How Higher Education is becoming more data-driven

The adoption of online learning methods in the light of Covid-19 makes an informed approach on how EdTech companies use our data vital.

This article is part 2 of a two-part series I’m writing about Higher Education, data and surveillance. Part 1 details instances of surveillance on campus and on online spaces. Part 2 discusses our increased dependency on data to drive education programmes.

During lockdown and social-distancing, students, academic staff and Higher Education (HE) administrators have completely changed the way university education is imparted. Everything we do related to university, like many other activities, is now done through online channels. Lecture capture software has helped many students study for their exams, Zoom and Teams meetings and tutorials have been held, and it’s likely that your university/college has a portal where you can access library resources, staff contacts, and so on.

Now that many institutions are considering a “blended approach” (a mixture of in-person and online delivery of their degree programmes) this fall, the HE sector is having to rethink the value of their academic offer and fundamentally restructure almost every single aspect of life on campus.

Education technology is one of them.

In a previous article, I’ve discussed how universities themselves monitor their students both online and on campus. Often, they use external providers to accomplish security-related and academic aims. The latter is where EdTech comes in.

The EdTech sector has been one of the fastest-growing sectors in the last few years. Things like collaborative whiteboards, interactive projector screens or platforms like Teams and Hangouts have drastically changed how we might think of “normal” student-teacher interactions, and potentially exposing staff and students to new problems.

I want to make clear that I don’t intend to come across as some sort of technophobe, or deny that specific technologies and data collection practices are positively received by the student body. For example, lecture capture software promotes inclusive learning, as it can help students with disabilities revise, and give more flexibility for mature/part-time students who may not be able to attend all of the lectures. Rather than the collection of data itself, we should be questioning how is this data being used, with what goals in mind, and what are our rights as students and staff members.

The “datafication” of Higher Education

The contributors of the latest issue of Teaching in Higher Education, a peer-reviewed academic journal, outline how digital technologies (encompassing a variety of processes, such as plagiarism detection, library resource management or predictive data analytics) were already changing HE before the coronavirus crisis. They call this “datafication”. Three key ideas permeate their analysis:

• Datafication puts performance of students, staff and degree programmes at the centre, encouraging comparison and evaluation between institutions. This reinforces worries about the marketisation of education and “efficiency”.
• Datafication entails the comparison and evaluation of a huge range of activities through automated systems that are often opaque and based on increasingly technical definitions of education.
• HE is being shaped to fit preferred ideals about what is “a measurably beneficial university experience” as a consequence.

Put simply, datafication reflects particular “practices, political priorities and business plans”: the idea that the way data is collected and used is neutral is repeatedly contested by the journals’ contributors.

Learning Management Systems: the case of Canvas

Roxana Marachi and Lawrence Quill examine Canvas, a Learning Management System (LMS) provided by Instructure. Canvas is predominantly used in American universities, but similar LMSs are used in HE institutions across the world (Blackboard or Moodle come to mind).

Some of Canvas’ features include quiz statistics, polls, both an iOS and Android app version, email notifications, the ability to link your social media accounts, web conferencing and personal profiles for students. While this online community and sharing of personal information can certainly improve distance learning and make students engage, we are often unaware that this is not like a closed physical classroom. Instructure relies on Amazon Web Services to host its LMS, meaning that our data is shared with Amazon’s servers, and potentially any third-party applications linked to Canvas.

It’s unclear what Canvas does or doesn’t do with the data they collect.

Instructure clearly states on its website that Canvas allows for third-party apps. Student information collected can include contact details, educational background or online reading patterns, which is valuable information for targeted marketing in the EdTech sector. There is little to no transparency when it comes to what these third-party affiliates do with the collected data. Furthermore, in their most recent privacy notice (2019), Instructure states:

Like most Internet services, we automatically gather this data and store it in log files each time you visit our Site, use our Apps, or access your account on our network. We may link this automatically-collected data to personally identifiable information.

The latest Common Sense Privacy Program evaluation of Canvas (2020) rates it at a 63%. While it ranks high in areas such as data rights (for e.g. users retain ownership of their data and give their consent), it scores a 30% for data security, as a lot remains unclear: whether the data is encrypted or whether it requires its third-party apps to have security measures in place. Furthermore, as a user, it isn’t clear if I can opt out of my data being sold/rented and with what purpose (research, product development…). Although data is shared in an anonymous format, it is also unclear if third parties can “re-identify deidentified information”, especially considering the statement (in bold) on their latest privacy notice.

It is worth noting that other LMSs such as Blackboard score higher (78%) in Common Sense’s privacy review. However, my issue is that as a student, you don’t choose which LMS is used at your university or school. Accepting Canvas’ Terms & Conditions is virtually useless, because if a student chooses not to use it, it is likely they will miss out on key information related to their course. This includes assignment deadlines, timetables, location of rooms, or staff emails.

A move towards predictive analytics

Recently, Instructure’s CEO, Dan Goldsmith discussed Instruture’s strategic move towards data analytics, and predicting certain phenomena, such as student performance.

What’s even more interesting and compelling is that we can take that information, correlate it across all sorts of universities, curricula, etc, and we can start making recommendations and suggestions to the student or instructor in how they can be more successful. Watch this video, read this passage, do problems 17–34 in this textbook, spend an extra two hours on this or that. When we drive student success, we impact things like retention, we impact the productivity of the teachers, and it’s a huge opportunity.

I guess that decades of research in education, psychology and cognitive science is not enough to tackle learning and course retention challenges faced in HE.

Sarcasm aside, I guess it depends on how much you value the human factor in education. As a humanities/social science student I am probably biased in saying that data models dictating my homework does not really sit well with me. Moreover, the data security and privacy issues I’ve already discussed remain the same.

“Crudely, what can no longer be measured is no longer relevant” Marachi & Quill argue. The reliance on data to measure learning can result in a simplistic definition of optimal performance- after all, some types of learning and aspects of coursework quality are not quantifiable (think about critically evaluating sources for a history essay for example). Creative thinking and problem-solving might be undermined if we are trying to fit a standardised mould of success.

Additionally, technology is being used by policy-makers as a cost-effective solution to underlying issues in HE, namely lack of funding, poor teaching, large classroom sizes and an increase in absenteeism. Under the pretense that Canvas and other LMSs offer a personalised learning experience, simultaneously regimenting and regulating a lot of the interactions a student would have with their teacher or professor, they can avoid tackling these problems.

The changes happening in the way university education is delivered are not new, but like with many other things, the coronavirus has accelerated the process.

It is clear that without universities’ digital resources, many of us would have not been able to continue our education. Technology is now a necessity, and universities can no longer afford to not offer their students a range of digital resources, especially if they want to compete in terms of student intake.

Ideally, adopting new technologies, such as lecture capture or LMSs should be done in an informed manner, and be student and staff-driven, rather than imposed in a top-down way. But for that to happen, there needs to be a stronger push for digital literacy, and an awareness of our data privacy, security and ownership rights as consumers. At a managerial level, universities should recognise the potential interests for EdTech companies and be more critical about the supposed benefits their software actually confers their students.