The P scales: how well are they working? Francis Ndaji and Peter Tymms The National Curriculum of 1988 provided a common curriculum for all primary and secondary schools in the state school sector in England, Wales and Northern Ireland, and ensured that schools in all local authorities had a common cur- riculum. However, pupils with special educational needs had attainments that fell well below those that the National Curriculum was meant to measure. It was not until the publication of the P scales in 1998 that it became possible for schools in England to measure the attainments and progress of pupils whose attainment levels could not register on the National Curriculum scale. This study by Francis Ndaji and Peter Tymms, from the Centre for Evaluation and Monitoring at Durham University, examines a number of aspects of the validity of the P scales data. In many ways the results are reassuring. The teacher ratings dis- criminate the levels and show that each level is more difficult to attain than those below it. However, the subjects are not so well discrimi- nated because they measure the same attributes. The P scales do not show any gender bias, indi- cating that teachers apply the level descriptions to both boys and girls in the same way. However, they seem to be applied in different ways to pupils with different categories of learning difficulties. The study also found that an attainment level in one subject area can be equated to the same level in another subject area. Despite suggestions from teachers, there is no evidence of a bottleneck at P8, or indeed, at any level of the P scales. In general, the P scales are working. Key words: National Curriculum, special educa- tional needs, level descriptors, attainment. Introduction The National Curriculum, established by the Education Reform Act of 1988, provided a national curriculum for all primary and secondary schools in the state school sector in England, Wales and Northern Ireland and ensured that schools in all local authorities had a common curriculum. The curriculum was associated with national statutory testing. However, some pupils with special educational needs have attainments that fall below those measured by the National Curriculum. Prior to the publication of the P scales in 1998, schools used the code W (working towards Level 1) during the collection of statutory end of Key Stage test/task results to describe the attainments of such pupils. The code W would not permit any discrimination in the attainment or progress of pupils in that category because W would not tell how far below Level 1 the pupils had attained. In addition, using the code W would not allow schools to meet the statutory requirement of setting performance targets that came into effect in 1998 as part of the drive to raise educational standards (Ndaji & Tymms, 2009). Therefore a new system of assessment was required that would enable teachers to measure the attainments and progress of pupils who are working below Level 1 of the National Curriculum. The P scales (Performance Scales) criteria, first published in 1998 by the DfEE/QCA (Department for Education and Employment/Qualifications and Curriculum Authority) as they were then known, consist of descriptions of what pupils are expected to do in order to attain a given level in different subjects. The scales were revised in March 2001 and pub- lished in a booklet entitled Supporting the Target Setting Process: guidance for effective target setting for pupils with special education needs (DfEE/QCA, 2001). One of the highlights of the revision was the subdivision of the three lowest levels of the scales (P1, P2 and P3) into six sub- levels, namely, P1(i), P1(ii), P2(i), P2(ii), P3(i) and P3(ii). This subdivision of P1, P2 and P3 was intended to increase the sensitivity of the scales at these lowest levels of attain- ment and progress. A minor revision was also carried out in June 2004. The P scales criteria currently describe attain- ments below Level 1 in all subjects of the National Curricu- lum, including English (five strands: Speaking, Listening, Speaking and Listening, Reading, Writing), Mathematics (three strands: Using and Applying Mathematics; Number; Shape, Space and Measures), Science (four strands: Scien- tific Enquiry, Life Processes and Living Things, Materials and Their Properties, Physical Processes), Information and Communication Technology (ICT) and PSHE, among others. Currently these criteria are published on the QCDA website (www.qcda.org.uk/resources/6058.aspx). A fuller examination of the P scales and their working can be found in Ndaji and Tymms (2009). From 1999 until 2004 QCDA organised annual collection and analysis of P scales assessment data through the Centre for Evaluation and Monitoring (CEM) at Durham Univer- sity. The aims of the annual data collection were (a) to collect enough data in order to present a national picture of the performance of pupils working below age-related expec- tations; and (b) from the resulting dataset, to prepare feed- back for schools to help in their self-evaluation and target setting. Currently the P scales data collection project is being run as a CEM project at Durham University with schools paying annual fees. RESEARCH SECTION © 2010 The Authors. British Journal of Special Education © 2010 NASEN. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. DOI: 10.1111/j.1467-8578.2010.00481.x