The use of teacher judgement for summative assessment in the USA

Studies of the use of teacher judgement for summative assessment in the USA are considered in two general categories. (1) Studies of teacher classroom summative assessment, that is, teacher grading practices, have historically and currently emphasised the lack of validity and reliability of these judgements. (2) Studies of how teacher judgement accords with large-scale summative assessment, most often standardised tests, have found teachers' judgements explain approximately half of the variance in student achievement, with wide variation among teachers. Not surprisingly, then, accountability policies in the USA have preferred to use standardised tests instead of teacher judgement measures. This article summarises research in both categories, describes some notable exceptions and makes suggestions for future research.