Application of Principal Component and Cluster Analysis to Performance Indicators of Federal Universities of Brazil

The Decision No. 408/2002 of the Federal Court of Audit determined that the Federal Institutions of Higher Education (IFES) should incorporate in their Management Reports nine indicators to monitor the performance and improve the management of federal institutions. The objective of this research is to analyze the performance indicators of the Brazilian federal universities in 2015 through the multivariate statistical technique Principal Component Analysis and Clustering, thus allowing a reduction of the number of study variables and the grouping of the federal universities of the Brazil according to their performance indicators. Through the results of the Principal Components Analysis (ACP), it was possible to form three main components that, together, explained 74.20% of the total data variation. The first principal component accounted for 39.22%, the second, 22.86% and the third principal component accounted for 12.12% of the total data variation. It was also observed, through the Analysis of Clusters, the formation of nine groups, or clusters, of the analyzed federal universities. These clusters were explanatory in relation to ACP, joining the Federal Institutions that are similar in their values for each of the principal components analyzed.