Behavioural Learning Analytics was the third theme to be presented at the Social Learning Conference. We were honoured to have Prof. Peter Reimann to share the growing trend of analytics in predicting future success and facilitating improvements.
Prof. Peter is a Professor of Education at the University of Sydney and a senior researcher in the CoCo Research Centre and a Scientific Coordinator of Next-Tell, a large research project funded by the European Commission in the area of educational technology.
His primary research includes cognitive learning with a focus on educational computing, multimedia-based and knowledge-based learning environments, e-learning, and the development of evaluation and assessment methods for the effectiveness of computer-based technologies.
Here are the top three highlights to recap his keynote at the Social Learning Conference 2017.
1. Self regulation of learning on the individual level
Prof. Peter explained the three cycles of self-regulation in an academic setting.
- Forethought: Analyse what the learning task is, develop a high-level plan and how we would tackle it, motivation management such as dealing with the emotions that comes up that we might not be able to do based on experience and beliefs.
- Performance or Volitional Control: Engaging in self-observation and self-control activities to find information for the required task. Activities can come in form of finding resources by reading a book, searching on the internet. Students will take actions if learning time is getting too long — they will proceed to take a break or do another activity to lower stress level or to increase confidence.
- Self Reflection: Looking at the bigger picture on how to better manage a task next time and exploring own strength and weaknesses.
According to him, self regulation is important for educators as:
- Self regulation is important to analyse because scores on self regulation scales correlates with performance and achievement.
- Differences in self regulation behaviour correlate with differences in performance and achievement
- Self regulation can be learned and learning is sensitive to differences in instruction (quality and extent)
- Regulation does not develop ‘spontaneously’, at least not much.
2. Using process mining to analyse and visualise temporal characteristics of individual self regulation
Prof. Peter created a method to track and analyse when students’ learning are self-regulated.
- Process Discovery: This is when students are ‘thinking out loud’ and do their own self-guided information search. Educators can record or take note of this activity.
- Group Comparison: Information received from the previous stage will be tallied into a frequency table. Look at differences in actions between successful and unsuccessful learners. Educators can also look at transition diagram to analyse their students.
Other methods can also be used, such as Process Modelling, Representation of Sequence, Dotted Chart Analysis, PROM Heuristic Miner, Representation of Sequence, etc.
To capture the processes involved in self regulation, data needs to be fairly fine grained — such as in the form of seconds and minutes.
3. Socially shared regulation of group work and group learning
Socially shared regulation consists of co-regulation and socially shared group-regulation.
Co-regulation is when peers help each other, through encouragement and motivations with their work when they have a problem or frustration with their assigned tasks.
Socially shared group-regulation means that a group of students share the same goals they want to achieve, have the same expectations, monitoring their group behaviour and processes, deciding what to do and learning from the activities mentioned.
According to Prof. Peter, groups are most complex to analyse, as groups face three tasks: a) to complete group projects, b) to fulfill members’ needs, and c) to maintain group integrity.
He introduced a few methods to analyse and track socially shared regulation of group work and group learning, such as the Wattle Tree, Network Analyses / Visualisations, Perceived Usefulness, Decision Function Coding Scheme, and Category Frequencies. You can learn more about these methods here.
How might you use learning analytics to inform your teaching and design practice? How do you think you can encourage more regulatory behaviour from your students?
Share a few thoughts in the comments section below!