Hacker News Reader

Yeah, it entirely depends on the timeline of the poisoning.

If the bad results were uploaded just prior to the exam, and the results were cloned, people likely cheated. I think this trap is fine. Hopefully he had his bad results pulled afterwards too, so the site only contained flawed data during the exam period.

If the bad results were uploaded several days prior to the exam we run the risk of people studied, and couldn't fathom why their answers didn't match the "official results", but learned the wrong result was the correct answer. It may have stuck in their heads simply because it was an unexpected answer. Perhaps these people didn't even go back to the site during the exam at all, but the weird results from the studying phase were what they could recall.

In this second case, the trap isn't good at all. It may have caused people to recall the incorrect answers learned from study, not from cheating during the exam. This is a horrible thing to do to your students. It may have caused some people to doubt themselves during study, and pushed them to cheat during the exam because they clearly didn't understand something. This is unfortunate. Prof and student share some blame here. How to classify this case is difficult. Without the poisoned study material, these students may have actually known the material, felt confident, and aced the exam. We'll never know.

Thank you, this is what I was trying to describe. I don't intend to argue this necessarily happened (I'd expect it probably did not), just point out the possibility. It doesn't seem to me the professor really did enough to determine the actions of these students and there is potentially good reason for them to believe the 13 or so students who claim to have not cheated. Certainly the "one student who is right ‘on the bubble’" could be given more benefit of the doubt.

I guess it's moot given the actual policy against looking up tests without the professor's permission. I can see how it's a shortcut to find the correct answers without necessarily understanding the material. But I still see a reasonable perspective that the professor's decision potentially caught people who were genuinely intending to learn.

>Hopefully he had his bad results pulled afterwards too,

Also remember anything you put on the internet is forever... Pretty much this guy may be doing the 'bullshit generation problem' we are starting to accuse AI of doing.

Everything you say might be true. But it seems that the only "studying" that really stuck was when they went to quizlet. Read the rest of it where most of them admitted to cheating -- which probably means pulling it up and copying the answer.

Cheating is rampant in general and a lot of people bend over backwards to rationalize it.

In short, they were given numerous valid resources and chose chose not to use them. Then they went completely outside the class and used unverified information in place of learning the material.

Assuming it wasn't an "open internet" test, then there is nothing wrong with this.

Out of all those people studying all those wrong answers author doesn't mention a single one of them bringing it up during office hours. I'm assuming they didn't. They all thought they were getting away with something and had an edge over actually studying. They were wrong.

> They all thought they were getting away with something and had an edge over actually studying.

My point is that these people didn't (necessarily) do something wrong. The only reason they had to think they were doing something wrong is that the rules say they're doing something wrong.

If a professor provided the questions and answers for flash cards to be made, is that functionally different from the students getting such information from a third-party? It's "cheating" from a rules-based definition but I don't see how a person is necessarily not learning from that.

>It's "cheating" from a rules-based definition

Is there any other kind of cheating?

Assuming there's a reason for a rule, an action might not go against that reason even if it goes against a literal interpretation of the rule. In such a case, it should be determined that there's no reason to be against the action.

The ostensible reason to be against cheating is that it devalues the degrees the university gives out if it can't be trusted that a person earned said degree through merit. I'm saying that a person who found the exam and learned from that material isn't going against this reason to be against cheating; they did, in fact, pass the exam on their own merit. At least, I don't see why I should consider it differently.

They did not, in fact, pass the exam on their own merit; they failed it miserably. Furthermore, in situations where they find the correct answers to the questions that will be asked, or even just the specific questions that will be asked, and then passed the exam, they would not have demonstrated that they could do so on their own merit, as an examination can only be a smallish sample of the knowledge they are expected to have in order to justify a passing grade.

In this case, they cannot even appeal to that perennial cheater's excuse: "what matters is being able to get the right answer, not how I got it." Here, they demonstrated utter incompetence at that task! Their knowledge was so impoverished that they could not even identify blatant errors.

It's clear from the story that no "learning" from the exam occurred. Students just copied the answers. Remember, this is a multiple choice exam, so this means question 42 - D, question 43 - A, etc -- nobody is looking up what this even means. According to the teacher, the answers should obviously wrong to anybody who knows the material.

Although it's very funny that this happened on an ethics course, the "is it ethical to cheat on an exam if you learned the material" question doesn't even apply here imo

> It's clear from the story that no "learning" from the exam occurred. Students just copied the answers.

I don't see this so clearly. It seems to me that this is assumed. Why do you think it is so clear that this is what the students did?

The point of an exam is that it’s a crude instrument to determine a students mastery of the subject matter. If that is not discernible because some students were able to memorize answers vs. others that actually learned the content what’s, the value of the exam is diminished.

From TFA:

> My University has an academic honesty policy that explicitly says that looking at other tests without the instructor’s permission counts as cheating (Although had I know it would be this much of an issue I would have been explicit about that in my syllabus as well, rather than just linking to the policy, an oversight I plan to correct going forward.)

If they went to Quizlet and viewed previous tests, yes, they were cheating by definition. No assumption necessary.

The answers were purposely wrong though and supposedly obviously so. Something I have seen in every multiple choice test I have ever taken.

I don't think it's possible anyone legitimately studied using easily determined false information. If they did, failing the course seems appropriate anyway.

They're on a website which is telling them that they're getting real exam questions and answers. It's reasonable to expect the questions and answers are not false and make, e.g., flash cards from them.

I'm not so sure I'd agree that failing the course is really appropriate. These people are paying to learn and the teacher provided wrong answers to students who may have legitimately been looking to study. Of course, 2/3 of the students admit it, so it's moot in those cases; at least some of those who claim innocence may have a reasonable claim.

Making flash cards from the exam's answer sheet is somewhat hard to defend on it's own. But when you're getting tested for your knowledge of a subject, and you make flash cards from an exam's answer sheet with blatantly incorrect answers that demonstrates a lack of knowledge, I have a hard time feeling sympathy.

To oversimplify, it's hard to defend a CS exam where a student selects (D) here and their excuse was that a flash card told them to do it:

How would you import a module in javascript?

A) import * from 'foo'

B) from 'foo' import *

C) import { * } from 'foo'

D) for i in `seq 1 10`; do echo "$i"; done

Note, there are a tonne of questions upstream from this like the validity of requiring these students to be taking this class, and requiring them to take this exam, and this format of testing students, etc. But keeping the question of whether the student's outcome was reasonably impacted by the decision to replace the answer sheet with a fake answer sheet on Quizlet, I have a hard time believing it did.

I don't specifically disagree that the student should have failed the course when they clearly do not understand the material. I know that's contradictory to what I said above; from one perspective (the university's; rule-based) they should fail and from another (the one for which I am arguing) they should not.

> whether the student's outcome was reasonably impacted

Here's where I have a hard time. It's very easy for the outcome to change from a failed test to a failed academic career. The method by which the professor determined who was cheating is dubious enough that it may have caught non-cheaters (from an ethical perspective; assume that the person who found the test on the website is genuinely intending to study the material which they believe is correct). I guess it's moot from the university's perspective if it's against their cheating policy. But on this ethics topic, I'd have to argue for the professor to have more doubts about their decision.

I see your point about a student attempting to pass the exam, studying the incorrect material, and failing the course. However, I see that situation as morally equivalent to the following scenario:

Sam has been having trouble understanding their evolutionary biology coursework all semester. It's coming close to the final and they are frantic to find any help they can get. Finally, they encounter the Answers in Genesis website and use it to make a set of flash cards. They drill those flash cards relentlessly in their best effor to learn the material. Come the final, Sam answers every question with "God did it and evolution is a lie".

I will agree with you that Sam has not behaved unethically and has not cheated on the exam, just as you believe that some students did not cheat on the ethics exam. That said, I do not believe that Sam should pass the final, as they have no understanding of the material. Even though Sam has put hard effort into studying, they have studied the wrong material and put zero effort into vetting their sources, so they deserve to fail the course.

Do you believe that Sam deserves to pass the course or am I missing an important difference between my hypothetical and the actual events?

They (at least the hypothetical ones who were not cheating in this philosophical sense) should not be allowed to pass the final and should not be expelled from the university for what they did. Admittedly, that enters the realm of personal opinion; the university’s policy is that a student using such a resource is cheating. The teacher also seems to be considering a lighter punishment so I don’t expect this case will carry any such injustices. (Really, I expect most or all of the students were cheating in the moral sense. The possibility that one wasn’t makes me question the policy.)

That being said, there does seem to be a certain assumption about the intentions of any student who would use this service. It’s not hard for me to imagine someone who has to and wants to bust their ass learning getting caught up in unfortunate circumstances (like the stress of wanting to pass a final causing the poor decision to use this study resource coinciding with their teacher poisoning said resource coinciding with the aforementioned stress causing them not to question the incorrect information they found ultimately resulting in their expulsion for otherwise not-necessarily-unethical behavior) because it’s assumed the professor was adequately tricky with their questions.

(Maybe I just need an hour off work to explain myself properly on a Friday.)

These days I find it fairly risky to think that it's reasonable to believe just anything you read on the internet because a website says so.

The academic integrity policy can forbid access or use of test solutions. A student who studied from that resource isn't innocent.

You want to tune your policies so that students will bring that extra test that was lost in the printer tray back to you.

> A student who studied from that resource isn't innocent.

I guess it's more understandable if that's the position being taken. However, that makes me want to ask a different question: why is that bad? A student who uses said resource to study for the test during usual study hours but walks into the test without any extra material is taking the test on their own merit.

I'm going to assume you're giving this hypothetical student the benefit of the doubt, so we can discard the "ACCACA DABACCA" version of test prep (Which is what the professor in the linked article is alleging) https://youtu.be/0cyS4HZ_Dxc?t=56

With that out of the way: Why is it bad for a student to study for a test from an unofficial test bank?

It's really stinking hard to write a good test question. It needs to be possible, clear, and not exactly* like problems you've done in class and homework. There are a few exceptions to this for "canonical" problems/derivations. Capacitance of parallel plates, for example.

It's very different if a student chooses a problem to study, and that problem is on the test, versus a student who is given a problem to study that is from the test. For example, if I give a "derivation of cylindrical capacitor" in my test prep packet, and a student looks at that and also studies the spherical and parallel plate geometries, that's great for them! They thought deeper about what was going to be on the test, and directed their attention accordingly.

Because the test can't reasonably cover all of the course material in depth. The idea is that the test takes a representative sample of the course material, so your understanding of the material on the test is indicative of your understanding of the course material overall. Specifically studying the material you know is on the test invalidates this premise, and by extension the results of the test

The article explicitly states that the school policy disallows the use of previous tests without the professor’s approval. So any use of the test as a study aid is cheating.