We can have tests that are trustworthy
Meanwhile, research conducted in England by the University of London's Institute of Education and King's College and the survey conducted by the NUT of its members in schools has now been completed*. The results demonstrate convincingly that the Government has a unique opportunity to recommend arrangements for 1997 onwards which will ensure that assessment of the national curriculum can both diagnose pupils' educational needs and secure accurate information on pupil achievement while at the same time gaining the confidence of the teaching profession. How can that be done?
The research and the results of the survey indicate some key issues in teacher thinking. For example, at key stage 1 the organisation of the tests still tends to disrupt the operation of schools and the planning of the curriculum while supply cover made a significant positive impact on the administration of the tests. At key stage 2 the external marking system came in for much criticism, with teachers feeling that it was unfair and unreliable. In key stage 3 English the tests, the mark schemes and the quality of external marking were a major issue.
A central thread running through both the findings and the research is that teachers need to "own" assessment. For example, it is apparent that, for many teachers at key stage 2 and at key stage 3 English, external markers have not resolved the issue of "ownership".
Professional issues remain to the fore. Commitment to teacher assessment remains as strong as ever. However, this view was further analysed by the research. At key stage 3 there were wide variations in attitudes. Teachers of English found that the tests hindered learning, teachers of mathematics found that the tests supported learning and science teachers felt a bit of both. Science teachers at key stage 3 were also more likely to believe that school-based tests were more important than ongoing assessments.
A clear message from the research is that teachers do not oppose testing but are in favour of assessment arrangements which support teaching and learning. Whether external tests are used is not the central issue. If tests provide high quality information and can be seen to enhance assessment the debate begins to edge towards consensus on their purpose and use.
In this context, the findings at key stage 1 are significant: Almost half the teachers (44 per cent) said that there was some diagnostic information to be derived from the tasks and tests, while 47 per cent said that their focus on assessment had been enhanced since their advent.
While the disruption caused at key stage 1 remains an issue, the greater element of teacher control over administration and marking has clearly helped, as has the ability to exercise more professional judgment over implementation.
The research and survey emphasise the case for change in 1997. The union believes that SCAA can bring forward proposals which reflect their message and the profession's commitment to assessment. Ongoing teacher assessment should provide the basis of national curriculum assessment at key stages 1 to 3, with safeguards on workload and quality assurance to support consistency and accuracy.
Teachers themselves should take part in that quality assurance. Tests have a role in confirming teacher assessment, but used according to professional discretion. Formative assessment should be the basis of reporting to parents.
Teachers must also be involved in drawing up national guidelines on assessment and in the creation of tests used in quality assurance. Sufficient resources should be specifically identified and allocated to schools.
Key stage 3 English marking, the effect of the tests on the operation of schools and the adequacy of supply cover are all real issues which can be addressed for 1996. After that it is up to the Government to seize the opportunity now offered.
*Reports available from the NUT Doug McAvoy is general secretary of the National Union of Teachers.