Abstract

This paper tries to disentangle the relative importance of family and school inputs on a child's cognitive achievement as measured by her percentile score on a mathematics test. We replicate a study by Todd and Wolpin (2007) in the United States with Canadian data. In contrast to their work that uses state-level indicators of school quality, we estimate our model with data from Statistics Canada's National Longitudinal Survey of Children and Youth (NLSCY) which provides micro-level information on the family and school history of the child. The sample used for the analysis is based on the 7- to 15-year old longitudinal children who have completed at least two consecutive math tests. As in Todd and Wolpin, we conclude that cognitive outcomes are determined by current and past family inputs. Contrary to them, who find no impact of school inputs, we find that the quality of schools has a positive impact on achievement in mathematics.