ABSTRACT
In the USA, research on teacher judgement for summative assessment has generally fallen into one of two categories: (1) studies of teacher classroom summative assessment, that is, teacher grading practices; and (2) studies of how teacher judgement of student achievement accords with large-scale summative assessment, most often standardised tests. This article presents evidence that both kinds of studies have found mixed results. Perhaps not surprisingly, then, accountability policies in the USA have generally preferred to use standardised tests instead of teacher judgement measures. A brief first section sets the context for the two perspectives, classroom and large-scale, on teacher judgement.
