Reminds me of over a decade ago I was in a high school AP course. At the start of the course, the teacher recommended we get a AP prep book to study throughout the year, and recommended 2 brands. She specifically said not to get one brand, noting that it was of inferior quality. I had already got the prep book she didn't recommend as I had found it on sale before that remark was made. I later learned that she used that book for all of her test questions, sometimes literally copying the test from the book with minimal editing. I got a 100% on a test out of the blue after averaging in the 80s and then had to make it less obvious.

Similarly, with a slight twist, a prof I had used to recommend we read the entire text book, not just the chapters required for the class. It was an open book exam so people figured they could annotate the pages or whatever and didn't bother to follow his adcice. He would often say, 'everything [we] ever want to know is in that book!'

He was right. The final exam was verbatim the practice exam given at the end of the book. As luck would have it, I had discovered the practice exams a week earlier. Once I realized they were identical to the final, I literally copied the answers and handed it in. It was a 3 hour exam and half the class was done in 15 minutes.

I had one where on the first day of the semester, the very first thing the minute the class started, the professor casually said "if you do every homework problem in the book, you'll get an A on the final."

Not everyone was there on time, some students were going to drop & add the class and most people weren't paying attention.

He casually joked for a couple minutes then settled down to get started.

There was never another hint.

When the final came, each question was verbatim from homework problems that were not assigned.

I had an exam once where we were told to read the entire question paper first.

At the end it told you to ignore all previous questions and just do this one.

2nd favourite exam question after an economics multiple choice exam where we had to calculate the probability of passing the exam just by guessing.

I once had a sociology professor whose final was multiple choice and every answer (out of 50) was a B, except for the last, which was a C.

That’s a fun anecdote, but horrible teaching.

Here's my little related story. I was in one of my fave used book stores (Baldwin's Book Barn in West Chester, PA) which is not exactly close to where I was living at the time. After browsing for an hour or so and picking out a few books, I spied a title on a top shelf that I thought was interesting but couldn't remember ever seeing before. I pulled the book down and began perusing it, and on the inside cover there was a familiar-looking stamped bookplate that said "Property of <My Name>." I was a little freaked out because I couldn't remember ever owning the book. After thinking about it a few minutes, I dimly recalled lending the book to somebody years ago and I guess it was never returned to me. When I went downstairs to the checkout counter, I showed the older man working that day my ID and the bookplate while telling my story, saying I never got a chance to read this book. He very kindly told me to just take the book and didn't charge me for it, saying "I'm just glad it found its way back to you."

What a lovely story, thank you for sharing.

It depends what the class was for. I had a sort of similar experience with a physics class where the professor would take a break mid-exam and just give everyone the answers, but it was a class that was a prerequisite for engineering students who didn't care and really didn't need to, and also a stepping stone for the passionate physics students who did care, but would do the heavy lifting later. The prof was well-regarded for spending his time more on helping with true understanding for the students who cared.

That describes my entire education.

Wouldn't you rather learn the material?

Sure. Not a lot of learning going on during exams, I'm afraid.

Cheating on the exam means it wasn't learned.

My chapter of Tau Beta Pi (engineering honor society) got so tired of how unfair it was that people would share around old tests and decided to do something about it. Since there is no way to stop people from doing this, we got permission from the dean to start a test and study sheet bank and made it publicly accessible on a website.

Over his whole career as a professor, my father was so concerned about test sharing that he kept a file of every test he ever gave and was careful to never ask the same question twice.

> my father was so concerned about test sharing that he kept a file of every test

I finished 2nd in my high school class (humble brag) because my best friend cheated (and got a zero) for an important biology term paper.

Story:

- My friend hated biology and he decided he would save time by submitting his older brother's bio term paper as his (his brother was 4 years older).

- A month after we handed in our term papers, the bio teacher asks my friend to step into his office, whereupon he shows him a copy of his brother's term paper. Big fat zero.

- It dinged my friend just enough that (even though he was more talented) he finished 3rd in the class.

- I ended up getting a tuition waiver to our state university (where we both went).

That laziness cost him about $10K in tuition (in 1980s dollars).

edit: the person who finished 1st in our class went to Yale. She was more talented than both of us combined.

Respect, the amount of work to do that is mind boggling.

Its easy to spot the pattern and useful to know what wont be asked.

Errr... many questions can just vary numbers a bit. There's very little value to knowing that the inverse of 3 mod 7 won't be asked this exam - it's trivial to change the numbers and test the same knowledge with a slightly but sufficiently different question.

I imagine there's a somewhat similar type of questions for non-CS disciplines (same case description, different key clue, hence different answer).

It's not entirely trivial, it takes work. If you are simply changing the number but ask the inverse of x mod y every time you didn't win the game. 2 out of 3 times changes nothing.

I had the most fun with multiple choice tests. I took the average A,B,C and D answers from previous tests. The teacher not paranoid would have a lot of B, slightly less A and C and below average D answers. The paranoid teacher would have perfect distribution. The supper paranoid used some kind of RNG which also comes in patterns.

One of each type came to ask me how I did it at the end. The last one was funniest. He had one test with a trick question that everyone got wrong. It just didn't have enough C. One of my answers was more likely wrong than right. I had to go over the A,B,D answer candidates a few times. Reduced the list to 3 potential trick questions. Figured out that 2 of them were correct. Don't ask me why but the remaining one must have been C.

He explained how over his [long] career he started out with to few D answers. I name the other teachers still doing that. He then went for the perfect distribution. I name the teachers still doing that. And now he thought he had the perfect formula. How am I suppose to fix this?? He asked. I'm like, I'm not doing your homework but I do foresee a very interesting meeting with the other teachers.

edit: I did study but cheating was just to much fun. When they switched to tests on computers I made a good bit of money as administrator. I swapped some of the answers around so that the wrong answer was right and everyone got poor grades. They would purchase the list of correct answers from me. Then after they aced it I would reduce their grade to something sensible for the student. When they protested I told them that they paid for answers not for grades. Grades are very expensive as -hey- they could potentially ruin this good thing I got going here!

Am I understanding this comment correctly that you were administering tests and rigging them to cause people to fail and then charging people money to get the rigged answers for the tests so they could pass?

If so, this is some real unethical and criminal behavior.

But I'm inclined to think it's all made up considering the first part of the comment seems to imply that you were doing well on tests simply by guessing the answers based an A,B,C,D distribution. Which is silly as the best you are going to do with that strategy is maybe get a slight hint for a few of the more difficult questions.

> Am I understanding this comment correctly that you were administering tests and rigging them to cause people to fail

They got many answers wrong on the test but I adjusted their grades later. The grades do have to be credible for the show to continue.

> and then charging people money to get the rigged answers for the tests so they could pass?

People who didn't bother to study purchased the answers but if they always had low grades they would forever have low grades.

> If so, this is some real unethical and criminal behavior.

The only thing that really bothered me was this: Modifying the log files left a log entry with a terminal number in the log. I provided the teaches with a fair clue by changing the test answers and indeed eventually they smelled a rat. The cat and mouse game couldn't go on forever, eventually they would figure it out. I picked a scapegoat, a girl who couldn't hurt a fly. It couldn't have happened to a nicer person.

The teachers ganged up around her pointed and shouted then suspended her for a week.

I thought they acted ridiculous so I erased everything from the server and arranged for backups to be formatted when attached.

A few years later I ran into her and apologized but she didn't recall the event.

The teachers were very happy to return to paper based tests.

> unethical and criminal behavior

Ah yes, was I ever that young? We got computers in school and all we were allowed to do was tests. I assure you I've learned quite a lot. For example: Backups should have write protect implemented in hardware. Who in their right mind....

Yeah. Tests are broken :-)

Oral examination is way superior but you need a healthy “students to prof” ratio for that

My discrete mathematics course had an oral exam for the final exam. I took it as a summer course and the books didn’t arrive until 1/3 into the class’s.

I felt like I was BSing my entire oral exam but got an A.

I was never the same after that course. 10 hours a day of logical homework fundamentally changed how I thought.

To this day I still cringe when people use phrases like everyone or all without actually meaning the set of everyone in existence.

Many of my CS classes posted pdfs of a bunch of previous exams on the course resource site to study from.

I did the same, shared with my fellows, got denounced, got a written warning by the university and the faculty hated me afterwards for "hacking".

Gods I miss the halcyon days of pulling the entire file directory down and rooting around for interesting stuff like that.

> slightly changing constants

As a maths lecturer once said: some of the problems will be similar with some of the numbers changed. Pi, however, will remain the same. ;)

My digital logic professor was so lazy he reused previous year homework questions on exams and exams were completely open notes. The old questions were available with answers on old versions of the course website that were still online. Everyone just brought the answers to each exam.

He may not have been lazy; he may not believe in the system.

He was probably still believing in the salary he was drawing from the system.

Exams are for you to evaluate your progress, a professor doesn't give a flying duck about your grades, maybe only in the statical sense. You are only lying to yourself.

Your GPA may be what gets your foot in the door, but your knowledge and your ability will determine the long arc of your career.