Ensaios sobre diferencial de desempenho escolar entre alunos de escolas rurais e urbanas no Brasil

Abstract

This dissertation is composed of two articles, the first of which is entitled "Factors that determine the differential of school achievement among students from rural and urban areas in Brazil", analyzed the factors that contribute to the differential of income of students attending schools in rural and urban areas. For that, we used information from the students evaluated in the 5th year of Elementary School in the Brazil Test of 2015, in the Portuguese and Mathematics subjects. The rural-urban school achievement differential was estimated by the method of decomposition based on regressions of refocused influence function (FIR), proposed by Firpo, Fortin and Lemieux (2007), combined with the re-weighting procedure inspired by Dinardo et al. (1996), which allowed the application of the traditional Oaxaca-Blinder decomposition method (1973) in distribution quantiles. The results pointed to a difference between students from rural and urban schools, showing an advantage for urban school students, where a large part of the difference between groups is due to the characteristic effect of the schools and the student's family, increasing over the years. Quantiles The second article was entitled "Rural-Urban Educational Migration: Analysis of the School Achievement Differences in Secondary School" and analyzed how the achievement of high school students who lived in rural areas was influenced by the school where they studied, whether In the countryside itself or when they were transferred to urban schools. A combination of two methodologies was used: the first consisted of a pairing method through observable characteristics among students from different schools, so that the only difference between one and the other was only the "treatment" (the zone of residence or Location of the school), the Coarsened Exact Matching (CEM) algorithm. And, for the decomposition, the method proposed by Oaxaca (1973) and Blinder (1973) was applied, which identified the variables that contributed the most to explain the differences in educational outcomes between the two groups. From the results, it was observed that, even controlling the individual and family characteristics of the student, the rural student who studied in an urban school presented a higher achievement than his peers who attended schools in the rural area. Such an occurrence indicates that it is better for rural students to study in urban schools, since these denote better infrastructure and more qualified teachers. As for the breakdown of differentials of students living in rural areas compared to those living in the urban environment, studying in similar schools, it was observed that even studying in similar schools they still had disadvantages in comparison to their colleagues due to individual and family characteristics.