The University of Pretoria (UP) is a multi-campus public research university in South Africa, teaching over 50,000 students across seven campuses. With an 83.5 percent pass rate and 93 percent of graduates employed within six months of qualifying, UP is one of Africa’s top universities.
On a mission to further improve student success rates, the UP was looking to advance its analytics capabilities. “We implemented an evidence-based approach to support our number one strategic goal to improve student access and success, therefore using student data to intervene in time, not after the fact, if a student is underperforming,” explained Dolf Jordaan, Deputy Director of eLearning. “We needed an integrated cloud-based solution data eco-system in our approach to improving student success—you can’t just put words on strategic documents focusing on student success without using data.”
Analytics from the Blackboard Learning Management System (LMS), which both students and lecturers use, provided reports on each course and numerous performance course reports to help identify students who were at risk of falling behind. But a more holistic institutional view was needed, one that could pull in other data sources, including an institutional resource of historical data used for national reporting, student surveys, and the PeopleSoft Student Information system that held campus data.
When some systems and data sources were migrated to the cloud with Amazon Web Services, there was an opportunity to develop a more sophisticated scalable approach. The goal was to provide Faculty Student Advisors and other key stakeholders with an intelligence platform for digging deeper into descriptive and predictive analytics. As a Blackboard partner, the Pyramid platform was a natural fit for UP in 2013.
The Pyramid Decision Intelligence Platform combines the best of data preparation, analytics, and data science, three areas that UP was looking to improve. As the first African higher education institution to implement Blackboard Analytics for Learn, UP was ready to build on its capabilities. As a Blackboard partner, Pyramid was perfectly positioned to supplement course analytics from the LMS with a holistic institutional view, drawing on other data sources.
Pyramid’s Pulse engine enables UP to pursue a hybrid approach, pulling in data from the cloud and on-premises databases. Data modeling features integrate Blackboard, PeopleSoft, Higher Education Data Analyzer (HEDA), and Student Academic Readiness Surveys (STARS) for granular analysis. Combining historical and real-time data on the Pyramid platform means descriptive insights on grades and course activity are now supplemented with cohort analytics that can be used to identify students at risk.
The flexibility to model more data sets was matched by a license that covered 250 active users and allowed for report bursting, where reports not dynamically tied to underlying data can be created, scaled, and made available to stakeholders, including deans, deputy deans, and heads of department.
Modeling in Pyramid has been fundamental for helping the executive leadership team. For example, informing the executive leadership through real-time data during the COVID-pandemic, and reporting the university’s teaching and learning performance to national stakeholders.
The big win for UP is a more holistic overview of students, with deep insights into learning lifecycles based on historical and real-time data that inform institutional change. “The Pyramid Decision Intelligence Platform has become the foundation for our student success and institutional data-driven strategies,” said Jordaan. “The fact that we’ve included data-driven decision making in our strategic goals is an indication of how we have changed the way we view the importance of data.”
UP has taken data analytics to another level, despite the acknowledged challenge that higher education institutions often face regarding access to highly specialized data scientists. “It allows users to carry out sophisticated analytics and integrate different data sources to answer their teaching and learning questions,” he said.
Features and functionality that stand out for Jordaan are Pyramid’s data modeling tools, its presentation and publication capabilities, the Pulse node for enabling hybrid data collection, and bursting, for making reports quickly available. He praises the platform for its ease of use and ability to carry out rudimentary and advanced analytics, meeting many user needs.
He described “speed to answers” as a significant benefit, the way the analytics team can provide stakeholders with course insights in minutes that would previously have taken them days. One project focusing on High Impact Modules has broadened performance analysis around specific courses. No longer limited to insights from Blackboard LMS, it now encompasses institutional data on previous and current cohort analysis to support student success rates and future teaching and learning models in these courses.
Data is also used to monitor teaching LMS engagement which helps to manage a broad adoption of analytics. “We would not have been able to deliver what we have and move ahead of many other national universities if we hadn’t partnered with Pyramid,” said Jordaan. “Pyramid is critical for us to be able to continue with what we’re doing and achieve our institutional goals.”
Identifying inactive students and intervening when they had limited access to devices was costly, but the ROI is student success. The Pyramid Platform has become the foundation for our student success and institutional data-driven strategies.