The rise in the proportions of pupils achieving grades A*-C has also been prompted by two clear “jolts” to the system - the introduction of the GCSE and the changed climate of expectations in which schools now operate - according to Professor John Gray from Homerton College, Cambridge, and Professor David Jesson from Sheffield University.

Their research shows that in under 10 years, the proportion of pupils gaining five or more A-C exam grades has increased from 26 per cent to more than 44 per cent.

It also reveals that between 1992 and 1995 no fewer than one in four secondary schools achieved 10 per cent rises in the percentage of pupils getting over the five A-C hurdle.

Many schools are now entering pupils for at least one more exam each, and both pupils and teachers are encouraged to work harder, the researchers claim.

In addition, many schools have identified “borderline” pupils who could significantly boost the school’s performance on the key five or more A-C grades indicator if their work improved.

These pupils often receive additional counselling, individual support and, on occasions, extra tuition.

In some schools, one or two departments have shopped around for exam boards whose curricula are more closely aligned to what they are teaching.

The researchers say that before performance tables were introduced in 1992, schools had been entering their pupils, on average, for between seven and eight subjects.

Within a single year the average jumped by about one entry per pupil to between eight or nine subjects. The researchers said: “Pupils who were entered for more examinations appear to have achieved better overall results.

“General exhortations to schools to perform better, in combination with evidence - however crude - of their performance in the league tables, may well have stimulated schools to explore the extent to which they had developed a culture of achievement, both among teachers and pupils.

“Pupils may well have been encouraged to work harder as a result. They, and their teachers, have certainly been encouraged to work more strategically towards whatever outcomes have been examined.”

The researchers say that while competition between schools has not increased performance per se, it has focused attention on the results achieved by different schools. It has also convinced some schools that they should try to recruit more able pupils.

The research, funded by the Economic and Social Research Council for the Improving Schools project, shows that the most rapid period of year-on-year rises occurred between 1988 and 1992.

Since 1992, schools have been encouraged to develop their own approaches to school improvement, the research shows.

Professor Jesson said: “It is disappointing that the year-on-year increases since 1992 have slowed down. The average rate between 1992 and 1996 is running at about half the level of that which was achieved between 1988 and 1992. ”

The study also found that many schools did not sustain improvements for more than a year or two at a time and that only one in 10 was steadily doing better.

The researchers warn there is no room for complacency despite the recent increases in schools’ exam results. Professor Jesson said: “Few schools to date have succeeded in building self-sustaining cultures of improvement.”

Pupils often achieve their best grades in vocationally-directed GCSE courses such as business studies, electronics, horticulture and commerce, says a new review by the Office for Standards in Education (OFSTED).

Overall, such courses were found to be “an appropriate and manageable introduction to the world of work”.