Corruption Levels and Country Cluster

Abstract

The objective of the chapter is to classify countries into different categories of corruption levels using the K-Means algorithm for 2012 and 2021. The sample is made up of an aggregation of 195 countries of different levels of income using data obtained from the World Development Indicators compiled by the World Bank and Transparency International. The optimal number of clusters is chosen at the elbow point, which is the point of the graph where the curve starts to flatten out. The optimal number of clusters chosen is three, which can be qualitatively interpreted as groups of countries with high, medium, and low corruption levels. The results obtained show that corruption has a directly proportional relationship with the number of adolescent pregnancies, gender inequality, overall inequality, and maternal mortality. While corruption levels have a negative relationship with levels of human development and the average years of schooling.