Everybody can take part in democratic election by giving a vote on the candidate most suited for the task. Still, many choose not to. The non-voters are rarely randomly spread out in an urban area but form geographical patterns. The aim of this thesis is to see which demographical factors that correlate and explain these patterns in Bogotá, Colombia. The method used is a mixed quantitative and qualitative analysis. Statistics about election turnout and demographical factors in the 20 districts of Bogotá are presented in scatter plots which show the characteristics of the correlations. The findings are thereafter explained in the qualitative part, which is based on interviews with 18 people living in different parts of Bogota. The theoretical framework explains what researchers say about electoral turnout and demographical factors, among others Norris (2002) and Milbrath and Goel (1977). They have found that education, age and socio economic class are influential for the electoral turnout and the correlation is positive. Tested on the percentage of blank ballots, the factors mentioned have a negative correlation. However their studies have a western perspective. In the case of Bogotá, a large urban area in a developing country, the correlations are different. Education, age, and socio economic class have weak and negative correlation with electoral turnout. Instead, the correlations of the demographical factors and blank ballots are strong and positive. The gatherings from the interviews show a situation where vote buying is a common element at the elections, and heavy campaigning in the poor parts of town before the elections. These phenomenon seem to the base for the voting patterns in Bogotá.