Sorting and inequality in Canadian schools

Researchers and educators often argue that a student's peers strongly influence his or her educational outcomes. If so, an unequal distribution of advantaged and disadvantaged students across schools in a community will leave many students doubly disadvantaged and amplify existing inequalities. We explore the relationship between the degree of sorting by socioeconomic characteristics, ethnicity and language across schools within a community and inequality as measured by the variance of standardized high school exam scores within the community. Simple cross- sectional estimates suggest a direct relationship between sorting by ethnicity and the variance of test scores, but no direct relationship between sorting by income or primary parent's education and the variance of test scores. We then implement a fixed effects estimator to control for endogeneity in the extent of sorting: the results indicate that sorting by ethnicity does not affect the variance of test scores, but that sorting by home language and primary parent's education does.social interactions, peer effects, sorting, classroom effects