Aplicação da análise de Cluster à avaliação do desempenho municipal

Abstract

This paper exams the application of data cluster techniques to the classification of the performance of the cities in the state of Rio de Janeiro. This investigation uses the municipal performance values and criteria adopted in the calculation of the Municipal Quality Index MQI (Índice de Qualidade Municipal IQM) developed by the CIDE Foundation (Rio de Janeiro Data and Information Center) and reported by CIDE (2005). More specifically we use MQI s citizenship dimension (CID) which considers the fulfillment of the population s basic needs in each city (health care, education, security, justice and leisure). Different measurements (Euclidian Square Distance and Cosine) as well as grouping forms (Center, Ward, Between-groups, Within-groups, Nearest, Furthest, and Median) are employed in the data cluster analysis of the 92 cities of the state of Rio de Janeiro, generating classifications that were then analyzed and compared among each other. The cities are grouped according to their particularities, pointing out those cities that are above or below average in the following criteria: enrollment in elementary school and population between 5 and 19 years old; number of beds available on basic healthcare units affiliated to the public health system; number of civil and military police officers; number of theater houses; movie houses; libraries and museums; presence of institutions that deal with issues related to citizen rights: PROCON and Public Attorneys; number of houses reached by running water, sewer, and garbage collection. The cities that are part of each group were identified, and the similarities amongst the groups formed were pointed out by altering the methods used. At the end of the work we present suggestions for future developments within the field of this research