Homicides and Social Vulnerability: A Relational Study of the Brazilian States

Abstract

The objective of this study was to design the relational space of homicides in the states of Brazil. To this end, we employed two statistical techniques: the first, Pearson correlation, which seeks to identify and measure connections between variables, highlighting their interrelations. The second, Principal Component Analysis (PCA), transforms an initial set of variables into a reduced set of independent variables that maintains the essence of the information from the original set. The data show that states with favorable circumstances in certain years tend to have lower homicide rates. However, states with high homicide rates also face challenges such as illiteracy, social vulnerability, and a higher number of mothers as heads of families. PCA shows that locations/years with high drug seizures, higher incarceration, quality education, more employed people, and fewer mothers as heads of families with teenage pregnancies have lower homicide rates. It also indicates that, despite increased income equality, social vulnerability remains high. The results demonstrate that the issue is historical and complex, and that governmental actions have the capacity to influence the scenario of homicidal violence. This research supports the assertion that, for reducing homicide rates, preventive and repressive governmental actions are essential.