Research Summary · Education
What the examination selects for
Assessment reform routinely targets published criteria. Evidence from thirty-four schools suggests the influential channel is somewhere else entirely.
That high-stakes assessment shapes teaching is not in dispute. The mechanism has been surprisingly under-examined, which matters, because reform efforts target the mechanism they assume.
The standard assumption is that influence travels through published documents: the syllabus, the assessment criteria, the specification. Reform accordingly consists of rewriting them. The results have been mixed to poor, in ways the assumption does not explain.
The alternative account. Classroom observation and teacher interview evidence point to a different channel. Teachers form expectations about marking behaviour, and they form them principally from exemplar materials and from informal professional networks rather than from published criteria. Where exemplars and criteria diverge — which is more common than assessment authorities generally recognise — practice follows the exemplars.
This has a direct and slightly deflating implication. Rewriting criteria without changing exemplar material and examiner training produces negligible practice change; one system studied over two years showed effectively none. The document that governs classroom practice is not the document the authority thinks it has published.
What follows. The finding is mechanistic rather than prescriptive, and it does not indicate what assessment should look like. It indicates where the leverage is: in exemplars, in examiner training, and in the informal channels through which teachers infer what will be rewarded.
It also suggests a diagnostic worth running. Any assessment authority can ask what its own exemplar materials imply, read as instructions rather than illustrations, and compare that with what its criteria state. Where the two diverge, the exemplars are what the system is actually teaching.
Corrections? Contact the desk.