The Sol Plaatje University Centre for Teaching, Learning and Programme Development (CTLPD) has steadily increased its use of learning analytics to measure, collect, analyse, and report on the data about learners and their contexts, with the aim of understanding and optimising learning and the environments in which it occurs.
The main goal of using learning analytics is to generate actionable insights to impact decision-making and make data-driven decisions, bearing in mind that learning analytics will only bear fruit if there is action.
The CTLPD uses learning analytics for:
• Providing data and visualisations relating to programmes, courses, assessments, student learning behaviour and performance
• Building data preparation workflows for analysis and automation of insights
• Providing information and support on using data to enhance learning and teaching in courses and programmes
Hence, the need to identify and assist students by providing the support they need most, be it academic or computer literacy support. SPU identified the need to develop an Automated Analytics System (AAS) for students at risk to provide temporary or ongoing intervention for academic success.
It helps to address those students in high-risk modules who need individualised support and an intentional intervention to increase proficiency in digital and literacy skills and improve their understanding of the specific module.
The project aims to build a machine learning dashboard for identifying at-risk students for early intervention and enhance the academic, digital and information literacy skills of students studying modules the system detects to be at-risk.
The impact will be measured through improved academic performance and the proportion of students that pass a minimum of 70% of the modules registered.
The AAS development started in 2021 and is funded through the DHET University Capacity Development Plan.
The project aligns with the institution’s strategic plan and its “deepening the academic programme and its orientation towards quality teaching and learning”.
Other types of learning analytics projects conducted in CTLPD:
1. Student Performance
The project aimed to use the historical and real-time data to analyse the performance of each school, programme, module, and SPU in general. It also looked at the number and rate of those who qualify for postgraduate study, for the purpose of providing student academic support. CTLPD has built helpful dashboards to monitor student performance per residence and there is a process to deploy residence-based academic support given actionable insight from the dashboards
2. Survey Data Projects
Accurate, actionable survey data is key to understanding, anticipating, and fulfilling students’ ever-changing needs and wants. The analytics can help improve how the University supports and communicates with students. Insights gained from surveys allow the University to enhance student well-being, student retention, learning design, courses, and teaching practices. Various surveys look at academic orientation, tutorship experience and academic psychosocial issues.
3. Module Evaluation
Module evaluation involves creating reports with minimum human intervention. These evaluations aim to improve the efficiency, reliability and speed of monitoring and evaluation tasks that were previously performed by CTLPD staff.
4. Google Trends Dashboard
The project utilises Google Data to analyse searches about SPU, university tech tools, universities and bursaries to identify the latest trends, monitor marketing performance (for example, SPU’s search interest), recognise areas and regions of high interest, choose the best performing keywords (for example, bursaries) and competitor checks (other SA universities).
5. Moodle Analytics
The purpose of Moodle Analytics dashboard is to monitor student engagement on Moodle and display information about the activities students engage in at an institutional and school level. Future improvements include merging Moodle data with student performance to study the impact of activities and engagement on academic performance.
Possible challenges with learner analytics:
• It can be overwhelming to figure out where to start
• Coordinating with different functions such as IT
• Ensuring expertise in eLearning, such as instructional designs, LMS, and analytics
Future projects include:
• Merging Moodle data with student performance data for diagnostic analytics to identify KPIs and metrics
• Social media analytics (Facebook/Meta) to generate insights for SPU and SPU marketing team/department
• At-risk factor dashboard for identifying at-risk students (pending)
• Probabilistic machine learning dashboard for at-risk student identification on a module level (pending)
• MS Teams meeting text data analytics for staff and students
• Report and dashboard automation opportunities
• SPU attendance dashboard revamp