Proctoring software is a nightmare for students. It doesn’t have to be this way.
Jay Serrano, Editorial Director
October 15, 2020
As you all know: COVID. In response to the lack of in-person interaction, many colleges and universities have begun to use proprietary software to ensure students do not cheat during exams, most often ProctorU, Proctorio, and ExamSoft. I take 3 issues with this development.
1. This is spyware.
When you require students to install software that literally watches them, that is spyware.
“Spyware describes software with malicious behavior that aims to gather information about a person or organization and send such information to another entity in a way that harms the user; for example by violating their privacy or endangering their device’s security.” (Wikipedia)
Modern tech’s propensity for obsessive surveillance has become increasingly difficult to combat and virtually impossible to avoid. However, one would hope higher institutions would advocate for things like data privacy and personal agency. Instead, the director of academic testing services at Utah State University lightheartedly described Proctorio as “sort of like spyware that we just legitimize.” (Washington Post) The University of Arizona’s assistant director of technology insisted students don’t mind because “they know this is an expectation because their professors put it out there.” Perhaps unsurprisingly, the student body says otherwise. (The Verge) ;Additionally, the chief executive of Proctorio reflected on the situation with a dystopian, “we’re the police.” (Washington Post)
Suffice to say that no academic software should ever be comparing itself to law enforcement. That’s probably not something we want conflated. Violating student privacy is a pattern—schools force us to engage with abusive proprietary software every day. Whether it’s opting us into a relationship with Google via school Gmail accounts, forcing students to have accounts with Adobe Creative Cloud as a requisite for even being able to take a course, or holding office hours via Microsoft Teams, there is an insidious drip of our data that is all being funneled through people who want to profit from it. All of these companies have been revealed to be astonishingly abusive with data. Google alone would take an entire new post to cover (4 lawsuits and counting).
I don’t expect universities to be a beacon of free and open-source software, especially given how frankly inconvenient most FOSS is. But I also don’t expect them to pretend there is no risk to handing over their students' data to private businesses. Proctor software requires a webcam to view (and, usually, tour) a student’s living space and often uses biometrics to track their physical motion; it often features facial recognition and eye tracking. It also records the event and human proctors may be able to remotely control the student’s machine. (Washington Post) It seems almost absurd to have to explain the Orwellian nature of this type of surveillance, but in case this wasn’t clear: allowing for-profit companies to record and monitor students in their private living spaces because they might look up a Calculus formula is absolutely unhinged.
2. It isn’t an effective measure for cheating and does not account for students with disabilities or, really, the majority of people.
One of the most infamous features of this type of software is that it tracks eye movement and physical motion. These are, perhaps, pretty easy behaviors to latch onto as signs of academic dishonesty. But, as is often the case, the easiest path is also the laziest and least thoughtful. The assumption that darting eyes and excessive motion are indicators of dishonesty is a lazy one that perpetuates ableist beliefs and assumptions. Students with ADHD may have a difficult time sitting still or staring directly at the monitor. Students with anxiety may need periods of time to readjust, perhaps closing their eyes to re-center. A student on the autism spectrum may need to stim during an exam. Students with chronic pain and/or fatigue may need to take breaks to stretch or struggle with uncomfortable seating. As one student reported, she struggles with tics, particularly in stressful situations (such as exams), which puts her in a situation where she is being recorded in a vulnerable moment as she struggles with her disability, which she describes as embarrassing.
Even neurotypical students often fidget (clicking a pen, shaking a leg, etc.) It’s a very common response to stress and hyper-concentration. Several peer-reviewed studies indicate that motion can be an effective tool to aid memory retrieval and clearer cognition. There is no reason to flag this as a suspicious or negative behavior, either in person or virtually. The only reason to discourage this behavior is for their benefit–it is much easier to identify any behavior other than the strictly prescribed one than it is to actually prioritize all students’ learning. Conventional academic settings are notoriously unfriendly to neurodivergent students and are often directly detrimental to the professed goals of teaching and learning. This is very much an institutional problem. It is just even more glaring and naked when distilled in this way–when given the choice between letting students learn comfortably (requiring some recalibration of course material) and forcing disabled students to be recorded by a software that is trained to view them as inherently suspicious, universities chose the latter.
To refocus and summarize: This software strips students of effective coping tools to take a test and hinders their academic performance.
So far, we’ve identified two ways this software works to the detriment of students and have identified zero ways it works to our benefit. At this point, we must ask: “Who does this serve?”
3. This is a byproduct of institutional laziness that does not value its undergraduate students.
We have access to all the information we could ever need to perform our tasks competently, rendering many old testing styles archaic and impractical. Of course, we should have some working knowledge, but most of us will not be in situations where we have 2 minutes to recall the types of fault lines of the North American plate.
It demonstrates a broader issue: universities take their undergraduate students for granted; they fleece us for money we don’t have under the pretense that good education costs good money, then refuse to intervene when they do not deliver on that promise. We’re forced to spend inordinate amounts of money on textbooks—an 88% increase between 2006 and 2016 (Vox)—and additional equipment like clickers (which are usually just used to take attendance). We have little recourse when our professors (especially tenured professors) implement abusive practices. But we make these institutions run. Without undergraduate students, every single one of these universities would go under. The institutional arrogance and entitlement seems to grow every day, becoming harder and harder to ignore. But we–and more importantly, they–know college is the single most important tool for upward class mobility. As the casualties of late stage capitalism’s death rattle, we have no choice. It’s why they do it–they know they’ll get away with it. They know we have nowhere else to go.
In this specific context, I understand the burden of reconfiguring a course is not an easy one to shoulder and I do not expect professors to suddenly have all the answers. However, by introducing this software, the professor shifts this burden to this student–again. It is not our burden to bear–again. We’re struggling as well—there is no need to make it worse.
Where do we go from here?
Some of my fellow Cicadas pointed out I left this on a fairly depressing note. Although I am determinedly cynical, I don’t think there’s any harm in sharing some ideas.
Proctoring software is generally used for summative assessments, which evaluates student learning at a given benchmark, like a midterm or a final exam. These are high stakes, which means there is a high incentive to cheat, hence the proctors. Formative assessments are lower stakes, things like a quick summary of a lecture or a mini-quiz. Formative assessments aid learning and summative assessments measure learning. Conventional wisdom says both are necessary. A trickle of research has indicated that this may not be the case and this teacher makes a very compelling case as for why summative assessments might not even be necessary anymore.
That in mind, the most logical way to resolve this proctoring issue would be to eliminate time-based, closed note summative tests. There are many ways to achieve this
Solution #1: More (formative) testing.
I think almost everyone can identify with the “cramming for a test” experience. You sit down at 11:00 PM to engage with the material for the first time before your 8:00 AM exam. If you’re like me, maybe you’re only just now reading the textbook (oops). You open Quizlet and stare at the screen till your eyes hurt. Is it too late to email the professor a clarification question? You sleep for 3 hours, remorsefully wobbling your way through the test as you desperately chug the dregs of your coffee. You leave the room and feel overwhelming relief. You pass the test and learn almost nothing.
Henry L. Roediger III, a famous cognitive psychologist known for his research on memory, asserts the following: fast learning leads to fast forgetting. Cramming is popular because it works. At least, long enough to get through the test. His study reveals that self-testing is an incredibly effective tool for learning, but that it is not leveraged in a productive way. He elaborates on a concept known as the “testing effect” and studies better testing practices, all of which you can find here.
Basically, he asserts that one day of intense formative assessments was so effective for learning that it enabled the student to survive a summative assessment. In other words, many times, a cramming situation occurs because the formative assessments either did not happen or they were not effective,
How to implement/Examples:
- Quizzes can be embedded into lecture videos using Canvas. Every lecture couldbe split into multiple videos, each one with graded, embedded quizzes.
- This could be a weekly quiz that goes over lecture material. Maybe this quiz has 2-3 attempts and records the highest score.
Solution #2: No memory-based testing.
If summative exams are really necessary, there are other ways to measure mastery of the material. One could argue that assessments such as recitals and other performances require a component of memory, but generally, performance-based summative assessments are an accumulation of all you’ve learned and retain the pressure of a traditional exam without requiring a proctor.
Have you ever taken notes so desperately you didn’t actually absorb what was said? Have you ever just listened to a lecture and been surprised at how much you absorbed? Our fear of not remembering something we’ll need on an exam can be extremely distracting. However, if you can focus on the lecture completely without being distracted, you can have a more meaningful recollection of the material. Maybe you don’t remember Crime and Punishment was published in 1866, but you do remember that it was published in a serialization for 12 months in the 1800s.
How to implement/Examples:
- Essays can supplement traditional exams. Instead of a time-based huntthrough the treasure trove of young adult memory, a student can take their time to sort through the information they’ve been presented and create a unique response. This does, of course, have its own host of challenges and should be treated carefully, but essays could just as easily measure mastery and comprehension.
- Perhaps a class could be conducted almost entirely through discussions and direct engagement. After everysingle lecture, you post a summary of what you learned with 3 questions. This is a type of formative testing that could replace mini-quizzes and other memory based assessments.