40

Today I got accused by my professor of cheating and copying two reports I submitted. I had a formal meeting with the professor where he tried to push me into admitting I cheated by showing me the surprising similarities. Even though I continuously denied it, he went forward to go to a formal hearing in the university where a panel will judge the outcome. I am currently freaking out. I know I did not cheat from anybody and I feel like the similarity could be due to a similar thought process (which sounds ridiculous). Should I just say that I did not cheat and stick with it? Or should I ask if I can redo the reports (though it sounds suspicious)? The reports were 68 percent similar and I feel like the odds are against me even though I did not do anything.

Edit: the software used was turnitin and I do not hav access to the original files to get time stamps. I literally can only deny the allegations. The only reasonable explanation is that the professor taught us in a similar manner and our report writing manner was similar due to the fact we have taken the same courses a year apart and had the same professors who molded our report writing skills.

  • 55
    You're best off being honest, and sticking to your story if it's true. If you start trying to make amends then that just makes you look guilty. – Nico Burns Nov 1 '19 at 13:31
  • 2
    Did you discussed with anyone your assignments? Did anyone tutor you? – Nick S Nov 1 '19 at 14:31
  • 8
    "The reports were 68% similar." Is this an arbitrary calculation, or was 68% of your report word-for-word identical with the previous year's report? Is this true of both the reports you submitted. – DJClayworth Nov 1 '19 at 15:17
  • 9
    Are the reports so similar because they are from identical experiments (during a prior term)? There are only so many ways to record results, and if the labs were both performed correctly, it seems entirely plausible you got close-to-identical results to at least one person from a prior term. Even more, has the professor instructed you to write up your lab reports in a specific way? If so, that may add to your evidence that the similarity is coincidental. – BradC Nov 1 '19 at 20:07
  • 2
    Is the report free-form or you were filling a lab report template? Did you include any mandatory (boilerplate) content like the description of the experiment or safety directions? – Džuris Nov 1 '19 at 21:13

10 Answers 10

59

In this kind of hearing, evidence will be paramount. "I know I did not cheat" isn't sufficient - that's just a denial. You need more, e.g.:

  • Can you show that you could not possibly have known about these 2 reports?
  • Do you have drafts of the report, preferably with time stamps, as you were writing it?
  • Did you work with anyone else? If so, can they vouch for you?

Offering to redo the reports won't work unfortunately - at this point, it's about academic integrity, not about the grade you get. It's likely advisable to get advice from someone familiar with your department and the academic integrity process, e.g. a faculty member you trust, or student counselors if available.

You might be interested in related questions such as this and this.

  • 19
    Yes, evidence is the key here. Drafts, notes, proof you accessed library resources on certain dates, the change log on any files, etc etc. Thinking about the last thing I wrote, I could produce an email where I laid out the idea, the library record where I accessed a digital version of the main source, my notes on the main source with time stamps, the record showing me checking out that physical book, a WhatsApp conversation where I talk about what I'm working on, and the file of the writing in various stages (thanks to the auto backups taken along the way). Gather that level of proof. – GrotesqueSI Nov 1 '19 at 7:59
  • 16
    Ah, but evidence might be impossible to obtain. Good if you have it, but who is so paranoid that they collect it for assignments. – Buffy Nov 1 '19 at 13:36
  • 46
    Isn't "innocent until proven guilty" the norm? I surprised that the accused needs more evidence than the accuser. – vsz Nov 1 '19 at 14:14
  • Let us continue the discussion about theories of law in chat. – cag51 Nov 3 '19 at 19:25
25

Firstly, use version control. It's quite likely that you've produced your report in either MS Word, Libre Office or Google Docs. All of these programs store old revisions, although Google Docs stores all of the revisions while Word and Libre Office only a limited number. If you're able to produce evidence of partial work, it'll be clear you've produced the report yourself instead of copying it from previous work.

See for example https://www.dummies.com/software/microsoft-office/word/view-older-version-word-2016-document/ for accessing old versions of a Word document. Warning: only a limited number of revisions are stored, so don't save the document again as it will overwrite the oldest versions! If you exported your work as a PDF multiple times, look in the Trash Bin - there may be additional files there.

If you used LaTeX, you might have to use a digital forensics program, such as Autopsy, to recover deleted PDFs and scrub through a sea of deleted documents it recovers to find a previous version of this document.

Secondly, remember that you're innocent until proven guilty. It's the professor who needs to convince the panel that you cheated and not the opposite. Do not accept the professors framing of the situation.

  • 9
    Most of your suggestions don't work. It is now too late. And innocent until proven guilty is only in, for example, the US court system. But, no, don't accept the framing if it is incorrect. – Buffy Nov 1 '19 at 13:35
  • 10
    If you use LaTeX, then you're actually much better off because it's better suited for using a proper dedicated VCS like Git, which is guaranteed to store the whole history. (Of course this doesn't help after the fact, it's a decision that needs to be done up front. Everybody should use VCS, unfortunately it's not commonly taught except in CS.) – leftaroundabout Nov 1 '19 at 14:27
  • 1
    @leftaroundabout: Not only intentional version control software, but implicit version control via cloud storage, can work well with any of these formats. Sure, text-based markup like LaTeX makes diffs much more readable, but for the purpose at hand just being able to retrieve prior versions is sufficient. – Ben Voigt Nov 1 '19 at 21:22
  • 3
    @leftaroundabout Just saying (as someone who wrote all their papers with LaTeX, checked into a git repository) that in this particular case Google Docs would actually be a pretty fool proof way to show that you did the work yourself. Although imaging writing a paper in Word or gdocs makes me shiver. – Voo Nov 2 '19 at 18:01
  • 4
    @einpoklum-reinstateMonica that's getting off-topic here. I actually put a (single!) line break after every or every second sentence in my LaTeX files, one of the reasons being precisely that it makes the diffs nicer. – leftaroundabout Nov 2 '19 at 18:48
16

I've read the above answers, and they seem to give pretty good advice, but I have some more for you that isn't about procedures of college, but on personal etiquette and experience when being questioned by those who have authority.

In the military we say, "He who yells first has lost the argument."

When you go into this, have your facts ready, and find some way to keep your calm. If you are freaking out as you say, this can easily be misconstrued as acting in a guilty manner. When you are in front of them, treat it like a soldier would at a military board. Listen to their questions carefully, do not interrupt them, answer their questions truthfully, support your statements if possible, and correct any inconsistencies in their facts as you find them.

In short you are having a logical argument with them, and if you can find a weakness in their argument you can use that to unravel their argument or cast doubt on it's validity.

Most importantly though, you cannot do this if you are freaking out.

Also, remember, if one of them yells first or gets upset, you've won the argument. You kept your cool, and now have a means to attack their argument, or show that it is not logically sound, but emotionally driven.

  • 3
    Students can freak out even if they are not guilty. They are less experienced than the plagiarism hunters – Captain Emacs Nov 2 '19 at 0:35
  • 12
    Imagine spending 10s of thousands of dollars, investing thousands of hours, to be told that a computer says you cheated and you are under investigation by a huge organisation. It is completely normal to be freaked out. You can't win with logic and composure because this is not an argument between peers. The people who accuse you are the ones you have to convince are innocent. If they lose their cool and shout, then they are sufficiently emotionally invested that you have already lost. Have you won a logical argument with someone who is yelling at you and whom also gets to decide who wins? – user115829 Nov 2 '19 at 1:42
  • @jgn Yes (not yelling, but reacting badly). – Captain Emacs Nov 2 '19 at 12:58
  • 1
    @CaptainEmacs Glad to hear that I guess, I wouldn't rely on it though! – user115829 Nov 2 '19 at 13:19
  • 1
    If a private goes into a review with an NCO who claims he did wrong and his commander, and the NCO starts to try to yell in order to make the private admit he did wrong, often times the commander who did not witness the scene will pick up on things the NCO let slip. If the private continues to keep his bearing, this can only reflect well for him. In an academic situation, you could use this fact that the other party was yelling and badgering you to press that he is trying to bully you, to either the others present, or those above them after, like the dean. – Chthonic One Nov 3 '19 at 15:50
13

Find an advocate to support you. Some universities have an office (ombudsman) for this. In other places a student organization provides help. You may even need a lawyer if you are able to afford one. Another professor who trusts you might be able to help.

Papers were similar. It happens. I assume that whoever did the previous work had the same teacher, with the same lectures, and the same written materials (books, etc). Students were actually encouraged to think in a certain way.

Instructors should make sufficient changes from year to year that this sort of situation is impossible. Lazy professors lead to spurious charges.

If you have evidence, of course, present it. Insist on your innocence. Complain to the department head or the dean.

If you have to ask for the chance to do it over, ask that someone else set the assignment and evaluate it. Ask the department head.

  • 3
    @ScottSeidman, you are being quite speculative here. We haven't seen the assignments and have no knowledge of the scope. – Buffy Nov 1 '19 at 14:04
  • 5
    @ScottSeidman, you haven't seen the assignment or the solutions. So 68% is just noise here. Nor do we know who made the assessment of that number. – Buffy Nov 1 '19 at 14:09
  • 5
    -1 for demands to get a lawyer and that all instructors must not re-use assignments or be charged with laziness, etc. Way off base. – Daniel R. Collins Nov 1 '19 at 15:29
  • 3
    I agree with @DanielR.Collins but finding a representative is the single most important piece of advice in this thread and the only thing that will work. – user115829 Nov 1 '19 at 16:51
  • 5
    "Instructors should make sufficient changes from year to year that this sort of situation is impossible." That's not an appropriate note in this case. It's completely normal to have the just a single kit for the Compton scattering, Frank-Hertz experiment or measuring Brewster's angle. Students will do the same experiment with the same guidelines and filling the same form year from year. The only error here is to put the lab report in a plagiarism checking tool. – Džuris Nov 1 '19 at 21:11
13

You don't have the option of resubmitting the reports -- and haven't since the day you handed in the reports. If a hearing process has been started, it will be finished.

68% similarity is a high number, and unlikely to happen by pure coincidence. I suggest you examine your process of writing your lab reports with a fine tooth comb, and be ready to convey this process to the hearing board.

In preparation for the hearing, you should ask for the document which you are accused of copying in order for you to prepare. As I said, 68% is a high number, but there could be explanations. For example, if both papers cite the same sources, then the citations in the bibliography could boost the similarity percentage, and the prof did not examine the similarities carefully enough to see this issue.

Another possibility is that you didn't copy the older lab report, but that both you and the student who wrote the old report used the same sources for their preparation, and there was insufficient citation of the sources you used. In this case, both you and the student you're accused of copying from plagiarized from the same source, and the hearing board will likely still find you responsible for a violation of an honesty policy.

If this is the case, and you insufficiently cited the same sources that the other student used, I'd encourage you to have the information in-pocket for your hearing. There is subtle difference between finding an old lab report and copying it, or boosting text from a wikipedia listing without citing it, because you don't understand how to do it correctly. Both are violations, but convey different intention and might differ in terms of severity.

You may also be faced with whether the two reports are treated as TWO violations or pooled together as one violation. If the case is improper citation, and not copying from an old report, you might ask for some mitigation of penalty for the second offense, as it would have been dandy if the prof could have identified your issue before you repeated it.

Go into the hearing realizing that the committee isn't out to get you, but to correct you and make sure you learn the right way of doing things.

Good Luck with your hearing.

  • "In preparation for the hearing, you should ask for the document which you are accused of copying in order for you to prepare." That is unlikely to be shared with the OP. At my university we cannot and will not share another student's work. At most, the OP may be shown excerpts of the previous text. Maybe. – GrotesqueSI Nov 1 '19 at 14:58
  • 8
    @GrotesqueSI your student's union needs to be doing a better job then, IMO. Trying to punish a student on the basis on secret evidence would have caused a fit, where I went to school. – mbrig Nov 1 '19 at 15:50
  • Let us continue this discussion about theory of law and the GDRP in chat. – cag51 Nov 4 '19 at 15:47
7

I am a professor. Even if you don't have any "hard" evidence that you wrote it yourself, one thing that would be convincing to me, if I was on this committee, was that you knew the material you handed in. E.g. you could take an oral exam on the material. So, make sure you really know what you handed in and why you wrote it the way you did.

And as written on another answer, try to stay calm. Have some faith that the committee will find the truth.

5

In addition to other excellent answers, I suggest you use make the following argument:

Rare events are likely in large populations

... to explain why it is reasonable for your version of events to have occurred.

The panel would be thinking: "It's extremely unlikely for independently-written reports to have such a high degree of similarity - @Ballislyf must be lying to us."

And you tell them: Yes, it's extremely likely if you're looking at just two report-writers. But in both this and the previous semester we had at least n students taking this lab; so actually, we have n * n pairs of students. And whenever some pairs of students all have non-similar reports, that only increases the probability of the remaining pairs to produce similar reports. Thus the probability of finding at least one pair of students with similar reports is in fact much much higher.

If you know a bit of math - and you probably do - you can work out an upper bound of this probability and present it.

Something to consider: The analogy of winning the lottery, or being struck by lightning. If a student misses a deadline saying they were struck by lightning and their laptop died, you would probably think they're just lying. But - some people do get struck by lightning; and some of them are students; and they often have homework on their computers. So - this must happen sometimes.

There are few problems with this argument:

  • Students aren't randomly-sampled from anywhere, nor are their reports uniform samples from the space of possible reports.
  • You don't know how slim the "faux-probability" of similar reports is.
  • To have two separate lab reports be similar between the same students is less probable than just one, and seems more damning. That might need extra explanation.
  • 1
    And, again, see "birthday paradox": the probability of two things being the same in a population tends to only need square-root-of-N population, rather than N or N/2, where there are N possibilities. – paul garrett Nov 2 '19 at 18:45
  • @paulgarrett: Well, it's not exactly like the birthday paradox since you take one person from each group. I could have said "two-sided birthday paradox" but that seems a bit too much... – einpoklum Nov 2 '19 at 19:28
  • 1
    As a professor, I don't think it should be the student's place to make these kinds of arguments. It is the job of the committee to understand the strengths and weaknesses of turnitin and to evaluate the reports and the documents which are supposedly 68% similar. – jerlich Nov 3 '19 at 5:05
  • 1
    @jerlich: While the committee should realize this, it doesn't mean that it will. – einpoklum Nov 3 '19 at 8:44
  • 2
    And combine that with the fact that in a few areas there are only a few ways in which something can be correctly expressed. If you ask 1000 mathematicians to independently define the derivative, you will get approximately one answer. (excluding "I forget", perhaps) – Buffy Nov 3 '19 at 13:33
4

From turnitin.psu.edu:

False Positives

A false positive, that is a paper with a high similarity score that is not the result of a student committing plagiarism, can occur for the following reasons.

If a rough draft is stored in the Turnitin repository, you may get a false positive for a final draft.

If a student submits a paper to Turnitin.com independently of the course to check for plagiarism, the version uploaded for the course will be a “copy.”

It is possible your student is being plagiarized. The Originality Report identifies matches, but does not necessarily identify which is the “copy” and which is the “original.”

A student could be expanding previous research for a new assignment, but copying text from the original research.

Does any of this apply to you? If so, you can use this as your defense during the hearing.

  • 2
    This is actually a denial by Turn-it-in of having any actual false positives... – einpoklum Nov 3 '19 at 8:45
  • @einpoklum-reinstateMonica: ... which they like any system (or human) will inevitably have in reality. (And seeing that they basically deny this doesn't improve my trust in that claim) – cbeleites supports Monica Nov 3 '19 at 21:03
  • The second item on the list (rough drafts) shouldn't occur if the system is set up properly - but does. The third is a bigger issue as the student submitting independently shows up as a different user. The final is often accounted for in the submission instructions, and there are workarounds for that (I think it may involve not checking against work submitted by the same student, but I don't configure the system) – Chris H Nov 4 '19 at 10:13
1

Here's a little more detail on Turnitin's process. It's a little too much for a comment, and offers some hints as to what's going on.

Turnitin (produces a "similarity score" using a black-box algorithm, but the marker can see duplicated passages, and in many cases their sources*. In the reports I marks, a similarity score of around 20% is common for reports on projects that run every year - titles, references, standard terminology, etc. Over about 35% generally seems to indicate some use of choice phrases from the literature, often combined with a limited vocabulary that ends up repeatedly using the same wording. To reach 68% would require many sentences in which only a few words are changed from the sources.

If the score is 68% I'd expect the marker to do some digging into what the "copied" passages match. If you match an earlier interim report of your own - no problem. Here we have guidance and systems to take care of that. But that should be commonplace, so I doubt it's what's going on here. The software is only a tool, and one that has to be used carefully; so far I see no evidence from what they've said that the necessary checks have been carried out, but that's something that should come out in the investigation.


*Published sources and sources within the same institution are identifiable; coursework from other institutions is only identified by institution name (and perhaps year).

0

The other answers seem to have summed up the main part. I'd only add that, depending on where you are, threatening a lawsuit might actually be a good course of action as well. Where I'm from (in Eastern Europe) many professors would probably drop the whole thing if you get them to believe you will actually sue the university as it's not worth the trouble and risk to them. In the US trials seem to be a somewhat trivial matter but where I'm from the professor could get fired if you do in fact win the trial. Going through the trouble of getting a lawyer involved also seems to make people think you actually believe that you're right.

Do keep in mind it could backfire as if he does decide to pursue he will do his absolute best to prove you cheated and you will have lost time that could have been spent preparing evidence in your favor. Think of it as a last resort. In case you go that route, even if bluffing, make sure you document as much as possible just in case you do decide to sue.

  • 1
    this is poor advice. he hasn't been found guilty of anything yet. – jerlich Nov 3 '19 at 4:55
  • @jerlich It's not really about the trial itself but the threat/pressure of one. Pretty much a last resort bluff. The part about documentation is in case the threat makes the professor emotionally invested enough to do something unprofessional enough to provide legal leverage. – Lanolderen Nov 3 '19 at 11:33
  • You are implying that a lawsuit would be a good action even if they were actually guilty. – DJClayworth Nov 4 '19 at 14:09

Not the answer you're looking for? Browse other questions tagged or ask your own question.