Intelligent support in group work in online collaborative learning environment
The increased demand for higher education has made online learning popular and appealing to many stakeholders including working staff and students. Though online learning has gained popularity, it is still being criticized for being a faceless medium that does little to support social interaction. Social constructivist argue that knowledge is constructed through social activities and therefore, knowledge developed using collaboration is more than what can be achieved by an individual alone. Online learning, if supported by a good collaborative strategy like discussion forums, can be at par with social constructivist view of learning in terms of learning achievement. Learning Management Systems such as Modular Object-Oriented Dynamic Learning Environment (Moodle) supports online tools that include discussion forums, chat rooms, e-mails, newsgroups, workshops, etc. These tools provide new opportunities for students to collaborate online and construct knowledge through peer learning. Despite the pedagogical advantages of collaborative learning, online learners can perceive the collaborative learning process as challenging. Although a number of challenges have been mentioned in the literature, they do not have empirically grounded evidence and therefore overcoming these challenges remains an issue. In light of this and the increasing demand for online learning, this research aimed at investigating the current status of online collaborative learning in Higher Learning Institutions in Kenya, identify perceived challenges, and explore strategies for improving online collaborative learning through intelligent support techniques such as machine learning. To that end, this research was designed using a Multi-Methodological approach in order to develop and validate a prototype which provided a novel approach through intelligent support techniques for group formation based on students’ collaboration competence level and a platform to provide immediate feedback in Moodle. The first part of the methodology was a cross-sectional survey which was used to carry out a pre-study to investigate the current status of online collaborative learning and students’ perceived challenges in an online collaborative learning environment v in Higher Learning Institutions in Kenya. This pre-study informed the system development and the validation processes for the prototype. The second part of methodology was the system development methodology which guided the development of the prototype. The final part of the methodology an experimental design that was carried out to evaluate the effectiveness of the intelligent module on the formation of diverse groups and the impact of the group formation method on group performance in an online collaborative learning environment. In this study three groups were used, where in group one students were assigned into groups using grade point average scores , in group two students were assigned into groups using intelligent grouping algorithm and in group three students were assigned into groups using the random method. The novel approach for group formation using machine learning techniques was found capable of forming heterogeneous groups that tended to perform effectively and efficiently at the same level as the random method and the grade point average method. There was no statistical significance in differences associated with the three methods of group formation and performance in a group task. Furthermore, all the three groups had similar positive learning experiences and had few common challenges. With the understanding that random assignment method only increases the likelihood of heterogeneity in the group and grade point average method involves the instructors and it is not dynamic, our proposed intelligent grouping algorithm has the advantage of guaranteeing heterogeneity based on learner’s collaboration competence level, dynamism in grouping students and less instructor involvement. Due to these advantages, instructors are more likely to adopt our intelligent grouping technique. Further studies should also be conducted to compare the intelligent grouping with other group assignment methods which were not studied in this study.