Speak to us
The Thought Bubble:
Exclusive content
Predictive analytics and student support: how to move from proof of concept to consumable data
In this exclusive content for Thought Bubble readers, we take a look ‘under the hood’ of OES’s Student Engagement Tool, which helps institutions to effectively identify students who need targeted support to engage, progress and succeed in their learning.
Is your university using analytics to inform student support, either by academic staff or centralised support teams? In our conversations with institutions, we’re hearing that the proof of concept is ready. Many have already done this work. The big hurdle is turning the data into a productionised tool that delivers useable data to the teams who need it.
Here, OES Director of Analytics Services Pavan Gamage shares the OES approach.
Step 1
Step 1
Integrate the data
The first step is to create a standardised data ecosystem, combining information from multiple sources. “Creating a standard data layer brings all the pieces together that are relevant for student support, from your LMS, CRM and SMS platforms,” Pavan explains.
Step 2
Step 2
Use machine learning to identify ‘at risk’ or disengaged students
The next step is to extract the data required to build and continually retrain a machine learning model. OES’s predictive modelling leverages students’ behavioural, academic and demographic data: engagement metrics collected daily, performance metrics collected per assessment and student characteristics collected during admission. “The intention is to identify and reach out to students with meaningful support, before they fail, defer or drop out.”
Step 3
Step 3
Getting the analytics into the hands of teaching and central support staff
“To be functional, analytics need to feed into the work teams are already dealing with students,” says Pavan. “The OES tool is embedded into the LMS, showing student risk profiles in each unit. Academics can take immediate action via phone calls or pre-scripted SMS and email interventions.” For support teams, OES has created a standalone web tool that aligns with CRM platforms. “Advisors can access a record of all previous interventions and add notes to inform future efforts. This of course leads to better student experiences and outcomes.”
Step 4
Step 4
Test, quantify impact and learn
Continually measuring the impact of analytics-led support will inform future interventions. OES trialed its interventions against control groups of high risk students from past or similar units with no interventions. In the trial units, 25% of students were identified as candidates for intervention. “Overall we saw a 9% uplift in pass rates and a 7% increase in progression to the next study period, compared to the high risk students who did not receive interventions.”
Step 5
Step 5
Resourcing
There is no question, universities require a comprehensive resourcing mix to move from concept to usable analytics. “Data scientists build the predictive modelling and data engineers work out how to deliver the risk metrics in a useable, sustainable way,” Pavan says. “A back-end application developer makes sure the data integrates with the institution’s existing technology, while a front-end application developer looks at user experience, often working with a dedicated experience designer.” Rolling quality assurance testing is also required.
To learn more about the OES Student Engagement Tool, get in touch with us.
Hot topic
How an Agile approach supports fast, flexible and collaborative learning design
You’ve likely heard the term ‘agile’ in relation to learning design. The agile approach has been borrowed and adapted from the world of software development, where it revolutionised the industry by enabling quality at speed, through genuine collaboration and continuous improvement.
Let’s look at how those same benefits can be applied in higher education, to improve the student and academic experience and solve challenges for universities and VET providers.
Our take
Overcoming data privacy and ethics concerns to uplift student experiences and outcomes
The power of learning analytics to optimise student experiences and study outcomes is broadly recognised, yet many institutions are yet to move forward.
Previously implementing the successful and ethical use of learning analytics, Sacha Nouwens, Executive Director, Student Experience and Insights, offers her perspective on how institutions can balance data privacy and ethics concerns against student experience and service expectations.
Solutions spotlight
Solving soft skill learning and assessment challenges
With graduate employability on every university’s radar, teaching and assessing soft skills has never been more important. So, we’re excited to share how we’re using new technologies to realise immersive experiences, helping our university partners to effectively teach and assess soft skills such as communication, negotiation and presentation.
Read more about how Simulations and VR are being used in different courses and disciplines.
Industry report
Exemplary use of AI for learning analytics
The 2022 EDUCAUSE Horizon Report (Teaching and Learning edition) is a round-up of the latest trends, technologies and practices shaping the future of higher education.
Based on the perspectives and expertise of a global panel of leaders from across the sector, the report showcases OES’s application of predictive analytics to proactively identify and support at-risk students as an exemplar approach to drive student persistence and uplift pass rates and progression.
Out and about
Universities Australia Conference
(6 – 7 July)
Platinum sponsors
We’re proud to support this unmissable event on the higher education calendar. Join us in Canberra, visit the OES booth or find us at networking events including the DVC Dinner on 5 July, co-hosted by OES and Studiosity.