Effect of groups size on students' learning achievement, motivation, cognitive load, collaborative problem‐solving quality, and in‐class interaction in an introductory AI course

• Published in: Journal of computer assisted learning Vol. 38; no. 6; pp. 1807 - 1818
• Main Authors: Zhan, Zehui, He, Guoqing, Li, Tingting, He, Luyao, Xiang, Siyu
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Inc 01.12.2022; Wiley; Wiley Subscription Services, Inc
• Subjects: Academic Achievement; Achievement motivation; Artificial Intelligence; artificial intelligence education; Behavior Patterns; Classroom Communication; Cognitive Ability; Cognitive load; Collaborative learning; Computer Science Education; Context; Cooperative Learning; Empirical analysis; Group Dynamics; High School Students; Instructional design; Introductory Courses; Learning; Learning Motivation; Problem Solving; problem‐solving ability; Robotics; Sequential analysis; Student Motivation; Student organizations; Students; Teaching Methods
• ISSN: 0266-4909; 1365-2729
• DOI: 10.1111/jcal.12722

Background Group size is one of the important factors that affect collaborative learning, however, there is no consensus in the literature on how many students should the groups be composed of during the problem‐solving process. Objectives This study investigated the effect of group size in a K‐12 introductory Artificial Intelligence course by comparing the students' cognitive load, learning motivation, collaborative problem‐solving quality, and in‐classroom interaction between two‐ and three‐student groups. Methods Forty‐eight high school students were randomly assigned to two kinds of groups (i.e., the two‐student group, and the three‐student group, each consisting of 24 students). During the experiment, Xiaofei robots were used to teach the theoretical and practical content of five AI topics over 6 weeks for 1.5 h each week. Results The ANOVA results indicated that group size mattered in the AI course, the two‐student group was more effective in terms of enhancing students' learning motivation and collaborative problem‐solving quality, as well as imposing more cognitive load than the three‐student group. The advantage was more obvious in the practical problem context. The Lag Sequential Analysis results indicated that more collaborative learning behavioural sequences existed in the two‐student group than in the three‐student groups. Contribution This research provides empirical evidence and potential guidance for group design in Artificial Intelligence Education. Although group size did not affect students' learning achievement, it affected learning motivation, cognitive load, and problem‐solving quality and processes. Two‐student groups work better than three‐student groups in the AI course. Lay Description Group size affected students' learning motivation, cognitive load, and problem solving process, but not their learning achievement. The two‐student group was more effective in terms of enhancing students' learning motivation and collaborative problem‐solving quality, while imposing more cognitive load than the three‐student groups. The Lag Sequential Analysis results indicated that more collaborative learning behavioural sequences existed in the two‐student group than in the three‐student groups. The tendency was more obvious in the practical problem context than in the theoretical problem context. Overall, in terms of problem‐solving and learning, group size should be considered in instructional design of AI courses.