Aristotle University Of Thessaloniki (AUTH)
Analysis and visualization of learning interactions in collaborative web environments
Abstract
dc:descriptionThe influence of web-based learning environments has proliferated throughout various fields of study, including education. Although there are many web-based software programs currently available for educational purposes, the use of educational blogs in Computer-Supported Collaborative Learning (CSCL) contexts is to this day favored by a large number of educators, since blogs allow the implementation of textual and asynchronous communication to support the social interactions of students online, under the supervision of teachers. Nonetheless, if not adequately competent in regulating their learning, students may find blog-based collaborative learning environments a source of disorientation. In these cases, awareness and metacognition can support regulating and controlling one’s actions in blog-based learning activities. Awareness and metacognition can be further explored through research into the field of computer-based Interaction Analysis (IA), regarded as one of the sub-directions of the modern Learning Analytics (LA) field. Taking into consideration the above, the author of this dissertation has developed a prototype IA toolkit for blogs, called GIANT (Graphical Interaction ANalysis Toolkit). GIANT analyses and graphically visualizes, in real-time, the complex interactions that take place during blog-based collaborative activities, by applying computer-based IA techniques that take into account quantitative usage data. The toolkit was designed to allow students (and teachers alike) to use the automatically produced visualized IA indicators, and support monitoring, awareness, and (self) regulation during the online collaborative learning process. The current dissertation aims at facilitating the adoption of GIANT-powered blog-based collaborative learning environments in education, as well as researching into and providing further insight into the impact visualized GIANT IA indicators have on several aspects. The results of this research were positive, building up to the outcomes of previous studies regarding the promising impact of IA tools on collaborative learning activities. The research results show that the visualized GIANT IA indicators evidently affect participant behavior, as they function as an additional motive that increases the participation ratio. Moreover, collaboration and dialogue improve, because of participation refinement, while students seem to understand the structure and the goals of the blog-based collaborative learning activity better. Most importantly, there is evidence that the visualized IA indicators, presented to the students as dynamically produced feedback information, assist them in reflecting upon the level of their activity, as well as the overall activity, allowing them to self-regulate their actions. On the other hand, the research findings of questionnaires and interviews indicated that teachers not only regarded GIANT as satisfactory in terms of its use and ease to control, but it also proved to be useful and reliable in evaluating and grading the contribution of the students at the end of the activity.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Michailidis, Nikolaos
- Μιχαηλίδης, Νικόλαος
Subjects
dc:subject × 14- Συνεργατική μάθηση υποστηριζόμενη από υπολογιστή
- Αυτορρυθμιζόμενη μάθηση
- Ανάλυση αλληλεπιδράσεων
- Συνεργατικά περιβάλλοντα διαδικτύου
- Computer supported collaborative learning (CSCL)
- Self-regulated learning
- Interaction analysis
- Collaborative web environments
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Επιστήμη ηλεκτρονικών υπολογιστών και Πληροφορική, άλλοι τομείς
- Computer and Information Sciences
- Natural Sciences
- Computer and Information sciences, miscellaneous
Rights
dc:rights- Statement dc:rights
-
- BY_NC_SA
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/49147