Using data to improve student success


In this Guest post Paul Prinsloo research professor in open distance learning at the University of South Africa discusses how South African Universities are starting to use learning analytics to improve student learning. The post first appeared in World University News.

Digital technologies and online learning environments permitting harvesting, analysis and use of student data is nothing new in higher education. They open up a number of opportunities and equally a number of perils – creating the need for higher education institutions to find ways to protect the harvested data.

As more and more faculty and students embrace the affordances of digital technologies and online learning environments, data’s velocity, scope, variety and resolution have increased exponentially.

Within the context of changing social norms regarding privacy and sharing, the scope of available data has become more exhaustive than ever before to include richer pictures of individuals’ interests, concerns, passions and different networks and relationships.

On 15 September 2014 in Pretoria, the South African Higher Education Learning Analytics, or SAHELA, held a one-day workshop on learning analytics during the 21st annual conference of the South African Association for Institutional Research.

Data based strategies

Higher education has to critically consider the promises but also the perils in the harvesting, analysing and use of student data. Traditionally, higher education has always used aggregated student data to report on student success and retention and to inform institutional strategy and the allocation of resources.

Student data (aggregated and individual) becomes even more important in the context of the impact of increasing funding constraints and scrutiny by a range of stakeholders and quality assurance regimes and ranking systems.

Having access to accurate digital and real-time data on individual student performance is becoming the basis to inform institutional strategies to increase student success and retention, as well as providing a rationale for securing funding and maintaining or improving institutional reputation.

In online learning environments, as students increasingly engage with their materials, each other and facilitators of learning in online environments, they leave a range of digital footprints or ‘breadcrumbs’.

These ‘breadcrumbs’ can be harvested, analysed and used to not only get a better sense of how and what they are learning, but also to provide real-time information of specific moments where students may need additional support or encouragement.

During the first International Conference on Learning Analytics and Knowledge (2011), learning analytics was defined as “the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs”.

Higher education institutions now have the ability to not only track real-time student engagement and performance, but also to combine data with other information sources such as demographic and location data, as well as data generated by students on social media platforms.

On the one hand, if this data is used in ethical and transparent ways, students may benefit as teaching and support staff will be able to respond to identified student needs, often in personalised ways.

On the other hand, there are a number of concerns regarding not only the ethical and privacy issues inherent in student surveillance, but also with regard to the potential for harm and ineffective and inappropriate storage, governance and use of student data.

Learning analytics projects

There are already numerous interesting learning analytics projects at higher education institutions such as the Signals Project at Purdue University, USA.

In the context of increasing concerns regarding the ethical harvesting, analysis and use of student data, the Open University in the United Kingdom recently adopted a policy to guide the ethical use of learning analytics.

The policy includes principles such as learning analytics as a moral practice, and the responsibility of all stakeholders to use and extract meaning from student data for the benefit of students where feasible.

Students are furthermore not wholly defined by their visible data or our interpretation of that data. There is also the need to be transparent regarding data collection, and to provide students with the opportunity to update their own data and consent agreements at regular intervals.

In the context of concerns regarding higher education becoming yet another Big Brother, students should also be engaged as active agents in the implementation of learning analytics through, inter alia, informed consent and participation in ensuring effective, individualised and appropriate support.

Any harvesting, analysis or use of student data also has to take cognisance of the inherent dangers and bias in data collection.

Safeguarding the process

In the South African higher education landscape, learning analytics is still very much an emerging issue and practice. In most South African higher education institutions most of the student data harvested and analysed are aggregated data and still mainly used for reporting and quality assurance processes.

Since the first SAHELA seminar in 2013 it is, however, clear that South African higher education institutions are increasingly using students’ digital data to inform teaching practice and student support interventions.

There are an increasing number of institutions that implement learning analytics interventions focused on the timeous identification of students who are at risk in order to offer effective and appropriate support.

While there is no doubt that higher education institutions should use student data to inform teaching and support practices, there are also a number of concerns and practical issues to consider.

Issues include ensuring that while the amount and fluidity of student data are increasing, we guarantee that the harvesting, analysis and use of student data are done ethically and transparently.

We must also be constantly aware of the unintended impacts of harvesting, storage, analysis and use of student data and provide frameworks and systems to safeguard students and institutions.

Within the context of international concerns about the increasing invasiveness of surveillance on individual privacy and the use of personal data to provide or withhold services and information, higher education institutions need to engage in critical, ethical and caring ways when harvesting, analysing and using student data.


Paul Prinsloo is a research professor in open distance learning at the University of South Africa. He tweets as @14prinsp.


About Author