Committee at Middlebury College Suggests Allowing Exam Proctoring
Plus, "this isn't fine." Plus, fifty arrested in Pakistan for trying to cheat on medical school exams. And why that's a good thing.
Issue 312
Subscribe below to join 4,132 other smart people who get “The Cheat Sheet.” New Issues every Tuesday and Thursday.
If you enjoy “The Cheat Sheet,” please consider joining the 16 amazing people who are chipping in a few bucks via Patreon. Or joining the 42 outstanding citizens who are now paid subscribers. Paid subscriptions start at $8 a month. Thank you!
Middlebury College Committee Recommends Scrapping Honor Code, Starting Exam Proctoring
A student-led committee at Middlebury College (VT) is suggesting the school change its Honor Code to allow professors to proctor exams. The reason — too much cheating, not enough enforcement.
I have some thoughts.
I’ll start with some context. Middlebury, like several other elite institutions has what I call a “capital H Honor Code” in that it’s not just a pledge students sign promising to be good. The “capital H” codes forbid professors — or anyone really — from most forms of exam and assessment security. In some schools, professors leave the room for exams. These codes rely nearly exclusively on student-to-student reporting of misconduct, as well as adjudication and penalties decided (mostly) by students.
The weak, lower case honor codes have not worked for some time. There’s plenty of research to prove that — see Issue 108 for one example. But the Capital H kind were rare and data were inconclusive. Though recently, some schools with the student self-policing practices have moved to abolish them, Stanford University most notably (see Issue 209).
Stanford, Middlebury and others have started to reassess and take apart their historic and prestigious Honor Codes for the same simple reason — they are not working. Cheating at these school has been rampant. They are not alone, of course. But schools with Honor Codes have been largely unable to do anything about it without changing policy.
For a view of the situation at Middlebury, see Issue 204, in which cheating at the school was so bad that a student wrote an article for the school paper begging the school to do something, begging teachers to start proctoring their exams and protect their honest students.
Here, please pardon my stating the obvious. When students are begging a school to crack down on cheats, there is a problem. But again, there simply isn’t much that the school can do in a true Honor Code system. Hence the report and suggestion to change it.
From the coverage of the report and news coverage thereof:
The draft report identified five overarching trends regarding the current Honor Code system: students regularly violate the Honor Code, holding students accountable to report their peers for violations is unrealistic, faculty should have the ability to proctor their exams, the Honor Code adjudication process is widely misunderstood, and the emergence of generative AI has complicated the line between authorized and unauthorized assistance.
To summarize: lots of cheating, students won’t report it, faculty should have some power to stop it, the process is murky, and generative AI has made all this worse.
That sounds about right if you ask me.
Quoting the report, via the coverage of it, it says:
the reality of daily practice suggests that the Honor Code has ceased to be a meaningful element of learning and living at Middlebury for most students
Yikes.
But this is the part I really want to underline. And yes, I know, if I wanted to highlight it, I should not have waited until paragraph 23 to mention it. My bad.
Nonetheless, consider this quote:
“I think that it would be respected if people saw punishment for breaking the Honor Code that was harsh,” said Zeke Hooper ’25. “If people were caught and were expelled — this would prevent a lot of cheating.”
Let me translate. Students know when a school takes cheating seriously and when it does not. They equate punishment — catching and expelling — to seriousness. And they are not wrong. There are rivers of research showing that the likelihood of being caught, and the severity and certainty of consequence, are significant factors in whether someone will try to cheat or not.
Let me try it again. If you want to limit cheating, you have to do something when cheating happens. When you do not, you get more cheating.
The paper also says that at least one professor is in favor of being able to proctor exams — a Professor of Mathematics and Statistics whose name is, I kid you not, Emily Proctor. Amazing.
To change the Honor Code, the news coverage says the new policy needs:
two-thirds of all undergraduate students to vote in a referendum and two-thirds of voting students to approve the changes.
And I am sorry, but I just don’t see that happening. I just don’t see two-thirds of students voting to make it harder and less worthwhile to cheat. I want to be wrong, but I will not wager on any cadre of 20-somethings voting to revoke their own free lunches. Especially since, the news coverage says:
Just one month prior [to the report], 65% of Middlebury students self-reported having broken the Honor Code
Given what we know about self-reporting and cheating, 65% is really like 90%. At least. And if cheating is that common, as we have been told it is, I just don’t know where you’re going to get two-thirds of the people doing it to vote for actually doing academic work again.
This Isn’t Fine
Joseph Thibault, a friend of “The Cheat Sheet,” has started a valuable substack newsletter called “This isn’t fine.”
It’s great and what it’s doing is important — the kind of actual action on academic integrity that I have rarely been able to do.
Rather than have me mess up explaining it, I asked Joseph to do it. Here’s what he wrote:
"This isn't fine" is a new weekly(ish) newsletter focused on disrupting the essay mill industry by enforcing existing platform advertising policies: we flag ads that promote cheating. Each week, we leverage the Essay Mill Database to identify at least one essay mill whose ads blatantly offer "academic fraud" as a service (test taking, essay writing, homework completion, etc.). The newsletter format is a quick summary of the data collected on a specific mill, info about the ads we've seen, and a direct link to flag the ad. Direct action, every week, with just ~1 minute to read and a few clicks to flag as "academic fraud."
These companies prey on students through search and social media advertising. Without advertising, they reach fewer students and they make less money. Period.
Essay mills exploit the lax paid ad filters (and students). We exploit the stopgap through direct human intervention.
The Substack offers a paid option to support the research and data entry required to maintain the Essay Mill Database.
In other words, they find and flag ads for academic fraud services offered by essay mills and ask you — us — to notify the ad platforms about the illicit marketing, which usually violates the platform’s policies.
It’s fast. It is easy. And it’s something everyone can do. I’ve used Joseph’s newsletter to flag ads. I hope you will as well.
More than 50 Medical School Entrant Examinees Arrested for Cheating in Pakistan
According to news sources in Pakistan, some 50 people taking the Medical College Admission Test (MDCAT), were arrested for cheating. Two test administrators were also arrested.
From the coverage:
The local authorities raided several examination centres, finding students using Bluetooth devices to cheat.
Keeping in mind that no one cheats in college, or on a medical school admissions exam, for the first time, that ought to be terrifying. If they cheated to get into medical school, what do you think they will do once they’re there?
But as is always the case, good for Pakistan for at least arresting cheats. Or suspected cheats. I mean, as the article points out, Pakistan shuts down the Internet on national test days to stop cheating. It puts signal jamming devices in test centers. They have police raid exam providers and arrest people. I significantly prefer that to a country — say, to pick a random example, the United States — that pretends there’s no cheating.