Abstract:
This article is a written version of a plenary presentation at the CLESOL Conference in Wellington in July 2014. It considers how we can conceptualise quality in language assessment through exploring different meanings of the term "standard", and in particular the contrast between standardised tests and standards-based assessment. The criteria of validity and fairness are introduced and the applied to three measures of academic writing ability: Task 2 of the IELTS Academic Writing Module, the integrated writing task in the internet-based TOEFL, and the NZQA EAP unit standard on writing a crafted and researched text. Each has its strengths and weaknesses in terms of validity and fairness, so that there is no single best way of assessing the construct.