Maybe I'm very cynical, but I think almost anybody would cheat if it was made easy enough and they were given the chance. At the end of the day, people are there to come out with a degree and a good grade, not feel good they didn't take the same opportunity available to others. It'd be trivially easy to entrap students.
Say, for example, if the professor put the answer up on Blackboard, seemingly by accident, which came in the form of an incorrect answer to a real exam question. They could then give everyone who gives the wrong answer an XF. I'd wager that would be most people, barring the few smartest who noticed the error. Could you really blame people for wanting to get it right and not handicap themselves relative to the other students?
Given that, the university then has a responsibility to make it as hard as possible to cheat. Fail grades for cheating work because they create an incentive to avoid cheating. Except in this circumstance the incentive effect could not work because the student had no idea they'd be able to be caught.
> Maybe I'm very cynical, but I think almost anybody would cheat if it was made easy enough and they were given the chance.
This attitude is so strange to me. I would assume that cheaters are a minority of students. (Perhaps a large minority, sadly. I've heard some third-hand stories.)
My school had a policy of an automatic 9 month suspension for a first offense. Professors were officially discouraged from exercising any discretion, and they were supposed to report all incidents to the disciplinary committee. For a second offense, the penalty was a permanent expulsion. I did see the system fail a few times when I was there, but it worked a lot of the time.
having such a severe penalty for the first offense probably causes professors to look the other way if they don't agree that it's reasonable. imo, the severity of the offense can vary a lot. cheating on a homework might not be as bad as cheating on an exam (worth a lot more points), wholesale copying of answers is much worse than getting a little too much help from your study group.
as far as how widespread cheating is, that probably depends on the culture at your particular institution. my school had a very strict policy for programming assignments: you were only allowed to discuss them with the instructor, a TA, or in public on piazza. and yet, every time I went into the cs lab, I would see everyone huddled around each other laptops debugging code for the latest assignment. I'm not sure whether they simply didn't understand the rule or didn't realize how obvious it was what they were doing, but essentially everyone in that room was bending the rules a little.
> At the end of the day, people are there to come out with a degree and a good grade
I’ve been listening to the History of Philosophy podcast, and one thing that struck me is that early philosophers seemed to consider virtue the end goal of philosophy. That is to say, we should care less about if our physics students can do physics, and more if doing physics had made them more ethical human beings. It’s quite a reversal we’ve ended up with...
True story:
One day, an assistant professor (and a first time teacher) came to class and announced that there would be a "surprise test" on so-and-so day. Not just that, the questions asked would be 5 out of 8 possible pre-announced questions.
On the day of the test, the professor wrote the questions on the board and then left the room. Almost every student in the class cheated. Some came with chits to copy from. Some copied from others. Legends though, they came with pre-written answer sheets for each of the 8 questions and attached answers for the 5 questions which were asked.
And in that case, (almost) everyone learns. The people who came with the pre written answer sheets studied, the people who wrote the chits studied, and the people who copied from others presumably learned _something_ in the process. Maybe some didn't, but that's always going to happen
I can't agree with this viewpoint at all. If you take a viva of one of these students who came in with a pre-written answer sheet copied from a friend, a vast majority of them would straight up fail.
As TA and prof, I have seen cheaters copy word to word, spelling mistakes included, incorrect grammatical structures as is, basic mistakes like 6/2=2 untouched, because they are literally Ctrl-C Ctrl-V with their hands, with no brain involvement whatsoever.
> At the end of the day, people are there to come out with a degree and a good grade, not feel good they didn't take the same opportunity available to others.
Some people, yes. Others are there to actually learn something. Either because they're intrinsically motivated to do so, or because they understand that while a degree may help get them in the door somewhere, career success in many fields does depend on actually knowing what you're doing.
There is definitely a point at which anyone will cheat, and one time I met that threshold.
There was an Econ exam that I hadn't prepared for, and I got there fully prepared to flunk it. That day there was a problem with the photocopier. We had to fill in a scantron, but they didn't have copies of the exam for everyone. The professor read the questions and answers.
And I watched people's heads go down.
> At the end of the day, people are there to come out with a degree and a good grade, not feel good they didn't take the same opportunity available to others.
This is an age-old question in education: are you trying to learn something, or get the credentials necessary to get a job? I wasn't doing a major in Econ, so the fact that I didn't get much out of the class put me strongly on the side of shrugging that off as "they pretend to teach me, I pretend to learn."
But for most of my coursework, I really did want to learn things and get recognition for having done so.
> Given that, the university then has a responsibility to make it as hard as possible to cheat.
I strongly agree. I know there were guys who cheated in various ways, I watched one guy who went around the labs asking people to do his homework for him. And because universities have allowed degrees to be devalued, I've spent hours in interviews demonstrating basic CS concepts.
I was lucky to spend my childhood and college years in peer groups where cheating was incomprehensible. Nobody would cheat in a math competition, for instance. That would just be a loud, lame declaration of stupidity.
When I went to the other side and started having to enforce the rules, I realized that while the top decile could be trusted, everybody else was suspect, and enforcement of the rules was incredibly hard. What do you do when you grade a string of tests containing the exact same silly mistakes? What about a group of students that suddenly jumps to high performance on one exam and then immediately crashes back down? Remote proctoring is even harder. I once saw a student fabricate an entire school with the aid of his parents, complete with website, address, and fake teacher information, so he and his parents could do the exam together. How are you supposed to vet hundreds or thousands of people like this?
I really hope that I get a faculty position where I can just teach low-pressure graduate-level seminars, and never have to learn the answer. Mass education is damn hard.
Educating and testing are separate issues. If you are testing to see what to teach next, and the student cheats, well they just cheated themselves out of learning what you were going to teach if they failed the test.
In isolation, maybe. But in reality students are competing for limited educational resources. Cheating can mean the difference between admission and rejection to courses, majors, scholarships, and matriculation at top schools.
It’s no different from cheating in sports. Yeah, the sports cheater is probably damaging their body and potentially harming their natural potential. At the same time, cheating may let them make the team and even the big leagues. That can change a person’s entire career, just as cheating to get into Harvard can.
Everytime, I end up with the same conclusion. Overpopulation and
imbalanced distribution of resources is the root cause for everything wrong with this world.
I keep doubting myself. The problem shouldn't be that visible. People don't seem to blame overpopulation... maybe it's not a problem but everytime, it is the problem. Look at education systems of smaller countries. They are way better than what you get as the number of people in your country increases despite bigger spending.
Stopping overpopulation shouldn't be hard. We haven't tried anything. How can people be so sure? Why not disincentivise having kids by making it too expensive? Maybe use a tax cut equivalent to someone's wealth so rich people are as much affected as poor people (albeit a bit less in the grand scheme). Free birth control for everyone - poor people in less developed countries are responsible for most kids. Force longer education period so fertility rate goes down by the time people settle down. There are so many ways you can push the average marriage age faster than it is happening naturally now. Why not do it for less developed countries? Isn't it better if there is less suffering collectively in this world?
How about paying people to not have kids and adopt? What's the problem with it? Taxes are used for so many other useless things.
</incoherent rant>
I am okay with downvotes but please provide a reason. I really want to know. Maybe my thoughts are unhealthy.
There will always be scarcity. That doesn't make all kinds of scarcity the same, morally or in other ways, nor does it mean there will be the same objects of scarcity, nor scope, nor distribution of scarcity at differing population levels.
Still, it will always be a "problem" but that also makes it an opportunity. Lack of opportunity or access to the things needed to even take an opportunity are worse than scarcity (in itself).
To me the problem is only injustice and distribution of resources, not how much or little we have in absolute terms. And if we must make up a measure as to who gets to have kids, maybe make it empathy, not wealth, which is a much worse indicator of the quality of a person than even physical strength would be.
I didn't say it's easily doable, I say while making up arbitrary factors, at least attempt to make it a good one. If you pick poverty as the criterion, convincing anyone isn't even a question, the coercion is a given, so there's no need to suddenly shift the goal posts for empathy, of all things.
> It can also be pretty subjective.
The comment I replied to literally mentions "everything wrong in the world" in the opening, it couldn't get more vague and subjective than that, but empathy is the problem? Overpopulation is subjective, misdistribution is subjective, quality of education is subjective (you may say knowing things and being able to use them well is "better", but that's subjective, too), even human extinction being a problem is subjective. As anything else relating to human ambitions and what we consider a good life.
Depending on the axioms you have defined for yourself. There is statistical evidence for poverty causing health problems or people to die early on average. Infant mortality rates among poor households or kids with malnutrition.
You could always think of unexpected death as unreasonable or objectively bad. By that line of thinking, you can conclude that poverty must be objectively bad. Anything that can result in unexpected death afterwards is objectively bad.
No matter what you end up with, you still need a fundamental axiom or moral position, I believe.
Like why is 3 greater than 2?
Because I attach a greater position value to 3 but what makes this objective really? Most people agreeing with it?
Can most people agree that this world needs less unexpected deaths?
Yes.
Does stopping the number of people born reduce that?
Yes.
Should you stop the demography with highest infant/kid mortality rate first?
Yes.
What's so subjective about it now? Unless you believe that unexpected deaths are a great thing, in that case I think you are fucked up.
I live in a country with 1.34 billion people. Surely, overpopulation isn't a problem when every place is crowded with pollution that people can't/won't be able to control due to sheer numbers and poverty.
Poverty is there for myriad of reasons. Improper distribution is one of those. And poverty breeds overpopulation which again breeds more poverty as people having the most kids can't properly pay for them. They don't pay enough taxes either. Tax theft is big in india.
You want to solve tax theft or thousands of other government problems?
Biggest democracy has got you covered there. Most people won't or don't care about it. It's too much friction to move things but it's there for a good reason. A big nation shouldn't disappear under its citizens feet in a night.
So now you have a slow corrupt bureaucratic system as a feature.
If you're looking for root causes, the place to look is human nature. What do people naturally do, without even thinking?
You get cities because people need things from other people, so there's value in living close together. Cities overstrain the environment because, well, you packed in a ton of people all doing things in close proximity.
You get inequality of wealth and power because of network effects. Why do people form networks? Would it really be better without these networks?
> And poverty breeds overpopulation which again breeds more poverty...
Yet many nations got out of that cycle, so it's not that simple. And poverty is consistently on the decline globally, so doing nothing is a great strategy at the moment.
> You want to solve tax theft or thousands of other government problems?
Tax avoidance is easy: lower the rates to what people will actually pay. Even when you can charge high rates, you still quickly hit a point where it's counterproductive. Governments, like all kinds of organizations, work best when they limit themselves to a well defined mission rather than trying to be everything to everyone.
Why are we stuck with governments? Again, it's human nature. It used to be that people were stuck with a dictator running their life: you'd work all day and send a portion of it to the dictator, who used that to bribe his lieutenants to support him, and they'd make sure some other invader didn't take over.
Democracy simply means that instead of having a dictator confiscate our wealth, we elect representatives who take our money to bribe our neighbors so they can stay in power.
We live with this protection racket of taxation because, fundamentally, power struggles are inherent to human nature. The best we've been able to manage is to set various groups at odds with each other so no one has too much power.
Do you honestly think there's education when a uni teaches cattle class of 200 people for calculus 101, and intends to flunk 80% of them?
This could easily be a set of videos recorded from a real lecture. Problems could be assigned from problem sets (plug n chug from schematics). What someone would pay for is 1 on 1 time from a professional.
And there's NO reason to fail someone, as in retake and pay a pound of flesh again. If you don't get it, go to a different degree that doesn't require it.
> I realized that while the top decile could be trusted, everybody else was suspect,
That is until the top decile is stuck in a required course they're not very good at and aren't interested in, then suddenly they have no problem handing in the same, slightly modified assignment the rest of the class handed in that was emailed by 'some classmate' who enjoyed stats programming in R.
Definitely. That's another type of course that I hope to avoid having to teach. The ideal course has only students that are relaxed, engaged, well-prepared, and who actually want to be there.
Because for 99.999% of new faculty, your experience will be as follows:
Adjunct (part-time) teaching of introductory courses to earn your stripes (students definitely don't want to be in this class, and you will need a second and/or third job).
Full-time teaching of introductory courses after 5-7 years of part-time, because now you're not tenure track, and you still have to earn your stripes.
Full-time, tenure track teaching of introductory courses with a couple of 300 levels thrown in to test your merits during the 3 year tenure process.
Tenured full time faculty status, teaching where you finally get to teach the classes that students want to take.
I can agree with the "it's a garbage system" sentiment. I think its even worse in engineering fields because many of the non-tenured faculty are incentivized to be that good -- if they were good, they would be working in industry making at least double of what they make now.
In high school I ended up the "president" of a team involved in academic competitions. We worked on the assumption that at this level no one would cheat. Every week we'd do a contest in the group, among other things. The scores on the contest determined who made the A team that competed with other schools. One student was a rampant cheater and made it onto the A team, which we didn't discover until the intermural competition began and this student dragged us all down. I guess she got something to put on her college applications, though.
Telling students about your anti-cheating measures seems like a health inspector telling restaurants in advance the dates when he'll show up. The problem is that most professors are not going to go out of their ways to employ anti-cheating measures; by catching cheaters, you are doing a service to the larger college community by removing them.
Ironically when I worked in a deli, we actually were told the dates the health inspector would turn up. You can imagine the day before was spent cleaning and scrubbing!
That's not what the example "shows". The standards aren't raised, but lowered. First, let's just agree on some terminology health codes are the "standard". If you only clean before an inspector arrives (and you are sure that that's the only time to do so) your standard is lower on average over periods of time or altogether, depending on the offense. If you don't know when the inspector is going to arrive, you are incentivized to keep to the code at all times, which is the reason you aren't supposed to know.
When I was a teenager I worked at a Shell gas station. Once I had a customer who asked about the advantages of using Shell nitrogen-enriched gasoline. I didn't know anything about it and found their question really odd.
As I recall, they came back later and revealed that they were a secret shopper for Shell, and we failed the random inspection. All because of a question that no real customer ever asked in my experience. ;-)
But would all those cheaters still have become cheaters if the act of cheating had been more difficult?
I don't think it's unreasonable to believe that someone who started cheating because they saw an easy opportunity to do so at some point would never have cheated at all if that easy opportunity had never been there to begin with.
"Oppurtunity makes the thief" as the old saying goes.
I'm not a teacher, so maybe I should just stay out of it, but I'm of the opinion that teachers should try to help cheaters the same way they help failing students. Cheating is really just a proxy for failure, so why not just treat it the same as failing?
That being said, some kinds of cheating don't deserved to be punished. The given scenario is bad for the cheater, but some students put so much effort into writing notes to cheat with that they just end up doing the amount of effort required to pass.
Cheating is only rarely a proxy for failure. It has been my experience as a TA and instructor that cheating is more often considered an “accelerator” or “advantage” than an act of desperation. It’s not something that can be solved by academic support because the most brazen cheaters usually feel entitled to cheat—they view it as equivalent to doing the assigned work because it reaches the same end (a passing grade). For example, the student in the linked article admitted cheating, but had no contrition about it.
No. Cheating is not a proxy for failure. They are in fact much more frequently inverse to one another. Cheating is intentional circumvention seeking reward. Failure is an outcome that comes more frequently from low effort, unethical effort, or honest effort with low capacity or skill.
In big courses at my university we'd print 4 colors of exams, with 2 actual versions. It was a little extra work upfront, but we almost never had to deal with cheaters.
Bottom line, my job was to teach, grade homework and proctor exams. The last thing I want to do is play piggie -- I don't wanna stay up at night worrying over my student getting expelled / deported. The easiest way to keep focused on my actual job is to take some precautions that make cheating feel risky to the students.
So I really don't understand the other professor in this anecdote. Your students are stressed, sometimes as a result of bad life choices, sometimes as a result of external issues like family death&drama, sometimes as a result of sadistic asshole professors... typically it's a combination of the above. So for each 200-person exam, you'll frequently have students for whom this is literally the worst week of their brief and sheltered lives. They might make a choice they'll regret. So you have a choice: add risk to scare all but the very most dedicated cheaters (and fuck those kids), or lay a trap for the vulnerable kids that the skilled cheaters will skate through.
I had a professor that printed in like 3 colors, and each exam had a different code on it, like 'X3-67' so we thought he made several versions (of course it was graduate math and he told us we would all get an A in we showed up and participated so we wouldn't have cheated anyways). Later we compared our exams after the course ended (because he kept them until then and would just review the problems we missed with each of us in office hours) and they were all the same problems :)
The purpose of learning is to understand the topic at hand. But school's purpose isn't for learning, primarily. It's for accreditation and getting a paper saying you learned certain things. And then you use that paper to get a better job, make more money, and get things like retirement and health insurance. (It used to not be that way - things changed after the GI bill... but I digress.)
If the school was non-profit and did as such, failure would not be wasted money or a permanent grade against you. Failure would only be lost time... But that's not how it is. Failure haunts you the whole time. The negative grades stay in the form of ever-approaching GPAs. But if you buy more classes, you can buy the average a slightly bit higher IFF you get better grades than the average.
The moral dilemma is that schools are using the threat of continuous harm of failure(GPAs) and harm to the pocketbook over taking a class and not doing well in it. Schools and their teachers aren't on the hook for bad teaching. It makes sense when possible to use other methods to make sure your grades and wallet aren't harmed.
I believe it would be different if schools were actually on the hook for teaching what you paid them to do. That calculation would certainly change - we wouldn't have "weeder" classes... Well, not in anywhere near the same way.
(Wonders just how much schools make on failing students. I'm sure there's an Atlantic or Vice writeup in that question.)
Indeed, the real problem is that content of courses are not aligned to the goals and interests of students. Higher Education is used as a filter in many countries. Student's know the game.
The comments here are interesting. My experience has been that cheating is too much trouble to deal with, but I can see that's others have had more success. Might depend greatly on the personalities involved.
Once I was a TA for a class and a student came to me asking for a regrade on part of one exam. They claimed that I unfairly gave them zero on a particular problem. I looked at it and yes, I did give them a zero. But I never give zeros on a larger problem unless nothing or almost nothing is written down. I think that I'm quite generous for gibberish even. There is no way I would have given this student a zero for this problem if he had what was present at my office hours. I told him that I believe he is trying to cheat by doing the problem after the exam. I didn't report him but I did make it clear that I didn't believe his story.
The student went to the lecturer, who was furious with me. The lecturer said that I had no clear evidence that he didn't do the problem; me saying that I never give a zero unless the problem is not attempted wasn't good enough for the lecturer.
I spoke with my advisor, who said that to avoid problems like that he puts a line through any problems that were not attempted.
The lecturer didn't like the line solution after I graded the next exam. He was furious with me again. He told me "You don't trust your students!" I don't trust some students, sure. Since putting the line through empty parts I haven't had anyone claim anything similar, though.
The lecturer was apparently so disappointed with me that he told my advisor to kick me out of his research group (or something along those lines). I didn't learn that until years later. My advisor told me at the time that the lecturer had a reputation for being a jerk to his TAs, so I figure my advisor just thought this lecturer wasn't worth listening to.
(To his "credit", the lecturer had another issue with me that he got furious over: He had a very particular system for writing grades on exams. As I recall on the back page on one of the top corners he wanted the total points received written over the total on the exam. On about 5 out of about 100 exams I wrote the wrong denominator by accident; the numerators were correct in all instances as far as I recall. Now, I'm sorry to have made this mistake but it wasn't on a number that mattered and it was only on a small number of exams. This issue combined with the cheating is all rather minor and not worth raging over.)
I did keep putting lines through empty parts when grading, at least after I stopped TAing for this class. But this experience made me reluctant to give a zero for cheating, much less report the student to the university.
I had a similar situation when I was a TA, except for the issues with the prof. We handled it by Xeroxing that student's exams before handing them back. If he did it again, we were going to have evidence to nail him.
Grades are useless in the face of cheating, so just tell everyone they'll get to choose their own grade at the end, then they'll focus on learning and even if everyone chooses an A, who cares? Does the A mean anything today? I've met 4.0 students in hard college courses that were terrible in the workplace and I've met 2.1 students who had a political science degree be awesome software developers, so the only reason that number exists is to make people feel good.
Maybe I'm naive but I believe cheating can be eliminated by allowing students to have a notecard when taking tests and where they're allowed to write down formulas or whatever is necessary for them individually before the tests.
I know a misconception exists in education "about memorization being something every student can perform optimally for succeeding" in getting top marks. That ignorant belief is simply not the case for everyone and I find it cruel we penalize children that could have succeeded if they're allowed to use their organization skills optimally. Otherwise they're hindered by their genetics or neglectful/abusive home-life that resulted in development issues with memory.
I know I'm not alone as a programmer using a trusty note taking app for multiple things I rather not memorize. Obviously I don't use it for syntax of the language I'm writing. Thus, brings me to my last point of how students will still be unable to cram all the subject material on a notecard and for completing a test within the time limit given. That makes it so the student gains a real life useful skill of organizing and breaking down material being taught. Further helping the longterm understanding than just memorize a few nights before the test and all is forgotten later depending on the subject being taught.
We got that in signal processing. Two sides of one sheet of engineering paper, anything you can fit, prof gets to examine it as you enter the exam. Some people would put their sheets through the printer and go for 8pt font... I used pencil
In middle school I had a social studies teacher that allowed a 3x5 note card for tests. Put it in my laser printer and printed my entire few pages of notes in 4pt font.
I've had classes where we were allowed to bring our textbooks and notes into the exams. For most Mechanical Engineering classes, if you haven't learned the material, cramming during the exam isn't going to help you. Or, if you can figure out how to do it during the exam just by using the textbook, you're going to be fine in the real world.
It's a pretty niche job if you need to do thermodynamics calculations under time pressure without any reference material.
I had a machine organs exam where the teacher allowed to brink laptops even. I did not bring it but i would fail just the same. The only way the laptop would help would be if somebody was solving it for you, but this would be overt cheating.
> For most Mechanical Engineering classes, if you haven't learned the material, cramming during the exam isn't going to help you. Or, if you can figure out how to do it during the exam just by using the textbook, you're going to be fine in the real world.
I had a similar experience in my better CS classes. Some professors clearly realized that, if you hadn't been paying attention and learned how not to screw up, you wouldn't be able to figure it out, even given pretty wide ranging access to information. So they'd hand us a take-home exam, due in a day or two, and we'd be allowed to use all relevant documentation.
I always liked these, as they seemed the most similar to what would happen in industry, unlike some other professors who expected you to have memorized obscure library functions and be able to write perfect code on paper under time pressure.
First: the more support you allow, the harder you have to make the test.
Second (especially relevant for the equivalent situation in job interviews / hiring): the student can hire someone to do the exam for them. (Or more casually: a friend can help.) There's no guarantee that on the job you'll always have that friend available, or are willing to shell out for hired help, nor even allowed to leak company secrets like that. So that hired help doesn't approximate real world conditions as allowing to look stuff up on Googe definitely does.
you can make this less effective by giving more complicated problems with different paths to the answer. it doesn't take a genius to notice that the four people who sit together in lecture turned in very similar exams. students who "change it up a little" tend make superficial changes that almost make it more obvious what they were doing.
My concern with take-home exams is it becomes a question of, how much else is competing for your time? A student who is taking a more intensive course load, or who has to work part-time to afford tuition, is at a disadvantage over their peers.
This applies to all assignments of course, but they tend to have further out due dates, giving students flexibility to budget their time.
You make a good point. At the very least, I think there's a way to better implement in class exams for CS and similar disciplines so that studying isn't just about trying to memorize stdlib functions that, in the real world, you would just have documentation for.
Memorisation is also just not that useful in my day to day as a programmer. I mean, sure, it’s great when I know something off the top of my head, but I have autocomplete and reference material and an interactive shell and whatnot.
In uni, we had one particular class where we were supposed to study something like ten or twelve different concurrent algorithms in some weird concurrent language that I’ve forgotten and not seen since, and would be asked to write four of them out in the test and talk about them a bit. We didn’t know in advance which four, so most people tried to memorise as many of them as they could and hoped that the four that came up would be among them.
Me, being stubborn, thought that was too much effort. Instead, I learned the bare minimum about the language so I could wing it in the test and write it up myself. I already programmed as a hobby so thinking critically about problems wasn’t a big deal. My code was a bit weird, because, for example, I only learned the syntax for while loops and not for loops, so had to make do with what I did know, but it didn’t matter which four problems came up. The problem description gave me enough to figure it out. I got really good marks in that class and put substantially less effort into preparing for the test than my classmates did.
My point, I guess, is that memorisation is a stupid thing to optimise for (I guess it’s harder to design good tests that can’t be best beaten by memorisation), isn’t that useful in “real life” and is a lot more work than if you were thought the actual building blocks so that the students have a toolbox they can use to approach problems.
> Maybe I'm naive but I believe cheating can be eliminated by allowing students to have a notecard when taking tests and where they're allowed to write down formulas or whatever is necessary for them individually before the tests.
I think there's different "cheating by writing down formulas / information we're supposed to have memorized" and "cheating by copying someone else's answers".
In the article, he actually describes a situation where there were identical problems, but with slightly different numbers between seats. One of the students had test version A, but the correct answer from version B. This clearly falls into the latter category; reading the full story, it's really not clear that having a notecard would have helped him.
> 2) Tell: Dealing with cheating distracts from our mission of teaching so best to be preventative so it does not happen.
The implicit assumption here is that telling the students is more effective prevention than not telling them. I don't think that's true, at least on a wider scale; secret anti-cheating countermeasures and rigorous punishment are surely going to be better than published countermeasures.
You can only cheat at a game, which is what the accreditation system we call education is.
If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
That aside, I think the act of cheating is a call for help, even if the callee doesn’t realise it. The pressure to do well coupled with a persons lack of self confidence or self belief come together to make cheating a reasonable route.
If you believed you could never get x, that getting x would massively improve your life, social status, parents acceptance etc you might be prepared to do anything to get x.
Not everyone though, some are probably just resigned to never getting x. They too are calling for help imo. But we weren’t talking about them.
The kids at Stuyvesant who cheated are now bankers and lawyers -- depends what they were learning to do. I figure if the invested the same energy in studying than cheating they would have had equal grades.
You can only cheat rules/agreements/contracts. No rules, no cheating.
I'd agree that learning has no rules. There are some natural laws I guess, like the "forgetting curve", but those cannot be cheated (hence natural laws).
> I think the act of cheating is a call for help
Or a lack of respect for the rules, or sheer curiosity. Maybe there are more reasons.
Somehow this question of cheating and should we tell feels like the wrong question to ask but I unable to precisely state why I feel so. I believe the answer to this question would also vary among cultures ... Like what would students in China feel about it, for example. After schooling in a competitive environment, kids are suddenly asked to solve problems together. If that's the requirement in society, shouldn't teaching process reflect that?
> If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
Nonsense.
There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used.
For example, if I were an engineer designing an elevator, it must meet its rated load limits times a hefty safety factor. That safety factor is a test, and if I were unscrupulous then I would have every incentive to "cheat" to falsely certify compliance. Under ordinary circumstances, that safety factor is left unused when the elevator is operated as per instructions, but the cheating covers up a latent problem with potentially catastrophic consequences.
An academic degree isn't about a worthless piece of paper; it's ostensibly certification that the student has an acceptable mastery of the course material. Much of that material won't be used on a day-to-day basis, but it's still there, and it's still part of the credential that the graduate holds out to the world.
But even if you're unconvinced by that, there's also the character signal at play. The "kind of person" who will cheat on an academic test is intuitively also the "kind of person" who will cheat in other areas of their professional life.
Why not instead modify the test so that it more accurately measures the capabilities of interest instead?
Such as in your elevator example, make it so that the operation of the elevator is tested with a heavy load approaching the desired safety factor, and evaluating the performance under this heavier than normal load? The idea being that the only way to "cheat" is to design a better elevator that can still meet the requirements.
Of course the reason this isn't done is because of the extra cost of performing a more comprehensive test, but given how much students are paying for education, surely we could make some improvements over the status quo.
I can't help but draw comparisons to SpaceX's starship development process vs that of the more traditional aerospace manufacturers. There isn't any way for the engineers to cheat because their work is tested under the target conditions and either passes[0] or fails[1][2]. The goal being to achieve the most efficient design that still meets requirements.
> Such as in your elevator example, make it so that the operation of the elevator is tested with a heavy load approaching the desired safety factor, and evaluating the performance under this heavier than normal load?
Yes, this is a reasonable way to test an elevator. "Cheating" in this context might mean surreptitiously swapping out the load for a lighter load, so that the elevator appears to pass the test while not actually supporting the heavy load during real operation.
>> If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
> Nonsense.
> There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used.
So stop issuing tests.
Or more importantly, the schools could stop holding a record of students grades. Students could still takes tests to confirm their understanding, but their results wouldn't benefit (or hinder) them in any sort of hiring situation. If you sit through the classes, assuming 1) you've put in the work 2) the classes were taught well and 3) you posses sufficient intelligence/aptitude, you should finish with some useful skills. However, no one "graduates" in the sense that they currently do: no fancy paper slip to mount on the wall, no gold stars. School then has nothing to do with career outcome, at least not directly.
This might seem unreasonable -- the employers of the world need some way to judge their fresh candidates!
This is easily solved: people new to a field would start out as apprentices. With a year or more of experience on their resume, they may now switch employers as readily as anyone else does now -- employers, in my experience, don't often check their candidates' formal education at that point.
But what about people who, at that point, would manipulate their way into a rewarding career without yielding any real value in return? At that point, we've just shifted the "game" to "get paid while being lazy and/or useless", right?
Well, maybe, but nothing has really changed. People already brown nose. Companies are already rife with cronyism/nepotism/etc. Lazy and incompetent employees already exploit past association with big name companies to coast through their career and into cushy retirement. None of that has changed.
The only things that really change in this hypothetical: universities are no longer the gatekeeper to people's future, people no longer feel the need to take on massive debt to study disciplines that they're not even sure they'd enjoy, and employers would be expected to invest a tiny amount (modest pay; the overhead of letting them shadow established employees) in new hires making their first entrance into the working world.
I bet you this kind of policy would have the exact opposite effect you want it to. Unless your goal was to restrict access to fields to people who already have connections and experience even more. Because this is what happens in a "wild west" market where you can trust no credentials. All that's left is the personal recommendations of people you trust. University grades are a relatively-objective way of "democratizing" the system. You don't need nepotism, you can just work hard. Even if it isn't a perfect system (nothing is of course).
Costs are a problem but that's driven by prestige whoring, not irrationality. The right majors/degrees are still worth the debt.
Why don’t we democratise nepotism instead (or as well)?
Personal recommendations are used because they work and outsource trust. Yes, they can be abused but this hurts the recommenders social status which reduces their incentive to recommend.
that's s pretty bleak outlook. some people have the integrity to not cheat on tests because it is simply wrong. to claim that everyone has an incentive to cheat is a very low opinion of humanity. trust noone. they are all out to get you.
(responding to a deleted comment, since it explains a downvote, which i appreciate, and therefore like to take the opportunity to clarify)
> to claim that everyone has an incentive to cheat is a very low opinion of humanity. trust noone. they are all out to get you.
I don't believe "everyone has an incentive" was ever stated, but regardless of that, if we could trust everyone with everything laws wouldn't need to exist.
maybe this is a language issue. i understand 'there is an incentive' as something that is there generally motivating, as if it asks for that to happen. like advertising is an incentive to buy, and a stop sign is an incentive to stop or a green light is an incentive to go.
but just like an open door may be an incentive to enter someones home, it is not an incentive to steal, and therefore i reject the idea that the existence of a test is an incentive to cheat.
but to me "There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used." says exactly that.
most laws are created as a reaction to missbehavior. it's not about needing to trust everyone, but designing the system so that everyone can learn and assess their learning without needing to fear that failing that assessment will leave a permanent record.
this is the problem with graded tests. they don't so much help me assess what i know and need to catch up on, but they grade how well i keep up with the classes progress. there is no way to catch up and have that test redone, showing that i have learned the material, even if i did.
a top grade does not indicate that i learned everything, but also that i learned it at the pace the teacher wanted it. if i fail the first test, i have lost any chance to make a top grade for the rest of the year, even if i catch up and maybe surpass others.
That assumes that everyone believes cheating is "simply wrong". Who is harmed by cheating? Unless you manage to cheat so well you blow the curve, you're not impacting anyone elses grades.
And of course everyone has an incentive to cheat. Higher grades and lower time investment are the incentive. Those are basically universally appealing to college students. Having an incentive doesn't mean everyone is going to take the opportunity, as some feel it is wrong and others feel the risk doesn't justify the reward.
The word cheating pretty much means unethical practices by definition of the category. So obviously the people who make the rules consider it wrong. And someone needs to make a common set of rules for grades to mean anything. The student handbook is basically a contract which sets out the ethical system you agree to abide by.
As for victims, ultimately you would probably defraud future employers presumably, by taking their money under false pretenses, and damage the school's reputation. Perhaps also kill people when the plane you designed crashes or something.
Whoever is making decisions based on the outcome of the test. A school or professor risks it's reputation by graduating incompetent students. An employer who hires based on a false GPA risks wasting its money and time.
> If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
Have you ever heard of people who wait until the last moment to start work on an assignment, or who study intensively a few days before an important exam?
It's always seemed to me one of the big 'features' of college education you can't get with a library card or internet connection is the unstoppable steamroller of deadlines - slowly but inexorably advancing, reducing your options from 'study now, study later or fail' to only 'study now or fail'
Do you think that exams and assignments provide such a role, and if so would your proposal address that need?
In the short term there would be nothing to fill that role, except maybe, for some, the fear of your peers leaving you behind.
A generation later, society will have gained a fuller sense of the need for discipline, self motivation, and personal responsibility in any individual who wants to succeed at life. And so there will be many more people in the world trying to cultivate those virtues. It might turn out to be a great benefit to the human race as a result.
>> That aside, I think the act of cheating is a call for help, even if the callee doesn’t realise it. The pressure to do well coupled with a persons lack of self confidence or self belief come together to make cheating a reasonable route.
That's fine for the SATs. The logic holds in a system where people are competing against large groups of others. There you can treat it as a call for help and understand it as a reaction to pressure due to lack of confidence. But what about more competitive situations, times when say 10 people are fighting for one job in an all-or-nothing competition?
Think astronaut training. Only one person is going on the rocket. Second place, staying on the ground, is meaningless. Confidence doesn't matter. Everyone is confident. Trying hard is irrelevant as everyone is trying 100%. Work meaningfully harder than everyone else might cause you to collapse into injury or breakdown. In such situations "cheating" may be the only way to differentiate one's self from others.
I think that is why when professional athletes cheat they demonstrate little remorse. At their level it isn't a failure of confidence or morality, instead it is a reasoned choice made of necessity. That's why we need to closely watch people in such competitions. Morality, right and wrong, isn't enough.
>> If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
I think the OP is saying that if the "reward" for education is that you know more -- and that brings it's own intrinsic value on a cultural and personal basis -- then there will be no need to "score points" on any test or report card. Therefore, no "game" necessary. And that follows that there will be no cheating (why would you cheat if you, and your culture, are happy with the reward of education in and of itself?).
imo, it's will take a quantum leap for us as a culture to reward education for education sake. To the point that when you interview, simply saying, "yes I understand X concept" will be met with belief by the interviewer/teacher (vs/ now when you have to "prove" your knowledge by a game-able "score").
As for personally, imo, people can do it now: love to learn for the sake of learning. No point in cheating on the test if you're learning the subject for your own edification. The test will simply tell you where you are still weak.
There are always people who want the easy way out of anything. And there are always people who want to follow strict protocol even if it makes no sense.
I had classmates who just wanted to pass by any means necessary and take that diploma back home and get a job there —their competition back home was usually subpar since these students had a leg up with their English and had a diploma from a US school, no matter how far down the US News ranking a system would be.
I doubt very many cheaters lack self confidence, as cheating takes some balls. They're the entitled and lazy but usually bright. Which is why they suck, because they could master the material, they're just too lazy.
> That aside, I think the act of cheating is a call for help, even if the callee doesn’t realise it. The pressure to do well coupled with a persons lack of self confidence or self belief come together to make cheating a reasonable route.
Sometimes in life to grow you need to deal with a situation or goal that produces anxiety.
Also sometimes your goals are simply not in line with your abilities or level of dedication, and you need to fail.
Sure and the next time you have a surgery the surgeon might be really, really bad at cutting inside you, but passed their exams because the other students helped out a lot.
Honestly, an exam should control the success of learning, nothing more. Passing an exam should always finish a course or topic. Failing an exam, though, should be quite normal and just lead to more repitition.
> If the student had been a coherent thinker, he probably wouldn't have needed to cheat.
This is a key point, but it may not be formulated the right way. Here is my take on it, as I explain it to my students:
If you are able to cheat so well that I can't catch you, it means you don't actually need to cheat. Corollary is that if you need to cheat and you do, I will see it, because you'll be bad at that too.
are your students studying to be Ninjas? Otherwise I don't see how needing to cheat to pass Spanish Language as an example would translate to not having the cheating skills to do so.
That would be awesome, but no, they're studying compsci. Most of what I grade is code. It's quite difficult to hide that you have copied code from elsewhere, in particular when you're not able to understand — let alone explain — how it works.
I guess this is another difference between the academic and business worlds then, often I ask people why something is the way it is and they start using some a mixmash of jargon pulled from design patterns, agile manifesto, and business goals of some sort.
I don't necessarily think it's because they have copied the code and they don't know how it works, I just think they don't know how to speak straightforward about code or are afraid of looking stupid and as a result I either have to accept stuff is the way it is, write down what they said and go try to decipher it at my leisure, or gird myself for battle.
I think that not telling students about different versions of tests is better for deterrence. Of course, not if applied once, but each test they will think twice before copy&paste of an answer.
...
Cheating is a form of deception. The person in the story wanted to cheat an exam, and then later - manipulate OP that they are the victim. One shouldn't fall for that.
> I think that not telling students about different versions of tests is better for deterrence.
No. Deterrence is not putting up a hidden camera. Deterrence is when you tell people the area is monitored by cameras. People will slow down their car when they know the police is around, not when they don't know.
If you want to catch cheaters, you set up anti cheating. If you want deter cheaters, you warn them about anti cheating measures.
Well, if you monitor a street, people slow down in this particular area and go back to their speeding once they leave the designed area.
If students know which exams have extra measures, they won't cheat on them, and will get a free pass everywhere else.
If you don't, sure - at first there is no deterrence, but later (after a few students get failed and exposed for cheating), it will deter others from trying.
I learned a lot reading the book "the honest truth about dishonesty" by dan ariely.
He did a whole bunch of really interesting experiments, like self-reporting how many you got wrong on a test before it is fed into a shredder and similar.
fascinating book.
How many of you work exactly 8 hours a day with your employer?
Not quite a fair comparison because most software engineers are salaried, not hourly (at least in the United States).
I’m trying hard to think about what cheating would be in the software world. Here’s a few things that I see:
* Grabbing an open source package without a license check
* Pushing to production without running all the integration tests (or worse, just SSHing into production)
* Accepting a PR without actually reviewing the code
* Writing code without unit tests
Its hard for me to come up with these - most involved skirting controls, which were usually there for a reason.
Data science seems much more obvious:
* Lying about performance on the test set
* Using highly correlated features which leads to instability
* telling people the model has broader applicability than it does (e.g. a speech to text model was trained on complaint phone calls and claiming it can be just as good for in person meetings)
This could be because I spend more time in the ML world, but I’m curious what other people think.
In my experience the primary way knowledge workers get into trouble is over-reporting progress on a task, then half-assing it at the end or making excuses.
This might be due to procrastination, personal issues, distraction, other job pressures, lack of direction / guidance on the task or many other reasons, some of them good.
The problem is succumbing to the temptation to report progress when there really hasn't been any.
> The problem is succumbing to the temptation to report progress when there really hasn't been any.
Just today at stand up I said I was going to "wrap up" a story I had been working on for several days. In reality I have not written a single line of code, so by "wrap up" I meant "start and finish coding". Here's how I rationalized it though:
It's not like I've done nothing, the past few days I've been mentally digesting the story and trying to figure out how all the pieces fit together. I've been researching different solutions online, reading through GitHub and StackOverflow. I thought about it a lot, even when not at work (i.e. in the shower or laying in bed). Now that I think I've fully worked out the solution in my mind, actually coding it is the easy part. Sometimes I hit a snag, but more often than not, I can just iteratively breeze through the implementation.
A bit of automation can help with most of the 'cheating in the software world' you mention.
For example, writing code without tests can be caught with a coverage checker.
In my ideal setup, you would be allowed to bypass the coverage checker with a simple comment-annotation in your code.
The point is not to be bullet-proof, but to have the automation remind you to do the Right Thing; and to make doing the Wrong Thing something you explicitly have to go out of your way to commit to. Not just something that happens because you are absent minded.
(And there are lots of legitimate reasons why you might want to bypass a coverage checker or a linter. So an easy, but explicit bypass is useful.)
Isn't cheating a symptom of the failure in how the education system works? This^ very interesting video of Ken Robinson produced by RSA was recently posted by someone on this forum. It diagnose this problem very well.
Trying to put kids in a system that is boxed due to constraints of age, time, subjects and grades is very harmful for the real development of individual kids. Kids understand this and get disengaged from the system but still have to force themselves to follow along because they are not provided with an alternative.
Look, we live in the real world with scare resources distributed to different needs, one of them education. It is sad that a lot of times students will have suboptimal teachers and teaching. Even at the best schools and universities in the world, there will be certain parts which will be taught badly.
Secondly, in the real world, some high value creating jobs can be extremely tedious and boring. And in the real world, some things you need to learn to do either these boring jobs or even interesting jobs, can be very boring.
As an educator, I am all for making education better (like a lot), but for the above reasons, parts of education will always be boring. And I understand that some students will cheat because of this boredom.
But what I observe is that the rate at which students cheat is much higher than the fraction of their education that is boring or badly done. Students cheat because it is easier than working hard, students cheat because it is cool to cheat and brag about it, students teach because they don't care about learning at all.
Really? Chegg was like the first textbook source that existed for college students. That used to be (maybe it still is) it's number 1 business. You could rent books from them for cheap for courses you didn't need to keep the text.
Working in higher ed for a few decades, I've seen their reputation go from - hey, they're undercutting our bookstore - to hey, they're a super good resource for students - to hey, their 'homework help' is just cheating for a cost.
They're a garbage company right now, from the perspective of a higher ed professional. They've gone off the rails with what could be a good business model.
I once had an electrical engineering exam and had the textbook brazenly lying open on my desk and trying to find something. The teacher made his rounds through the class, me assuming that he just hadn't noticed. After the third round when he walked by he put his hands on the book and said "If you haven't found it by now you might as well stop".
That's it, no repercussion.
That's something I will remember for the rest of my life. I do believe that people will always try to find an easier solution for a problem, that does not mean the punishment has to be too excessive.
If you notice that cheating attempts get out of hand take correctional steps before hand, by for example taking them in shifts to increase space between the students, and interrupt cheating attempts when they happen.
Just using draconian measures to try and make everyone behave is a lazy solution for a social problem.
A joke in what way? I am not familiar with US exam standards but I stand by my opinion that there are better ways to deal with cheating than harsh disciplinary measures.
That's also the way I preferred it with my students.
No. This is how some cheaters actually think. Nothing matters as long as they don't face consequences. They always have some excuse, and if this one doesn't work that one does. It doesn't matter what the excuse is as long as it works for them and they don't have to face up to the fact that they are doing something wrong.
Then there's just the ones who know they are cheating and are just okay with it. That's your grade A evil there which comes in two varieties: The ones who pretend they don't know, and the ones who do it openly.
So you have your cheaters who do it openly.
The ones who lie about cheating.
And the ones who have lied to themselves.
I don't think the message I wanted to send was received as intended.
I am not for allowing students to cheat, I would not have endorsed that for my students either. But I have had lots of professors and teacher colleagues who did a good job dealing with this without simply increasing the punishment until the intended result is reached.
Excessive cheating is not the problem itself but the symptom of an underlying rot in the system. Trying to increase the penalties is only treating the symptom not the cause.
So here's a plausible expansion of the scenario described:
The student actually does the problem on his own sheet, and does it correctly. But he then just happens to glance over at his neighbor, who he knows is actually really good, and sees a completely different answer. Concerned, he spends a bunch of time going over the problem, and can't figure out where he went wrong, nor how the other student got the answer they did.
So, in a fit of self-doubt, they copy the other student's answer.
I can see how, in that situation, a student might feel like things were unfair: They did try to do the question themselves, and if they'd known there were different exams, they would have just stuck with their own answer.
Of course, in the story as repeated, the student didn't say anything about trying to do the problem themselves first. But you can certainly imagine that happening; so that comes down on the side of, "Tell your students about nearly-identical problems."
I do not think that this make it better for the student. They still cheated, only instead of acquiring the actual answer they obtained the information whether their answer was correct. (Based on their knowledge of the situation.) So they acted unethically to gain an advantage, but instead sabotaged themselves.
Thing is, it literally does not matter if the neighbors exam was the same or different - they intentionally made a decision to cross the taboo line and act in an unacceptable, punishable manner.
"just happens to glance at his neighor" and his answers is unacceptable, and it's fair to punish acting on that information. It does not matter if they tried doing it themselves. It does not matter which answer was right. It matters that they decided that copying is an appropriate response, which is was not.
If the student considers "So, in a fit of self-doubt, they copy the other student's answer." as plausible course of action instead of automatically discards it as unacceptable, then this is a behavioral issue that needs to be addressed and corrected; and addressing and correcting that behavior is arguably more important than whatever they learn or don't learn in that course.
The whole point of this "dilemma" is that if the student feels that this is unfair then this indicates a fundamental mistunderstanding of the criteria. To fix this, it might be useful to have a followup test for all the students in this class with the following questions:
What circumstances justify copying another student's answer:
A) Seeing a different answer than yours;
B) That other student being really good;
C) It being a full moon;
D) None of the above, nothing justifies copying another student's answer.
If anyone - especially the culprit - answers anything than D then they should be blocked from taking any tests until they complete some program on the University rules and ethics. If after this incident the student feels that they have a reason that justified their copying, then they're wrong and need to be educated until they know the correct answer. And if they refuse to accept that there is no valid justification and that they must obey the no-copying principle no matter what justification they have, then it's reasonable to expel them from the university.
It didn't seem like much of a dilemma to me. If it appears that it was a clear case of cheating, then the only response to the student is that they shouldn't have to be told not to cheat. Where's their inner moral compass?
How severe should the penalty be? I give an instant oral quiz, and if it looks like they could have solved the problem anyway, I let them off with a warning.
Speaking from a really longterm idealized answer, the answer is "both".
You should communicate to the students that you take several measures to make cheating difficult like passing out multiple versions of the same exam.
You should also mention that you're consistently inconsistent. Sometimes you have 4 exam kinds, sometimes just 2. You do your best to mention it, but you know, people forget things.
Forgive an unkind analogy, but when you're training a dog you don't want to reward it _every_ time. You actually want to be inconsistent with rewards so that it doesn't intentionally stop the behavior the first time it doesn't get a reward.
The goal is to train students to not cheat by communicating that the cost of cheating is very high. At the same time, actually hunting down possible cheaters is hopefully not valuable - there's always some freeloaders but they should be at a low enough percentage to not influence very much.
In academia, professors are quite concerned with cheating right now, when all teaching happens online.
Some professors are considering using software to randomize the exams, so no two exams are identical. This obviously increases the grading effort. Some are asking students to take a pledge, like "I solemnly swear to not google the problem". Some are just trusting the students to not cheat.
There's no easy and universal solution. Ideally, technology could solve the issue, but it's not that easy to identify and implement a technology solution in a matter of weeks.
If the goal is to gather as many valid evaluations of the students' knowledge as possible, then the students should be made aware that there are multiple versions of the tests.
Students who might have tried to cheat if all the tests were identical would probably not attempt to cheat if they knew that there were multiple versions.
By reducing the frequency of cheating you increase the frequency of valid evaluations of students' knowledge, and that should be the goal of testing I think.
Making it easier for students to cheat just makes the tests more and more of an evaluation of the students' morals, not of their knowledge about whatever the test supposed to be about.
TL;DR: Sometimes "opportunity makes the thief", so telling students that there are multiple versions should reduce the amount of cheating by making them aware of how difficult cheating would be.
143 comments
[ 54.0 ms ] story [ 3431 ms ] threadSay, for example, if the professor put the answer up on Blackboard, seemingly by accident, which came in the form of an incorrect answer to a real exam question. They could then give everyone who gives the wrong answer an XF. I'd wager that would be most people, barring the few smartest who noticed the error. Could you really blame people for wanting to get it right and not handicap themselves relative to the other students?
Given that, the university then has a responsibility to make it as hard as possible to cheat. Fail grades for cheating work because they create an incentive to avoid cheating. Except in this circumstance the incentive effect could not work because the student had no idea they'd be able to be caught.
This attitude is so strange to me. I would assume that cheaters are a minority of students. (Perhaps a large minority, sadly. I've heard some third-hand stories.)
My school had a policy of an automatic 9 month suspension for a first offense. Professors were officially discouraged from exercising any discretion, and they were supposed to report all incidents to the disciplinary committee. For a second offense, the penalty was a permanent expulsion. I did see the system fail a few times when I was there, but it worked a lot of the time.
as far as how widespread cheating is, that probably depends on the culture at your particular institution. my school had a very strict policy for programming assignments: you were only allowed to discuss them with the instructor, a TA, or in public on piazza. and yet, every time I went into the cs lab, I would see everyone huddled around each other laptops debugging code for the latest assignment. I'm not sure whether they simply didn't understand the rule or didn't realize how obvious it was what they were doing, but essentially everyone in that room was bending the rules a little.
I’ve been listening to the History of Philosophy podcast, and one thing that struck me is that early philosophers seemed to consider virtue the end goal of philosophy. That is to say, we should care less about if our physics students can do physics, and more if doing physics had made them more ethical human beings. It’s quite a reversal we’ve ended up with...
On the day of the test, the professor wrote the questions on the board and then left the room. Almost every student in the class cheated. Some came with chits to copy from. Some copied from others. Legends though, they came with pre-written answer sheets for each of the 8 questions and attached answers for the 5 questions which were asked.
As TA and prof, I have seen cheaters copy word to word, spelling mistakes included, incorrect grammatical structures as is, basic mistakes like 6/2=2 untouched, because they are literally Ctrl-C Ctrl-V with their hands, with no brain involvement whatsoever.
Some people, yes. Others are there to actually learn something. Either because they're intrinsically motivated to do so, or because they understand that while a degree may help get them in the door somewhere, career success in many fields does depend on actually knowing what you're doing.
But I don't see where the responsibility goes from hard to "as hard as possible".
There was an Econ exam that I hadn't prepared for, and I got there fully prepared to flunk it. That day there was a problem with the photocopier. We had to fill in a scantron, but they didn't have copies of the exam for everyone. The professor read the questions and answers.
And I watched people's heads go down.
> At the end of the day, people are there to come out with a degree and a good grade, not feel good they didn't take the same opportunity available to others.
This is an age-old question in education: are you trying to learn something, or get the credentials necessary to get a job? I wasn't doing a major in Econ, so the fact that I didn't get much out of the class put me strongly on the side of shrugging that off as "they pretend to teach me, I pretend to learn."
But for most of my coursework, I really did want to learn things and get recognition for having done so.
> Given that, the university then has a responsibility to make it as hard as possible to cheat.
I strongly agree. I know there were guys who cheated in various ways, I watched one guy who went around the labs asking people to do his homework for him. And because universities have allowed degrees to be devalued, I've spent hours in interviews demonstrating basic CS concepts.
When I went to the other side and started having to enforce the rules, I realized that while the top decile could be trusted, everybody else was suspect, and enforcement of the rules was incredibly hard. What do you do when you grade a string of tests containing the exact same silly mistakes? What about a group of students that suddenly jumps to high performance on one exam and then immediately crashes back down? Remote proctoring is even harder. I once saw a student fabricate an entire school with the aid of his parents, complete with website, address, and fake teacher information, so he and his parents could do the exam together. How are you supposed to vet hundreds or thousands of people like this?
I really hope that I get a faculty position where I can just teach low-pressure graduate-level seminars, and never have to learn the answer. Mass education is damn hard.
It’s no different from cheating in sports. Yeah, the sports cheater is probably damaging their body and potentially harming their natural potential. At the same time, cheating may let them make the team and even the big leagues. That can change a person’s entire career, just as cheating to get into Harvard can.
Everytime, I end up with the same conclusion. Overpopulation and imbalanced distribution of resources is the root cause for everything wrong with this world.
I keep doubting myself. The problem shouldn't be that visible. People don't seem to blame overpopulation... maybe it's not a problem but everytime, it is the problem. Look at education systems of smaller countries. They are way better than what you get as the number of people in your country increases despite bigger spending.
Stopping overpopulation shouldn't be hard. We haven't tried anything. How can people be so sure? Why not disincentivise having kids by making it too expensive? Maybe use a tax cut equivalent to someone's wealth so rich people are as much affected as poor people (albeit a bit less in the grand scheme). Free birth control for everyone - poor people in less developed countries are responsible for most kids. Force longer education period so fertility rate goes down by the time people settle down. There are so many ways you can push the average marriage age faster than it is happening naturally now. Why not do it for less developed countries? Isn't it better if there is less suffering collectively in this world?
How about paying people to not have kids and adopt? What's the problem with it? Taxes are used for so many other useless things.
</incoherent rant>
I am okay with downvotes but please provide a reason. I really want to know. Maybe my thoughts are unhealthy.
Please explain more.
Still, it will always be a "problem" but that also makes it an opportunity. Lack of opportunity or access to the things needed to even take an opportunity are worse than scarcity (in itself).
Every other metric I can think of will be hard to convince and implement. It can also be pretty subjective.
> It can also be pretty subjective.
The comment I replied to literally mentions "everything wrong in the world" in the opening, it couldn't get more vague and subjective than that, but empathy is the problem? Overpopulation is subjective, misdistribution is subjective, quality of education is subjective (you may say knowing things and being able to use them well is "better", but that's subjective, too), even human extinction being a problem is subjective. As anything else relating to human ambitions and what we consider a good life.
Reasonable people will agree that poverty is bad.
Depending on the axioms you have defined for yourself. There is statistical evidence for poverty causing health problems or people to die early on average. Infant mortality rates among poor households or kids with malnutrition.
You could always think of unexpected death as unreasonable or objectively bad. By that line of thinking, you can conclude that poverty must be objectively bad. Anything that can result in unexpected death afterwards is objectively bad.
No matter what you end up with, you still need a fundamental axiom or moral position, I believe.
Like why is 3 greater than 2? Because I attach a greater position value to 3 but what makes this objective really? Most people agreeing with it?
Can most people agree that this world needs less unexpected deaths?
Yes.
Does stopping the number of people born reduce that?
Yes.
Should you stop the demography with highest infant/kid mortality rate first?
Yes.
What's so subjective about it now? Unless you believe that unexpected deaths are a great thing, in that case I think you are fucked up.
Those can't be the roots, since both are symptoms.
I live in a country with 1.34 billion people. Surely, overpopulation isn't a problem when every place is crowded with pollution that people can't/won't be able to control due to sheer numbers and poverty.
Poverty is there for myriad of reasons. Improper distribution is one of those. And poverty breeds overpopulation which again breeds more poverty as people having the most kids can't properly pay for them. They don't pay enough taxes either. Tax theft is big in india.
You want to solve tax theft or thousands of other government problems?
Biggest democracy has got you covered there. Most people won't or don't care about it. It's too much friction to move things but it's there for a good reason. A big nation shouldn't disappear under its citizens feet in a night.
So now you have a slow corrupt bureaucratic system as a feature.
If you're looking for root causes, the place to look is human nature. What do people naturally do, without even thinking?
You get cities because people need things from other people, so there's value in living close together. Cities overstrain the environment because, well, you packed in a ton of people all doing things in close proximity.
You get inequality of wealth and power because of network effects. Why do people form networks? Would it really be better without these networks?
> And poverty breeds overpopulation which again breeds more poverty...
Yet many nations got out of that cycle, so it's not that simple. And poverty is consistently on the decline globally, so doing nothing is a great strategy at the moment.
> You want to solve tax theft or thousands of other government problems?
Tax avoidance is easy: lower the rates to what people will actually pay. Even when you can charge high rates, you still quickly hit a point where it's counterproductive. Governments, like all kinds of organizations, work best when they limit themselves to a well defined mission rather than trying to be everything to everyone.
Why are we stuck with governments? Again, it's human nature. It used to be that people were stuck with a dictator running their life: you'd work all day and send a portion of it to the dictator, who used that to bribe his lieutenants to support him, and they'd make sure some other invader didn't take over.
Democracy simply means that instead of having a dictator confiscate our wealth, we elect representatives who take our money to bribe our neighbors so they can stay in power.
We live with this protection racket of taxation because, fundamentally, power struggles are inherent to human nature. The best we've been able to manage is to set various groups at odds with each other so no one has too much power.
This could easily be a set of videos recorded from a real lecture. Problems could be assigned from problem sets (plug n chug from schematics). What someone would pay for is 1 on 1 time from a professional.
And there's NO reason to fail someone, as in retake and pay a pound of flesh again. If you don't get it, go to a different degree that doesn't require it.
That is until the top decile is stuck in a required course they're not very good at and aren't interested in, then suddenly they have no problem handing in the same, slightly modified assignment the rest of the class handed in that was emailed by 'some classmate' who enjoyed stats programming in R.
Because for 99.999% of new faculty, your experience will be as follows:
Adjunct (part-time) teaching of introductory courses to earn your stripes (students definitely don't want to be in this class, and you will need a second and/or third job).
Full-time teaching of introductory courses after 5-7 years of part-time, because now you're not tenure track, and you still have to earn your stripes.
Full-time, tenure track teaching of introductory courses with a couple of 300 levels thrown in to test your merits during the 3 year tenure process.
Tenured full time faculty status, teaching where you finally get to teach the classes that students want to take.
It's a garbage system.
Used alongside the odd random visit I would imagine this to be incredibly effective.
You missed this bit.
As I recall, they came back later and revealed that they were a secret shopper for Shell, and we failed the random inspection. All because of a question that no real customer ever asked in my experience. ;-)
My boss didn't care, fortunately.
"Oppurtunity makes the thief" as the old saying goes.
That being said, some kinds of cheating don't deserved to be punished. The given scenario is bad for the cheater, but some students put so much effort into writing notes to cheat with that they just end up doing the amount of effort required to pass.
Bottom line, my job was to teach, grade homework and proctor exams. The last thing I want to do is play piggie -- I don't wanna stay up at night worrying over my student getting expelled / deported. The easiest way to keep focused on my actual job is to take some precautions that make cheating feel risky to the students.
So I really don't understand the other professor in this anecdote. Your students are stressed, sometimes as a result of bad life choices, sometimes as a result of external issues like family death&drama, sometimes as a result of sadistic asshole professors... typically it's a combination of the above. So for each 200-person exam, you'll frequently have students for whom this is literally the worst week of their brief and sheltered lives. They might make a choice they'll regret. So you have a choice: add risk to scare all but the very most dedicated cheaters (and fuck those kids), or lay a trap for the vulnerable kids that the skilled cheaters will skate through.
The purpose of learning is to understand the topic at hand. But school's purpose isn't for learning, primarily. It's for accreditation and getting a paper saying you learned certain things. And then you use that paper to get a better job, make more money, and get things like retirement and health insurance. (It used to not be that way - things changed after the GI bill... but I digress.)
If the school was non-profit and did as such, failure would not be wasted money or a permanent grade against you. Failure would only be lost time... But that's not how it is. Failure haunts you the whole time. The negative grades stay in the form of ever-approaching GPAs. But if you buy more classes, you can buy the average a slightly bit higher IFF you get better grades than the average.
The moral dilemma is that schools are using the threat of continuous harm of failure(GPAs) and harm to the pocketbook over taking a class and not doing well in it. Schools and their teachers aren't on the hook for bad teaching. It makes sense when possible to use other methods to make sure your grades and wallet aren't harmed.
I believe it would be different if schools were actually on the hook for teaching what you paid them to do. That calculation would certainly change - we wouldn't have "weeder" classes... Well, not in anywhere near the same way.
(Wonders just how much schools make on failing students. I'm sure there's an Atlantic or Vice writeup in that question.)
Once I was a TA for a class and a student came to me asking for a regrade on part of one exam. They claimed that I unfairly gave them zero on a particular problem. I looked at it and yes, I did give them a zero. But I never give zeros on a larger problem unless nothing or almost nothing is written down. I think that I'm quite generous for gibberish even. There is no way I would have given this student a zero for this problem if he had what was present at my office hours. I told him that I believe he is trying to cheat by doing the problem after the exam. I didn't report him but I did make it clear that I didn't believe his story.
The student went to the lecturer, who was furious with me. The lecturer said that I had no clear evidence that he didn't do the problem; me saying that I never give a zero unless the problem is not attempted wasn't good enough for the lecturer.
I spoke with my advisor, who said that to avoid problems like that he puts a line through any problems that were not attempted.
The lecturer didn't like the line solution after I graded the next exam. He was furious with me again. He told me "You don't trust your students!" I don't trust some students, sure. Since putting the line through empty parts I haven't had anyone claim anything similar, though.
The lecturer was apparently so disappointed with me that he told my advisor to kick me out of his research group (or something along those lines). I didn't learn that until years later. My advisor told me at the time that the lecturer had a reputation for being a jerk to his TAs, so I figure my advisor just thought this lecturer wasn't worth listening to.
(To his "credit", the lecturer had another issue with me that he got furious over: He had a very particular system for writing grades on exams. As I recall on the back page on one of the top corners he wanted the total points received written over the total on the exam. On about 5 out of about 100 exams I wrote the wrong denominator by accident; the numerators were correct in all instances as far as I recall. Now, I'm sorry to have made this mistake but it wasn't on a number that mattered and it was only on a small number of exams. This issue combined with the cheating is all rather minor and not worth raging over.)
I did keep putting lines through empty parts when grading, at least after I stopped TAing for this class. But this experience made me reluctant to give a zero for cheating, much less report the student to the university.
I know a misconception exists in education "about memorization being something every student can perform optimally for succeeding" in getting top marks. That ignorant belief is simply not the case for everyone and I find it cruel we penalize children that could have succeeded if they're allowed to use their organization skills optimally. Otherwise they're hindered by their genetics or neglectful/abusive home-life that resulted in development issues with memory.
I know I'm not alone as a programmer using a trusty note taking app for multiple things I rather not memorize. Obviously I don't use it for syntax of the language I'm writing. Thus, brings me to my last point of how students will still be unable to cram all the subject material on a notecard and for completing a test within the time limit given. That makes it so the student gains a real life useful skill of organizing and breaking down material being taught. Further helping the longterm understanding than just memorize a few nights before the test and all is forgotten later depending on the subject being taught.
Plus of course it puts everyone on the same page with people who are going to sneak in cheat sheets anyway. They always are there.
It's a pretty niche job if you need to do thermodynamics calculations under time pressure without any reference material.
I had a similar experience in my better CS classes. Some professors clearly realized that, if you hadn't been paying attention and learned how not to screw up, you wouldn't be able to figure it out, even given pretty wide ranging access to information. So they'd hand us a take-home exam, due in a day or two, and we'd be allowed to use all relevant documentation.
I always liked these, as they seemed the most similar to what would happen in industry, unlike some other professors who expected you to have memorized obscure library functions and be able to write perfect code on paper under time pressure.
First: the more support you allow, the harder you have to make the test.
Second (especially relevant for the equivalent situation in job interviews / hiring): the student can hire someone to do the exam for them. (Or more casually: a friend can help.) There's no guarantee that on the job you'll always have that friend available, or are willing to shell out for hired help, nor even allowed to leak company secrets like that. So that hired help doesn't approximate real world conditions as allowing to look stuff up on Googe definitely does.
This applies to all assignments of course, but they tend to have further out due dates, giving students flexibility to budget their time.
In uni, we had one particular class where we were supposed to study something like ten or twelve different concurrent algorithms in some weird concurrent language that I’ve forgotten and not seen since, and would be asked to write four of them out in the test and talk about them a bit. We didn’t know in advance which four, so most people tried to memorise as many of them as they could and hoped that the four that came up would be among them.
Me, being stubborn, thought that was too much effort. Instead, I learned the bare minimum about the language so I could wing it in the test and write it up myself. I already programmed as a hobby so thinking critically about problems wasn’t a big deal. My code was a bit weird, because, for example, I only learned the syntax for while loops and not for loops, so had to make do with what I did know, but it didn’t matter which four problems came up. The problem description gave me enough to figure it out. I got really good marks in that class and put substantially less effort into preparing for the test than my classmates did.
My point, I guess, is that memorisation is a stupid thing to optimise for (I guess it’s harder to design good tests that can’t be best beaten by memorisation), isn’t that useful in “real life” and is a lot more work than if you were thought the actual building blocks so that the students have a toolbox they can use to approach problems.
I think there's different "cheating by writing down formulas / information we're supposed to have memorized" and "cheating by copying someone else's answers".
In the article, he actually describes a situation where there were identical problems, but with slightly different numbers between seats. One of the students had test version A, but the correct answer from version B. This clearly falls into the latter category; reading the full story, it's really not clear that having a notecard would have helped him.
The implicit assumption here is that telling the students is more effective prevention than not telling them. I don't think that's true, at least on a wider scale; secret anti-cheating countermeasures and rigorous punishment are surely going to be better than published countermeasures.
You can only cheat at a game, which is what the accreditation system we call education is.
If we stopped making it a game, without points to earn and rewards for points, people wouldn’t cheat.
That aside, I think the act of cheating is a call for help, even if the callee doesn’t realise it. The pressure to do well coupled with a persons lack of self confidence or self belief come together to make cheating a reasonable route.
If you believed you could never get x, that getting x would massively improve your life, social status, parents acceptance etc you might be prepared to do anything to get x.
Not everyone though, some are probably just resigned to never getting x. They too are calling for help imo. But we weren’t talking about them.
You can only cheat rules/agreements/contracts. No rules, no cheating.
I'd agree that learning has no rules. There are some natural laws I guess, like the "forgetting curve", but those cannot be cheated (hence natural laws).
> I think the act of cheating is a call for help
Or a lack of respect for the rules, or sheer curiosity. Maybe there are more reasons.
Nonsense.
There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used.
For example, if I were an engineer designing an elevator, it must meet its rated load limits times a hefty safety factor. That safety factor is a test, and if I were unscrupulous then I would have every incentive to "cheat" to falsely certify compliance. Under ordinary circumstances, that safety factor is left unused when the elevator is operated as per instructions, but the cheating covers up a latent problem with potentially catastrophic consequences.
An academic degree isn't about a worthless piece of paper; it's ostensibly certification that the student has an acceptable mastery of the course material. Much of that material won't be used on a day-to-day basis, but it's still there, and it's still part of the credential that the graduate holds out to the world.
But even if you're unconvinced by that, there's also the character signal at play. The "kind of person" who will cheat on an academic test is intuitively also the "kind of person" who will cheat in other areas of their professional life.
Such as in your elevator example, make it so that the operation of the elevator is tested with a heavy load approaching the desired safety factor, and evaluating the performance under this heavier than normal load? The idea being that the only way to "cheat" is to design a better elevator that can still meet the requirements.
Of course the reason this isn't done is because of the extra cost of performing a more comprehensive test, but given how much students are paying for education, surely we could make some improvements over the status quo.
I can't help but draw comparisons to SpaceX's starship development process vs that of the more traditional aerospace manufacturers. There isn't any way for the engineers to cheat because their work is tested under the target conditions and either passes[0] or fails[1][2]. The goal being to achieve the most efficient design that still meets requirements.
[0] https://twitter.com/elonmusk/status/1259344535991140352
[1] https://www.youtube.com/watch?v=QlDBjHa0NkU
[2] https://www.youtube.com/watch?v=wFXQ5SRCy74
Yes, this is a reasonable way to test an elevator. "Cheating" in this context might mean surreptitiously swapping out the load for a lighter load, so that the elevator appears to pass the test while not actually supporting the heavy load during real operation.
For a real-world example, see https://en.wikipedia.org/wiki/Volkswagen_emissions_scandal
There is unlikely to be a silver bullet that can make any test completely immune to cheating.
> Nonsense.
> There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used.
So stop issuing tests.
Or more importantly, the schools could stop holding a record of students grades. Students could still takes tests to confirm their understanding, but their results wouldn't benefit (or hinder) them in any sort of hiring situation. If you sit through the classes, assuming 1) you've put in the work 2) the classes were taught well and 3) you posses sufficient intelligence/aptitude, you should finish with some useful skills. However, no one "graduates" in the sense that they currently do: no fancy paper slip to mount on the wall, no gold stars. School then has nothing to do with career outcome, at least not directly.
This might seem unreasonable -- the employers of the world need some way to judge their fresh candidates!
This is easily solved: people new to a field would start out as apprentices. With a year or more of experience on their resume, they may now switch employers as readily as anyone else does now -- employers, in my experience, don't often check their candidates' formal education at that point.
But what about people who, at that point, would manipulate their way into a rewarding career without yielding any real value in return? At that point, we've just shifted the "game" to "get paid while being lazy and/or useless", right?
Well, maybe, but nothing has really changed. People already brown nose. Companies are already rife with cronyism/nepotism/etc. Lazy and incompetent employees already exploit past association with big name companies to coast through their career and into cushy retirement. None of that has changed.
The only things that really change in this hypothetical: universities are no longer the gatekeeper to people's future, people no longer feel the need to take on massive debt to study disciplines that they're not even sure they'd enjoy, and employers would be expected to invest a tiny amount (modest pay; the overhead of letting them shadow established employees) in new hires making their first entrance into the working world.
Costs are a problem but that's driven by prestige whoring, not irrationality. The right majors/degrees are still worth the debt.
Personal recommendations are used because they work and outsource trust. Yes, they can be abused but this hurts the recommenders social status which reduces their incentive to recommend.
> to claim that everyone has an incentive to cheat is a very low opinion of humanity. trust noone. they are all out to get you.
I don't believe "everyone has an incentive" was ever stated, but regardless of that, if we could trust everyone with everything laws wouldn't need to exist.
maybe this is a language issue. i understand 'there is an incentive' as something that is there generally motivating, as if it asks for that to happen. like advertising is an incentive to buy, and a stop sign is an incentive to stop or a green light is an incentive to go.
but just like an open door may be an incentive to enter someones home, it is not an incentive to steal, and therefore i reject the idea that the existence of a test is an incentive to cheat. but to me "There is an incentive to cheat any time there is a test, where capabilities are measured that are not routinely used." says exactly that.
most laws are created as a reaction to missbehavior. it's not about needing to trust everyone, but designing the system so that everyone can learn and assess their learning without needing to fear that failing that assessment will leave a permanent record.
this is the problem with graded tests. they don't so much help me assess what i know and need to catch up on, but they grade how well i keep up with the classes progress. there is no way to catch up and have that test redone, showing that i have learned the material, even if i did. a top grade does not indicate that i learned everything, but also that i learned it at the pace the teacher wanted it. if i fail the first test, i have lost any chance to make a top grade for the rest of the year, even if i catch up and maybe surpass others.
And of course everyone has an incentive to cheat. Higher grades and lower time investment are the incentive. Those are basically universally appealing to college students. Having an incentive doesn't mean everyone is going to take the opportunity, as some feel it is wrong and others feel the risk doesn't justify the reward.
in my university there were no grades. only pass or fail. anyone who was not at risk of failing had no incentive to cheat whatsoever.
As for victims, ultimately you would probably defraud future employers presumably, by taking their money under false pretenses, and damage the school's reputation. Perhaps also kill people when the plane you designed crashes or something.
Whoever is making decisions based on the outcome of the test. A school or professor risks it's reputation by graduating incompetent students. An employer who hires based on a false GPA risks wasting its money and time.
Ah, I see. Do you also believe in ‘good’ and ‘bad’ people?
In my experience, all people are capable of cheating and given the uniquely require circumstances, most people will cheat.
Integrity is a well supported, secure persons value.
Have you ever heard of people who wait until the last moment to start work on an assignment, or who study intensively a few days before an important exam?
It's always seemed to me one of the big 'features' of college education you can't get with a library card or internet connection is the unstoppable steamroller of deadlines - slowly but inexorably advancing, reducing your options from 'study now, study later or fail' to only 'study now or fail'
Do you think that exams and assignments provide such a role, and if so would your proposal address that need?
Now I'm way more motivated to do stuff as early as possible since I can clear them off my to-do list as soon as they're done.
A generation later, society will have gained a fuller sense of the need for discipline, self motivation, and personal responsibility in any individual who wants to succeed at life. And so there will be many more people in the world trying to cultivate those virtues. It might turn out to be a great benefit to the human race as a result.
That's fine for the SATs. The logic holds in a system where people are competing against large groups of others. There you can treat it as a call for help and understand it as a reaction to pressure due to lack of confidence. But what about more competitive situations, times when say 10 people are fighting for one job in an all-or-nothing competition?
Think astronaut training. Only one person is going on the rocket. Second place, staying on the ground, is meaningless. Confidence doesn't matter. Everyone is confident. Trying hard is irrelevant as everyone is trying 100%. Work meaningfully harder than everyone else might cause you to collapse into injury or breakdown. In such situations "cheating" may be the only way to differentiate one's self from others.
I think that is why when professional athletes cheat they demonstrate little remorse. At their level it isn't a failure of confidence or morality, instead it is a reasoned choice made of necessity. That's why we need to closely watch people in such competitions. Morality, right and wrong, isn't enough.
I think the OP is saying that if the "reward" for education is that you know more -- and that brings it's own intrinsic value on a cultural and personal basis -- then there will be no need to "score points" on any test or report card. Therefore, no "game" necessary. And that follows that there will be no cheating (why would you cheat if you, and your culture, are happy with the reward of education in and of itself?).
imo, it's will take a quantum leap for us as a culture to reward education for education sake. To the point that when you interview, simply saying, "yes I understand X concept" will be met with belief by the interviewer/teacher (vs/ now when you have to "prove" your knowledge by a game-able "score").
As for personally, imo, people can do it now: love to learn for the sake of learning. No point in cheating on the test if you're learning the subject for your own edification. The test will simply tell you where you are still weak.
I had classmates who just wanted to pass by any means necessary and take that diploma back home and get a job there —their competition back home was usually subpar since these students had a leg up with their English and had a diploma from a US school, no matter how far down the US News ranking a system would be.
Sometimes in life to grow you need to deal with a situation or goal that produces anxiety.
Also sometimes your goals are simply not in line with your abilities or level of dedication, and you need to fail.
Honestly, an exam should control the success of learning, nothing more. Passing an exam should always finish a course or topic. Failing an exam, though, should be quite normal and just lead to more repitition.
But the current system in the USA will put the history calculus person in the OR.
This is a key point, but it may not be formulated the right way. Here is my take on it, as I explain it to my students:
If you are able to cheat so well that I can't catch you, it means you don't actually need to cheat. Corollary is that if you need to cheat and you do, I will see it, because you'll be bad at that too.
Well, it depends which skills you want to test with your exam..
I don't necessarily think it's because they have copied the code and they don't know how it works, I just think they don't know how to speak straightforward about code or are afraid of looking stupid and as a result I either have to accept stuff is the way it is, write down what they said and go try to decipher it at my leisure, or gird myself for battle.
...
Cheating is a form of deception. The person in the story wanted to cheat an exam, and then later - manipulate OP that they are the victim. One shouldn't fall for that.
No. Deterrence is not putting up a hidden camera. Deterrence is when you tell people the area is monitored by cameras. People will slow down their car when they know the police is around, not when they don't know.
If you want to catch cheaters, you set up anti cheating. If you want deter cheaters, you warn them about anti cheating measures.
If students know which exams have extra measures, they won't cheat on them, and will get a free pass everywhere else.
If you don't, sure - at first there is no deterrence, but later (after a few students get failed and exposed for cheating), it will deter others from trying.
He did a whole bunch of really interesting experiments, like self-reporting how many you got wrong on a test before it is fed into a shredder and similar.
fascinating book.
How many of you work exactly 8 hours a day with your employer?
I’m trying hard to think about what cheating would be in the software world. Here’s a few things that I see:
* Grabbing an open source package without a license check * Pushing to production without running all the integration tests (or worse, just SSHing into production) * Accepting a PR without actually reviewing the code * Writing code without unit tests
Its hard for me to come up with these - most involved skirting controls, which were usually there for a reason.
Data science seems much more obvious:
* Lying about performance on the test set * Using highly correlated features which leads to instability * telling people the model has broader applicability than it does (e.g. a speech to text model was trained on complaint phone calls and claiming it can be just as good for in person meetings)
This could be because I spend more time in the ML world, but I’m curious what other people think.
This might be due to procrastination, personal issues, distraction, other job pressures, lack of direction / guidance on the task or many other reasons, some of them good.
The problem is succumbing to the temptation to report progress when there really hasn't been any.
Just today at stand up I said I was going to "wrap up" a story I had been working on for several days. In reality I have not written a single line of code, so by "wrap up" I meant "start and finish coding". Here's how I rationalized it though:
It's not like I've done nothing, the past few days I've been mentally digesting the story and trying to figure out how all the pieces fit together. I've been researching different solutions online, reading through GitHub and StackOverflow. I thought about it a lot, even when not at work (i.e. in the shower or laying in bed). Now that I think I've fully worked out the solution in my mind, actually coding it is the easy part. Sometimes I hit a snag, but more often than not, I can just iteratively breeze through the implementation.
- I like to solve
In fact, reading my comment will probably help you get in the zone to write the code just before deadline!
For example, writing code without tests can be caught with a coverage checker.
In my ideal setup, you would be allowed to bypass the coverage checker with a simple comment-annotation in your code.
The point is not to be bullet-proof, but to have the automation remind you to do the Right Thing; and to make doing the Wrong Thing something you explicitly have to go out of your way to commit to. Not just something that happens because you are absent minded.
(And there are lots of legitimate reasons why you might want to bypass a coverage checker or a linter. So an easy, but explicit bypass is useful.)
Trying to put kids in a system that is boxed due to constraints of age, time, subjects and grades is very harmful for the real development of individual kids. Kids understand this and get disengaged from the system but still have to force themselves to follow along because they are not provided with an alternative.
^ https://www.youtube.com/watch?v=zDZFcDGpL4U
Secondly, in the real world, some high value creating jobs can be extremely tedious and boring. And in the real world, some things you need to learn to do either these boring jobs or even interesting jobs, can be very boring.
As an educator, I am all for making education better (like a lot), but for the above reasons, parts of education will always be boring. And I understand that some students will cheat because of this boredom.
But what I observe is that the rate at which students cheat is much higher than the fraction of their education that is boring or badly done. Students cheat because it is easier than working hard, students cheat because it is cool to cheat and brag about it, students teach because they don't care about learning at all.
The most interesting part I noticed is Chegg which I hadn’t heard of before.
Working in higher ed for a few decades, I've seen their reputation go from - hey, they're undercutting our bookstore - to hey, they're a super good resource for students - to hey, their 'homework help' is just cheating for a cost.
They're a garbage company right now, from the perspective of a higher ed professional. They've gone off the rails with what could be a good business model.
That's it, no repercussion.
That's something I will remember for the rest of my life. I do believe that people will always try to find an easier solution for a problem, that does not mean the punishment has to be too excessive.
If you notice that cheating attempts get out of hand take correctional steps before hand, by for example taking them in shifts to increase space between the students, and interrupt cheating attempts when they happen.
Just using draconian measures to try and make everyone behave is a lazy solution for a social problem.
Then there's just the ones who know they are cheating and are just okay with it. That's your grade A evil there which comes in two varieties: The ones who pretend they don't know, and the ones who do it openly.
So you have your cheaters who do it openly. The ones who lie about cheating. And the ones who have lied to themselves.
I am not for allowing students to cheat, I would not have endorsed that for my students either. But I have had lots of professors and teacher colleagues who did a good job dealing with this without simply increasing the punishment until the intended result is reached.
Excessive cheating is not the problem itself but the symptom of an underlying rot in the system. Trying to increase the penalties is only treating the symptom not the cause.
The student actually does the problem on his own sheet, and does it correctly. But he then just happens to glance over at his neighbor, who he knows is actually really good, and sees a completely different answer. Concerned, he spends a bunch of time going over the problem, and can't figure out where he went wrong, nor how the other student got the answer they did.
So, in a fit of self-doubt, they copy the other student's answer.
I can see how, in that situation, a student might feel like things were unfair: They did try to do the question themselves, and if they'd known there were different exams, they would have just stuck with their own answer.
Of course, in the story as repeated, the student didn't say anything about trying to do the problem themselves first. But you can certainly imagine that happening; so that comes down on the side of, "Tell your students about nearly-identical problems."
"just happens to glance at his neighor" and his answers is unacceptable, and it's fair to punish acting on that information. It does not matter if they tried doing it themselves. It does not matter which answer was right. It matters that they decided that copying is an appropriate response, which is was not.
If the student considers "So, in a fit of self-doubt, they copy the other student's answer." as plausible course of action instead of automatically discards it as unacceptable, then this is a behavioral issue that needs to be addressed and corrected; and addressing and correcting that behavior is arguably more important than whatever they learn or don't learn in that course.
The whole point of this "dilemma" is that if the student feels that this is unfair then this indicates a fundamental mistunderstanding of the criteria. To fix this, it might be useful to have a followup test for all the students in this class with the following questions:
What circumstances justify copying another student's answer:
A) Seeing a different answer than yours;
B) That other student being really good;
C) It being a full moon;
D) None of the above, nothing justifies copying another student's answer.
If anyone - especially the culprit - answers anything than D then they should be blocked from taking any tests until they complete some program on the University rules and ethics. If after this incident the student feels that they have a reason that justified their copying, then they're wrong and need to be educated until they know the correct answer. And if they refuse to accept that there is no valid justification and that they must obey the no-copying principle no matter what justification they have, then it's reasonable to expel them from the university.
How severe should the penalty be? I give an instant oral quiz, and if it looks like they could have solved the problem anyway, I let them off with a warning.
The severity of punishment, student morality, ect are context only. hence "A non-moral dilemma about cheating"
You should communicate to the students that you take several measures to make cheating difficult like passing out multiple versions of the same exam.
You should also mention that you're consistently inconsistent. Sometimes you have 4 exam kinds, sometimes just 2. You do your best to mention it, but you know, people forget things.
Forgive an unkind analogy, but when you're training a dog you don't want to reward it _every_ time. You actually want to be inconsistent with rewards so that it doesn't intentionally stop the behavior the first time it doesn't get a reward.
The goal is to train students to not cheat by communicating that the cost of cheating is very high. At the same time, actually hunting down possible cheaters is hopefully not valuable - there's always some freeloaders but they should be at a low enough percentage to not influence very much.
Some professors are considering using software to randomize the exams, so no two exams are identical. This obviously increases the grading effort. Some are asking students to take a pledge, like "I solemnly swear to not google the problem". Some are just trusting the students to not cheat.
There's no easy and universal solution. Ideally, technology could solve the issue, but it's not that easy to identify and implement a technology solution in a matter of weeks.
Students who might have tried to cheat if all the tests were identical would probably not attempt to cheat if they knew that there were multiple versions. By reducing the frequency of cheating you increase the frequency of valid evaluations of students' knowledge, and that should be the goal of testing I think.
Making it easier for students to cheat just makes the tests more and more of an evaluation of the students' morals, not of their knowledge about whatever the test supposed to be about.
TL;DR: Sometimes "opportunity makes the thief", so telling students that there are multiple versions should reduce the amount of cheating by making them aware of how difficult cheating would be.