Discovering abstentionist profiles in 2015 Catalan elections

Abstract

One of the main democratic act is to elect representers for a country Government. Voters have to choose the representers of political parties that best fit their interests. However, it is also possible to abstain, i.e., to give up choosing the representers. Lastly, the percentage of abstention has increased in many democracies and, consequently, many studies have been carried out in order to discover the causes of the abstention. Causes such as alienation and indifference for politics have been found and characterized, but it is not clear who is abstaining. Some studies in specific countries have related aspects such age, education or wealth with voting. Nevertheless, all these studies are based on questionnaires that could not be completely reliable. In the present paper we use decision trees, an artificial intelligence technique, to extract patterns that characterize the profile of abstentionists. Differently from other studies, we used a data base elaborated from public data bases with both socio-economic and electoral data. In particular, we focused our study in Catalonia and the electoral results to the Catalan Parliament held in 2015. We also propose the use of decision trees to analyze the database and to extract patterns to characterize different profiles of abstentionist people.