Consensus moderation for quality assurance of assessment: Overcoming the illusion of consensus.

Abstract

Research at Griffith University is exploring the use of 'consensus moderation' (Sadler, 2009, 2010, 2011) as a process that can help to ensure consistent and appropriate academic standards when assessing the quality of students' learning outcomes. In principle, consensus moderation achieves agreement among markers about what comprises quality, and about the symbols (marks or grades) that are used to represent judgments about the quality level of students' work. The most directly observable result is marking consistency. Unfortunately, the mere existence of marking consistency does not necessarily signify consensus about the judgments of quality of the students' work. Marking consistency can, and often is, achieved in other ways. Consequently, the illusion of consensus may be created. This paper describes different ways in which consensus moderation processes may fail to achieve consensus. Understanding this is necessary for genuine quality assurance to be possible through appropriate policy and practice.