Returns to initial years of formal education: how birthdate affects later educational outcomes
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The primary school entry policy in New Zealand is different from the policies of other developed countries like most of the European Union, the United States of America (USA), the United Kingdom, and Australia. In most countries, children start school at a fixed date in contrast to New Zealand where there are rolling admissions and children can start school right after their 5th birthday. Schooling from ages 6 to 16 is compulsory for every child; primary school term 1 in New Zealand starts in February; and the primary education system runs from Years 1 to 8 (Ministry of Education 2015). If a child’s birth date is between January and May, the young student will typically spend the year he/she turns 5 in Year 1 and the next year in Year 2. If a child’s birth date is between June and December, the student will spend the year he/she turns 5 in Year 0 and start Year 1 the following February. This means that the date of birth of the child affects the number of months/years spent in primary school and may further result in different educational outcomes.
My thesis uses a three-pronged approach to exploit the unusual features of the New Zealand system to test whether additional time spent in school raises subsequent achievement. First, I replicate a Dutch study by Leuven et al. (2010) where a similar system is in operation. In the Netherlands like in New Zealand, schools have a rolling admissions policy and children can start school right after their 4th birthday instead of 5th as in New Zealand. Dutch children with birthdays during, before and after the summer holidays are placed in the same class. Leuven et al. (2010) indicated that these two features of the Dutch schooling system create adequate exogenous variation in children’s enrolment opportunities to identify the effects of additional early formal education on later test scores. My thesis replicates Leuven et al. and finds some notable differences. This replication, in addition to being of interest in itself, serves as a useful starting point for analysing the New Zealand school system as the Dutch system is very similar.
Second, I apply a similar approach to New Zealand data and analyse the effects of school start on long-term educational achievement in New Zealand. Specifically, I focus on National Certificate of Educational Achievement (NCEA) and University Entrance (UE) results. Controlling for demographic and socio-economic characteristics, I find that an additional month of schooling (before the start of Year 2) increases the probability of achieving NCEA level 1 by 2.2%, NCEA level 2 by 4.2%, NCEA level 3 by 6.2%, and UE by 5.2%. Thus, differences in the timing of birth – and hence in school start – seem to have large effects on achievement even years later, in high school.
Finally, I subject my main results to a series of robustness and falsification checks. I also investigate whether the effects are homogeneous across socio-demographic groups or whether they are concentrated in certain sub-populations. I find that the effects are the strongest among male, Māori, and decile 5-7 students.