Harvard Forced Dozens to Leave in Cheating Scandal | Hacker News Reader

Harvard Forced Dozens to Leave in Cheating Scandal

nytimes.com

35 pointsfahrbach47 comments

I find it a little disingenuous that students who are subject to "forced withdrawal" are allowed to re-enter after 2-4 semesters. This sends a horrible message to students - that you can cheat, and the worst case scenario is taking a couple semesters off. No mar on their transcript of starting a degree at Harvard, and suddenly transferring to another, probably academically inferior, school. No explaining to employers why they suddenly decided to pack up and leave during a large cheating scandal. Just a forced vacation.

I'm also pretty appalled that students supplied the same answer down to typological errors. When the scandal first broke out, it sounded much more mundane - that there was little guidance, and that students asked their TAs for help. At least then, it's possible to rationalize the situation by saying that students genuinely did not believe they were doing something wrong. I don't think there's any way to justify multiple students submitting the same typos. This indicates blatant cheating, which any Harvard student should readily recognize as being morally wrong, and clearly violating any reasonable academic honesty standard.

If so many kids, proportionally, are cheating, the reasonable conclusion is not that there's an abundance of 'bad seeds' but perhaps that Harvard isn't very good at getting students to take ethics seriously.

At your typical university, cheating on take-home exams is common, pressure is high, and ethics are underemphasized--is it your place to say a student reacting in the obvious and ordinary way deserves expulsion? Are kids trying to sink or swim in such an environment really deserving expulsion? Besides, more effective cheaters are less likely to be caught. The moral Harvard sends its students is not "don't cheat on a take-home exam" but "don't be caught by copy-pasting answers". The people who work together and tweak their answers to look different get A's.

Honestly, I think it's an overreaction; an institution that has been so poor at imparting academic ethics to its students is now harming the same in order to save face.

More generally, it is an anti-pattern to paint large groups of people as wrong and deserving of punishment, even if you feel the underlying ethics are correct.

perhaps that Harvard isn't very good at getting students to take ethics seriously

The current financial crisis would support that hypothesis.

Harvard is not going to bake ethics or morals into students during their time there, try or not.

Additionally, your blithe assertion that "but everyone was doing it!" is a valid defense is... unsettling.

I've also wondered if the incentives with regards to cheating have changed. It's certainly easier to cheat. And if grades are perceived as having a greater effect on standing, the rewards have increased as well.

My understanding is that forced withdrawals at Harvard work like this:

1. If you want to come back you have to leave Cambridge and get a job for something like 6 months, and it has to be a real job, not a job at your Dad's company or something.

2. Any letter of recommendation that Harvard issues will note that you've been required to withdraw.

I don't see this as a forced vacation. The intent is that you have to have some understanding and appreciation of the real world before you are allowed to come back. Compared to many other schools, Harvard's penalties for cheating are actually very harsh.

I was under the impression that the de-facto penalty for cheating was expulsion. Without coming back. Without ANY "recommendation letters." You cheated, fuck you, get out.

Then again, I went to a school which is struggling to raise itself up in the ranks ( or into the ranks at all ), so maybe Harvard doesn't have to take cheating seriously for people to take their graduates seriously.

Actually, when I talk to people who teach or TA at other universities, they tell me that Harvard's policies are comparably harsh. If you plagiarize a paper at many universities, you are likely to get a zero for the assignment. That means that you might lose a letter grade in one class. At Harvard, professors were until recently required to report students to the Ad Board which generally led to being forced to withdraw.

On the flip side, most US universities simply turn a blind eye to blatant cheating, and actively punish professors for even reporting it. I don't know if Harvard is better or this case was simply too egregious.

I'm not sure how cheating is handled outside my university, but they take it very seriously in my CS major. If we're caught cheating on any assignment we generally fail the course and have a note placed in our academic file, and are potentially removed from the major.

I have seen it even worse in medical and dental school. And because there is little ability to replace the lost revenue from expelling a student the schools are typically lenient.

This is because Harvard prides itself on its graduation rate; any decline in that rate would cause it to fall in the popular rankings of colleges.

This is a silly remark. Harvard doesn't care at all about the popular rankings of colleges, because no one picks Harvard just because they read it got a good rating in US News. And Harvard isn't proud of itself because it has a high graduation rate; Harvard is proud of itself because, whatever anyone else may think, Harvard thinks it's the best university in the world.

This seems to be a matter of confusion over the rules more than anything (though it sounds like there were some clear cut cases as well). After all, it sounds like they were able to use everything but each other for the exams, which seems like an atrocious way to prepare students for the "real world".

The real world isn't a series of read and regurgitate cycles, and allowing students to use resources shifts the test focus from memorizing details to understanding.

In the same vein, I also find interviewing by nitpicking to be poorly reflective of real world work -- when you are actually programming, you have access to tools like linters and google to help you when you are chasing down a bug (and its not entirely clear how asking very subtle issues in a stressful situation like an interview represents the realities of working for the company)

I meant more that in the real world you are not only expected but often required to work with your peers.

> In the same vein, I also find interviewing by nitpicking to be poorly reflective of real world work -- when you are actually programming, you have access to tools like linters and google to help you when you are chasing down a bug (and its not entirely clear how asking very subtle issues in a stressful situation like an interview represents the realities of working for the company)

In my recent interviews I've been encouraged to google in front of my interviewers if I didn't know something because they considered it crucial. So perhaps this is changing?

I think duaneb is saying that not being able to work with the other students is not like the real world, rather saying that than being able to use all the other resources is unrealistic.

So how do you suggest they test students personally, without assholes taking advantage of the nice students who do all the work?

> So how do you suggest they test students personally, without assholes taking advantage of the nice students who do all the work?

If that was their intention, why did they allow open book, open internet, etc? Far better would be in-class timed tests and essays.

EDIT: Well, there is also merit to the approach they used, so perhaps they just should have been far more explicit about what collaboration is OK and what isn't. Come to think of it, I'm not quite sure where to draw the line myself, so I just err on the side of caution and don't discuss anything that could be graded until after deadlines have passed.

So half of 279 (=~140) students cheat, and 70 are forced out.

Gosh. It makes you wonder what sort of privilege the other 70 cheaters invoked. Who was sacrificed? Who was spared?

Selective enforcement of the "law" is no law at all.

I don't think that's a fair assessment of the situation. The article didn't say that half the students were cheating. It said that half were suspected of cheating.

Of course, considering that it's an Introduction to Congress class, you could argue that the cheaters grasped the essence of Congress better than the non-cheaters.

This isn't selective enforcement, just evaluating each case individually.

Those enforcing a law are typically free to find different answers (punishments) to similar questions (crimes) as long as they can distinguish the case.

The big dangerous problem, rule-of-law-wise, is when a court deviates in its method of judging, not in its judgements.

Half were suspected of cheating.

If to you, that sentence is synonymous with "half were cheating". I'd say it's your interpretation of the law that is suspect.

You don't know if the law was overlooked in those cases, or if they were found to be not guilty or significantly less guilty, or if they were simply given less harsh sentences. Discretion given to specify different sentences for the same crime is sensible.

A cynic might predict Harvard's building plan that was suspended for the financial crisis will soon be back on track.

>Selective enforcement of the "law" is no law at all.

Actually it's very much a law. And it's how the law works in the real world. It might not be "justice", but it's law alright.

From the kid who the local cop knows his family and lets him walk on something that would land someone else in juvenile prison to the rich guy literally getting away with murder. And there are stuff that cops selectively enforce everyday, e.g ignoring someone selling dope small-time, because he snitches bigger fishes for them.

A law can still be effective, if the selective enforcement prevents the majority of the people of breaking it.

>One implicated student, who argued that similarities between his paper and others could be traced to shared lecture notes, said the Administrative Board demanded that he produce the notes six months later.

If I were accused of cheating and I argued that my notes would exonerate me, I would have no problem producing them six months later. Heck, I still have notes from a class I took in 1982.

Not all of us keep all of our notes forever. If the expectation is that they should be kept the University should note it (and maybe they do). Mine did not.

If you've been accused of cheating and you think your notes would exonerate you, you would be a fool to throw them out.

Good god! Final exams at Harvard are 'take home'? I really thought that at least Harvard was a little more testing than that... Aside from the standard that sets, it is naive to the extreme to suppose that students allowed to take an important exam home will still write it just off the top of their head.

Take home exams can actually be quite brutal, but are rarely encountered in large (i.e. 100+ student) classes because they usually result in an correspondingly brutal amount of marking. I've had them only in a few senior undergrad courses and grad school.

Why are they brutal when you have your notes, books, the web, etc. available? They're usually not based on spitting back answers you were given or can look up. You actually need to use what you've learned to produce something novel or, at least, very esoteric. e.g. I've had exams where I was asked to look up a paper and reproduce the unpublished proof to get from equation (x) to equation (y). Such proofs are often not published because the authors are freaking geniuses who view the derivation as obvious, and getting to the point where you can do the same proof with intense effort is one point of the course! These exams are often fiendishly difficult and/or open-ended, so you will be working like a crazy fool from the moment you get the exam until it must be turned in. This sometimes means pulling an all-nighter (or more). (I despise all-nighters as the last resort of poor planners, and hate being forced into them.) Time is such a constraint that, if you didn't study and absorb all the course material thoroughly prior to the exam, you will likely not be able to get much done in the time allowed. Frequently nobody in the class will actually finish, and it's really just a matter of how much you can do relative to your peers, and how well.

Collaboration during preparation was never forbidden in my field, although I hear other disciplines (especially humanities) can be quite different. Collaboration during the test is usually discouraged, but obviously students are on the honor system. If any actual copying is detectable from what is handed in you can bet it was egregiously bad. If Harvard was able to make charges stick to 70 students, you can safely bet that twice that number cheated their hineys off, but many were just a bit smarter about it.

Not so much difference to exercises in difficult courses I encountered. But we called that homework. (Germany)

Which is why German universities are consistently at the top of international rankings, right? Oh, wait...

I thought the international top 20 equal the U.S. top 20.

Seriously, think about it, it just cannot make sense. Nobel price winners come from all countries. A university does not need to belong to the U.S. Ivy League to have smart alumni.

However... The Ivy League is really expensive and well-known. In fact people like George W. Bush studied there. It is almost like a self-amplifying loop.

This is true. When the professor asks, if you want a take-home or an open book it's usually a trick question, and you always supposed to ask for a closed book test. Grad school, doubly so.

This is definitely true. I remember walking in to a computer architecture course, felt prepared, and then the teacher said it was an open book final right before handing it out. He had a stack of books for those who hadn't brought theirs as well. Needless to say, one of the scarier tests I've taken and while half the class seemed to sigh with relief, my stress level jumped...and rightly so. Open book or take home exams done right are downright scary.

Bingo. I had an Automata class where the final was a 5 question take home exam. It took me about 15-20 hours to complete all 5 proofs, and I still only received a 98. I jumped over what I thought was an obvious step in one of the proofs :/

The only stipulations for the exam was no collaboration and show all work.

In truth, this is the way it should go

University should not be about a memory test. Not memorizing all the formulas, etc but building on knowledge, doing something you would be expected to do 'in real life', and of course with all the resources you have.

Unfortunately it seems this kind of work is more appropriate to math/engineering/science type of problems.

Most of work in other areas is bibliographical research.

How can you prevent group-work in a math exam?

The reality is, there's a few "twists" (non-intuitive steps) in every proof. If you know the "twists" it cuts the time down to 1/10th. If you're working in a group, there's a clear advantage, but no way to prove anyone cheated (unless students are dumb enough to copy it out line by line).

A take home exam can be as taxing as any other test you can take and often is more difficult than your generic test. In some ways more so, at least on campus you have access to resources and other students.

If students, act according to an honor code, then professors can get on with the work of teaching.

If people have to spend all their time fighting cheating, which is a non trivial task, it creates a very different environment. I would go so far as saying that it ends up obviating the entire point and purpose of college.

----

This is what colleges were originally built around: Students came because they wanted to learn X subject, and the best place to do that was with the professors who were essentially THE experts in their field.

You want libraries and data sources bar none? Universities had them. Want to deal with and learn from like minded students and teachers? Few other places you could be to get that.

People assumed that you wouldn't cheat on your test because that would be like cheating on a MIT open courseware test.

Now things are different - for a lot of people, colleges have become "The next step" to get a job. People assume that the correlation now goes the other way - college MAKES you better at getting jobs.

As a result your population is also now filled with a lot more students who just don't care. The degree is your only way to get a foot into the door of a stupid system.

To steal a phrase from finance, college must be decoupled from "financial success."

Note: A better insight is avaialble from Panos Ipeirotis' experience dealing with cheating in his class a year or so ago. IT was discussed extensively on HN and the blog post, now pulled, was called "why I will never pursue Cheating again". If you dig around a bit though you should be able to find a copy.

Panos' piece can be found here:

http://news.ycombinator.com/item?id=2774254

I had a take home exam in Combinatorics that was 9 questions. I had 20 days to turn it in and worked on it each day for about 8 hours. I finished 5 questions with a mountain of paper turned in. I was pretty depressed thinking I had blown it. Turns out someone answered 7 (Math PhD), 6 (Physics PhD), 5 (two of us) and the rest 4 or under (6 or 8 students I think).

I never ever wanted a take home final like that again.

I think it's pretty common at "elite" schools in the US to have unsupervised, take home tests. Where I went to college we often had closed book, timed take home tests and I think compliance was actually extremely high.

aren't they elite, because the difficulty is elite as well?

Take home finals are not the norm at Harvard. Most final exams are proctored by old ladies from Cambridge who are very strict about everything.

Where take home exams are used, a) they're usually harder than proctored exams, and b) they're usually open-book, so you're not expected to "write it just off the top of [your] head".

The rule is you have to work alone, can't work with another student. That's the rule that these students broke, apparently.

Had to be Government students.

Not sure why they're being forced out, it sounds like they're getting excellent practice for the real world of the ambitious.

Reply on news.ycombinator.com