Perfil de aprendizagem escolar municipal: o desempenho dos municípios brasileiros na aprendizagem da educação básica

Abstract

The municipal learning profile can be represented by the school learning curve. It is the evolution of the learning of the students of a municipality throughout the basic school cycle, going from the 1st year of elementary school to the 3rd year of high school. It brings to the analysis of the profile of municipal education the longitudinal perspective of learning of school cohorts: students who started the cycle of basic education in the same calendar year. Two measures are drawn directly from the learning curve: the level of student learning, which is related to the school content learned by them, and the added value, being the growth of learning throughout the school years. The first one describes the contextual characteristics of the pupil and his/her learning history, while the second one indicates the contribution of the school system to learning in each year of the educational cycle. This work shows that to compare the performance among Brazilian municipalities, it is necessary to consider these two measures. There is a growth pattern of learning along the curve strongly related to the educational context of each municipality, represented by the level of learning prior to assessment. Municipalities that start from a lower level tend to add more learning in their students than the municipalities that depart from a more favorable context. Thus, to analyze municipal performance we used the Student Growth Percentiles (GSP), an indicator that standardizes the evolution of the curve between the evaluated units that depart from different educational levels. The results show that, despite the convergence of school learning between municipalities in different educational contexts, regional inequality tends to remain, because it is the more developed regions that concentrate the municipalities with the best performances. The construction of the municipal learning curve was carried out based on data from the SAEB and ENEM, both from INEP.