Year: 2025
Type: article
Abstract: Educational recommender systems (ERSs) traditionally prioritize prediction accuracy, often overlooking the impact of recommendation diversity on user satisfaction. This study aims to understand how recommendation diversity and user psychological traits, such as need for cognitive closure, affect user satisfaction and preferences in ERSs. In two experiments involving university students, we analyzed subjective perceptions of recommendation qualities—including accuracy, novelty, and usefulness—and evaluated the effects of psychological priming on user evaluations. The results reveal that user satisfaction depends not only on perceived accuracy but also on the interplay of diversity, perceived usefulness, and novelty. Furthermore, priming users to consider accuracy or diversity prior to using the system appeared to mitigate the influence of psychological traits, resulting in more consistent evaluations. These findings highlight the potential of task-based strategies, psychological trait personalization, and priming for designing ERSs that foster more effective and satisfying learning experiences.
DOI: 10.53761/ited/1.9
Authors: Robert Songer, Tomohito Yamamoto
Institution: Kanazawa Institute of Technology
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W4413900231