2020 a ‘baptism of fire’ for UCT’s data analytics project

The University of Cape Town’s (UCT) Data Analytics for Student Success (DASS) project, launched in January 2020, found itself on the frontline of the demand for urgent data analytics, which would impact the success of the 2020 academic year, and beyond into 2021.

As COVID-19 lockdowns forced the University of Cape Town (UCT) to abandon face-to-face teaching and go online earlier this year, the institution’s newly minted DASS project had to step up and urgently deliver data analytics, which would prove critical to decision-making around student support – and ultimately help shape some of the successes of the 2020 academic year.

DASS, which is led by Professor Suellen Shay, is a collaboration between UCT’s Centre for Higher Education Development (CHED), the Institutional Planning Department (IPD) and Information and Communication Technology Services (ICTS).

Ultimately, the work is intended to help students succeed on their path to graduation – one of the most pressing challenges currently facing South African universities. But within the climate of enormous uncertainty created by the COVID-19 pandemic, DASS leveraged the power of data and analytics to provide real-time insights that proved invaluable to university leadership tasked with supporting students to rapidly make the shift from traditional pedagogy to emergency remote teaching (ERT).

Professor Shay said that while UCT has mountains of data and “is home to amazing expertise”, the university has yet to take full advantage of the insights this can yield. The overall intention of DASS is to utilise analytics to complete the huge “jigsaw puzzle” that represents UCT’s data reserves.

Accurate statistical picture

Accelerated by COVID-19, the design and delivery of the crucial student access survey for the Teaching Online Task Team was among DASS’s first responsibilities. This produced an accurate picture of students’ ability to participate in online learning, which guided the provision of laptops but also informed decisions, such as determining the optimal number of ERT study hours per week. A follow-up student experience survey at the end of the first semester revealed the highs and lows of students’ experiences with emergency remote teaching, with a view to offer guidance for online course design in the second half of 2020.

DASS also conducted an analysis of first-semester marks in order to help the university leadership fully comprehend the impact of ERT, and to inform teaching, learning and assessment in the second semester, as well as into 2021.

Stephen Marquard, the acting deputy director in CHED’s Centre for Innovation in Learning and Teaching (CILT) who will pick up the baton from Shay to lead DASS in 2021, said that although the project is well under way, and the institutional capability exists, the next challenge is to connect the dots.

“A definite characteristic of the ERT transition was huge uncertainty, and our most important task was to help reduce that uncertainty,” he explained, adding that the student access survey and other cross-analytics that followed have served to build confidence through understanding actual impact.

Making it easier for students to learn remotely

“A lot of what this survey work has done is to help us all understand not only the student experience, but also the variations in that experience. The results highlight the different ways in which inequalities shaped how students could participate in ERT, but also led us to simple design changes that reduced friction and made it easier for students to learn and study remotely.”

But the more that data can be used to highlight the realities of individual experiences, the higher the likelihood that something realistic can be done to address the myriad challenges in ways that really count, Marquard stressed.

“Analytics will help us identify and smooth out obstacles that can trip students up.”

DASS is currently busy with further examination analysis, including measuring variations in assessments across a wide range of courses and tracking the extent of those differences.

Shay said that data from the experiences of 2020, for students and staff alike, will help the university build on its medium- and long-term vision. The questions the university must grapple with may well change in 2021, when more students have returned to residences, but the aim will remain the same: to provide a clear picture to support appropriate action.

“Next year we will be asking questions like whether we have lost students, examine whether we can track weekly engagement of students in ways that support teaching, and help identify students at risk early in the term. Analytics will help us identify and smooth out obstacles that can trip students up,” she said.

Identifying students’ vulnerabilities

In 2020 DASS found itself working to provide fundamental data about where students were and what access they had, in terms of its mandate to identify students’ vulnerabilities so the university could step in to ensure the teaching process continued uninterrupted.

“But now, as we look towards 2021, we hope to have something of a slightly sturdier landscape at UCT, which will allow us to shift focus from the immediacies of emergency remote teaching to some long-term aspects of DASS’s work,” Shay said.

Two long-term goals for DASS include making data and visualisations widely available to educators. This dovetails with the theory of analytics, which is to help lecturers better understand their students, so influencing teaching approaches and, in the longer term, also curriculum design and analysis that identifies so-called hotspots. This will focus on tracking student performance over time across programmes, with a view to addressing attrition.

“But ours is a work in progress, and it will take a couple of months … for us to build a complete picture of what happened this year, with the November and December assessment results still outstanding. We’ll have to wait until 2021 for the full story of 2020.”