Teacher turnover: does it matter for pupil achievement?

We add to a small literature examining whether teacher turnover affects academic achievement, focussing on age-16, state secondary school students, using a unique dataset of linked students and teachers in England. We advance previous work by: a) looking at entry rates and student achievement in subject groups across which there is unlikely to be non-random selective assignment; and b) by looking at a context where students study a curriculum for two years during which they will generally be taught by the same teachers. This allows us to estimate the effects of getting a new teacher mid-way through the teaching period. Our identification is based either on a school fixed effects design which exploits year on year variation in turnover in different subject groups, within schools, or a student fixed effect design where identification is based on cross-section variation in turnover in different subjects, in the same school experienced by the student. Both methods give the same results: a higher teacher entry rate has a small but significant negative effect on students' final qualifications from compulsory-age schooling, despite organisational responses which assign new teachers to less risky grades. This result is robust to wide range of identification and robustness tests. Our findings point to the general disruption and lack of continuity in teaching as the main mechanism through which turnover harms student attainment.