Exploring an Approach for Grouping through Predicting Group Performance from Analysis of Learner Characteristics

Wang, Jingyun; Kojima, Kentaro

Abstract

In this paper, we present a mathematical model for forming heterogeneous groups of learners under different teaching strategies. This model requires a formulation which can effectively predict the learning performance of cooperative learning groups. Therefore, we explore the correlations between learning performance and various learner characteristics including learning motivation, learning strategy use, learning styles and gender based on real-world data. By means of analyzing learner data of 157 students in a cooperative learning course, learner attributes irrelevant to cooperative learning performance are excluded from the formulation; this sharply decreases the workload of group formation calculation. In future work, a tool will be implemented based on this adjustable mathematical model and this tool will be used in daily teaching to evaluate its effectiveness.

Citation

Wang, J., & Kojima, K. (2018, July). Exploring an Approach for Grouping through Predicting Group Performance from Analysis of Learner Characteristics. Presented at 2018 7th International Congress on Advanced Applied Informatics (IIAI-AAI), Yonago, Japan

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2018 7th International Congress on Advanced Applied Informatics (IIAI-AAI)
• Start Date: Jul 8, 2018
• End Date: Jul 13, 2018
• Online Publication Date: Apr 18, 2019
• Publication Date: 2018
• Deposit Date: Jul 15, 2021
• Publisher: Institute of Electrical and Electronics Engineers
• ISBN: 978-1-5386-7448-2
• DOI: https://doi.org/10.1109/iiai-aai.2018.00062
• Public URL: https://durham-repository.worktribe.com/output/1139241