From data to insight : understanding students' metacognition through learning analytics using written reflections and learning traces
- Publication Type:
- Thesis
- Issue Date:
- 2024
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
Metacognition is a multi-faceted skill that allows students to develop their learning processes effectively. Despite extensive study over the decades, gaps remain in understanding how students employ metacognition in their studies. Leveraging contemporary learning analytics approaches to understand students’ metacognitive processes can bridge these gaps and add value to the existing pedagogical practices. This study aimed to explore students’ metacognition using data grounded in theory, utilising learning analytics techniques, such as epistemic network analysis, process mining, and natural language processing.
We have examined students’ written reflections and metacognitive awareness scores from various IT subjects to analyse differences between high and low-score students’ metacognitive processes using epistemic network analysis. Additionally, we employed Linguistic Inquiry Word Count to explore the linguistic features associated with IT students’ academic performance and metacognitive awareness. Moreover, we analysed the differences in students’ learning traces when metacognitive interventions were applied, using the process mining technique. The intervention involved metacognitive talk time and writing reflections. Data was collected over two semesters from higher education students at the Faculty of Engineering and Information Technology, University of Technology Sydney.
The results indicated no significant difference between high and low-score students’ metacognitive processes in their written reflections and metacognitive awareness. However, differences in the distribution of metacognitive phenomena were observed among cohorts from different IT subjects and levels of study. Additionally, students who received the intervention demonstrated varied interactions with the learning content, showing a higher presence of regulatory components of metacognition compared to those who did not receive the intervention. Finally, certain linguistic features, such as personal pronouns, time orientation, tone, emotion, and discrepancy, were significantly associated with students’ metacognitive awareness and their academic performance.
This research contributes to our understanding of metacognition in educational practices, highlights the importance of incorporating metacognition into subject design, and identifies possible ways to uncover students’ often-hidden metacognitive processes.
Please use this identifier to cite or link to this item: