Design principles and outcomes of peer assessment in higher education

TLDR

The study aimed to identify effective ways to organize peer assessment of written assignments in university history courses. The authors developed and implemented multiple peer assessment designs in courses, evaluating outcomes based on student revisions, grades, and perceived progress in writing skills. Students reported improved writing after peer feedback, teachers observed more structured interaction, yet no significant grade differences were found, and effective designs involved timing separate from staff assessment, reciprocity, and small feedback groups of three or four.

Abstract

This study was aimed at finding effective ways of organising peer assessment of written assignments in the context of teaching history at university level. To discover features yielding optimal results, several peer assessment designs were developed, implemented in courses and their learning outcomes evaluated. Outcomes were defined in terms of the revisions students made, the grades of the written products, and the perceived progress of products and writing skills. Most students processed peer feedback and perceived improvement in their writing as a result of peer assessment. Significant differences between grades of groups using or not using peer assessment were not found. Most teachers saw better‐structured interaction on the subject of writing problems in their classes. Important design features seemed to be the timing of peer assessment, so that it will not coincide with staff assessment, the assessment being reciprocal, and the assessment being performed in feedback groups of three or four students.