Gender representation and the coproduction of emergency preparedness: An experimental study and replication

A significant body of literature on representative bureaucracy has been advanced in the public management literature. Some studies have investigated basic descriptive patterns of passive representation (or the extent to which... [ view full abstract ]

A significant body of literature on representative bureaucracy has been advanced in the public management literature. Some studies have investigated basic descriptive patterns of passive representation (or the extent to which the bureaucracy’s workforce reflects the demographic profile of clients or citizens), while other studies have examined the effects of active representation (in which street-level bureaucrats use their discretion to respond to or advocate for the needs and interests of social groups, particularly women and racial minorities). But recent scholarship has suggested that passive representation itself may have substantive effects through the mechanism of enhanced trust and cooperation on the part of citizens, which is sometimes referred to as symbolic representation. The symbolic representation hypothesis leads to the expectation that citizens will be more willing to cooperate and thus co-produce important outcomes when their gender or race is represented in a government bureaucracy.

In previous experimental work (recently published in Public Administration Review), we found substantively large symbolic representation effects on women’s willingness to recycle and engage in food composting when a hypothetical city initiative included more female officials. In this new study, we seek to replicate and test the symbolic representation hypothesis in a quite different policy context: emergency preparedness. To do so, we conducted a survey experiment in which a national sample of US adults (n=603) were shown a description of Citizen Corp, a federal government program to encourage local communities to become better prepared for emergencies. Following the tactic of the prior study, the names of four public officials quoted in the announcement were randomly varied to be all male names, all female names, or a mix of male and female names. Respondents in each of the three arms of the experiment were then asked to indicate their willingness to volunteer, donate money, or give blood if requested by Citizen Corp.

In contrast to our prior results, women in the experiment did not exhibit any increased willingness to volunteer or donate money or blood when the quoted officials representing Citizen Corp had female names. Overall, men in the experiment also appeared unresponsive to the gender balance of the names of Citizen Corp officials. We interpret these null findings as evidence that symbolic representation effects may be limited to specific policy areas or issues. Importantly, our study illustrates the need for replication of experimental studies, which are becoming a more important part of contemporary public management research, in order to probe the generalizability and boundary conditions of previous experimental findings.