[frazwood]

I am a professor at a state university. Last spring semester, I caught two students cheating when they submitted approximately the same MS Excel file for a take-home exam. (note: no, these students were not very good at avoiding detection).

I caught these students by going to the properties and checking the Author in the properties. If the Author was "Bob Smith" and the student submitting the file was "Jane Doe", then I became suspicious. When I compared the file submitted by the actual Bob Smith with that submitted by Jane Doe... And they were identical... Then I accused both students of cheating and I gave them an "F" for the semester.

Fast forward to now... Two of these students are appealing my decision to flunk them... The strongest piece of "evidence" that these students state is that they have never met each other. Based on my people-reading skills, this seems like an honest statement, leading me to believe that other students are involved.

So, I would really like to figure out how many different students have touched this particular MS Excel file. Using the Properties document, I know when the file was first created (and by whom) and when it was most recently save (and by whom).

Is there any way for me to further mine the history of the file to determine how many different times the file has been saved and by whom?

[JBeaucaire]

No, not without access to each of the "versions" and where they are/were stored. There is not a "log" of all users who acted on the file unless the TRACKING feature was explicitly turned on prior to it being passed it around. I am 99% confident that did not happen.

My two cents are that these two people may "honestly" have not met each other, but the files submitted were statistically identical and provably from the same source.

In my college you were guilty of cheating if you:

1) Submitted work that was not your own (cheater)
2) Allowed your work to be turned in by someone else (helper)

I can't see how you need to prove anything other than what you've already done... the documents are statistically identical and provably from the same source. At this point, it is on them.

You don't have to actually "win" this appeal, just present the evidence and fail them. If the decision is overturned then you were still in the right and can move on confidently. At least the person whose name did not appear in the document properties will keep his "F", right?

[frazwood]

My theory is that one of the students gave his file away to another student who gave it to another student... etc... etc... etc... The student with the wrong name in the properties was the only one stupid enough to not cut-and-paste (which would have avoided this problem with the properties) or download the various pieces of software that can be used to change the information in the properties.

The appeal is a hearing in front of an impartial panel. The "standard" for cheating is "more likely than not" not "beyond a reasonable doubt" so I am really confident that one of the students will lose the appeal/hearing.

The other student is going to be more ambiguous... was he actually a "helper" or did someone "steal" his file? He has provided no evidence on his own behalf so far. I was hoping for additional evidence could be mined from the file itself, but apparently not.

Thanks!