Crescimento educacional brasileiro : uma análise da distribuição e disseminação dos efeitos espaciais

Abstract

This doctoral dissertation presents three essays in Economics of Education that use data from Brazilian municipalities between the years 2000 and 2013. In the first, the objective is to analyze the distribution of education, classified in quantity and quality. The methodology was based on the Growth Incidence Curve (GIC), in which educational data were inserted. Educational growth was identified in the period, and the main results showed that the growth of quantity and quality of education presents different types of distribution. While the quantity exhibits a pro-poor behavior; the quantity is "anti-poor". It was also observed that the income factor becomes less determinant in the educational growth rates, the higher the school performance of the municipalities. The second essay analyzes the relation of the educational quality of a municipality with the educational quality of the neighboring municipalities. The results identify a strong dependence, suggesting that spatial structure influences school performance. So that the performance of a municipality is positively associated with the school performance of the neighboring municipalities. It was also observed the effect of higher education in the municipality itself, and that this effect is associated with its quality and not with the fact of its existence. Other measures related to teacher training also have a strong association with school outcomes. Considering the relevance of education and the results obtained in the second test, the third essay considers the heterogeneity of the regions within the spatial approach. The results indicate that socioeconomic status, such as parental schooling and student color, is strongly associated with school performance. The teacher effect was also verified, that is, the better the concept of universities and the higher the proportion of teachers with post-graduation, the better the school result of the municipality. Spatial heterogeneity and autocorrelation were important in the education of Brazilian municipalities.