Students Cheating

[jaw19883]

Hey guys,



As educational technology increases, students are using it more and more to complete assignments. One thing that I am running into is that I will post an assignment/project in MS Word. What the cheaters will do is get a student who completed the assignment to give them an electronic copy of the assignment that they completed so they can hand it in as their own.

I can catch them sometimes by the 'last modified' option, but that is not reliable. I also see some documents that have an editing time of 1 or 2 minutes for an assignment that might take 40 minutes.

What can I do to have more concrete evidence? Is there a particular way to set up the word file before I assign it? what are the options?

Thanks....

[macropod]

The 'editing time' is quite unreliable, even for a student who doesn't cheat. A simple 'Save As' resets it to 0. Conversely, all a cheater need do to avoid detection via that is to not use Save As when saving the copied document.

[fumei]

First of all, there is no definitive way to make anything to do with a Word document truly, fully, secure. Someone who really knows Word can find a way to get past just about anything.

That being said, yes checking the editing time is a good clue. However, "editing time" ONLY means the accumulated time in a single session the document is open. If you open a Word document and do nothing, just wait an hour, the "editing time" is increased by one hour. So if someone knew that, they could cheat your check of time by simply leaving the document open for an hour. And, as macropod pointed out, the new person would likely do a SaveAs, which resets it.

You could grab some data on who and when the document is opened and store them in document variables. Again, if someone really knows Word they could fudge this. However, document variables are NOT visible to normal Word use, unless you look for them. You could put the following in the Document_Open event (in the ThisDocument module).

Private Sub Document_Open()
Dim strUser As String
Dim aVar
Dim num As Long
strUser = Environ("username")
For Each aVar In ActiveDocument.Variables
   If aVar.Name = strUser Then num = aVar.Index
Next aVar
If num = 0 Then
   ActiveDocument.Variables.Add Name:=strUser, Value:=Now
Else
   ActiveDocument.Variables(num).Value = Now
End If
End Sub

This gets the user logon name - environ (username) - and gives that to a variable. then it checks to see if there is an existing document variable of that name. If there is, it changes its value to Now (current date and time). If there is not - i.e. this is a NEW user opening the document - a document variable is added: Name = username, Value = Now.

To read who and when opened the document:

Sub GetAllVariables()
Dim aVar
Dim strListVars As String
For Each aVar In ActiveDocument.Variables
  strListVars = strListVars & _
     aVar.Name & vbTab & aVar.Value & vbCrLf
Next
MsgBox strListVars
End Sub

This displays a messagebox with the name and date the document was opened.

AGAIN, though, it is a very very simple thing to open up the VBA editor, see the above code and...delete it. Boom. then delete any existing variables. Boom. Gone.

Can you "sort of" catch cheaters? Maybe. Can you catch a cheater who knows Word? Probably not.

[macropod]

Originally Posted by fumei

AGAIN, though, it is a very very simple thing to open up the VBA editor, see the above code and...delete it. Boom. then delete any existing variables. Boom. Gone.

Indeed, the whole scheme will be defeated if the user has their macro security set to 'high' or doesn't allow the macro to run. It also relies on the cheater using the accomplice's document as their own. Students will soon wise up to the security when they find they're being prompted to save changes to the document after having done nothing more than opening it. Although you could add code to save the document immediately the variables are updated, the change in the file's date/time stamp would be its own give-away.

Gerry's approach will also be defeated if the cheater simply opens a copy of the accomplice's document to copy content from, pastes it into their own document, then deletes the accomplice's copy after closing (which is quite likely).

[Charles Kenyon]

This goes back to the question of how you can keep someone from copying something in Word. The answer, unfortunately, is to not let them see it. One other thing you could do would be to distribute the documents individually to each student, with the document variable different for each student. For this to be effective, it can only be revealed once.

Another method might be to run document compare on the submissions. Both of these are a lot of work, though.

[fumei]

Gerry's approach will also be defeated if the cheater simply opens a copy of the accomplice's document to copy content from, pastes it into their own document, then deletes the accomplice's copy after closing (which is quite likely).

LOL. Indeed, that IS most likely. That is the way it goes, and shows just how easy it is to defeat any approach whereby one person CAN get hold of the document. If someone acts as an accomplice...shrug, there is not much you can do about it.

I know someone who is an instructor and they figure the cheating level, or plagiarism level is close to 80%. A sad day really.

On the other hand, Word is a word-processor not an dedicated enterprise level document management system. Asking it to do this is, and will likely remain, futile.

[macropod]

At Uni, they use a product called TurnItIn (http://turnitin.com/) to check for plagiarism. Ironically, on assignment I tried it on returned an 80% or thereabouts likelihood of plagiarism because: (a) I was analysing a well-known ancient text; and (b) I cited my references. Every reference to the text and other discussions of it was reported as a 'hit'. Not much good at differentiating between legitimate referencing and plagiarism, it seems. Still, it might be useful for detecting cheating where a paper has been copied and both share a lot of phrases in common. I doubt it would be much good if one simply took someone else's ideas and expressed them differently.

[fumei]

Well a lot of secondary research is, in fact, expressing someone else's ideas. It takes a lot to come up with truly original ideas. It does seem a bit unfair that references and discussion of those count as "hits".

On the other hand if two documents share more than a few phrases, it does seem to be dancing into plagiarism. Maybe. Some phrases are of course attached to a topic and difficult to avoid. I suppose that is where a teacher needs to be attentive and reasonable. It is the sloppy hand-ins that are virtually identical to others that need to be caught. Although, it sometimes seems that even getting caught makes little difference.

[macropod]

I was just highlighting a limitation of such software. One can't automatically conclude that the material has been plagiarised just because the software identifies a high hit rate. One also needs to look at what those hits are and decide whether they fall into the plagiarism or legitimate referencing categories; I doubt software could do that with an acceptable degree of reliability.

[fumei]

That is why I wrote:

that is where a teacher needs to be attentive and reasonable.

No software is going to be faultless.