Examining the Validity of Assessment of Students' Academic Writing Using Comparative Judgment

• In Event: Roundtable Session 31
  In Roundtable Session: Supporting Valid Decisions at the Individual and Program Levels

Abstract

To measure competences, performance assessments are mostly relied upon. To judge these, Adaptive Comparative Judgement (ACJ) is increasingly used. This method asks judges to compare two pieces of work and decide, holistically, which piece of work is better with regard to the assessed competence. In this study, we aim to underpin the validity of ACJ by exploring how judges approach the comparative judgement task. In an assessment of academic writing at a Belgian university, judges wrote down justifications for choice for one of the two papers. Results indicate that although most arguments are relevant to the competence, judges differ in the aspects they focus on during the comparison process. Additionally, judges differ in how they approach the judgement task.

Authors

• Liesje Coertjens, University of Antwerp

• Tine Van Daal, University of Antwerp

• Marije Lesterhuis

• Sven De Maeyer, University of Antwerp

• Vincent Donche, University of Antwerp