Pupil Performance in Data Manipulation and Its Relationship to the Skill of Interpretation

Abstract

The ability to read data in graphical forms and to interpret it by looking for general patterns or trends is an important skill in children's science education. When viewed from a procedural basis, such skills become key areas of competence in science. The framework developed by the Assessment of Performance Unit (APU) recognised the importance of reading and interpreting data as elements of a ‘process’ (i.e. content‐independent) science curriculum. However, secondary analysis of APU performance data suggests that data manipulation should not be seen as an independent process skill, since children respond to assessment items in a number of ways, which are difficult to predict. Indeed, pupil performance on different components of the same data handling question is often very variable; indicating that these questions carry ‘hidden’ cognitive demands for pupils. It is argued that in order to tackle these hidden demands, pupils require an integrated understanding of graphical representations based on a sophisticated procedural model. Implications for changes of emphasis in science teaching and the reliability of summative assessment items are discussed in the light of these findings, since both issues impinge directly on the current National Curriculum arrangements in England and Wales.