Sustainability multivariate analysis based on the global reporting initiative (GRI) framework, using as a case study: Brazil compared to Spain and Portugal

The main objective of this research is to find the sustainability gradients of Brazilian companies and sort them as a function of economic, environmental and social components, using multivariate methods for dichotomous data, in order to establish the foundations for better knowledge of the trends and sustainability reporting habits of Brazil. The methodology used is a combined approach, comprising Principal Coordinates Analysis (PCoA) and Logistic Regression Model (LRM) to build an External Logistics Biplot (ELB). We found that the most sustainable companies are associated with great concern for biodiversity, conservation and air pollution prevention, as well as with the strong relationship with employees, human rights promotion and fighting against child labour and forced labour. We conclude that it is possible to obtain sustainability gradients not only for Brazil but also for any country that uses the GRI framework, applying the ELB. © 2017 WIT Press.