Another year, another row over whether rising A-level and GCSE standards mean schools are improving or whether politicians or the exams system are making it easier for students to succeed.
Consider the two main aims of GCSEs. First, they are meant to assess the qualities of the student. In this, GCSEs do fairly well. Despite concerns about errant marking, pupils with As and A*s have mostly achieved high performance and will tend to be better than those with, say, a B grade. But exam results are also meant to be a check on national performance. So, are standards improving, constant or falling?
Here, the exams regime is on less secure ground. Any system that tries to compare the performance of millions of pupils, across hundreds of shifting syllabuses, can offer at best only hesitant conclusions. Exam content changes, as does the support given by exam boards. Rules that allow pupils to drop certain subjects are likely to boost figures: the strongest take the exams and results go up. So the "standards" question cannot be answered on the basis of statistics alone. Yet this is the function demanded of the results in August.
Ministers may bemoan the fact that those who question the figures are undermining pupils' hard work. But they have chosen to hold themselves, teachers and schools accountable for school performance, as measured by exam results. Given the political capital any government will make from improved figures, questions will continue about what the data really mean.
If it were serious about changing this debate, the Government could do two things. First, it could refrain from publishing national data until well after pupils receive their results. The standards row would be postponed and pupils could digest their results in peace. Second, it could question whether the amalgamated GCSE and A-level results are effective measures of educational performance. If not, other approaches, such as setting an unchanging bank of tests to a small sample of pupils, might provide better information.
GCSEs and A-levels would then simply be a check on individual pupils' performance, rather than as an assessment of the abilities of politicians to raise educational standards. They would be de-politicised, and much of the poison from this debate would drain. Is anyone listening?