Análise do Sistema Permanente de Avaliação da Educação Básica do Ceará (SPAECE):um estudo comparativo entre o efeito escola e os resultados de proficiência em escolas da rede estadual nos anos de 2012 A 2014

Abstract

From the Coleman Report (1966), which was established in its results, widely published, indicated that the main determinant of inequality in student performance was not the difference in the structure of the schools, but socioeconomic differences in their families. Highly publicized, successful completion of the Coleman Professor of the research report stated that the school does not matter and it spread including Brazil. The objective is to highlight the relationship between student performance in Portuguese and mathematics tests administered by SPAECE in the years 2012, 2013 and 2014 and the contextual variables: type of school (academic or professional), the School Performance (high or low), student Socioeconomic Status, Family Context student, Pedagogical Practices professor, School Climate School, Academic Expectations / Professional student and Democratic Management. These variables were constructed based on micro SPAECE data. He worked with eight high school schools, four academic and four vocational education. In each type of school they were taken both with the greatest proficiency in test Mathematics and Portuguese and two with less proficiency in SPAECE applications in the years 2012, 2013 and 2014. It took up these schools a simple random sample of 3,708 students distributed in the three years of the application, with a 95% confidence and error 0.05, and their respective directors and teachers of Mathematics and Portuguese. From the bank of micro SPAECE data in the three years, the variables were built up together with the proficiency of students in Mathematics and Portuguese in a SPSS database - Statistical Package for the Social Sciences, version for Windows 20.0. Working with this database were three multiple linear regressions with independent variables Proficiency in Portuguese, Proficiency in mathematics and the Middle Proficiency in Portuguese and mathematics and as predictor variables eight contextual variables. We used the stepwise method (input and output) in each regression was obtained a final model with six predictor variables. Observing the standardized coefficients of each model obtained it appears that the three predictor variables as larger weights in the independent variable were: Type of School, School Performance and School Climate. The predictor variable "Family Context was not included in any of the three models.