Recency effects in pairwise comparisons during peer assessment of oral presentations

Peer assessment is widely used in higher education. In courses that include oral project presentations, peer evaluations are typically conducted sequentially, with students assigning scores as each presentation concludes. This study examines a peer-assessment mechanism in which evaluators are required to make immediate forced pairwise comparisons when identical scores are assigned. Using data from live classroom deployments, we analyse whether such comparisons are influenced by presentation order rather than judged performance. Across sessions, evaluators consistently showed a preference for more recently presented projects when comparing equally scored submissions. This tendency increased both with the temporal separation between compared projects and with the score assigned to them, indicating that order effects are strongest for highly rated work. These results suggest that order effects can shape comparative judgments embedded within sequential peer assessment and should be considered when interpreting peer-based differentiation among similarly rated submissions.