Deborah Wilson
Performance tables for UK secondary schools have been published annually since 1992. In 2003, for the first time, these tables additionally include a measure of the educational 'value added' by a school to its pupils. This paper provides the first large scale analysis of the likely impact of the new value added performance indicator on the rankings of schools in the resulting league tables. Our analysis employs a national dataset of matched exam results, recently released by the Department for Education and Skills (DfES), which includes the results of the cohort of pupils who sat Key Stage 3 (KS3) exams at age 14 in 1997, and GCSE (or equivalent) exams at age 16 in 1999; this yields data on over half a million pupils. Using this dataset we have replicated five performance indicators which have been or will be published in the UK. In particular, we focus on the key pre-2002 PI, the percentage of pupils gaining at least five GCSEs or equivalent at grade C or above (%5A*-C), and the new value added indicator (VAcap). At a national level, we investigate the relationships between both the indicators themselves and the rankings which result. We then focus on one LEA, Bristol, and show to what extent school positions in the league tables are sensitive to the PI employed. We find a low degree of correlation between %5A*-C and VAcap and the resulting rankings, both at national and local level. This is reflected in the degree to which Bristol schools' ranking positions change when different PIs are employed. We conclude that value added does provide a more accurate measure of school performance and hence should help parental choice. We provide evidence, however, which suggests that a single PI, representing a school average value added score, may not be sufficiently informative.
Published in Public Money and Management 24 (1) 37-45 (January 2004)Note: some of the documents on this page are in PDF format. In order to view a PDF you will need Adobe Acrobat Reader
