A determinação de indicadores sócio-econômicos e classificação de um município paranaense através da análise estatística multivariada
Resumo
Resumo: Este trabalho analisa a importância de variáveis sócio-econômicas na definição de áreas prioritárias de investimento. Tomando como referência o município de Pato Branco - PR para o desenvolvimento desta pesquisa, partiu-se do pressuposto contido nas informações do Censo 2000, onde priorizou-se temas relacionados ao objetivo maior deste trabalho: saneamento, renda, escolaridade, habitação e demografia. Nesses, depois de gerada a matriz "bairros do município x indicadores", foram utilizados dois métodos de Análise Fatorial: Componentes Principais e Máxima Verossimilhança, a partir dos quais identificou-se os índices mais representativos. Para definir o número de fatores a serem extraídos, seguiu-se o critério de Kaiser e um mínimo para a comunaiidade de 0.7. O número de fatores foi confirmado adotando a estratégia sugerida por Johnson e Wichern (1992) que verifica se os pesos fatoriais para uma fator particular concordam. Esta primeira análise permitiu selecionar os indicadores com maior poder de explicação da variância observada, com base nos quais, em nova Análise Fatorial, foram escolhidos os indicadores finais. A obtenção dos indicadores finais permitiu a classificação dos bairros do município em ordem decrescente de carências sócio-econômicas, fornecendo elementos para um planejamento municipal direcionado para suprir essas carências, priorizando os bairros com menores escores naquela classificação Abstract: This study analyzes the importance of socio-economic variables in the establishment of priority areas of investment. The town of Pato Branco - PR was our reference for the development of this research, and Census 2000 was the source of information mainly in relation to the utmost objective of this study. They are: sanitation, income, education, housing and demography. After generating the "neighborhoods of the town x indicators" matrix with them, two methods of Factor Analysis were used: Principal Components and Maximum Likelihood, which served to identify the most representative rates. In order to define the number of factors to be extracted, Kaisehs criterion was followed and a minimum of 0.7 for the communalities. The number of factors was confirmed by adopting the strategy suggested by Johnson and Wichern (1992) which checks whether the factor weights for a particular factor agree. This first analysis allowed the selection of indicators with the highest condition of explanation regarding the observed variance, and they also served, from a new Factor Analysis, to pick out the final indicators. The obtainment of the final indicators permitted to classify the neighborhoods of the town in a decreasing order concerning socio- economic needs, and to offer elements for a directed municipal planning which can help make up for these needs, with priority to the neighborhoods with the lowest scores in that classification
Collections
- Dissertações [167]