By making the richness of the learning process more visible, learners and teachers can access deeper insights into their shared experience. Data and models can provide a mirror for self-reflection and metacognition (Koedinger 2009). As Gašević (2015) reminds us, Learning Analytics are about learning. However, too little attention has been paid to the student’s role in data-rich learning environments (Kitto 2016).
This research will use probabilistic machine learning techniques in conjunction with other learning model approaches to produce interactive learning models (Millán 2015) that can be integrated in existing learning analytics systems. One such system will be shared with students in a module of a BSc in Computing degree course and a mixed-methods study of their experience conducted – with students having full control of their data.
1. Introduction
As Biesta (2009) notes, central to education’s purpose is ‘the coming into presence of unique individual beings’ and to facilitate this, education spaces must ‘open up for uniqueness to come into the world’. This is the ontological starting point of this research along with Paulo Freire’s (1968) emphasis on the student as an agent of praxis in their learning environment.
A key part of this individual development is the role of metacognition. As students encounter learning challenges, they can greatly increase their agency and personal development by learning about their own learning process and engaging in metacognitive activities (Koedinger 2009).
2. Goal of the research
Develop and apply machine learning and open learning models to support student metacognition in a pre-existing connected learning analytics systems
Modelling student learning along with machine learning techniques will be used to make learning more visible to students and facilitate metacognitive reflection on their learning process.
2.1 Ethical framework
These objectives will be grounded in a clear ethical framework for the management and governance of the data involved to ensure the protection of student privacy informed by Prinsloo et al (2013) and Daschler et al (2015).
2.2 Critical analysis of learning analytics approaches
A keystone of this research will be a critical analysis of how we ‘do’ learning analytics and how that impacts learning environments and learners. Perrotta (2016) notes that learning analytics are not objective and neutral. Embedded in them are societal and political power structures, on which we need to reflect. Learners should not be mere data.
3. Research Questions
The primary questions posed in this research are summarized as follows:
i. Can a Learning Analytics system provide an interface for students to engage in metacognitive activities around their own learning?
ii. Can we retool an existing learning analytics system using machine learning modelling and classifiers to provide this metacognitive interface to students?
iii. Can such a system help students visualize, track and reflect on their own learning and development goals and help them to improve performance?
4. How is this solution different, new or better than existing approaches?
- Grounded in the Student perspective
- Students as owners of their learning data
- Machine Learning with an emphasis on modelling and visibility as well as prediction
- Data literacy capacity building for students
References:
The list of references is too extensive to include here due to word limit but is available for the book of abstracts.
Topics: Learning analytics: research and practice , Topics: Students as Partners
