The COVID19 pandemic has caused major shifts in higher education. One of those shifts, which I am glad to see, is a more rapid adoption of online learning. However, the acceleration in online learning, when carried out by those that have not spent the time to educate themselves on online learning methodology and pedagogy/andragogy is a domain full of dangers. The recent class action lawsuit against Northwestern University is representative of that danger.
What is the case about?
The case involves North western’s use of biometric information in their online proctoring tools. Although this is being alleged as a violation of Illinois law (Illinois Biometric Information Privacy Act, BIPA; BIPA has been successfully used in the past, see Six Flags case), the implications of this lawsuit are much farther reaching than simply affecting Illinois institutions. A verdict in the students’ favor, will probably affect non-Illinois institutions that have students located in Illinois.
At issue is the use of Facial Recognition to verify the identity of a student taking a test by tools such as Respondus and Examity.
Why should you be concerned?
BIPA allows plaintiffs to sue and seek damages of $1000-$5000 for each violation of the law. If those tools are shown to have kept the students’ biometric information without their consent, each violation could result in that damage amount. A class action lawsuit also means that it would cover all students who are in the class (i.e. their biometric information was used and stored by an online proctoring tool) regardless of whether they are plaintiffs in this lawsuit, and so could reach millions of dollars. And if the lawsuit is successful, there will be many similar lawsuits coming down the pike.
Apart from the significant financial concern, higher education institutions as “centers of learning” should also take time to consider the ethical concerns involved in the use of these tools. Smarter people than me have written about these issues, ex: Colleen Flaherty (Inside Higher Ed), Drew Harwell (Washington Post), and Shea Swauger (MIT Technology Review).
As educators we should always be concerned about the health and welfare of students. Online proctoring is a temporary non-fix to a problem that needs a different approach.
Don’t just throw Tech at everything
This is a much larger discussion and one that extends far beyond higher ed alone. As a techie of 20 years, I feel the need to repeatedly say this: just because we can come up with a Tech fix, doesn’t mean that is the right solution to a given problem.
We are in a particularly dangerous time because it seems as though anything-and-everything is having AI thrown at it.
Academic integrity is a complicated issue and cannot simply be addressed by a tool. When transitioning to an online environment something that is forgotten by instructors and institutions is that you are no longer in a space that you own and control. You are now in someone else’s space, and you need to respect that fact. Facial recognition may seem like an easy fix to verify the identity of a student taking a test, but facial recognition is a non-fix because it does not work well, encodes bias, and violates a student’s privacy (they absolutely have that right when they are in their bedroom).
So: how do we approach academic integrity
The real question is: before we insist on student integrity, have we evaluated our own?
At the start of the COVID19 pandemic, and the forced transition to online that many had to undergo, I was shocked and dismayed to learn that faculty, nationwide, particularly in Computer Science, were mainly concerned about cheating on the final, not student wellbeing. There was a very simple fix available (depending on class size), replace your final with a different assessment (ex: final project). But that in turn leads us to the deeper problem, and that is that assessment in higher ed is sometimes not the best. There is research to show that formative assessments are better than summative assessments for student learning. The truth is formative assessment requires a lot more work. It is much easier, and I would say lazier, to put together a multiple choice test, than an assessment that while measuring student learning actually furthers it. Before we run towards the next Tech thing that will “solve” academic integrity, let’s spend a little time thinking about what we can do to improve our assessments so that not only does cheating become difficult, it often becomes a non-issue because students are engaged and want to assessed on their learning.
In the last 7 years, I have had exactly 1 case of cheating. Why? because I changed my assessment methodology. My exams are “open everything” so I never need to “lock down” a student’s browser.
Here’s an anecdote: when I was graduate student, everyone dreaded open book exams, they were in fact harder, often much harder, than closed book exams.
Also: why did it not occur to higher education institutions that there are problems aplenty in the use of those tools? Is it occurring to you now?