Presidential Elections and Municipal Development: The Colombian Case (1986-2014)

Can the probability of voting for a candidate based on the municipal conditions of violence and development be explain? This paper analyses the impact of economic development and violence at the municipal level on voting patterns in presidential elections in Colombia for the period 1986-2014. It uses Principal Components Analysis (PCA) methodology and generates a municipal voting index, which is used as main input together with municipal development and violence variables in a Bayesian beta model. In the end, it is concluded that with a higher level of education and access to the media, the probability of voting for right-wing candidates increases and the probability of voting for liberal candidates is less likely.