Human Rights Commission in Netherlands: Proctorio Did Not Discriminate Based on Race
Plus, record numbers try to cheat driving exams in Germany. Plus, two stories from the University of Connecticut.
Issue 247
To join the 3,663 smart people who subscribe to “The Cheat Sheet,” enter your e-mail address below. New Issues every Tuesday and Thursday. It’s free:
If you enjoy “The Cheat Sheet,” please consider joining the 14 amazing people who are chipping in a few bucks via Patreon. Or joining the 24 outstanding citizens who are now paid subscribers. Thank you!
Netherlands Commission: Proctorio Software Did Not Discriminate Based on Race
In a long-anticipated decision, the Netherlands Institute for Human Rights issued a ruling this week on a student complaint that her university and its remote exam proctoring provider, Proctorio, discriminated against her based on race.
Though it appeared that the quasi-judicial body initially sided with the student by issuing a preliminary finding requiring the school and Proctorio to prove they did not discriminate, the final judgement is a clear dismissal.
Please note: the Institute’s findings were issued in Dutch and I used Google Translate to move them to English. It’s that translation I am quoting here. Any errors in that conversion are unintentional.
The judgment found:
no prohibited discrimination on the grounds of race was made by using Proctorio software for online proctoring of exams during the corona pandemic
Pretty blunt.
The quick back story is that a student at Vrije Universiteit (NED) filed complaints that Proctorio had discriminated against her based on the color of her skin by having difficulty detecting her face during exam initiation, requiring her to shine a light in her face during the exam, disconnecting her from the exam, and disproportionally “flagging” her for suspicious exam activity.
These are, if you follow academic integrity and assessment security, complaints we have heard before. Frequently. And invariably asserted without evidence. And amplified - as in this story in Wired, about this exact case.
But now, a court-like body has examined the evidence and, in addition to flatly finding that the school and Proctorio did not discriminate based on race, the commission exposed some of the details and facts that are often discounted or discarded in these conversations.
They note, for example, that this student took several exams with Proctorio without incident before the difficulties that sparked her complaints. And, the Commission writes:
The Board concludes, based on the observations in considerations … that the applicant did not experience more login, verification or restart problems during her examinations than the other students. Insofar as the applicant took longer to complete the verification process during one of her exams and had to restart during the exam, it has been established that these problems were due to facts that were not related to her skin color.
Specifically, the student’s Internet connection was fritzy during that particular exam attempt and when the student was connected to Proctorio’s helpdesk:
The Board notes that this transcript shows that the helpdesk advised the applicant to take off her glasses. After the applicant has followed this advice, she passes the webcam check within 8 seconds.
Glasses. The student’s glasses, along with an unstable connection, caused the software issues.
As for the assertion that this student was “flagged” incorrectly or disproportionally due to her skin color, the Board found:
it does not appear that she was frequently awarded a 'flag'. The applicant once received a 'flag' because the connection with her was lost for about nine minutes during an exam. This concerned the exam in which the applicant had to restart. With regard to this event, the Board has already concluded that it was not related to the applicant's skin color, but to the unstable internet connection. The Board therefore concludes that the defendant has sufficiently refuted that the applicant was disadvantaged by the awarding of 'flags' as a result of her skin color.
The student did get one flag — because her Internet quit for nine minutes and she had to restart.
Then the Board repeats:
The Board is of the opinion that the defendant [the University] has not made any prohibited discrimination against the applicant on the grounds of race by using Proctorio software for online proctoring of exams
As sure as I am that everything about that is true, I am also sure it won’t deter people from calling remote proctoring solutions racist. Or abelist. Or whatever else they can think up.
I am also confident that, had there been any indication of discrimination, a body named the Netherlands Institute for Human Rights would have been eager to say so.
For the record, back in December, when the initial findings were released, I wrote that they were probably wrong (see Issue 173). I did not get the why exactly right. It’s not that I was incorrect, it’s that the things I cited were not the grounds on which this commission dismissed the complaint. I did not have access to the details.
But I did nail the outcome:
News this past week is that The Netherlands Institute for Human Rights has preliminarily found that it’s possible that an anti-cheating, remote proctoring algorithm may have “discriminated against a person” based on race.
That sounds bad. But it’s actually quite instructive. And probably not right.
And I also did correctly say:
Reading the tea leaves here, the student likely had a bad Internet connection. That would cause or exacerbate both issues.
I’m calling that a win.
The Board did, to be clear, decide that the school did not take the student’s complaint seriously enough and made it difficult for her to get a hearing. Maybe they told her early on that the issue was her glasses and her Internet connection, I don’t know. But complaints, even the unfounded ones, should get fast and fair hearings. I hope the school cleans that up.
But that is a secondary point. The headline here is that a student accused a remote assessment proctoring company of racial discrimination and an investigative body said no. They actually looked at the situation, examined the evidence and found unequivocally that:
no prohibited discrimination on the grounds of race was made by using Proctorio software
I don’t think there is any more to say.
‘Record Number’ of People Try to Cheat Driving Exams in Germany
Sticking in Europe, a local news report shared that:
Germany sees 'record number' of cheating cases on driving licence exams
According to the story, more than 2,700 people tried to cheat the driving tests - in the first nine months of 2023. That’s a 38% increase in attempted cheating over the past year. And those are the ones who were caught.
A third of the cases, authorities said, involved proxy test-takers. Another third involved the use of unpermitted resources such as cell phones.
For everyone who’s ever said, “Cheaters only cheat themselves” or “Tests are outdated when people can look up any information they need for free,” - I have two words for you: buckle up.
Two Stories from UConn
Here are two cheating-related stories from The University of Connecticut that are worth sharing.
The first one is this little fluff piece in the local news with the headline:
UConn's Husky Harvest Food Pantry gets boost from Paige Bueckers and Chegg
Chegg is donating $50,000 to the University of Connecticut for its Husky Harvest Food Pantry in Storrs.
It was also a TV story.
Bueckers is a college basketball star and was Chegg’s first “student athlete brand ambassador” (see Issue 111).
I am glad this food bank is getting money. And I don’t care that the source is illicit. Or quite possibly even illegal, since Connecticut is one of 17 states that has actually outlawed the selling of academic fraud:
No person shall prepare, offer to prepare, cause to be prepared, sell or offer for sale any term paper, thesis, dissertation, essay, report or other written, recorded, pictorial, artistic or other assignment knowing, or under the circumstances having reason to know, that said assignment is intended for submission either in whole or substantial part under the name of a student other than the author of the term paper, thesis, dissertation, essay, report or other written, recorded, pictorial, artistic or other assignment in fulfillment of the requirements for a degree, diploma, certificate or course of study at any university, college, academy, school or other educational institution
The statute makes an exception for “tutorial assistance,” which Chegg dishonestly claims it provides.
Food banks need money and people need food. If good can come from dirty money, I’m fine.
What irks me is the presumption of normalness that comes from standing with a popular athlete and giving away money. That’s obviously what Chegg is buying. And, as I’ve said many times before, these kinds of PR stunts featuring UConn and Chegg in the same headline make it more challenging for students and investors to understand that Chegg is not a legitimate academic player.
UConn, for whatever it’s worth, has not updated its academic integrity policy in 15 years.
Story two from UConn is this student opinion article in the college paper calling for — and I am not kidding — terminating the use of lock-down browsers in exams. No, really.
The student writes:
LockDown Browser prevents students from copying, pasting and printing exam questions for later use or sharing them with peers. The program effectively limits the window so students are unable to open any other applications or tabs unless the professor specifically permits them to.
Yes, pretty much.
The reason they need to go, the student says is that:
in this day and age, students’ access to multiple devices, most commonly a cell phone, undermines the system entirely. While this was initially used to prevent cheating during the pandemic, it seems almost useless given that students were at home with access to notes, other devices etc.
I love this logic. Well, a burglar could come in through a window, so let’s go ahead and remove our door locks. Smart. It should seem obvious that the answer to ineffective security — assuming it is — is more security, not less.
Anyway, the student also says lockdown browsers are unsafe. And that remote proctoring is “an incredibly invasive and extreme violation of privacy.” She says it’s a pain to use and has connectivity and combability issues. And:
Some schools have banned the use of it entirely due to its problematic nature, and plenty of professors within UConn don’t use it at all and still seem to combat cheating.
I am not aware of any schools that have banned the use of lockdown browsers. And I’d argue that professors who are not using it are not doing enough to “combat cheating.” But that’s me.
Mostly, I get that security measures are inconvenient. I hate TSA security screenings. It annoys me when a website has to text me a code or badgers me to change my password. And it would be way better if we did not need to do all that, but we do. People do bad things, including cheat. And if what we’re trying to protect is valuable at all, inconvenience is inevitable.
I know I made this point already but keep that frustration and inconvenience in mind the next time someone says that “Cheaters only cheat themselves.”