A science in the service of an art? : the use of 'value added' analyses of school performance to aid school improvement
The thesis is concerned to explore whether and how ‘value added’ data analyses can contribute to school improvement, and to identify some of the conditions under which this might be so. In the course of conducting the study, the author experienced a tension between the ‘research’ and the ‘development’ dimensions of her work, and this is used to inform the outcomes of the thesis. The thesis is underpinned by three related aims: first, to provide a historical and evaluative overview of how the idea of ‘value added’ came to enter and influence the debate on educational quality in England. The study of the literature demonstrates that the main principles of ‘value added’ were already well developed before the term was in common use; it also reveals that the ambiguities in the term are not merely reflective of disagreements about how best to calculate value added but actually central to how the idea has been made to function within a particular political agenda for education having to do with ‘raising standards’. Because of the relentlessness of that agenda, ‘value added’ measures of attainment have undergone considerable methodological development over the past ten years, to the point where sophisticated statistical data on pupils’ and schools’ performance is being generated and used by government agencies, LEAs and schools themselves to an extent virtually unknown elsewhere. However, this thesis indicates that the technical and conceptual issues involved in putting such data to practical use in schools are likely to test the interpretative and organisational skills of users. The literature search confirms that not much investigation has been done into how data is actually used, but that what there is suggests some important lines of inquiry. The second aim of the thesis is accordingly to explore, through a small-scale empirical study, the use by secondary school staff of value-added data as exemplified by the NFER’s value added service QUASE. The study was conducted in nine schools, with staff at senior and middle management levels, and focused on mathematics, English and science departments. The third aim of the thesis is to assess how far the case-study findings might shed further light on the issues entailed in using such data for school improvement. The evidence suggested that value added data are seen as complex and often ‘high stakes’ and that – at the time of this study – the uses of value added data were rather more limited than expected; furthermore, the meanings of value added would seem to be socially constructed by the political and institutional environment, and to be closely related to individual teachers’ values and attitudes. This in turn suggests that better insights into, and management of, ‘the psychology and sociology’ of how value added data are perceived and used are necessary. Nonetheless, the study concludes that there is potential for value added analyses to contribute to school improvement under certain conditions; crucially, the study indicated that these included a culture which emphasised self-evaluation – rather than external accountability – within the school or subject department, combined with input from a ‘champion’ or facilitator who understood the technicalities and significance of the data. Value added analyses seemed to be used most actively by,staff who were able to use them ‘heuristically’, that is, to pose informed questions about teaching and learning (rather than as literal truths or statistical fictions). The study argues that ‘value added’ measures of performance are accordingly better understood as a technology than a science – that is, a practical application of knowledge which interacts dynamically with its social and cultural environment.