Exploring the effects of gender in skills acquisition in collaborative learning based on the ontological clustering model

• Published in: Journal of computer assisted learning Vol. 40; no. 6; pp. 2484 - 2495
• Main Authors: Hadyaoui, Asma, Cheniti‐Belcadhi, Lilia
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Inc 01.12.2024; Wiley Subscription Services, Inc
• Subjects: 4CS skills; agglomerative clustering; Algorithms; Clustering; Collaboration; Collaborative learning; Communication; Composition effects; Context; Cooperation; Cooperative Learning; Data Analysis; Gender; Group dynamics; Inferences; Machine learning; Ontology; PBCL; Programming; Python; Semantic web; Skill Development; Skills; Students
• ISSN: 0266-4909; 1365-2729
• DOI: 10.1111/jcal.12852

Background Project‐based collaborative learning (PBCL) is a technique that supports knowledge and skill development through complex, real‐world projects. Understanding factors that influence group performance in PBCL, such as gender composition, is crucial. Objectives This study investigates the impact of gender composition on group performance in PBCL, specifically focusing on the critical thinking, creativity, communication, and collaboration (4Cs) 21st‐century learning skills. Methods A total of 312 students were divided into distinct Python programming groups for the experiment. Group performance was assessed using an agglomerative clustering algorithm guided by the ontological group learner model, which examined gender composition, communication patterns, collaboration strategies and project assessment outcomes. Results and Conclusion Gender‐balanced groups demonstrated positive outcomes in the 4Cs skills, with groups containing a greater proportion of women exhibiting superior performance in collaboration and communication, whereas majority‐male groups performed relatively poorer across all skills. Notably, in the context of PBCL, the presence of more women in small groups enhanced the 4Cs skills project assessment outcomes. Takeaways Future research should focus on providing effective support for male online learners in developing the 4Cs skills. The findings offer insights and recommendations for group formation in collaborative learning, facilitating the design of inclusive and impactful learning environments. Lay Description What is (not) known about the subject matter? Assessment may be influenced by diverse perspectives on group work practices, such as what constitutes valuable involvement in a collaborative process. In light of this, gender composition must be taken into account when interpreting evaluation results. Little is known about the effect of gender composition on group assessment results, particularly in the context of project‐based collaborative learning. What is the contribution of this paper? 312 pupils enrolled in our institution's introductory programming course took part in the experiment. The students were grouped into varied, small groups based on their Python programming abilities. We developed semantic web technologies to define the characteristics and accomplishments of groups in terms of the 4Cs of 21st‐century learning skills. We deployed ontological models to characterize the data we gathered about our learners' groups and their performance on assessments. Using an agglomerative clustering machine learning algorithm, we categorized groups based on the results of the assessment of the proposed collaborative project. What are the implications of the findings? The findings indicate that groups with a greater proportion of women achieve outcomes that are superior to those obtained by groups with an equal gender split, particularly in terms of collaborations and communications. The greater the number of women in a small group, the better the project assessment results for the 4Cs 21st‐century learning skills in a PBCL setting. In future research, we should make inferences based on the ontological models to reason about assessment data by executing logical rules and then analysing them in order to develop an intelligent assessment environment. Also, we should concentrate on how to provide male e‐learners with high‐quality 4Cs support. We might additionally examine the potential effects of gender variations on the performance of large groups.