Essays in Economics of Education

This thesis is an examination of how aspects of the English secondary school system affect attainment of pupils, particularly those who are disadvantaged. The analysis is based on administrative data for all pupils enrolled in state schools in England. The thesis includes three self-contained chapters. In the first chapter we study whether substituting family with school inputs in the education production function of high ability pupils with low socio-economic status has an impact on their achievement in the exams at the end of compulsory education. We consider a selective, well-resourced boarding school admitting an unusually high share of talented pupils from disadvantaged backgrounds and we estimate the effect of attending it with propensity score matching to obtain comparable control groups in selective day schools. Our main finding is that the probability of being in the top decile of achievement in the exams increases by about 17 percentage points compared to the baseline of 59% for controls.

The second chapter investigates whether gender segregation in secondary schools affects achievement and subject choice in non-selective schools in England. The empirical analysis is based on a value added model for achievement and a linear probability model for subject choice, both of which incorporate neighbourhood fixed effects. A robustness check based on a reasonable assumption about the relationship between the selection on observables and unobservables reveals that gender segregation has no effect on achievement of girls and a small effect on achievement of boys in english; in addition it does affect the probability of taking advanced science subjects at A-level for girls. My main finding is that girls from disadvantaged background who attend single sex schools are 2.6 percentage points more likely to choose an advanced science subject at A-level compared to a baseline of 7.3% in co-ed schools. Using a survey of students in England I find that girls and boys in single sex schools have less gender-stereotyped tastes and self-assessment of their abilities. These results support the hypothesis that girls in same-gender classes are less exposed to gender stereotypes, therefore more confident in their abilities in science and maths and more motivated to study these subjects.

The third chapter explores the effects of school competition on the academic per- formance of pupils. In the early 2000s the Labour government introduced academies, a new type of state-run school managed by a team of private co-sponsors. This reform broadened the choice of schools available to pupils and their parents increasing competitive pressure in the education sector. I use administrative pupil-level data to evaluate whether pupils in traditional secondary schools located near academies were affected by this new competition in the education market. Credible causal estimates of the short term impact of academies on neighbouring state schools are obtained by exploiting variation in both the timing and the number of academy entries. I find small positive effects on achievement in schools located within three miles from an academy: this finding suggests that increasing competition in the education market in England does not affect negatively the academic performance in less popular traditional schools and instead results in modest benefits particularly for more disadvantaged pupils.