Hi all,
This may or may not be tangential to our main interests on this list, but I'm interested in asking it here because I can assume a certain level of familiarity with the strengths and weaknesses of digital tools and algorithms.
Last semester I started using turnitin (http://www.turnitin.com/). Initially this was as a pure essay management and on-line marking system (for which it is very good, BTW). But the software was designed to test originality against a database of web and submitted sources, of course, and I found myself checking the originality reports on students' essays whenever something sounded fishy (you can read an initial discussion of my experience here: http://people.uleth.ca/~daniel.odonnell/Blog/digital-plagiarism).
What I found was that a small percentage of essays (6 in about 200 submissions) contained sections of material that had close parallels elsewhere and no acknowledgement of sources--in other words, likely were plagiarised to a certain extent. In a few cases, mostly the two senior papers that showed this problem, the question was fairly cut-and-dried: the students involved had clearly copied and pasted large chunks of text from on-line sources, leaving behind tell-tale clues like the use of an asterisk to mark a cross-reference in an encyclopedia entry.
The more disturbing examples were found in my first year classes and involved a very different kind of parallelism. Here the students in question seemed to confine their unacknowledged debts to two main types: the use of quite small sections (a couple of sentences at most) at key points of their argument (e.g. theses, key transitions) and/or the wholesale use of examples from a single source (i.e. all the quotations used were found in the same source, in one case a site that students could use to find "useful quotations" from a given work.
I interviewed some students (both those tagged by turnitin and others) and discovered that they were working in a way that is quite similar to how I would build a blog entry and which almost seems to invite this problem in essays (this is the point of the blog entry I wrote on the question, mentioned above).
My question for the digital literati among medievalists, however, is whether this second kind of "plagiarism" actually is plagiarism, and, if it is, whether it should be treated in the same fashion as the relatively certainly deliberate use of large passages from unacknowledged sources that I'm more used to catching?
On the one hand, I need to stress that this is minority behaviour--by far the largest number of essays I looked at last semester correctly documented their debts (including in several cases among my first years, debts to essays found at sites that sell essays to students!). Also, the examples I found have an extremely distinctive thumbprint in the software--they fade in and out of verbatim quotation in a way that stands out clearly through length and general pattern of endebtedness from examples of passages where people just happen to have said something quite similar to each other. On the other, I'm really not sure that the plagiarism is not the result of ignorance or relatively minor carelessness--a lost quotation mark during the editing phase, etc. Given the way these students seem to work, moreover, I'm surprised it doesn't happen more often: both the students who turnitin flagged and the others I spoke to worked in a way that is almost designed to encourage unrecognised borrowing: they copy large verbatim quotations from internet sources into their wordprocessor, and then reorganise and edit the quotations until they have a 'narrative' of sorts; then they add their own introduction and bits of text between the quoted material.
My question then is whether I should treat it the same way I do blatant, old-fashioned, copy-big-chunks-from-a-journal-article plagiarism: with a 0 on the course. Any ideas?
Any ideas or suggestions?