209
referencing concern academic theories of assessment, it is an aspect of the modem
educational Zeitgeist that cannot be claimed to affect the examining boards uniquely.
However I submit that a serious misunderstanding of the basis on which examination
grades are awarded has disturbed in an important but unacknowledged manner the
subsoil of the English social consensus around examination grades. The fallout from
the debate in terms of the breakdown in confidence in the reliability of examination
standards provided the justification for many of Dearing’s recommendations and the
whole of Guaranteeing Standards. This section will analyse how this has come about
and suggest that the confusion in public understanding as to how the examining
boards reach their decisions in awarding grades has been a definite factor in
undermining their reputation for producing results that are comparable, valid and
reliable. The doubt cast upon their performance has in turn facilitated the continuing
centralisation of control over them. I have identified this doubt, and the resulting
increased central control, as an effect of the underlying breakdown in social trust
which is at the root of the quest for accountability. Therefore I contend that this issue
is a fundamental one.
Any national qualifications system is deeply culturally embedded and dependent on
the society’s acceptance of it. When a series of technical developments in assessment
practice had the unintended consequence of rupturing the English consensus over the
basis of the qualifications system, the result was bound to be a destabilising factor.
Yet the norm∕criterion referencing issue is rarely highlighted in the regular debates
concerning the drive to maintain standards over time and through qualification
change. The crux of the issue is clear to academic analysts like Kathryn Ecclestone:
...reliability is still a central goal of quality assurance procedures used by
awarding bodies and the QCA. More recently this goal has been in conflict with
validity as a key feature of outcome-based systems [like]... NVQs and GNVQs.