Aplicação da análise multivariada no desempenho e hierarquização das escolas estaduais do Núcleo Regional de União da Vitória
Resumo
Resumo: A busca por melhores níveis de qualidade no ensino vem crescendo constantemente. A escola que deseja produzir ensino de qualidade deve primeiramente conhecer as necessidades de seus alunos e de sua equipe. Este trabalho busca identificar a importância das variáveis referente ao desempenho e a infraestrutura das escolas públicas estaduais do ensino fundamental pertencentes ao Núcleo Regional de Educação de União da Vitória (NRE-UVA). Para o desenvolvimento desta pesquisa, utilizou-se como instrumento principal a coleta de dados por meio de questionários que foram distribuídos a todas as escolas pertencentes ao NRE-UVA. Esses questionários foram divididos em três blocos: diretores, professores e alunos, com o objetivo de obter uma descrição quantitativa de informações. rganizaram-se os dados utilizando parte das variáveis correspondentes formando uma matriz para cada bloco, após utilizou-se para análise dos dados a técnica estatística da Análise Fatorial com o método da rotação Varimax que possibilitou obter uma estrutura linear do conjunto original de dados, por meio da sumarização da informação contida em um número de variáveis em um conjunto menor de fatores, a fim de obter os indicadores de maior poder de explicação em cada um dos três blocos. A partir desta análise utilizou-se sobre os indicadores revelados nos três blocos a análise de agrupamento (Cluster) usando o método de Ward para hierarquizar e identificar as escolas que melhor influenciam na qualidade do ensino através do desempenho e infraestrutura. Assim pode-se identificar a formação de dois agrupamentos de escolas na distância três do eixo das ordenadas, sendo o primeiro grupo formado pelas escolas que tiveram o melhor desempenho na indicação dos professores e o segundo grupo formado pelas escolas de melhor desempenho indicadas pelos diretores e alunos. Abstract: The search for better levels of quality in education has been growing steadily. The school wants to produce quality education must first meet the needs of their students and staff. This paper seeks to identify the importance of the variables on the performance and the infrastructure of public schools from elementary schools belonging to the Regional Education Center of the Union of Victoria (NRE-UVA). To
develop this research was used as the main instrument to collect data through questionnaires that were distributed to all schools belonging to the NRE-UVA. These questionnaires were divided into three blocks: the principals, teachers and students, in order to obtain a quantitative description of information. They organized the data using a portion of the corresponding variables in a matrix for each block, after it was used for data analysis the statistical technique of Factor Analysis with Varimax rotation method which allowed to obtain a linear structure of the original set of data, through the summarization of the information contained in a number of variables into a smaller set of factors in order to obtain indicators of greater explanatory power in each of three blocks. From this analysis we used on the indicators revealed in three blocks in a cluster analysis (Cluster) using Ward's method to categorize and identify schools that best influence the quality of education through the performance and infrastructure. Thus one can identify the formation of two clusters of schools in the three distance the ordinate axis, the first group of the schools that performed best on the suggestion of teachers and the second group of the best performing schools indicated by the directors and students.
Collections
- Teses & Dissertações [9330]