3 ms·

One thing that I really appreciated at MIT was that when we were given take-home tests or other take-home assignments, it was expected (and often encouraged) th

One thing that I really appreciated at MIT was that when we were given take-home tests or other take-home assignments, it was expected (and often encouraged) that we collaborate with our classmates. Writing down the exact same answer as another classmate was fine, so long as you documented who you worked with and who contributed what to the assignment. It provided an important lesson since collaboration has been a huge part of my professional career. Now if those Harvard students broke the rules, well, they broke the rules. But I can't help but think that if Harvard is restricting collaboration in take-home assignments, that they're doing it wrong.

danso 14y ago
Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her? Sure, you could say that's a lesson in life, but ideally, you want a passing grade in engineering mechanics to mean that that person has competency in engineering, not merely the politics and management of engineering. Of course, this risk is mitigated by having the final weighted much heavier than the rest of the coursework and not being take home. But that approach ends up going to the polar extreme of valuing collaboration
- Anechoic 14y ago
  Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her? There were charismatic students who tried that, but it would become apparently very quickly what was happening and that point our competitiveness would kick in and we would demand contributions from that person before we would assist any further. But yes, there is always someone who will abuse the system, and you're not going to get far trying to achieve 100% compliance all the time.
- ghshephard 14y ago
  The mechanism to avoid having a charismatic leader unfairly being granted credentials in courses not based on their knowledge is have a pass/fail (heavily protctored) exam at the end of the sememster drawing on knowledge from all the sections of the course. Fail the exam, fail the course. Pass the exam, you get the grade you earned throughout the semester. We had a few of those at SFU, and they were awesome - the only ones who ever failed them were the cheaters - The ones who had actually done all the work/exams finished them in about half the time allocated.
- danielweber 14y ago
  On assignments, that ends up with the charismatic leader not understanding what's going on come test time. (I am unfamiliar with collaboration on take-home tests.) The practice questions really are warm-ups for the test in most classes.
- victorhn 14y ago
  "Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her?" Funny, you mention that. There was a "charismatic guy" who passed trough my whole computer engineering program just by getting a dedicated and smart girlfriend who would help him in all the assignments and even give him copy during written tests. The guy could not program to save his life, and yet got his degree. At the end of the program he ditched her instantly and got a hot girlfriend instead, i kind of felt sorry for her girlfriend, but there is your life lesson.
michael_miller 14y ago
I agree that giving take home exams might not be the best idea, but I place 100% of the blame on the students for this. It is completely reasonable to give academic honesty instructions to students and request that they be followed. Maybe it's not realistic to expect that the instructions will be followed, but it is certainly reasonable to punish the students for violations of academic honesty. Honestly, I think Harvard would be well within their rights to expel any student who blatantly plagiarized an exam, take-home or not. It's a little scary (maybe not surprising) that 2% of Harvard undergrads were involved in this scandal. This incident, combined with the Stuyvesant scandal makes me lose so much faith in the national student body. I wonder what the solution to this apparently endemic problem is. Maybe the best idea is to accept that students are going to copy and collaborate, and give each student personalized assignments.
- Anechoic 14y ago
  It's a little scary (maybe not surprising) that 2% of Harvard undergrads were involved in this scandal. This incident, combined with the Stuyvesant scandal makes me lose so much faith in the national student body. You've lost faith "in the national student body" based on the actions of a tiny minority of students? Really?
  - sur 14y ago
    It seems like the GP is saying that based partly on Harvard's reputation, not just the incident itself.
tkahn6 14y ago
Now contrast that with Virginia Tech's honor code policy in use in all CS courses where absolutely no collaboration of any kind is permitted. The only thing you can discuss with classmates is the meaning of a particular requirement. Sigh. Only 2 more years...
danielweber 14y ago
Take-home assignments in course VI are ripe for co-work. But the take-home test in 6.046, in my day, required us to affirm that we had worked on it alone. It was a lot of fun (I'm an algorithm dork), and I felt really bad -- like I had cheated -- when someone looked over my shoulder at what I was working on and shouted out the answer.
lesterbuck 14y ago
As a counterpoint, at Caltech during the 70s (at least) it was pretty standard to have a take home final that was (1) no collaboration, (2) closed book, and (3) timed. In fact, I very clearly remember screaming out loud on hard tests as time ticked away. Thank goodness I was in a deserted campus building in the middle of the night. You could study/cram all your notes and texts that you wanted before you put everything away for the test, but, alas, it usually didn't help. They weren't that kind of test. For non-take home tests of any kind, the campus honor system specifically prevented the professor from staying in the room where the test was being given. They could force the students to collect in one place, but there was a strict rule against any kind of proctoring. The professor or TA with the test passed it out, hung around for a bit to answer any initial questions, and then left the room. They would typically return every fifteen minutes or so to answer any new questions, and then finally return at the time limit to collect the tests. The professor also could not prevent you from taking the test from the specified room and going, say, to the library to take the test. You missed any chance to clarify test questions, and you had to be back before time was up, but they could not force you to stay in a room packed with tense people. It was pretty refreshing when trust and presumed integrity like that was integrated into the educational system.
- mikexstudios 14y ago
  The majority of classes at Caltech still have take home exams like you described. For midterms and finals, the exams were handed out at the beginning of the week and due at the end of the week. More often than not, they were open book (although the joke was that those books weren't going to help you solve the problems anyway). The honor code at Caltech was taken very seriously and in my opinion, the students reciprocated with integrity. Cheating and dishonesty was a big social stigma there.
  - tzs 14y ago
    My freshman year (1977-78) I was what was then known as a "computroll": someone who spends far too much time in the computing center, usually to the detriment of their academic performance. I managed to flame out so spectacularly first and second terms that I had to petition to be allowed to stay for third term. I was allowed, on the condition that I pass all my third term classes. It was close. It came down to one exam where I knew that if I got this one question, I would pass, and if I did not get it, I would fail and be kicked out of school. I also knew exactly what equation I needed from the textbook to solve the problem. I could visualize what page it was on, and exactly where it was on the page--I just couldn't visualize the damned equation itself. I was alone in my room, and the book was sitting closed on my desk just a foot or two away. It would have taken less than 10 seconds to glance at the equation I needed and pass that exam. But it was a closed book exam and peeking would violate the honor system, so peeking was out of the question. So, I failed the exam and got kicked out. I spend the next year at my local community college, taking courses in psychology, anthropology, and economics to show that (1) I could actually do college level academic work (2) even in subjects that weren't in my main interests. I also took a computer course, to show that I could succeed in college course even when I have access to a computer, to show I could keep my computroll tendencies in check. Caltech let me back in, and I even was able to transfer enough credits from the community college courses to come back as a sophomore. And since I had graduated high school in three years, the year at the community college actually dropped me back to where my age matched my classmates instead of being a year younger. So it all worked out reasonably.

danso 14y ago

Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her? Sure, you could say that's a lesson in life, but ideally, you want a passing grade in engineering mechanics to mean that that person has competency in engineering, not merely the politics and management of engineering. Of course, this risk is mitigated by having the final weighted much heavier than the rest of the coursework and not being take home. But that approach ends up going to the polar extreme of valuing collaboration

Anechoic 14y ago
Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her? There were charismatic students who tried that, but it would become apparently very quickly what was happening and that point our competitiveness would kick in and we would demand contributions from that person before we would assist any further. But yes, there is always someone who will abuse the system, and you're not going to get far trying to achieve 100% compliance all the time.
ghshephard 14y ago
The mechanism to avoid having a charismatic leader unfairly being granted credentials in courses not based on their knowledge is have a pass/fail (heavily protctored) exam at the end of the sememster drawing on knowledge from all the sections of the course. Fail the exam, fail the course. Pass the exam, you get the grade you earned throughout the semester. We had a few of those at SFU, and they were awesome - the only ones who ever failed them were the cheaters - The ones who had actually done all the work/exams finished them in about half the time allocated.
danielweber 14y ago
On assignments, that ends up with the charismatic leader not understanding what's going on come test time. (I am unfamiliar with collaboration on take-home tests.) The practice questions really are warm-ups for the test in most classes.
victorhn 14y ago
"Perhaps, but isn't there the risk that a charismatic leader could get team members to essentially do the work for him/her?" Funny, you mention that. There was a "charismatic guy" who passed trough my whole computer engineering program just by getting a dedicated and smart girlfriend who would help him in all the assignments and even give him copy during written tests. The guy could not program to save his life, and yet got his degree. At the end of the program he ditched her instantly and got a hot girlfriend instead, i kind of felt sorry for her girlfriend, but there is your life lesson.

michael_miller 14y ago

I agree that giving take home exams might not be the best idea, but I place 100% of the blame on the students for this. It is completely reasonable to give academic honesty instructions to students and request that they be followed. Maybe it's not realistic to expect that the instructions will be followed, but it is certainly reasonable to punish the students for violations of academic honesty. Honestly, I think Harvard would be well within their rights to expel any student who blatantly plagiarized an exam, take-home or not. It's a little scary (maybe not surprising) that 2% of Harvard undergrads were involved in this scandal. This incident, combined with the Stuyvesant scandal makes me lose so much faith in the national student body. I wonder what the solution to this apparently endemic problem is. Maybe the best idea is to accept that students are going to copy and collaborate, and give each student personalized assignments.

Anechoic 14y ago
It's a little scary (maybe not surprising) that 2% of Harvard undergrads were involved in this scandal. This incident, combined with the Stuyvesant scandal makes me lose so much faith in the national student body. You've lost faith "in the national student body" based on the actions of a tiny minority of students? Really?
- sur 14y ago
  It seems like the GP is saying that based partly on Harvard's reputation, not just the incident itself.

tkahn6 14y ago

Now contrast that with Virginia Tech's honor code policy in use in all CS courses where absolutely no collaboration of any kind is permitted. The only thing you can discuss with classmates is the meaning of a particular requirement. Sigh. Only 2 more years...

danielweber 14y ago

Take-home assignments in course VI are ripe for co-work. But the take-home test in 6.046, in my day, required us to affirm that we had worked on it alone. It was a lot of fun (I'm an algorithm dork), and I felt really bad -- like I had cheated -- when someone looked over my shoulder at what I was working on and shouted out the answer.

lesterbuck 14y ago

As a counterpoint, at Caltech during the 70s (at least) it was pretty standard to have a take home final that was (1) no collaboration, (2) closed book, and (3) timed. In fact, I very clearly remember screaming out loud on hard tests as time ticked away. Thank goodness I was in a deserted campus building in the middle of the night. You could study/cram all your notes and texts that you wanted before you put everything away for the test, but, alas, it usually didn't help. They weren't that kind of test. For non-take home tests of any kind, the campus honor system specifically prevented the professor from staying in the room where the test was being given. They could force the students to collect in one place, but there was a strict rule against any kind of proctoring. The professor or TA with the test passed it out, hung around for a bit to answer any initial questions, and then left the room. They would typically return every fifteen minutes or so to answer any new questions, and then finally return at the time limit to collect the tests. The professor also could not prevent you from taking the test from the specified room and going, say, to the library to take the test. You missed any chance to clarify test questions, and you had to be back before time was up, but they could not force you to stay in a room packed with tense people. It was pretty refreshing when trust and presumed integrity like that was integrated into the educational system.

mikexstudios 14y ago
The majority of classes at Caltech still have take home exams like you described. For midterms and finals, the exams were handed out at the beginning of the week and due at the end of the week. More often than not, they were open book (although the joke was that those books weren't going to help you solve the problems anyway). The honor code at Caltech was taken very seriously and in my opinion, the students reciprocated with integrity. Cheating and dishonesty was a big social stigma there.
- tzs 14y ago
  My freshman year (1977-78) I was what was then known as a "computroll": someone who spends far too much time in the computing center, usually to the detriment of their academic performance. I managed to flame out so spectacularly first and second terms that I had to petition to be allowed to stay for third term. I was allowed, on the condition that I pass all my third term classes. It was close. It came down to one exam where I knew that if I got this one question, I would pass, and if I did not get it, I would fail and be kicked out of school. I also knew exactly what equation I needed from the textbook to solve the problem. I could visualize what page it was on, and exactly where it was on the page--I just couldn't visualize the damned equation itself. I was alone in my room, and the book was sitting closed on my desk just a foot or two away. It would have taken less than 10 seconds to glance at the equation I needed and pass that exam. But it was a closed book exam and peeking would violate the honor system, so peeking was out of the question. So, I failed the exam and got kicked out. I spend the next year at my local community college, taking courses in psychology, anthropology, and economics to show that (1) I could actually do college level academic work (2) even in subjects that weren't in my main interests. I also took a computer course, to show that I could succeed in college course even when I have access to a computer, to show I could keep my computroll tendencies in check. Caltech let me back in, and I even was able to transfer enough credits from the community college courses to come back as a sophomore. And since I had graduated high school in three years, the year at the community college actually dropped me back to where my age matched my classmates instead of being a year younger. So it all worked out reasonably.