A distinguished professor retired from a university a few years ago. On his last day in the university, a keen journalist, notebook and pen quivering, came to interview the great man. “After 25 years as head of this department, Professor Scroggins,” he began, “what would you say is your proudest achievement?”.

The professor mused for a moment. “I think,” he intoned, “my proudest boast is that, on the day I leave, the course is exactly the same as it was on the day I arrived.” The bizarre belief that pride is to be associated with improving nothing at all during a career is, fortunately, unusual.

During the last six years I have directed three Leverhulme projects looking at classroom processes in the primary school.

The first studied class management, questioning, explaining and other important classroom skills; the second monitored teacher appraisal during the vital 1992 to 1994 period. My current project is looking at primary school improvement, especially in the field of literacy.

Trying to make primary schools better is a major industry nowadays, and there is no shortage of consultants eager to relieve them of half their budget to help them “improve”. Yet much can be done “in house”, and within existing resources.

A great deal of time has been spent in recent years on structure rather than process, on how many boxes must be ticked, rather than how to explain concepts better. Instead of having time to improve what they do in their classroom, teachers have been buried under bureaucratic demands that sapped precious energy.

Professor John Gray, of Homerton College, Cambridge, has analysed the first batch of school reports from the Office for Standards in Education. He reported that their recommendations were more to do with structures - improving the school development plan, changing the organisation - than classroom skill.

There are at least four important ingredients for improving primary education. The first is teachers’ professional skills. A summary of research by American Herb Walberg and two colleagues found that what they called “distal” factors, the American equivalent of national or local policies, were less influential on pupils’ achievement than “proximal” factors, those within each school. Of the school factors, the teaching skills of the staff came top. The most important of these was effective classroom management.

In theory, teacher appraisal was supposed to improve teachers’ skills. In practice, as we found during the Leverhulme Appraisal Project at Exeter University, some 28 per cent of teachers were only observed teaching one lesson, for as little as 10 minutes, and slightly over half said they never changed what they did anyway.

The time and space to be able to help each other improve was what teachers most complained that they lacked. Teacher appraisal had to be done in a hierarchical way, with the big cheeses appraising the little cheeses. Then along came a giant mouse, called “Time Constraints”, which gobbled lumps of both big and little cheeses, as heads and teachers were interrupted by numerous distractions. This is the Emmenthal cheese version of appraisal. Small wonder that many of the teachers we interviewed could not even remember their targets.

Time must be found to allow people to sit in on each other’s classes or visit other schools, and there must be a focus. We found that very few appraisers knew much about classroom observation. Peer appraisal can only work if teachers know how to analyse teaching and learning. Improvement does not happen by accident. In our appraisal study, we found that if observers wanted teachers’ respect, they had to have a proper professional’s grasp of the age group and disposition of the children, the material being taught, and be able to offer fresh insights.

The second vital ingredient for improvement is the raising of aspirations and expectations. I often visit German schools, and do not share the uncritical adulation of German education, except in the fields of vocational education, which is superb, and the teaching of lower achieving pupils. One reason why German children outperform British children in number (though not in geometry), is the higher test scores from pupils of lower ability. German teachers’ expectations of lower achievers are higher than I see in Britain, where it is almost as if we are resigned to being duffers at sums.

Third, teachers’ morale and attitude to their craft is an important factor. It is hard to improve what you are doing through clenched teeth. There are numerous examples of excellent leadership in primary education, but what is happening in Birmingham is of interest. Chief education officer Tim Brighouse has worked closely with primary and nursery heads to give primary schools more resources and to support teachers. Baseline testing of school beginners and trusting schools to set their own targets have replaced the whip and scourge approach of the Government.

The fourth feature of school improvement is the climate within the school. I have called schools with a positive attitude to improvement “dynamic schools”, in which people look at what happens in classrooms, reflect on it, and then implement judicious change.

After the mad whirligig of recent years, the operative word is “judicious”. For Professor Scroggins, of course, “judicious change” was the same as “no change”, but as the 21st century dawns, that cannot be the right recipe for the children who will spend most of their lives in its rapidly changing environment.

* Professor Ted Wragg is the author of An Introduction to Classroom Observation , published by Routledge.

* On October 6, The TES will publish a special pull-out section on the international conference on school effectiveness at the University of London Institute of Education, which begins on that day.