Abstract
In this article, we report the results of an empirical study of the impact of school inputs on pupils’ performance in private (independent) schools in the United Kingdom. We use a new school-level panel dataset constructed from information provided by the Independent Schools Information Service (ISIS). We show a consistent negative relationship between the pupil-teacher ratio at a school and the average examination results at that school. Our estimates indicate that the relationship persists even when we are estimating ‘added-value’ models conditional on previous exam results. The results are noteworthy in comparison with studies for the state sector, relatively few of which have found a consistent and significant effect.