Olmo Silva (LSE): "School types: evidence on the relative effectiveness of different school arrangements" (with M. Bertoni, F. Gonshorek and T. Klein, with M. Bertoni, F. Gonshorek and T. Klein.)

To get invitation for Zoom, please contact the seminar organisers.



Within the public-school system, a variety of school types coexist – each with its own distinctive features. Such differentiation has emerged as part of a new paradigm that promotes choice as part of a quasi-market in education. Notwithstanding such trend, comparable evidence on the relative effects of alternative arrangements on students’ achievements is scant: disparate estimates have been obtained with different methods – making like-for-like comparisons difficult. In this paper, we study the relative effectiveness of several school types operating in the English school market using a unified empirical framework that yields equal-footing (comparable) estimates. In order to do so, we use an identification technique that combines the insights of research design with the principles of mechanism design. This allows us to estimate school-type effectiveness by comparing students near admission cut-off while synthetically accounting for the structure of their preferences and admission priorities at different schools. We apply these methods to data from Birmingham - the second largest city in the UK and home to a very ethnically diverse population. We find clear evidence that selective (grammar) schools significantly improve students’ attainments: every year of additional exposure to selective education increases students’ end of secondary school achievements by 5%-6% of a standard deviation. Conversely, we find little evidence that autonomous (academy) schools improve attainments – irrespective of whether these are ‘conversions’ or school ‘take-overs’. Finally, we find no evidence that religious education has a positive effect on students’ learning.