Although the general trend in exam results is upwards, few schools have yet found ways of improving consistently from one year to the next. This is one of the main conclusions of a research project directed by John Gray, David Hopkins and David Reynolds reported to the congress.
Three or four years of improvement seems to be a good run for a school before it starts to plateau. A "rapidly improving" school, they conclude, is finding ways of raising its pupils' exam performances by, on average, around one grade in one subject per pupil per year.
The researchers say their study offers a realistic view of how much and how quickly schools can improve over time. Drawing on data from all the secondary schools in three local authorities and using a rigorous definition of what part of a school's performance should count as "improvement", the project claims to be the first to look at such questions over the longer term.
In case studies of 12 schools, they found that teachers felt there had been substantial changes in the ways their school was run and organised, in its attitude and approach to planning, in the ways the curriculum was organised, and in its ethos or culture. Many schools had launched up to a dozen initiatives over the past five years and now felt they had done most of the "obvious" things to raise achievement.
What distinguished the more successful schools, the researchers said, was the way they had tackled issues at the classroom level.
Rapidly-improving schools had combined a number of short-term tactics to raise exam scores with strategies to change the quality of teaching and learning. The researchers say the latter area is where most schools should now be trying to concentrate their efforts.
The project is funded by the Economic and Social Research Council. A book on the research, The Improving School: Performance and Potential, will be published by the Open University Press in 1998.