18.421 plagarism checkers

               Humanist Discussion Group, Vol. 18, No. 421.
       Centre for Computing in the Humanities, King's College London
                   www.kcl.ac.uk/humanities/cch/humanist/
                        www.princeton.edu/humanist/
                     Submit to: humanist_at_princeton.edu

   [1] From: "Lisa L. Spangenberg" (51)
                 <lisa_at_digitalmedievalist.com>
         Subject: Re: 18.417 plagarism checkers?

   [2] From: "pjmoran" <noci_at_cox.net> (14)
         Subject: Re: 18.417 plagarism checkers?

   [3] From: Scott Sadowsky <lists_at_spanishtranslator.org> (18)
         Subject: Re: 18.417 plagarism checkers?

--[1]------------------------------------------------------------------
         Date: Mon, 13 Dec 2004 07:10:07 +0000
         From: "Lisa L. Spangenberg" <lisa_at_digitalmedievalist.com>
         Subject: Re: 18.417 plagarism checkers?

> Date: Wed, 08 Dec 2004 08:47:33 +0000
> From: Willard McCarty <willard.mccarty_at_kcl.ac.uk>
> >
>The mushrooming of plagarism has to be a problem for everyone these days,
>given the services of online businesses and the presence of competently
>written prose freely available online. When style takes a sudden turn for
>the better, unacknowledged use is easy enough to detect thanks to Google
>and its kind. It would be useful to hear from those who know about other
>styles of plagarism and other ways of dealing with them with the help of
>computing.

My school has, at some expense, licensed turnitin.com. Instructors submit
student papers, that is the file itself, via the web, through a link on a
class site.

The papers are returned with "plagiarized" text marked up, and a percentage
indication.

I'm not at all a fan for the following reasons:

1. Student's do not have to consent, nor does the school require them to.
2. Papers are generally submitted with the student's name and school ID on
them.
3. Papers are stored, permanently, in the Turnitin database, and used as a
base for comparing other papers, and for other Turnitin purposes, if they
so wish.
4. Students have copyright on their original work; this is ignored both by
my school and Turnitin. It also just seems wrong to me to use students'
work without their permission--even if they are plagiarists.
5. The report flags properly quoted and cited text, for instance, literary
quotations or properly documented source quotations.
6. The report is such, and the presentation of the link to submit is such,
that many faculty will misinterpret and misuse the data; that is, read the
percentage of unoriginal text at the top and stop there (we have faculty
who teach hundreds of students in a class).
7. In my admittedly naive texts (I'm neither a statistician nor an expert
in natural language parsing) all Turnitin does is match text strings, and
rather long ones (I'd guess eight or nine words) against it's database of
text, and the Internet. It's pretty easy to change enough text in minor
ways to get by.
8. I have better luck with spotting suspicious text, using Google, Jstor,
Project Muse, my own memory, and consulting with other T.A.s, and
particulalry, with faculty who have been teaching the class in question for
years, than with Turnitin.com.
9. As a graduate student, it seems to me, perhaps naively, that we should
be able to spot plagiarism on our own, rather than engaging in mass
submission of papers to a service (and there are legal ramifications of
only submitting in one or two students' work from a class)? Isn't
source-spotting part of what we should be able to do?

Lisa

--
Lisa L. Spangenberg             | Digital medievalist
Celtic Studies Resources   | http://www.digitalmedievalist.com
My opinions are mine         |not those of the University,
who is of the opinion           |that I should be writing my dissertation.
--[2]------------------------------------------------------------------
         Date: Mon, 13 Dec 2004 07:12:35 +0000
         From: "pjmoran" <noci_at_cox.net>
         Subject: Re: 18.417 plagarism checkers?
Teaching English composition via the Internet, I used to check suspicious
prose by cutting a chunk out of a paragraph in the middle and pasting it
into a  search engines.  I usually used Excite.  In two cases, when I wrote
back to the students and told them their essays were receiving failing
grades because of their plagiarism, they were shocked. The "bodacious"
nature of the current student population means all instructors need to do
(at least) this kind of perfunctory check.  Requiring outlines, rough
drafts, and incorporation of teacher's suggestions into text can cut down
on the number of questionable papers.  I believe the students count on our
being stressed and overworked by virtue of our occupations and on our being
less than competent Internet navigators by virtue of our age.  They count
on our not having time to check every single paper.  Patricia J. Moran,
Graduate Student, FSU, COE, Stone Building, Tallahassee, FL 32306.
850-243-4081.
--[3]------------------------------------------------------------------
         Date: Mon, 13 Dec 2004 07:13:10 +0000
         From: Scott Sadowsky <lists_at_spanishtranslator.org>
         Subject: Re: 18.417 plagarism checkers?
On 12/10/2004 03:58 AM, Humanist Discussion Group wrote the following:
 >It would be useful to hear from those who know about other styles of
 >plagarism and other ways of dealing with them with the help of computing.
Last semester, a student of mine handed in a very unusual paper indeed.  It
felt and sounded extremely unnatural, but didn't show up in Google, and
obviously wasn't hers.  Turns out she took a paper written in another
language, passed it through an automatic translator, cleaned it up
manually, and turned the resulting product in.
Through back-translating and trial-and-error, I tracked down the source and
the program used to translate it, but this was a far less routine job than
the usual 30 seconds of googling required to detect most plagiarism.
Cheers,
Scott
--
Scott Sadowsky
Profesor, Universidad ARCIS
ssadowsky_at_universidadarcis.cl
http://ingles.universidadarcis.cl/