Live data from Hacker News

My students cheated... a lot

crumplab.com

791–800 of 1001 posts

Re: My students cheated... a lot

#791
Have so far read up to part III (not inclusive)

I've read the story of a failing teacher, who set up his class in a way which normal sharing of experience and information among classmates would fall under the technical definition of "cheating". In order to "not cheat", students had to artificially isolate themselves, alienate themselves from their classmates. For a person of an earlier generation, who did not take online examples when also being in a group chat on their phone, that may seem like not much of a requirement (and even then I'm not sure.)

The responsibility for this debacle lies squarely on the teacher's shoulders - perhaps with some responsibility on the department/faculty for lack of oversight and guidance regarding expected dynamics for different ways of evaluating student achievements.

Rule of thumb: When most people fail, the failure is systematic.

One must also wonder about the maturity of these students. I'm wondering if, over the past couple of decades, a more childish attitude and behavior is not presented as acceptable for a longer period of time. Suddenly these, well, judging by the chat messages, children, come up against the wall of having to obey a semi-arbitrary rule, or else. It seems like they have not been directed along a path where they could take gradual steps to scale this wall. This is perhaps less of the teacher's responsibility, but he must still realize that's the kind of student cohort he is facing.

Re: My students cheated... a lot

#793
post #685
post #633

Earlier quoted context omitted.

My position on this has been pretty controversial when I've shared it before, but I still think it's correct: Measuring knowledge at scale is futile, harmful, and pointless. The fact that a lot of society has been arranged around the fiction that this is a feasible endeavor does not mean it has borne out in practice, and prioritizing assessment in this way has been gradually hollowing out most forms of pedagogy of th…

What you are saying is not at all controversial, but it is incomplete, which is probably why you have received pushback in the past. Criticising the existing system is easy. Giving an alternative is harder. Implementing that alternative and showing that it's actually better on some metric is MUCH harder than that. But you have not even given an alternative!

Are you aware that there are grade-free and exam-free schools out there and that they have been operating for decades?

Measuring the effectiveness of school systems is difficult because of selection bias, but I'm sure you could find some attempts (e.g. PISA) if you went looking.

Re: My students cheated... a lot

#794

Earlier quoted context omitted.

> HOW do you measure knowledge? And when you decide how, how do you scale it? I have experienced good tests and bad tests. I studied in France, tests were open book with no multiple choice questions, only problems to solve. This approach scales badly and is a lot of work for the professor grading but it measures knowledge. The problems were long, had few questions besides describing the problem and maybe a few questi…

I'm currently grading an open book test as you describe. It turns out that someone put their attempt at answers on chegg.com shortly after I posted the test. The temptation to use chegg is too great for students to resist. When chegg has the wrong solution (which is often the case), students will doubt themselves and will go with the wrong chegg answer. To be clear, the only goal of chegg.com is to help students chea…

My solution to this is to use version control and have them record an explanation of their work. If they copy from chegg they also have to forge a commit history, as well as explain the code line by line. I’d like to see them do that without learning anything.

Re: My students cheated... a lot

#795
post #525

Threads like these are always frustrating, because as usual people (programmers in this case) freely air their opinions on how schools are broken with phrases like “we need to fix the system”. As someone who studied pedagogy for years and quit due to an immense frustration with exactly this — how broken the system is — I would encourage you to entertain the thought that maybe, you as a person who is almost in all cas…

>In the case of testing it’s because you choose to focus on the obviously bad thing (current state of testing) rather than the very complex and difficult question behind it: HOW do you measure knowledge? And when you decide how, how do you scale it?

This sounds like an entirely different question. When you have a method for testing, you have at least two different measures of effectiveness:

- how well the test measures knowledge when it is taken honestly

- how likely is the test to be taken honestly vs. subverted

I thought this thread was about the second question, but you seem to be focused on the first. But these problems require different kinds of solutions, and crucially, it is much easier (but still not easy) to verify success or failure in addressing the second question (cheating) than the first (predictivity).

Re: My students cheated... a lot

#796
These students are ready to enter the workforce. Result driven, team players, pro-active, tech savvy. They are displaying all qualities required to be a productive employee. To them, and in practice, education is merely a hurdle. Just graduate them already.

Re: My students cheated... a lot

#797
I am reminded of a quote from fictional character Hagbard Celine (Illuminatus!)

"They are not in university to get knowledge, they are here to receive papers that say they are qualified for job xy"

Re: My students cheated... a lot

#798
post #525

Threads like these are always frustrating, because as usual people (programmers in this case) freely air their opinions on how schools are broken with phrases like “we need to fix the system”. As someone who studied pedagogy for years and quit due to an immense frustration with exactly this — how broken the system is — I would encourage you to entertain the thought that maybe, you as a person who is almost in all cas…

> Well, maybe we should start by rolling back this common conception that when it comes to schools, everyone’s opinion matters an equal amount

I have some thoughts about the education system, and despite not being a teacher or academic I like to believe that my opinions have some value because I'm an expert programmer that has worked in the field for over 50 years. I attended three different major universities and have degrees in Math, EE, and CS. I still code almost every day (my Emacs configuration is never finished!), and I have in the past taught or been a teaching assistant for both undergrad and graduate courses for four semesters. Cheating has always been a concern, but now things are different.

The original article highlights the scale of exam cheating during the pandemic, but for us, the readers of HN, there is another problem with university learning that happens because of the internet. I've seen this affect virtually all of my younger friends pursuing degrees in CS. Programming assignments in school are unrealistically difficult, and it causes everyone to cheat.

Here's a typical real-life example: after covering doubly linked lists in the undergrad data-structure's class the programming assignment is to write a GUI based text editor in Java using doubly linked lists. This isn't especially hard for a professional programmer, but this is the first programming assignment of the course. Students had to wrestle with Eclipse, learning the AWT/Swing interfaces, event loop programming, and how to translate low level pointer based data structures into non-idiomatic Java based imitations of linked lists that kind of simulated using pointers. Most of the students really couldn't do this on their own, but they didn't have to because they can find the solutions to this very problem right on the internet.

Why would professors give such a program to beginning programmers to write? Because every student turns in a solution, and this causes the professors to lose touch with how difficult their assignments are. Over and over again difficult assignments are given, but the students are seemingly keeping up. The bomb lab assignment is a great assignment for CS students[1], but I've seen it given out with far too few attempts allowed to solve it. Again professors feel like a small number of attempts is all the students should need, they keep turning in the answers. The reason they can is that the complete solution is available on dozens of public Github repos and web sites.

The consequence of such hard and challenging programming assignments is a kind of inflation of the difficulty. The high difficulty causes students to cheat more, since their fellow students are cheating by downloading, cutting and pasting, or simply sharing their programs. There are commercial web-sites like chegg.com that sell the solutions to virtually every homework problem found in CS textbooks. Why should an undergrad spend so much time working on their own homework solutions while other students work openly in big teams at table in the university library?

This kind of cheating is pervasive at the undergrad level. How do we prevent our students to being pushed into cheating to keep up? Graduate school is different, the classes are smaller and more interactive. In my grad school classes I've often had to go to the board to demonstrate my code or proof to the class. Professor Dijkstra used to give individual oral exams to his students. So small interactive classes would help.

I've also seen assembly language programming classes given that require all work to be done on lab computers. The lab computers weren't on the internet and students had to sign in with the lab proctor to use the machines for their assignments. This at least helped some with the problem.

If I was teaching a programming class now, I would require everyone to maintain a git repo that could be checked for realistic commits of the programs as they are written. This might discourage the simple copying of a solution from GitHub the day before the assignment was due.

[1] The text for the bomb-lab assignment (highly recommended by the way): Computer Systems: A Programmer’s Perspective, 3rd edition,Bryant and O’Hallaron, Prentice-Hall, 2016 (ISBN: 0-13-409266- X), a google search will return many bomb-lab assignments and solutions from colleges all over the world.

Re: My students cheated... a lot

#799
Discussion forums like Piazza can really help with stuff like this. By allowing students to communicate with each other in public, with TAs/instructors explicitly present answering questions, the need to get a group chat going diminishes and you avoid the temptation to cheat.

As an aside, I will say, weekly quizzes are one of the lowest-reward types of homework. They are mainly favored by instructors because they are automatically graded, and they do really little to make you excited about the material. Students recognize and respond to instructor effort (or a lack thereof); if there’s a written assignment every two weeks that students know you are going to take time to grade and give feedback on, then they are significantly more likely to invest the effort to do it well.

Re: My students cheated... a lot

#800
post #525

Threads like these are always frustrating, because as usual people (programmers in this case) freely air their opinions on how schools are broken with phrases like “we need to fix the system”. As someone who studied pedagogy for years and quit due to an immense frustration with exactly this — how broken the system is — I would encourage you to entertain the thought that maybe, you as a person who is almost in all cas…

Computer science is one of the harder subjects to assess knowledge through tests. In other industries, passing an accreditation exam gives you value in the market. In software engineering, certificates are seen as useless. This can explain the cynical attitude programmers have against testing.

I have degrees in math and physics. Those degrees gave me close to zero value in the market. Spending X years in school, then having to prove to an employer that you can do barely more than squat is a familiar experience in a lot of fields.

There are tests you can take in those subjects, such as the Graduate Record Exam. Those tests work to some extent because the subject matter is relatively mature, and consistent from one college to another. And yet there are entire fields of math and physics that I've never been exposed to. Their main purpose is to see if you're conversant in a body of knowledge that would prepare you for typical graduate study, not for a job.

Software engineering is a comparatively young field, with less standardization. There are even debates on HN as to whether software engineering is a real thing. There are places where every programmer has the title "engineer" regardless of their background.

I'm only employable because most people hate math and physics so much that they're relieved if anybody offers to do those things for them. That, and I'm pretty good at programming and electronics.

Post reply on HN