Gastos públicos estaduais em educação e qualidade do ensino médio

Abstract

Due to the small number of studies focusing on high school, this dissertation aims to investigate how government expenditures affects the high school educational quality (measured by math notes from Prova Brasil), through a panel data analysis using FINBRA and SAEB database, to the years of 2011 and 2013. Thus, this study main contribution comprises the usage of control variables that may influence the high school student performance, such as the average education of those over twenty-five and the percentage of teachers with higher education per student. Besides that, we considered some other relevant variables to evaluate the students’ performance – namely, the number of school establishments, the number of students enrolled in the public middle schools, average number of students per class, number of libraries per school and number of schools. In response to expenditure on education in 2013, which grew exaggeratedly, a dummy for this year and a variable of interaction between this binary and expenditures were included. We observed that the teacher quality is the main variable to determine the students’ performance. We also saw that the average education of those over 25 years old positively influences the quality of students – this evidence may display a knowledge spillover. Moreover, the results of this analytical exercise show a weak relationship between government spending and student performance on standardized tests. The logarithm of expenditure per student did not present a statistically significant coefficient. However, when analyzing the impact denoted by the interaction between the dummy and the variable spent in 2013, it is estimated that the parameter is significant and negative. Hence, the grades presents a decreasing of 0.006% when the expenditures per student increases by 1.000%. This demonstrates the existence of a negative relationship between public education spending and ANEB math grades. I.e., the increase in education spending by Brazilian states causes a slight reduction in student performance, which demonstrates inefficiency in the application of expenditures.