Access Type

Open Access Dissertation

Date of Award

January 2013

Degree Type


Degree Name



Political Science

First Advisor

Sharon F. Lean


A growing body of research suggests that institutional performance affects citizen participation (see Holzner 2010; Bratton, Mattes et al 2005; Hiskey and Bowler 2005). But despite this recognition of the importance of institutions, there is relatively limited research into the institutional determinants of political participation, particularly with respect to developing countries. This dissertation fills this gap by examining local political participation in Ghana and Liberia using an institutional approach. I employ qualitative analyses based on field work and primary source documents, as well as quantitative analyses at both the individual and contextual level.

At the individual leveI, I find the performance of local government matters for political participation. Whether or not citizens perceive their local government as transparent and accountable, and whether government is corrupt makes a difference in how citizens engage with the local political system. The more transparent citizens perceive their local government to be, the more likely they are to participate in local politics. Greater experience with corruption also leads to higher levels of local participation.

Analyses at the contextual level demonstrate that subnational institutions also matter for local political participation. Decentralization, the density of civil society, and the quality of local elections all affect participation. In a two-country pooled model, I show that higher levels of decentralization are associated with lower levels of local political participation, while higher density of civil society is associated with increased local political participation. Higher levels of spoiled ballots in national elections, a proxy indicator for electoral fraud, is associated with increased local political participation. Interestingly, these findings contrast with common claims that decentralization should increase participation and that poorly run elections should depress participation. They confirm expectations that an active civil society promotes participation.

My research provides an institutional model for assessing the determinants of local political participation that can be applied to cases beyond those considered here. It also holds lessons for public administrators who seek to encourage citizen participation in local decision-making. It highlights for public administrators how variations in the institutional capacity of local government can affect citizen-government interactions. Higher levels of decentralization may not be achieving the goal of greater citizen participation in local decision-making. Given the results on civil society, local government initiatives to partner with and support civil society are likely to be fruitful.