There is clear evidence that pupils with the same basic ability in maths are getting higher A-level grades than before, says a leading researcher in the field of academic assessment.

This probably means that standards have fallen, according to Professor Carol Fitz-Gibbon, who was speaking at a TES-sponsored debate on standards in maths at the annual meeting of the British Association for the Advancement of Science.

Professor Fitz-Gibbon attacked the Government’s “unrealistic” national training targets, and criticised the Office for Standards in Education for its “unreliable and unvalidated” inspection methods. She called for nationwide statistical monitoring of A-levels, GNVQs and degrees. “Candidates with the same basic skills are now getting higher grades in mathematics A-level than similar candidates obtained several years ago,” said Professor Fitz-Gibbon from Newcastle University.

“There has been a steady drift to higher grades. The interpretation of these differences has to be tentative. Candidates may be getting higher grades either because grades are easier to get or because the candidates are better taught or working harder.”

However, she said, if pupils were working harder, universities and colleges would be expressing satisfaction. “In contrast, there is considerable concern. ” At the same time, she said, it should not be forgotten that A-levels represent a high standard by international comparisons, nor that schools have social as well as academic outcomes.

Professor Fitz-Gibbon called for the re-establishment of the Government’s Assessment of Performance Unit which, she said, “disappeared without trace into the School Examinations and Assessment Council. Nothing has taken its place.

“The APU did the job properly using representative samples and items repeated over several years and which were consistently marked. The APU collected the necessary quantative and qualitative evidence and took account of . . . ever-changing syllabuses.”

The criticism of OFSTED follows last week’s publication of a report by the Organisation for Economic Co-operation and Development written by former TES assistant editor Caroline St John-Brooks. This suggested that the new inspection system is unlikely to raise standards without a more rigorous assessment of strengths and weaknesses and without better training for weak teachers.

Professor Fitz-Gibbon pointed to some possible factors contributing to the alterations in standards. The “unrealistic” national targets expect so much of average-ability students, she said, that there may be a downward pressure on standards. The targets “roughly require the average student who used to get a CSE grade 4 to perform at a level equivalent to the top of a grammar school.

“Universal acceptance of this political idea simply illustrates the lack of any sense of quantification among politicians and their advisers. What the educational system needs is realistic benchmarks, not arbitrary targets. It is unlikely the targets could be met without standards being lowered.” Other contributing factors included: * universities’ need to fill places; * pressure on exam boards to keep their customers - no board wants to have a “difficult” syllabus; * students’ belief that the best salaries are in business and finance rather than science and engineering; * lack of standards in degree courses. Universities, she said, should take part in national, blind marking of randomly selected scripts to see if standards are constant.

Professor Fitz-Gibbon also listed some factors which she said merited further research. These included teacher stress, lack of job opportunities and a stable teaching force, and the bleak graduate employment market.

A series of papers from academics presented at the TES debate give rapid and badly-planned curriculum changes which ignored the views of professionals a large share of the blame for the decline of maths. The papers urge the Government to work with academics, as well as professional scientists and engineers, to find a way forward.

“Over the past nine months, three separate reports by distinguished independent bodies have concluded that there is something very wrong with mathematics in schools,” writes John Hogan, professor of mathematics at Bristol and this year’s president of the mathematical sciences section of the BAAS.

He points to a study from the Engineering Council last spring which found that the majority of engineering lecturers “believe the mathematical knowledge of first-year undergraduate engineering students is weaker than 10 years ago”.

According to Dr Tony Gardiner, from Birmingham University’s school of mathematics and statistics: “Most of the changes were based on hope rather than careful ananlysis, discussion and testing.

“We are now in the embarrassing position where official reports continue to assert that everything is improving while mathematicians and teachers know from their daily contact with students that something has gone seriously wrong. ”

* The inspection system run by the Office for Standards in Education is “grossly unfair” and an attack on local democracy, according this year’s president of the education section at the British Association.

Chris Tipple, chief education officer in Northumberland, said in his presidential address that future educationists “could only see the system of the 1990s as a corrupt one whereby public monies had been removed from local authorities to be syphoned back to private providers, thus undermining the quality control function that the public had every right to expect from local authorities.”