How Citizens Respond to Outputs, Outcomes, and Costs: A Survey Experiment About an HIV/AIDS Program
通过一项调查实验，比较了向公民展示HIV预防项目的产出（服务人数）与结果（预防感染数）对项目支持度的影响，发现结果信息反而降低支持，成本信息则提升支持。
ABSTRACTGovernment performance measurement is often faulted for focusing on outputs, while citizens are said to demand more information on outcomes to hold government accountable. To compare the influence of these measures, we randomized 774 participants to receive outcome or output information about a real HIV prevention program, with or without cost information, in a survey experiment. Citizens expressed less support for spending on the program when shown outcomes (infections prevented), rather than outputs (people served). Showing participants the high cost of treating HIV/AIDS boosted support for program spending, but did not make outcome information more persuasive. We interpret these results as partly a reaction to the high per-unit cost of an outcome in an HIV prevention program. But it may reflect a tendency of citizens simply to misinterpret less costly outputs, including serving more people, as if these were outcomes. This bias has implications for performance reporting and accountability. NotesGiven the ordinal level of measurement of both dependent variables, we also used ordinal logistic regression. Results did not differ substantially from those of the ordinary least squares regression. We present the linear regression results for ease of interpretation.Additional informationNotes on contributorsAshley GrossoAshley Grosso (agrosso@healthsolutions.org) is a research scientist at Public Health Solutions. She received her PhD in Public Administration from the School of Public Affairs and Administration, Rutgers University–Newark. Her current research is focused on evaluating government and nonprofit responses to HIV and other adverse sexual and reproductive health outcomes. She also is interested in epidemiological studies with at-risk populations, with a focus on adolescents in sub-Saharan Africa and urban areas in the United States.Étienne CharbonneauÉtienne Charbonneau (Etienne.charbonneau@enap.ca) is associate professor of public management at the École Nationale d'Administration Publique. He received his PhD in Public Administration from the School of Public Affairs and Administration, Rutgers University–Newark. When he is not examining citizen satisfaction with public services, he delves into questions of how and why public managers integrate benchmarking into their decisions, and how benchmarking systems influence performance and accountability.Gregg G. Van RyzinGregg G. Van Ryzin (vanryzin@rutgers.edu) is professor in the School of Public Affairs and Administration, Rutgers University–Newark. His research employs surveys and experiments to understand how citizens evaluate and respond to public services and institutions. He has published widely in public administration and policy journals and is author (with Dahlia K. Remler) of Research Methods in Practice (Sage).