The furore about 2012 English GCSE grading continues. For many headteachers, rightly committed to their own students’ success, the concerns are about fairness and what the results say about the success of their own teachers and schools. For Sir Michael Wilshaw, Her Majesty’s Chief Inspector of Schools, the experience provides an opportunity to ask questions about the rigour and fitness for purpose of English examinations. For the Secretary of State, the issue is the long-term reform of examinations.
None of these matters is straightforward. As Tina Isaacs points out in her excellent blog post, the technical challenges of managing an assessment system are daunting. But there are underlying issues which emerge from the debates so far which will need resolving before assessment policy can be settled.
The 2012 results were, as is well known, made up of different elements, including controlled assessment in schools and terminal examinations. Grades in January’s controlled assessments appear, on the Ofqual report evidence, to have been set too leniently, so that, as the Chief Executive of Ofqual said in an interview, the January candidates “got lucky”. Not surprisingly, there are renewed calls to reconsider the place of controlled assessments and to focus more on end of course terminal examinations.
Such a move would address the long-standing concern that 15 and 16-year-old pupils are “over-assessed”. But controlled assessments in schools allow assessment of aspects of learning which are difficult in a two or three hour final examination: speaking and listening in English, practical performance in drama, sport, art and design, music performance, experimental work in science. Any assessment activity is partial: it always samples a candidate’s performance. In other examinations – the driving test, graded assessments in music – assessment of practical performance is core to the final result. The technical challenge is to get the balance right between controlled assessment and external examinations in ways which are fit for the different assessment objectives.
A second feature of the debate which has caused immense argument comes from the introduction of “comparable outcomes” following a 2010 Ofqual decision. ‘Comparable outcomes’, designed to ensure that outcomes in successive examination years are broadly comparable, underpinned the setting of overall 2012 grade boundaries, and thus began to move examinations toward the pre-1986 model of “norm” rather than “criterion” referenced assessment. Norm and criterion referencing hang over the debate. In norm referenced assessment, candidates are assessed against each other, and the highest performing are awarded the highest grades. In criterion referenced assessment, candidates are assessed against a defined standard and all who meet it are awarded the highest grades.
We know, of course, that the most powerful indicator of any candidate’s likely attainment in examinations is prior performance. Since the Key Stage 3 tests were abolished in 2008, the principal measure of prior performance has been the results of Key Stage 2 assessment, which allows for attention to comparable performance as a guide to the calibration of subsequent outcomes. The introduction of “comparable outcomes” is not, strictly, norm referenced assessment, though it is obviously closer to norm referencing in determining awarding practices.
The implications of this are profound for secondary schools and the way they are held accountable. If “comparable outcomes” put a brake on GCSE performance – because prior attainment is the major determinant of subsequent attainment – then the long-term performance of secondary schools is, to some extent, pre-determined. If England is to move more decisively towards norm-referencing GCSE, then the concept of floor targets– which now require each school to secure GCSE 5A*-C grades including English and Mathematics for at least 40% of students — becomes difficult to manage. There are questions, moreover, about school-to-school support – the government’s principal tool for school improvement — since, as one outstanding headteacher provocatively asked, “why would I help a school next door? Every student in my neighbouring school I help push over the C/D borderline reduces the chance of one of my students securing a grade C”.
Assessment issues are always complex. Looking forward from 2012 there are tough policy questions to be asked. What range of skills and competences do schools, awarding bodies and politicians want GCSE to assess? What assessment approaches are most likely to secure this? Should the examination system ration success or do all young people, appropriately taught, have the potential to achieve at the highest level?