Comment by john_strinlai
15 hours ago
huh, i had no idea princeton specifically disallowed proctors, and instead relied on an honor system. seems... like a poorly thought out system, especially given:
"29.9 percent of respondents reported that they had cheated on an assignment or exam during their time at Princeton. 44.6 percent of senior respondents reported knowledge of Honor Code violations that they chose not to report."
crazier is the people protesting by saying: “students should behave honorably, and that faculty and students should trust each other given the 1893 Honor Code compact.”. obviously that isnt happening if 1/3rd of the student body has admitted to cheating (meaning that the real percent of cheating is even higher).
A couple of my friends teach university classes. Mostly undergrad. I get to hear some of their interesting stories when we game together.
My impression is that there was a sharp shift around COVID. Doing classes over Zoom with a talking head broke the connection they had with their professors and other students. College felt closer to a video game operated through your screen than a community.
When I was in college not all that long ago, cheating was a scandalous thing. I knew a friend of a friend who cheated on an exam with some tricks and it resulted in suspension for a semester. There were rumors of someone hiring a service to write their papers for them and it was a wild story.
Now students have ChatGPT to write their papers and they've been practicing how to use cell phones without the teacher noticing for 10 years before getting to college. Combine that with social media grumblings about how college is "just a piece of paper" and doomerism about how they're never going to get a job or buy a house and cheating starts to look the only rational option to some.
The pattern is not contained to college. Every time the topic of cheating comes up on Hacker News there are more comments defending cheating than I would expect from this crowd. The usual justification is that the system is broken in a hand-wavey way and therefore nobody can be blamed for cheating.
>My impression is that there was a sharp shift around COVID. Doing classes over Zoom with a talking head broke the connection they had with their professors and other students.
I think this is part of a larger phenomenon than simply Zoom classes. COVID severely damaged the already decaying social contract of the US and we have mostly been trying to ignore that ever since. The most prevalent viewpoint of American life is now that the only thing that matters is the individual and therefore anything an individual can do to better their station in life is inherently justified. We can see this on so many levels from politics to rampant academic cheating to quiet quitting to prediction markets full of insiders. When we don't owe each other anything and the consequences are minimal, rarely applied, or completely non-existent, the only reason to not give into cheating, scamming, and corruption is your own personal morals.
If AI is going to steal all white collar work - why use AI to get a degree to do white collar work, paying both the AI and the college for it? Wild times.
For the legacy clout of the big college name?
1 reply →
> Now students have ChatGPT to write their papers and they've been practicing how to use cell phones without the teacher noticing for 10 years before getting to college.
Colleges will need to remodel the rooms where these tests are given to become large SCIF type rooms so that wireless communication is not possible. Let the students go back to writing on their arms, wearing eye patches, or shoving notes up their casts. Yeah, I've probably seen Spies Like Us a couple of times:
https://www.youtube.com/watch?v=iaSUOFleNRU
Note that this is how the FAA has been doing its written tests for years: you go to a proctor test facility and through a metal detector. The entire thing is videotaped.
Local models exists…
1 reply →
This fits with my priors. I was in grad school during covid and had some professors I was close to (and whose class I was taking) reach out asking for feedback on their exam because students were blatantly cheating despite the allowances the professors were making (up to being open to the internet, just no direct communication). They couldn't punish them, and they were perplexed why anyone would bother cheating on even trivial exams.
Even recently when I last spoke to them, the profs described how students were refusing to think for themselves even when given open ended projects. They were just having ChatGPT come up with the project idea for them instead of taking advantage of the freedom to do something they enjoyed.
Hmm, anxiety of making the wrong choice and have it on record? I’ve read that later gens are extremely aware of “the internet never forgets” and are terrified of any choice being the embarrassing and defining moment of the rest of their life
1 reply →
Wait, why can't the students be punished?
1 reply →
As someone who went there (albeit many decades ago) I can tell you FWIW when I was there folks took it seriously. I literally knew of no one who ever cheated on an exam. And I'm pretty sure that anyone I knew who observed cheating would have taken it seriously enough to bring it to the process. It was pretty much a fixture of how students thought about things. So it worked (near as I could tell) back then.
But institutions take awhile to adjust to new realities, and it while looks like Princeton may have been a bit behind the curve on this one, I can understand why they were reluctant to abandon this practice. Living in an honest community cuts a lot of extra effort out - crap that you don't even have to think about. Princeton will be a less productive place to learn going forward.
I'd wager the main difference between "many decades ago" and mid 2000s onwards is the perceived stakes of college. My time in college (around that time) was perceived by most as "make or break": either you did well in college, or you were doomed to a sub-standard lifestyle (not to mention the debt of college tuition).
Obviously, whether this was true or not is a whole discussion, but the attitude did lead to a lot more cheating (due to desperation) than I'd imagine past generations had.
A midterm being worth 25-33% of a grade, plus some classes only being offered in fall or spring semesters meant a bad test could roughly cost you tens of thousands of dollars, since the next time you could retake the class would be in a year, and it often was a prerequisite for another class. It just leads to an environment that encourages desperate "survival" behavior.
Which is bad, someone who cheats on a test or someone who rats out their friend for cheating on a test?
It really is a cultural thing, and that sort of culture is primarily passed down from upperclassmen to underclassmen. I went to a different college with an honor code (Harvey Mudd) and when I graduated in 2019 it was still doing relatively well, but from what I've heard COVID really killed students caring about / adhering to the honor code.
I'm in Mudd's class of '27 (and I was on the honor board for 2 years), and I do think the honor code system has gotten somewhat less functional over the time I've been here. But I think a majority of students and faculty still want to make it work.
> Which is bad, someone who cheats on a test or someone who rats out their friend for cheating on a test?
Obviously the first. How is this even a question?
11 replies →
The stats beg to differ. ⅓ admitted to cheating. Cheating was rampant at my uni and we also had an “honor code”
Bad argument. All countries have laws yet criminality rates varies a lot from country to country. It’s all about the culture.
7 replies →
recent stats.
1 reply →
I'd guess this is selection bias and naivete more than anything else.
I went to a school with an honor code and cheating was rampant among the premeds and future Obamas.
I wasn't at Princeton, but I remember blatant cheating going on and 'study groups' in CS classes that were mere passing around of completed code. (1997-2001)
I'd asked them what they expected would happen when they tried to get jobs or landed one. Like how do you fake work? They just said all jobs are group-based like their study group. (Keep in mind they were soliciting my code as their group was struggling to find solutions to assignments.)
The answer is a one of them works at a grocery store as a cashier, another one I saw now manages a bagel store (didn't know all of them). A waste of time, money, and effort to get a CS degree then just not be able to use it.
Maybe he's happier managing a bagel store rather than dealing with Kubernetes.
2 replies →
When I was at Georgia Tech back around 2002, the freshman Java CS course introduced a brand new and improved cheat detector, and they immediately caught over half of the class for cheating. It was disastrous. The penalty for being caught cheating was so onerous that they simply couldn't do it to the whole class.
To the school's credit, they responded as best they could. They considered each case, interviewed the students, and punished them on a sliding scale based on how much cheating, ranging from a zero on that one assignment up to suspension.
Then, for the next semester, they simply changed the rules. You were now officially allowed to cheat all you wanted on homework, but it now only counted for 5% of your grade. That was REALLY bad for people who weren't doing the homework, but it also sucked for people who were just lousy test takers.
And you will think less of the people who go there. 30% cheated!!
The history of the Honor Code system might be instructive: https://universityarchives.princeton.edu/2015/01/i-pledge-my...
Exames were previously proctored, and it led to a "us vs them" mentality that meant students banded together to
The Honor Code system, and removing proctors was a way to route around that—it made all of the students responsible for catching cheaters and turned the "Students vs Faculty" mentality into a "Honor vs Cheaters" mentality among the students.
Unfortunately, it seems like the "Students vs Faculty" mentality has seen too much of a resurgence due to outside factors, and the Honor Code is no longer a match for the current climate. That's what the article is about
Even more astounding is the reporting number.
If 44.6% of students saw an honor code violation and didn't report it, and 0.4% of students saw an honor code violation and did report it, that means that 99.2% of Princeton students that pledge to report honor code violations break that pledge. And that's only counting the voluntary reporters, meaning that the actual rate is presumably even worse!
But also, how would reporting a suspected honor code violation even work? There's intentionally no staff witnessing the exam, and you aren't likely to know the names and faces of your whole class, so what would the professor even do with that information? "Professor, I saw someone take his phone out, I think maybe he was cheating, I don't know his name." Okay, thanks Captain Non-Actionable. We'll file that in the circular academic integrity investigation bin.
Could just be that about half of them never saw violations
It is true that more than half of respondees reported that they never saw a violation.
However, the 44.6% was the category of "respondents reported knowledge of Honor Code violations that they chose not to report," and the 0.4% was the category of "had reported a peer for an Honor Code violation."
Everybody who saw a violation was in one of those two categories (did report or didn't report), so we can compare them to see what percent of people who saw honor code violations did, and the answer is that >99% of them failed to uphold their pledge.
4 replies →
The burning question for me is how does this compare to previous years?
Looks like they started doing this senior survey in 2022, so unfortunately there's no pre-COVID info.
2022 20.9% cheating, 31.5% non-reporting
2023 25.4% cheating, 33.6% non-reporting
2024 28.8% cheating, 42.0% non-reporting
2025 29.9% cheating, 44.6% non-reporting
So from this it seems like cheating has been increasing significantly over just the last few years
> 29.9 percent of respondents reported that they had cheated on an assignment or exam during their time at Princeton. 44.6 percent of senior respondents reported knowledge of Honor Code violations that they chose not to report
What is it at other universities? I went to a big public school, and remember cheating being halfway rampant. The penalty, moreover, was never expulsion.
Anyone caught cheating at my university, especially if they lied about it, was expelled more or less immediately.
Pretty much my observation. Professors could give lenience and a warning by not reporting it, but if they reported it to Academic Affairs, the Provost would probably end up throwing them out.
One time, several people cheated on physics homework (apparently in a very obvious way), and the professor took fifteen minutes out of the next lecture to basically say "you know who you are, you got a zero, and if I see it again, I'm going straight to the Provost."
Public schools are public schools. They're more or less compulsory and are just meant to try and get you to a point where you can contribute meaningfully to the society.
Princeton is very much optional and is a school for future elites. They're supposed to produce CEOs, politicians, and Nobel prize winners. So the standards should be different.
Of course, expectations are a part of the problem. Many kids go to Princeton or Stanford or MIT because they had wealthy parents who really wanted their kids to go there. And many of these kids are mostly interested in computer games, weed, and the opposite sex. A combination of unmotivated students and high academic standards lead to predictable outcomes.
Public universities (what Americans call public schools in the context of higher education) are optional to the exact same degree as private ones. In other words they are all schools that you apply to.
They also produce more "elites" than "elite" schools do if you go by executives at F500 companies and politicians.
Are we going to pretend that Berkley, Michigan, UNC-CH, UVA etc. do not produce world class educations from world class people?
When I was at Rice a quarter century ago, I can honestly say everyone I knew took the honor system seriously.
Same. I knew exactly one student reprimanded for plagiarism in four years. The idea of cheating on a test was absurd.
Ivy League has always been like this. Everyone gets goods grades. It’s a legacy of the good old boy network.
It’s good for the brand in general. It’s pretty easy to find a 3.8 GPA kid from Harvard. There’s no C students to dirty up the alumni network.
You’re mostly buying into a tribe. Other tribes do well too.
My guess is that the vast majority of those self-reported cases where relating to a take-home assignment (e.g., copying off a classmate's solution). Even without proctoring, you need to be a lot more brazen to cheat on an in-person exam.
Not just Princeton, my uni had a similar honor code and changed it a couple years ago to have proctors after a bunch of cheating events. I don't really get it either. Cheating has been going up exponentially since 2020 but it existed before then. I don't think it's COVID related strictly. Things moved online so cheating became easier and then LLMs became popular and from what I hear that's the most common way of cheating now. I have tested LLMs on undergrad level algorithms problems and was surprised it easily solved them so I think their use goes well beyond just coding assignments.
I've heard that it's the same at <other elite private university I don't want to name>, and people cheat, to the point where non-cheaters are suspicious that it's just a method of grade inflation
The honor systems is the correct system for an institution where learning is the goal as the tests are there to help you internalize the material and know how well you've absorbed it. Instead we've turned universities into vocational schools where the goal is the degree and testing is seen as a hurdle to overcome to attain it.
It's nothing crazy about it. Why do you study? To learn. The exams are there to benchmark your progress. If you cheat, everything falls apart for you.
It is possible that when the metric becomes the target(AKA Goodhart's law) cheating can be beneficial but this is failure of the institution because it means you are no longer there to learn.
>Why do you study? To learn.
mature students (25+, at my school) are indeed there to learn. the 18 year olds are mostly there because its what is expected of them, no more.
I think most are there to get a diploma, learning is a secondary benefit.
It's interesting that people can anonymously admit to cheating. It's a way of saying "don't hate the player, hate the game."
Note, "cheating on assignment or exam during your time at the institution" is a ridiculously broad net to cast. It includes everything from "merely asking one friend on one random night if they got the same numerical answer on the first freshman-year homework (despite both of you working independently and figuring out the entire derivations onto your own)" to "blatantly copying every answer on your final exams every single semester." The fact that they don't distinguish radically different things makes their 30% figure suspect.
AIUI, these schools see their mission as training the next generation of leaders and elites. They aim for people with strong abilities, and moral character.
And, the way you guide youth to act in a certain way is by treating them that way. If you want them to be trustworthy, you trust them. This is not a totally fringe idea.
>If you want them to be trustworthy, you trust them.
sure, but it seems exceptionally silly to continue to blindly trust them when a sizeable portion of them admit to not being trustworthy
Most of us have done something stupid once in our lives. That does not mean we do stupid things all the time, nor does it mean that we didn't learn from the experience. The goal of school is to help immature young adults grow into mature ones.
4 replies →
The next generation of leaders know there are no consequences to cheating, that's what I get out of this.
You can game theory it out and see that everyone gets to cheat and nobody reports is the best outcome for the group. Defectors must be punished in some way or perhaps the profs are not carrying through with punishments for cheaters.
Is it though if the value of the degree for the overall group is collectively diminished?
The worst people in society right now are immoral elites. Why would any elite be moral when it’s obvious that you get more by being immoral?
No, they frame their mission that way.
Clearly the actions were helpful for maintaining that illusion,
while also maintaining the illusion of academic excellence,
despite rigorous courses.
Seems like it's had the opposite effect.
No, they want to train the next generation of leaders and elites, and those elites and leader's parents pay a huge amount of money to get them there.
They know not to bite the hand that feeds them.
This has to be one of the most pretentious things I've ever read about post secondary education.
I'm completely flabbergasted to learn that an Ivy League holds students to a far different and much lower standard than I what I was held to at a regular university in Canada.
From now on I don't see how I can't be skeptical of the credentials of someone from Princeton knowing that their exams weren't proctored.
Ivy Leagues have always been about having money not academic rigor.
[dead]
Here in Canada, a housemate of mine used ChatGPT to cheat in all his courses. He got caught on only a single one because he scored 100%. Then he did it again and got caught again in the same course. For some reason, the professor never reported him to the dean or whoever is supposed to deal with this kind of stuff. He graduated but his degree is basically a degree in cheating.
The interesting thing is that cheating is much easier when done online. When I was a TA and we were in the process of moving quite a bit of the classes to online, we still mandated in person testing.
It was eye opening to find cheat sheets and other cheating materials obviously left behind by students. The majority of the stuff we'd find we either inaccurate and completely wrong. Like a half awake student copied something they thought was the right equation or solution, when in fact, it was for something completely different that wasn't on the test.
So I agree with your notion, but its one thing to try and cheat. Its a completely different one to do so successfully.
I think it really depends on how you view our high education system. As a middle-aged man returning to school to switch careers, my entire reason for going to school is to learn, and I'd never cheat because that would undermine my own goals. To me the purpose of school isn't the degree--I made an entire career already without one--it's to learn.
Students are at school for a lot of poorly-thought-out reasons: inertia, not knowing what else to do, because their parents made them go, etc. If they're not there to learn, you can't make them learn. No, not even by proctoring exams. The only purpose that achieves is to gatekeep.
And, gatekeeping for doctors and pilots is a good thing. We don't want to let just anyone become a doctor or pilot. But frankly, I don't give any shits about whether an AI programmer has made it through a gatekept degree. That stuff can be gatekept at other points--if they show up to work pretending and don't know anything, that will become obvious, and degrees maybe aren't the only or even best way to obtain that knowledge anyway.
All that's to say: if you view higher education as gatekeeping for further life options (i.e. a career) then proctored exams make sense. But if higher education is just for learning, it's stupid to put all this gatekeeping around it--that simply closes doors to interested learners, while allowing people who can "college" well to thrive without really learning. Let the cheaters cheat--they're only hurting their own learning--and I think it's often because you're forcing them to take some gen-ed thing that isn't useful knowledge to them (I'm looking at you, calculus--why was I forced to take 4 semesters of calc, when I always knew that the prob and stat classes I took as electives were more useful?).
> if you view higher education as gatekeeping for further life options (i.e. a career) then proctored exams make sense. But if higher education is just for learning, it's stupid to put all this gatekeeping around it [...] Let the cheaters cheat--they're only hurting their own learning
I wouldn't reduce student motivations to career vs. learning.
College can also be about aspiring to a better society, with the university as microcosm.
For example, a society in which people are honest, and have integrity.
> if you view higher education as gatekeeping for further life options
From the institution's perspective--or at least an "elite" institution like Princeton--that is what it is. When they confer a degree, they're conferring something valuable, even if its main value is as a status marker and ticket to future options. They can't afford to take the attitude of "let the cheaters cheat, they'll only hurt themselves", no matter how true it is, because it would destroy their brand.
are feelings more strongly felt more valid? the same things are happening at caltech - that is, just as much cheating - and they have an honor code. but they feel much stronger about their honor code, so it is more valid.
To people who have not grown up in extremely honor-bound societies and communities the idea sounds strange, yes. To those of us who did, however, events like this remind us of how fragile those systems are and that entry should be severely restricted.
[dead]