The Impact of Early Schooling on Subsequent Literacy and Numeracy Performance - Estimates from a Policy Induced 'Natural' Experiment

This paper exploits a policy-induced natural experiment that occurred in South Australia in the mid-1980s to generate a 'causal' estimate of the effect of schooling on the literacy and numeracy performance of school students in their middle years of secondary school (in Year 9 for most students). The Early Years of School policy changed the way that an identifiable subset of students progressed through junior primary school, causing them to obtain an additional year of schooling for any completed Grade or level compared with their predecessors. The impact of the policy change on the age-grade structure of student cohorts in South Australia is captured between two waves of longitudinal data. Based on the analysis of the impact of this policy change, it appears that an additional year of junior primary school increased the numeracy and literacy performance in mid-secondary school significantly, by around one half of a standard deviation. These effects were the same for boys and girls and were similar across the distribution of ability - they were the same for low school achievers as high ones.