How the learning environment affects the children’s school performance

Abstract

In this paper we estimate how the learning environment affects the children’s school performance. Our learning environment determinant is the percentage of the child’s peers having at least one parent with university education. In our regression we use Norwegian register data of all graduating secondary pupils during the years 2002 to 2007 and their parents. The results in our main analysis, using the final assessments as the dependent variable, indicate a negative peer effect of being in a good learning environment. This indicates that children’s school performance gets poorer when the parents’ education level at school increases. We carried out subsample analyses and controlled for school fixed effects in the main analysis to explore what kind of mechanisms causing our negative results. The results from the subsample analyses show that the well-performing pupils are more negatively affected being in a good learning environment, and the subsample reveals that there probably is no selection of well-performing pupils into schools. When controlling for school fixed effects in the main analysis the estimates became close to zero. This indicates no peer effect. The results suggest that the negative effect we found in the main analysis was due to school specifics as for instance the school quality or the teachers’ grade setting. To examine the school specifics more thorough we did an analysis using the examination grades as the dependent variable. The results from this analysis revealed a positive peer effect. This indicates that the teacher’s grade setting probably was the reason for our negative estimates in the main analysis. We also controlled for school fixed effects in this analysis and found no peer effect. This indicates that a child’s peers at school do not influence the child’s own school performance. The learning environment, as we measure it, does not affect the children`s school performance.