One of the functions of money, according to economists, is to act as a unit of measurement. Countless hours of economists’ time is wasted by constant adjustments to measurement to take account of the changing value of money – as a result of inflation or deflation. On the other hand, endless hours of family amusement have been caused over the years by reflections that ‘I used to be able to buy a Mars Bar for less than 10p’, and so on. On balance, most would agree that an absence of inflation, and the consistency of measurement would be a good thing.
Similarly, in the UK at least, the examination of school pupils and the grading of their efforts, has been subject to a form of inflation. Are grades in this year’s exams consistent with the same grades in the same exams in previous years? This kind of longitudinal consistency is vital if exams are going to be meaningful – for example as a guide to the probability of success in other academic courses, or as a means of comparison of the ability of those who take their exams in different years.
It’s clearly been a mistake that a larger and larger percentage of pupils have been gaining higher grades since GCSEs were first taken in 1988. The change in the proportion has been greater than the rise in ‘standards’ in schools and so a B grade in English Language GCSE 2011 means something different from a B grade in the same exam twenty years earlier. Of course this has to be balanced against a greater accountability in schools, an inspection regime which has teeth, increased scrutiny of what teachers do in the classroom. It is possible that more pupils deserve their B grade in English Language – just because more people have climbed Everest in the last twenty years does not mean that doing so has got easier, or that Everest has got shorter.
It is even possible that what is meant by proficiency in English Language has changed over a twenty year period – for cultural, technological and practical reasons. But this is not the point – a B grade (or any other) should mark a basic level of proficiency which is intelligible to those who will interpret it – in education, or in the workplace.
So, the solution to two decades of grade inflation is not to reverse the trend, and gently reduce the percentage of pupils gaining higher grades. To do this introduces a second layer of complexity in interpreting the results. Two decades from now, who will remember in which year grade inflation was reversed, or by how much. These changes make the nature of the signal given by the grade even less clear – the measurement is even more ‘noisy’. It would have been much more helpful to freeze the value of the GCSE at the 2011 value, and insist that any deviation from that point could only be because there had been a significant change in the capability of the educational infrastructure. This would be the examining equivalent of joining the Euro – or for historians of joining the Gold Standard.
If one does this, how does one discriminate at the top level? By introducing – once and for all – a new grade. Perhaps an ‘S’ level at GCSE, for the top 2-3% of candidates would accomplish this, although this is only worthwhile if the examinations themselves reward the most able pupils, but that is a whole new can of worms. The only tenable alternative is to shift to a whole new grading system, as Michael Gove proposes, on a numerical scale. On this score (alone?), he is absolutely right. If it is to have longitudinal integrity, however, there will have to be clear regulatory protection of the standard required for each grade, so that grade inflation – or its more recent cousin, grade deflation – becomes a topic in history, rather than in politics.
Click here to read the blog.