I study post-conflict peace-building and the democratization process in Sub-Saharan Africa. Specifically, I am interested understanding the causes, the consequences, as well as the prevention strategies of political violence, in particular during the electoral process.
I employ qualitative, quantitative, and experimental methods.
I employ qualitative, quantitative, and experimental methods.
Threatening your own. Electoral violence within ethnic groups in Burundi and beyond (dissertation)
I propose and test a theory of intra-ethnic electoral violence, arguing that intimidation and violence can be used to control the electoral support of coethnics by demobilizing coethnic opposition candidates and by coercing the support of coethnic opposition voters. I evaluate this argument quantitatively (by an econometric analysis of sub-national patterns of violence in Burundi, based on original 2,500+ questionnaires and subnational statistics) and qualitatively (through electoral campaign text analysis and fieldwork interviews).
I show that political parties in Burundi targeted predominantly their own coethnics (using violence to signal strength and their ability to protect supporters, punish defectors, and coerce opponents) and violence was greatest when the majority ethnic group was larger and politically polarized. I further show that electoral violence in Zimbabwe and Sri Lanka followed the same pattern as in Burundi.
How abolishing school fees increased support for the incumbent in Burundi (African Affairs (2017) 116 (462): 101-124)
I investigate voting behaviors following the provision of basic public services in Burundi, showing that the incumbent's vote share significantly increased between 2005 and 2010 in localities where access to, and quality of, schooling improved during the same period.
The study employs econometric analysis of subnational statistics and voting data, a local observers' questionnaire, as well as qualitative analysis of campaign messages.
Malawi: widespread tension, limited violence (book chapter, in "Electing Peace: Violence Prevention and Impact at the Polls" USIP press, 2017)
I investigate the conditions under which models of preventing election violence failed or succeeded in recent Malawian elections.
The study employs fieldwork interviews and focus group discussions across the country.
Civic Education and Electoral Observation in Violent Elections: Evidence from Côte d’Ivoire’s 2015 Election (under review, with Leonardo Arriola, Justine Davis, Aila Matanock)
We explore the impact of civic education on voters' attitudes and behavior in Côte d'Ivoire. We implement a randomized control trial of a civic education program in neighborhoods of Abidjan followed by a representative survey in treatment and control areas. We find that exposure to civic education induced voters to fear more violence during the election and lowered their trust in political institutions.
Coercing the co-ethnic vote: violence against co-ethnics in Burundi's 2010 elections (working paper)
I theorize on why and under which conditions pre-electoral violence occurs within ethnic groups, then discuss motivations and triggers of sub-national patterns of state violence in Burundi, showing that it intensified when close to the winning threshold and spiraled between the largest parties competing for control of the same ethnic group.
Does pre-election violence impact voting behavior? Evidence from Burundi (working paper)
Exploiting territorial variation of state violence in Burundi's 2010 elections, I show that electoral coercion was most effective (in enhancing participation) when perpetrated selectively and when the incumbent relied on strongmen.
Who Fears Violence in African Elections? (working paper, with Leonardo Arriola)
We seek to understand what drives insecurity and fear of violence among African voters today. We statistically analyze the data of several nationally representative surveys (Afrobarometer surveys). We uncover that violence is consistently feared by citizens who support political opposition parties and that believe their votes are monitored.
The sub-national origins of insurgency in Africa (working paper ,with Benjamin Pasquale)
We collect an original geo-coded dataset of the geographic origin of armed groups in Sub-Saharan Africa. We statistically analyze these data and find that civil wars are predicted by economic shocks at the national level, but not by local economic conditions.
Building peace with institutions? Revisiting institutional engineering in divided societies using experimental methods (work in progress)
I implement a laboratory behavioral experiment to study the impact of power-sharing institutions on ethnic conflict. I find that that proportional systems strengthen ethnic-group attachment and preferences, and that, in contrast to general expectations, power-sharing arrangements do not always ensure cooperation between elites from different ethnic groups.
I propose and test a theory of intra-ethnic electoral violence, arguing that intimidation and violence can be used to control the electoral support of coethnics by demobilizing coethnic opposition candidates and by coercing the support of coethnic opposition voters. I evaluate this argument quantitatively (by an econometric analysis of sub-national patterns of violence in Burundi, based on original 2,500+ questionnaires and subnational statistics) and qualitatively (through electoral campaign text analysis and fieldwork interviews).
I show that political parties in Burundi targeted predominantly their own coethnics (using violence to signal strength and their ability to protect supporters, punish defectors, and coerce opponents) and violence was greatest when the majority ethnic group was larger and politically polarized. I further show that electoral violence in Zimbabwe and Sri Lanka followed the same pattern as in Burundi.
- Nominated for Best Dissertation in Comparative Politics award (APSA Almond Award)
- Awarded National Science Foundation Dissertation Improvement Grant
- Awarded Stanford Center on International Conflict and Negotiation Goldsmith grant
- Awarded Dean’s Dissertation Writing Fellowship, New York University
How abolishing school fees increased support for the incumbent in Burundi (African Affairs (2017) 116 (462): 101-124)
I investigate voting behaviors following the provision of basic public services in Burundi, showing that the incumbent's vote share significantly increased between 2005 and 2010 in localities where access to, and quality of, schooling improved during the same period.
The study employs econometric analysis of subnational statistics and voting data, a local observers' questionnaire, as well as qualitative analysis of campaign messages.
Malawi: widespread tension, limited violence (book chapter, in "Electing Peace: Violence Prevention and Impact at the Polls" USIP press, 2017)
I investigate the conditions under which models of preventing election violence failed or succeeded in recent Malawian elections.
The study employs fieldwork interviews and focus group discussions across the country.
Civic Education and Electoral Observation in Violent Elections: Evidence from Côte d’Ivoire’s 2015 Election (under review, with Leonardo Arriola, Justine Davis, Aila Matanock)
We explore the impact of civic education on voters' attitudes and behavior in Côte d'Ivoire. We implement a randomized control trial of a civic education program in neighborhoods of Abidjan followed by a representative survey in treatment and control areas. We find that exposure to civic education induced voters to fear more violence during the election and lowered their trust in political institutions.
- Awarded National Science Foundation Rapid Response Research Grant
Coercing the co-ethnic vote: violence against co-ethnics in Burundi's 2010 elections (working paper)
I theorize on why and under which conditions pre-electoral violence occurs within ethnic groups, then discuss motivations and triggers of sub-national patterns of state violence in Burundi, showing that it intensified when close to the winning threshold and spiraled between the largest parties competing for control of the same ethnic group.
- Awarded Honorable Mention for Best Graduate Student Paper, by the American Political Science Association's African Politics Conference Group
Does pre-election violence impact voting behavior? Evidence from Burundi (working paper)
Exploiting territorial variation of state violence in Burundi's 2010 elections, I show that electoral coercion was most effective (in enhancing participation) when perpetrated selectively and when the incumbent relied on strongmen.
Who Fears Violence in African Elections? (working paper, with Leonardo Arriola)
We seek to understand what drives insecurity and fear of violence among African voters today. We statistically analyze the data of several nationally representative surveys (Afrobarometer surveys). We uncover that violence is consistently feared by citizens who support political opposition parties and that believe their votes are monitored.
The sub-national origins of insurgency in Africa (working paper ,with Benjamin Pasquale)
We collect an original geo-coded dataset of the geographic origin of armed groups in Sub-Saharan Africa. We statistically analyze these data and find that civil wars are predicted by economic shocks at the national level, but not by local economic conditions.
Building peace with institutions? Revisiting institutional engineering in divided societies using experimental methods (work in progress)
I implement a laboratory behavioral experiment to study the impact of power-sharing institutions on ethnic conflict. I find that that proportional systems strengthen ethnic-group attachment and preferences, and that, in contrast to general expectations, power-sharing arrangements do not always ensure cooperation between elites from different ethnic groups.