The role of aggregation in estimating the effects of private school competition on student achievement

This paper investigates possible reasons for the disparity in results in the private school competition literature. In particular, the focus is on the data set, the grade range, and level of aggregation of the competition variable, and on the choice of OLS or IV estimation strategies. The results show that the size and significance of the competition variable depends on each of these attributes, although the grade range of the competition variable has a slightly smaller impact than the others. Private school competition does not have a consistently positive, significant effect on student achievement.