plagiarism detection?

[jqj]

What practices to other OJS journals and journal editors follow to detect and avoid plagiarism in submissions?

Some commercial journals are now routinely screening all submitted manuscripts with online software, and after only a few months of using such a tool one journal editor I know has found quite a few examples of plagiarism, mostly fairly benign (scattered passages of self-plagiarsm), but a few have been quite severe. Building plagiarism detection into a journal workflow seems like it would be a great idea if feasible.

There is an OJS plugin for iThenticate that accomplishes this, but requires a separate iThenticate contract. I don't have pricing, but conventional wisdom is that iThenticate is expensive, possibly too much so for a low-budget OA journal such as ones I might support as publisher.

What are other OJS journals doing about plagiarism? Is anyone actually using iThenticate (and if so, would you be willing to compare notes on pricing)? Are you using some other approach to systematically check journal submissions?

JQ Johnson
Director, Scholarly Communications & Instructional Support
University of Oregon Libraries
1299 University of Oregon
Eugene, OR 97403-1299
email: jqj@uoregon.edu
http://journals.oregondigital.org

[cbdilger]

Isn't peer review supposed to deal with this?

Plagiarism detection software is quite problematic. It generates hurtful false positives and depends on a business model where its own access to intellectual property is unfettered (unlike those it is pointed toward). The notion of "self-plagiarism" is troubling, at the least, if not completely absurd. And standards for citation differ markedly between cultures--heck, even between disciplines. In sum, I don't see a one size fits all solution here.

I'd love to hear more about the commercial journals using PDS. Which journals? Why? Who is paying whom? What improvements in scholarship have been realized? At what cost?

[mcrider]

Hi all,

I wrote the iThenticate plugin for a client, and though I won't divulge their information without their permission, if you like I could ask them to comment on the subject.

iThenticate does have a flat annual fee plus a usage-based cost system. I'm not sure of how much it costs, but its probably prohibitive for a journal without a fair amount of income... I won't say anything for sure though, as you have to contact them to get a quote, and they may very well give price breaks to low volume/open access journals (you are, after all, strengthening their corpus of papers).

While I'm also cautious about using these types of services (when I was in university, the students nearly rioted when teachers started using turnitin), for a journal that receives a high volume of submissions, I can totally understand the reason behind it... Publishing a plagiarized article could be disastrous, and you can't expect your editors to catch everything.

Also, one more comment -- iThenticate at least doesn't have a yes/no response on whether a paper is plagiarized; It gives you a chance in percentage that it could be plagiarized. If its high enough, then the editor would come in and confirm themselves whether a paper could be a copy.

Cheers,
Matt