<p>I consider the accusation of cheating a serious enough one that the standard of proof has to be high. Perhaps the standard of proof is such the statistical evidence is considered positive and there is no appeal. That does upset me, but if that is the fact, then there is nothing that can be done. But if this is an emerging standard, I would challenge it. In court, if it came to that. If I were accused of this and were innocent, I would press it all the way I could.</p>
<p>Look at the example in the paper. They didn’t even factor in the same wrong answers and the probability that it wasn’t cheating was 0s after the decimal place to 19 digits.</p>
<p>
Wait, so the most important criterion for evaluating a standard for judging guilt or innocence is that it produce some guilty verdicts? </p>
<p>If it’s impossible to reliably determine how cheating took place or who was actually guilty of it, then no one should be penalized. It’s never OK to punish the innocent because that’s the only way of getting at the guilty. If the proctoring doesn’t allow you to identify and punish the guilty, improve the proctoring.</p>
<p>At least in the three test results published in post #108, the scores were not high. Is OP’s son actually at the top of this class ? If not, if is less clear why this particular pair cheated other than bilateral choice.</p>
<p>IANAL, but my opinion prior to seeing the integrity policy is the burden of proof that a student is a victim of cheating and not a willing participant falls on the student.</p>
<p>Just an aside, I cannot quite explain to myself why the better student would have so many wrong answers that were quite different that the rest of the class. As others have mentioned, it is usually pretty easy to reliably exclude at least one answer, and often two. So most of the wrong answers should have been 50% concordant with the rest of the class – assuming four multiple choices per question.</p>
<p>
</p>
<p>Nobody said there was no appeal. There in fact was. If the student declined to drop the course, the professor would refer the case to the Dean of Students and the student would have full due process to make the case at that point.</p>
<p>The student (or family) could present the statistical evidence that were was a 1 in 274 billion chance that they turned in indepedent exams or present the notes that the two students had taken that had incorrect information (that others in the class did not take) or that he was cheated from but that it was not collaborative.</p>
<p>The professor is allowing them to not face formal disciplinary action by allowing them to drop with a WF.</p>
<p>It’s like when you do something bad at work and your boss gives you a chance to resign instead of being fired.</p>
<p>The problem is now the other child has officially taken the WF and withdrawn from the class. He will be off to college this fall and never look back. My child, the junior, the one they basically are saying the other child cheated off of, is left to fight this alone. Now he has to prove he didn’t cheat if we fight this. I have looked at the phone records I have on a couple of the test dates and my child did receive one text, but sent none during the class time. I will have to call verizon to get the other date since I don’t have that bill yet. I have a meeting with an attorney Friday.</p>
<p>
</p>
<p>Did you call him or his parents? Do they know what is going on for your S?</p>
<p>The other student won’t be able to forget about it if you fight it. He can get subponead by attorneys, he can get letters from the Dean of Students, and other things can happen. Academic dishonesty is a serious deal. This shouldn’t wait to the fall anyway, the Dean should pursue it this summer.</p>
<p>(You may not be able to leave the other student out of the mess if you do fight it, depending on how these things work. I don’t have experience with the Dean of Students so I don’t know.)</p>
<p>Using the methodology described in the paper that BCEagle referenced earlier in this thread, I calculated the odds of the results on the first test being nothing more than a coincidence at less than 1 in 10 billion. That is for the first test alone! This calculation doesn’t account for some reasonable similarity in the answers due to common study notes, and it does assume that the questions are completely independent events, which they likely are not. But the odds are still staggering. </p>
<p>This is a very interesting thread, because it taps into the desire we all have to believe and protect our children. If this were me, given these odds I’d be looking for the out that would result in the least damage, because short of the other student testifying that he was the lone party to the cheating and explaining how he did it, which he is not motivated to do, I don’t see how your son can prove he didn’t participate. A phone log can only prove that he didn’t use a phone to relay answers, not that he didn’t use some other method to relay answers. I’m sure this has occurred to you, but it’s much easier to think of ways that this cheating could have occurred with the two of them working together than it is of the other student pulling this off by himself. </p>
<p>Very difficult situation. Good luck to you.</p>
<p>tjd, when you did that calculation, did you include the questions that both got right, or only the ones that they got wrong? Also, you might take a look at what would happen if both students had narrowed the questions that they got wrong down to two possibilities, and chose the wrong one each time. And what happened with the 7 test where they didn’t appear to have excessively similar answers? These should be factored into a statistical analysis, too.</p>
<p>Despite these statistical qualms, it does look bad for the other student, at the moment, based on his own decision to accept the WF to avoid being accused of cheating. It is unclear whether smom123’s son was complicit or not, but if not, I still think he should not simply accept the WF.</p>
<p>At this point, we don’t want to accept the WF. I have not talked to the parents of the other child. He told my son today he didn’t want to have to deal with it and he was tired of it. I do know his mother is a teacher, but that’s all I know. I’m not exactly sure how the grading works (for the person who questioned getting so many wrong), but I believe it’s similar to how AP classes are. They actually get a higher grade than they earn because of the difficulty of the classes. College courses are the same for the high school kids.</p>
<p>Quant, the paper I’ve referenced purports that the wrong answers are enough to calculate the probabilities of cheating. And the narrowing down you ask about is not relevant to the calculation, I think. </p>
<p>I think the calculation is really pretty simple given the info supplied by the OP. To summarize, we know that on the first test, 6% of the students chose the same wrong answer on one question as both of the students in question. So the odds of any two students, chosen at random, of both choosing this same wrong answer to this question is .06 squared. Do this for each of the questions, multiply the probabilities, and the result is 1 in 10 billion for the first test alone. Not inclusive of the caveats I listed, and undoubtedly others. </p>
<p>What to do with the info is another matter, but the university has likely calculated these odds too.</p>
<p>^^ huh.
I am not strong in statistics, to say the least. But I remember this example from memory: the chance of two people in a class of 30 having the same birthday is about 1/3 – quite a bit different than 1/365 squared.</p>
<p>I agree that the results posted thus far are overwhelmingly pointing to cheating – I just am less sure about the actual chances.</p>
<p>None of these calculations prove that the OP’s son was involved in cheating. The other student may have somehow cheated without the OP’s son being aware of it. Apparently the proctor was not aware of any cheating. How is the professor going to prove that the OP’s son cheated? I don’t think that statistics prove that the OP’s son cheated, although they might prove that at least one of the students cheated.</p>
<p>Apples and oranges, Eric. Two people having the same birthday, and two randomly chosen, are different things. The 1 in 3 odds are the former. The valid analogy here is the latter.</p>
<p>
</p>
<p>Correct me if I’m wrong, but don’t you get billed for a received text only when you accept the text? If that’s the case, he WAS using his phone during class time. Can you figure out who the text was from?</p>
<p>
</p>
<p>Correct me if I’m wrong, but don’t you get billed for a received text only when you accept the text? If that’s the case, he WAS using his phone during class time. Can you figure out who the text was from?</p>
<p>You’re right, nem, none of these calculations prove that the OP’s son cheated, but do suggest it’s highly likely somebody did. But does the prof have to prove that he cheated? I don’t know this, who has the burden of proof. If it is with the student, that seems a very daunting prospect.</p>
<p>This test rings a bell: [Chi-square</a> test - Wikipedia, the free encyclopedia](<a href=“http://en.wikipedia.org/wiki/Chi-square_test]Chi-square”>Chi-squared test - Wikipedia)</p>
<p>smom, what does your s wants to do himself? he is a junior after all and seemingly intelligent as well. Is he ready to take the WF and let it go at that? Is he fighting for his reputation, or does he let you do all the work? has he contacted the other student and asked him what’s up? This is not just your fight, it actually is his. What I mean is, if he wants to fight, them help him, don’t do it for him.</p>