Evaluating Creativity and Accuracy in Humour-based Language Use

Abstract

Digital tools are increasingly used to deliver language assessments; however, the alignment between peer evaluations and instructor judgements remains underexplored, particularly in creative, humour-based tasks. This mixed-methods case study examined (1) the alignment between peer and instructor scores and (2) how peer and instructor comments aligned with or diverged from the assessment rubric. Thirty undergraduate English-major students, working in seven groups, produced 3- to 5-minute humorous video commercials for a Technology Enhanced Language Learning course. The videos were assessed by the instructor and peers using the same three-criterion rubric: Language, Audio and Visuals, and Task Fulfilment. Peer and instructor scores were compared using Pearson’s correlation and a paired-samples t-test, while 147 comments were analysed through thematic content analysis (K = 0.84). The results revealed a strong positive correlation between peer and instructor scores, r (5) = 0.873, p = 0.010, and no significant difference in their mean scores, t (6) = 0.526, p = 0.618. However, their comments revealed divergent priorities. The instructor emphasised linguistic accuracy (71.7%) and technical conventions, whereas peers prioritised creativity (73.6%), humour (68.9%), and audience relatability (55.2%). Interpreted through Amabile’s Componential Theory of Creativity, the findings tentatively suggest that combining instructor attention to technical accuracy with peer sensitivity to audience engagement may support a more balanced assessment of creative multimodal assignments. Given the small, single-institution sample, the proposed framework remains exploratory and requires validation in larger and more diverse contexts.