Published online by Cambridge University Press: 07 February 2017
This article reports on the effects of domestic election observers on electoral fraud and violence. Using an experimental research design and polling station data on fraud and violence during Ghana’s 2012 elections, it shows that observers reduced fraud and violence at the polling stations which they monitored. It is argued that local electoral competition shapes party activists’ response to observers. As expected, in single-party dominant areas, parties used their local political networks to relocate fraud to polling stations without an election observer, and, in contrast, party activists relocated violence to stations without observers in competitive areas – a response that requires less local organizational capacity. This highlights how local party organization and electoral incentives can shape the manipulative electoral strategies employed by parties in democratic elections.
Asunka: The William and Flora Hewlett Foundation (email: email@example.com); Brierley, Golden and Ofosu: Department of Political Science, University of California Los Angeles (firstname.lastname@example.org, email@example.com, firstname.lastname@example.org ); Kramon: Department of Political Science, George Washington University (email@example.com). The authors are grateful to Daniel de Kadt, Jennifer Doherty, Barbara Geddes, Danny Hidalgo, Theresa Kuhn, John McCauley, Galen Murray, Daniel Posner, Michael Ross, Michael Thies, Daniel Treisman, Lynn Vavreck, and Adam Ziegfeld for comments. They also wish to record their gratitude for useful comments from audience members at various workshops held at Stanford, the APSA in Chicago, UC at Berkeley, UCLA, Montreal and Gothenburg 2013–15. They gratefully acknowledge the collaboration of their research partner in Ghana, the Centre for Democratic Development, as well as Ghana’s Coalition of Domestic Election Observers. They also thank their 300 research assistants for data collection. Funding came from the United Kingdom’s Ghana office of the Department for International Development and a U.S. National Science Foundation Grant for Rapid Response Research (RAPID) SES–1265247 (Miriam Golden PI), as well as the UCLA Academic Senate, none of which bears responsibility for the results reported here. This research was approved by the University of California at Los Angeles IRB#12-001543, 2012. Data replication sets available at https://dataverse.harvard.edu/dataverse.BJPolS and online appendices are available at https://doi.org/doi: 10.1017/S0007123416000491.