Strategies for Teaching with Online Tools
Bedford Workshops on Teaching Writing Online
Nick Carbone, New Media Consultant
Bedford/St. Martin's
ncarbone@bedfordstmartins.com
Workshop Home
Tools 

E-Mail

Discussion Boards

File Sharing

Real-Time Discussions
 
 

Topics 

Peer Review

Research

Plagiarism

Workload
 

Composition@Bedford/St. Martins

Start

Using Portfolios to Avoid Plagiarism in Your Classes

The argument in this workshop is that research portfolios --careful record keeping, reflective essays, research logs, copies of sources used, annotated bibliographies, and so on-- are essential skills and practices for researchers to follow and therefore are essential to teaching research. If you require these activities, set incremental deadlines (collecting items as the weeks progress) and help students to keep their portfolio organized, accurate, and useful, the research portfolio then does a number of things for you as a teacher:

  1. It helps you teach students how to manage time and set a series of smaller goals and deadlines. Even small research projects, where you require only a handful of sources, benefit from teaching these time management and project planning skills. Setting deadlines helps students stay on course, and if you make it explicit that you're setting deadlines to model the kind of planning they need to learn to do when they engage research on their own, it helps them to see the value of planning the project that way.

  2.  
  3. It lets you tie the use of the research portfolio to the issue of plagiarism and honesty. Make it clear and explicit that such a portfolio benefits students should there ever be any question about the origins of their work. You can use, as an example, the case of Michael Bellesiles, author of Arming America, who is being accused of making up research that showed colonial Americans did not own as many guns as thought. Other researchers have not been able to find the same sources Bellesiles claims to have consulted, and Bellesiles says his copies of those records were destroyed in a flood. Research portfolios help students show the work that goes into their writing, and that helps them should there ever be a question about their honesty. And of course, by collecting the work incrementally and seeing how students are doing as they go, it helps you to teach source management (avoiding inadvertent, technical plagiarism) and makes it harder for students to cheat (avoiding intentional plagiarism).

  4.  
  5. One of the best ways to start any research project is to ask student to create a knowledge of inventory: a list of everything they know or think about  their given topic, as well as tracing where they learned what they know or how they derived the opinions they have. From this, students can generate two very useful things: some early essays that help them structure their own initial thinking, and as well research leads. There are two benefits to this strategy:

  6.  
    Essay Drafting Benefit: Writing before research helps students get their voice on paper. It gives them a document that they can weave their sources and research results into. This helps avoid the get-a-stack-of-sources-cobble-quote-cite-and-then-patch-a-paper-together-thing that often results in voiceless, bland, unengaged research writing. By writing first, you can help students move back and forth between research and writing. They write to get started; research to explore their thinking, answer questions, solve a problem or learn more. They work what they've learned from research into the evoloving draft, as REsearch helps them REvise their thinking. 

    Research Leads Benefit: If students remember having read an article in the dentist's office six months ago, or recall hearing something on the news, had a textbook a few years back which addressed the issue, or took another course where a teacher discussed the topic, and so on, then they have leads, places to begin their research.  They can look up those articles and call or e-mail those teachers.

  7. An incrementally collected portfolio lets you see how students research, and how their research and writing strategies are unfolding. These views into their process let you intercede and find more teachable moments than can arise when you only get a paper on a due date.

  8.  
  9. In WAC environments, where teachers may not focus as much on commenting incrementally, collecting work in increments still helps students because it gives them deadlines. All a teacher needs to do is skim and read, looking for broader issues that might be addressed to the class as a whole or in consultation with a research librarian This is also a good way to find class-wide issues that might be addressed by consulting your writing center and arranging a workshop on some particular research writing skill. That is, collecting portfolios does not necessarily have to entail a lot of extra work reading and responding.  But still the benefits are immeasurable and merely skimming opens up opportunities for working with your local writing center and library.

  10.  
  11. Seeing work from the start -- brainstorming, searching for topics, finding a thesis, and so on -- makes the chances of plagiarism "vanishingly thin" (see http://bedfordstmartins.com/technotes/hccworkshop/plagiarism.htm  to see where that quote comes from).

  12.  
  13. By collecting work in increments --and regularly checking research logs-- whose entries can be delivered as emailed memos or or entered in a threaded discussion board at required dates-- teachers can sooner find struggling students. If a student's not doing the work, is falling behind, you know that student is in some trouble. Students who get behind and in trouble are often the ones tempted to cheat. The chance to intercede and help them sooner is much more useful  -- and less agonizing -- than trying to catch them at cheating later.

  14.  
  15. Failure to do the research portfolio and/or to meet its incremental deadlines, can be a reason to refuse a student's final paper or to consider the research project incomplete. Thus, if you feel students have plagiarized, but can't find the source materials they may have used, and they don't confess to plagiarizing, despite all the circumstantial evidence you present before them, you can still give them a low, perhaps failing, grade because they don't have a research portfolio. To do this, however, you have to make it clear in your syllabus and research assignment that a complete, and incrementally developed reserach portfolio is part of the project. And it needs to be part of of the assignment's articulation that missing --or grieviously incomplete-- research portoflios will result in no credit for the research project. Whether you do this is up to you, but it's an option that helps enforce the importance of the research portfolio on the one hand, and gives you manuevering room and discretion on the other. Because relying on search engines to find the plagiarized source will often fail (see reviews below), having a policy which explicitely links portfolio completion to assignment completion gives you an alternative approach to handling plagiarism.


What About Non-Reseached Writing?

For papers that don't involve research, all of the above applies. In the workshop given, I tried to make the connection that with online tools especially--email lists, threaded discussion, MOO's and/or chats for real time discussion, and so on--it's easy to raise topics and responses to reading that get written as conversation and then you can design writing assignments or request essays based on those discussions. This lets students cut and paste their conversations --their emails, discussion posts, items from chat transcripts-- into a word processor or html editor for first draft stuff. That is, all writing begins in the class and evolves from the class.

And again, the process of collecting drafts, reading stuff even if you can't comment on it all, setting incremental deadlines and so on, not only stresses valuable writing practices and teaches students how to work in drafts, but coincidentally cuts down on the likelihood of plagiarism.

O.k., I Collected Drafts and Required a Research Portfolio, But I Still Think Some Writing Might be Plagiarized, What About Searching the Internet?

I've tried four different search methods on a test paper.

Here's what I did: I used 'the simpsons' and 'the hobbit' as search terms in Google, and found some writing, both at free paper mills and on regular Web pages, about both. I only used sources from the first two pages of hits, stuff that's easy to find, not at all obscure, and readily available on the Internet to anyone. I then cobbled together a paper on each topic, and plagiarized extensively. I then tried to find the plagiarism, using the resources listed below.

Let me begin with Turnitin.com, which failed miserably as plagiarism search tool, and then describe alternatives that worked better.

Turnitin.com -- Causes More Problems than it Solves
http://www.turnitin.com/
This expensive --and aggresively marketed-- service failed to deliver. But the key thing is, that given their heavy advertising, promotion, an d promises they make to solve the plagiarism problem, along with the appearance of their reports, their failures  to find easily findable plagiarized sources can actually help students get away with plagiarizing.  To see a review (and rejection) of Turnitin.com's self-proclaimed pedagogical logic, go to "Turnitin.com, A Pedagogic Placebo for Plagiarism" at http://bedfordstmartins.com/technotes/techtiparchive/ttip060501.htm. To see how Turnitin.com failed to find obviously plagiarized Simpson and Hobbit papers, go to http://bedfordstmartins.com/technotes/hccworkshop/missedplagiarism.htm.
 

Google --Very Highly Recommended as a First Place to Check
http://www.google.com
Google is a free, and considered by many, the best World Wide Web search engine there is. It's fast, and offers links to cached copies of pages in case the link you try is to a page that's been removed. Google ranks returns by popularity: the pages that are most linked to or chosen are put first. If you start with Google, you may not need to try other options. You'll need to experiment with phrase length; I wouldn't go over four or five words. To do a phrase search, put the words in "quotation marks like this" and Google will search for the exact phrase.

Yes, this wasn't a fair test, quite, since I found the source papers in Google to begin with. But in the test, I used phrase searching, taking longer passages, five to six words, verbatim from my plagiarized papers. Google put the source papers at the top of its results, in the first or second place, every time.

EVE2 -- Highly Recommend; a Great Value for the Price
http://www.canexus.com/
EVE2 succeeded where Turnitin.com failed. EVE2 is downloadable software that anyone can try for free for 15 days. After that, it only costs $19.95, and it can be used an unlimited number of times -- you own it. Unlike Turnitin.com's expensive service, EVE2 comes at a teacher friendly price. A copy or two on shared computers in a given department is a good alternative to a Turnitin.com account. Also, EVE2 offers volume pricing and other service to colleges and departments. Further, because papers are submitted on the local machine, copies of student work aren't archived on the WWW and used by EVE2 to sell their service, thus violating the student's copyright over their own work. EVE2 works by parsing a sampling of the submitted paper into searchable chunks. It doesn't submit the entire paper, and it didn't find all the plagiarized quotes in my test, but did find enough to make a case, and their random sampling method takes enough to perform far better than Turnitin.com. EVE2 is an excellent choice for teachers who read their students work, and will only submit the occasional suspect paper. It's all that most teachers need.  
 

MyDropBox -- Very Effective and Better on Student Copyright
http://mydropbox.com
Instead of putting all student work into an international database, and basing their service's efficacy on that database as Turnitin.com does (which has lead to copyright concerns on some campuses using Titin), mydropbox.com creates institutional databases which the school controls. Big plus: Offers a draft feature so papers can be checked, but aren't permannently stored in campus archive. This allows for tool to be used for teaching more efficiently that Turnitin.com. Paper can be checked, corrected and resubmitted and you won't get false positives from second draft matching up to copy of first draft that was loaded earlier.
 

Glatt Plagiarism Services
http://www.plagiarism.com/INDEX.HTM).
I haven't tried it, but here's the gist of how it works (quoting from their Web site):

Based on Wilson Taylor's (1953) cloze procedure, the Glatt Plagiarism Screening Program eliminates every fifth word of the suspected student's paper and replaces the words with a standard size blank. The student is asked to supply the missing words. The number of correct responses, the amount of time intervening, and various other factors are considered in assessing the final Plagiarism Probability Score. I will say this about Glatt. Theirs is the only site to offer some tips and advice on how to teach about plagiarism and the mechanics of citing. All the others listed above center on catching. According to their Web site, they claim never to have had a false accusation with this method. I haven't had a chance to see what they do with student work vis-a-vis copyright, but I suppose one could always take a student file and eliminate every fifth word on one's own. Then you could meet the student in your office to see how many of the missing words they can fill in. Next, you could compare the test to their draft. There won't be perfecting matching, but if there are a preponderance of terms that are off, you have strong circumstantial evidence that paper is likely plagiarized. If the student also failed to hand in the research portfolio work in increments, or the paper changed topics at the last minute, the out of nowhere nature of it, and lack of research history is also good evidence.

The Limits of Finding the Smoking Gun

Remember any search tool you use only finds text matches, much the way a spell checker doesn't read or think, but only matches words in a document to words in its dictionary. Searches don't prove or disprove plagiarism; and as you saw when considering how a plagiarized paper made it through Turnitin.com, they miss obvious plagiarism. They can't discern fraud and deliberate cheating from learning mistakes in handling sources. They're at best crude tools. That's why, even if you use these tools, a research portfolio for research papers, and drafts and writing portfolios for any other kinds of project, are essential tools for helping you to discern the nature of your students' use of sources.

While different search strategies might help you find a plagiarized source, keep in mind that none of these services or programs can search proprietary databases such as InfoTrac, Lexis/Nexus, ProQuest, and other digital collections or e-books your campus library might offer. Each of those entities would need to be searched separately. Also, these services cannot find papermill content behind passwords or papers written to order. Nor can they find material cribbed from print sources.  


One of the dangers of relying on search services, and one of the sad fallouts from the kind of aggressive marketing of Turnitin.com, is that they send this wrong message: that you need to have direct proof of plagiarism to do anything about it. You do not need to have the original source in hand to bring a plagiarism charge. Before electronic text databases, when the world was print, there was plagiarism. Instructors didn't roam their library stacks for hours, skimming page after page trying to find that obscure passage in a forgotten book. You don't need to do that either.  

In fact, with research and writing portfolios --or more likely the lack thereof-- you have the basis for making convincing circumstantial cases should you need to. But the wonderful thing about requiring portfolios and using them to help teach essential research and writing skills, is that they make it much less likely --"vanishingly thin"!-- that you'll even need to make a plagiarism case. They reduce plagiarism caused by fraud to almost nothing (because to carry off the fraud, the writer would have to incrementally create a complete research and/or writing portfolio), and they make it possible to correct plagiarism caused by learning mistakes, the mishandling of sources, the stuff that comes with learning how to quote, paraphrase, summarize and tell the difference among them all.

by Nick Carbone