citizen outcomes. The context of the study is a program

in which young job seekers without vocational training were assigned to a job counselor for one-on-one meetings. The

article finds that gender matching has a positive effect on citizen outcomes and that this effect is partly explained by

an increased level of effort on the part of the citizen. This implies that citizen outcomes can be improved by matching

bureaucrats and citizens of the same gender or by providing training to bureaucrats to enable them to better serve

citizens of the opposite gender .

Evidence for Practice

• Matching street-level bureaucrats and citizens by gender may be an inexpensive method of improving

outcomes for citizens in one-on-one service areas in which there is organizational gender diversity.

• Street-level bureaucrats tend to spend more time on citizens of their own gender, and citizens who are

matched with street-level bureaucrats of the same gender tend to make a greater contribution to the delivery

of the service.

• The additional time use does not necessarily result in improved citizen outcomes, whereas the increased level

of the effort on the part of the citizen does.

• Training bureaucrats to enable them to better motivate citizens of the opposite gender—by improving

communication, for example—may improve citizen outcomes regardless of organizational gender diversity.

Thorbjørn Sejr Guul

Aarhus University

Thorbjørn Sejr Guul is a PhD student

in the Department of Political Science,

Aarhus University, and affiliated with the

TrygFonden ’ s Centre for Child Research.

His research focuses on representative

bureaucracy, discrimination, and street-level

bureaucrat coping behavior.

E-mail: tsg@ps.au.dk

I n recent years, the theory of representative

bureaucracy has been studied extensively and

with great progress (e.g., Andrews, Ashworth, and

Meier 2014 ; Gade and Wilkins 2013 ; Hong 2017 ;

Riccucci, Van Ryzin, and Li 2016 ). The theory ’ s key

claim is that passive demographic representation

among bureaucrats will affect active representation

and, ultimately, outcomes for the citizens represented.

A bureaucracy is passively representative if the

demographic composition of bureaucrats is similar to

that of the general population. Active representation

occurs when bureaucrats use their discretion to

eliminate discrimination and pursue the interests of

the group of citizens they represent (Mosher 1968 ,

12). For example, the share of female police officers

may affect the rate at which sexual assaults against

women are reported and offenders arrested (Meier and

Nicholson-Crotty 2006 ).

Despite the volume of studies, most have been carried

out at the organizational level, which makes it difficult

to determine why these effects occur (Andrews,

Ashworth, and Meier 2014 , 19; Atkins, Fertig, and

Wilkins 2014 , 504; Bradbury and Kellough 2011 ,

164). Is the mere presence in an organization of

bureaucrats with specific characteristics enough, or

is it necessary that these bureaucrats interact directly

with the citizens they represent? A recent study of

racial representation shows that direct interaction has

the greatest impact (Nicholson-Crotty et al. 2016 ).

Other studies of bureaucrat behavior, however, have

demonstrated that nonwhite caseworkers discriminate

just as much against nonwhite clients as their

white colleagues (Schram et al. 2009 , 415), which

contradicts the proposed effect of direct interaction.

In addition, most studies of the effect on citizen

outcomes focus on service areas such as schools and

the police force, in which one-on-one interaction

between bureaucrat and citizen seldom occurs. Many

street-level bureaucrats (e.g., caseworkers and health

workers), however, do deliver their services on a one-

on-one basis. While it has been shown that passive

representation in these settings affects the values

of representative bureaucrats (Riccucci and Meyers