Probing for Plagiarism in the Virtual Classroom

• 04/29/03

By Lindsey S. Hamlin and William T. Ryan

[Editor's note: Ryan and Hamlin will participate in a panel session, "Academic Integrity in the Virtual Classroom" at Syllabus2003 on July 30. The original article posted May 1, 2003 has been expanded by conference proceedings submitted by the presenters. The udpated article is posted here.]

Virtual learning in higher education has seen enormous development in both public and private universities. In 2000, about 47 percent of U.S. colleges offered some form of distance learning. This figure will increase to almost 90 percent by the end of 2004 (Flisis, 2001).

Educators who are making the transition into online teaching are skeptical about the preservation of academic integrity in the virtual classroom. They often assume that Internet technology and online classrooms are providing students with additional opportunities to cheat. In reality, the probability that a student will cheat in an online course is about equal to the chances that a student will cheat in a traditional course (Carnevale, 1999). In fact, with the Web sites and software now available, educators have the ability to detect and battle plagiarism and cheating in virtual classrooms. Also, the various types of online assessment tools, assignments, and activities available within a virtual course (i.e. threaded discussions, virtual chats, quizzes, group presentations, etc.) are, by their very nature, a deterrent for cheating.

Virtual vs. Traditional Cheating
Unfortunately, cheating has always existed and will continue as long as there is temptation to do so. In 2002, 47 students at Simon Fraser University turned in nearly the same economics paper (Black, 2002). According to a 1999 study conducted by the Center for Academic Integrity at Duke University, of the 2,100 students surveyed on 21 campuses across the country, "more than two-thirds of the students admitted to one or more instances of serious cheating, such as copying from another student on a test, plagiarizing or submitting work done by another student" (Muha, 2000). Although these statistics show that cheating remains a serious academic problem, it is unclear as to whether the Internet has really changed the percentage of students who cheat.

Online Exams
While giving an exam in the traditional classroom, educators look for roaming eyes and cheat sheets to identify cheaters. Yet, in an online classroom instructors do not have the benefit of visually monitoring students during an exam. To compensate for this problem, instructors can place time restrictions on exams or require that exams be proctored by a college testing center or library. While creating an online exam, instructors have the option of restricting the amount of time the student has to complete the exam. When the time limit has expired, the exam is automatically submitted to the instructor. This technique is most successful at deterring cheaters when it is used in short, multiple-choice quizzes. For example, if an exam consists of 10 multiple choice questions and a student has nine minutes to complete the exam, the chances of the student looking up each answer is slim because of the time restriction. Instructors that require exams to be proctored are gaining the same benefit of the visually monitored exam in a traditional classroom. However, many testing centers charge students a fee of up to $15 per hour for use of their facilities. In addition, if the class perceives proctoring as an indication of the instructor's mistrust in his/her students, it can markedly damage the student/teacher relationship. Some educators have criticized proctoring as a "violation of the spirit of the honor code" (Young, 2001).

Discussion Boards
Virtual skeptics have criticized online education for worsening a student's sense of "isolation and anonymity" (Carnevale, 1999). From an outside perspective this may seem true, however, virtual learning offers ways of communicating with students and assessing their knowledge that extends beyond the traditional classroom. Threaded discussions allow instructors to post discussion-type questions to which students can respond. These discussions are not asynchronous, which allows students to at their leisure within a given time period. Threaded discussions encourage students to communicate, discuss, and debate topics with each other. They also provide instructors with countless examples of a student's writing style, which can be very useful in determining if a student has plagiarized a paper. The benefit to these discussions, as opposed to live, in-class discussions, is that students have the time and opportunity to research and thoroughly structure their responses. Many times students will even provide Web links within the discussion for other students to view. Because of this continuous research that takes place during the threaded discussions, the quality of responses tends to be much higher than that of traditional, classroom-based discussions. Along with threaded discussions, instructors can assess students through virtual chat or synchronous discussions. By requiring groups of students to meet at a predetermined date and time in the online chat room and discuss a specified topic, students begin to form working relationships with their classmates and instructor. The time commitment required to participate in virtual chats and threaded discussions is also a deterrent to online cheating.

Online Plagiarism
With the increasing number of online term-paper mills, such as Schoolsucks.com and Cheater.com, students have an even greater temptation to plagiarize. Instead of copying text out of books or journals by hand, students can now find an array of term papers online and can copy and paste blocks of text right into their word processors (Heberling, 2002). Deceitful students may also copy papers from Web sites of conference proceedings or well-intentioned academics. As of March 2003, the Kimbel Library at Coastal Carolina University had identified 250 active Internet term-paper and essay Web sites (Fain, 2003). A national survey conducted by Donald McCabe, a professor of management at Rutgers University, found that 54 percent of students admitted to plagiarizing from the Internet; 74 percent of students admitted that at least once during the past school year they had engaged in "serious" cheating; and 47 percent of students believe their teachers sometimes choose to ignore students who are cheating (Stricherz, 2001).

In the May-June 2002 issue of the Journal of College Student Development, Patrick M. Scanlon and David R. Neumann of the Rochester Institute of Technology reported their research findings on Internet plagiarism. Surprisingly, their research indicates that the proliferation of Internet plagiarism may not be as extensive as many may assume. The professors polled 698 undergraduate students at nine institutions of higher learning. Some 16.5 percent of the respondents reported plagiarizing "sometimes," while 50.4 percent claimed that their peers "often" or "very frequently" committed plagiarism. Slightly more respondents said they plagiarized conventional text more than online documents and almost 100 percent agreed that their peers plagiarized conventional text. Scanlon and Neumann concluded that more conventional plagiarism is occurring, while the growth of online plagiarism may not be significantly contributing to the growth of plagiarism in general (Kellogg, 2002). These statistics show that online access to papers has increased plagiarism in both the traditional and online classroom. However, because papers are submitted electronically in the virtual classroom, it is easier for online instructors to detect plagiarism by running student-submitted papers through plagiarism-detecting Web sites or software programs (Heberling, 2002).

Plagiarism-Detecting Web Sites
Plagiarism.org maintains a database of thousands of digitally fingerprinted documents including papers obtained from term-paper mills. According to Plagiarism.org, when an instructor uploads a student's paper to the site, the document's "fingerprint" is cross-referenced against the local database containing hundreds of thousands of papers. At the same time, automated Web crawlers are released to scour the rest of the Internet for possible matches. The instructor receives a custom, color-coded "originality report," complete with source links, for each paper. For a fee, this service will detect papers that are entirely plagiarized, papers that include plagiarism from different sources, or papers that have bits and pieces of plagiarized text (www.plagiarism.org). However, educators must remember that even though plagiarism-detecting software can identify plagiarized text, it may not highlight the quotation marks surrounding the text or the reference to the text within the paper. An overzealous professor could hastily accuse a student of plagiarism by running their paper through plagiarism-detecting software and then fail to revisit the paper to verify whether the identified text was referenced.

Internet detection services, both fee-based and non-fee-based, are on the rise. Many educators would find this growth positive, however, a March 2002 article in The Chronicle of Higher Education reported that two plagiarism detection Web sites, PlagiServe.com and EduTie.com, appear to have ties to Web sites that sell term papers to students. Apparently, the companies that were checking student papers for plagiarism were then selling those same papers through its term-paper mills. Although the allegations were denied by both companies, the possible conflict of interest is a reminder to educators to be cautious in submitting student papers to unsubstantiated sites (Young, 2002).

Plagiarism-Detecting Software
Many software companies have developed innovative programs for detecting plagiarism. Glatt Plagiarism Services Inc. produces the Glatt Plagiarism Screening Program, which eliminates every fifth word of the suspected student's paper and replaces the words with a blank space. 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. This program is based on Wilson Taylor's (1953) cloze procedure, which was originally used to test reading comprehension (www.plagiarism.com).

Internet Search Engines
Educators may also find the more popular Internet search engines to be a useful tool in plagiarism detection. Google, Yahoo, Excite, AskJeeves, HotBot, GoTo, AltaVista, and MetaCrawler are just a few of the search engines that can aid an instructor in detection. When an instructor suspects a student of copying text or notices an inconsistency in a student's writing style, he or she can enter the suspect phrase into the search engine. The search engine will return a listing of all websites that contain an exact match of the entered text. Instructors can broaden their results by searching a few different search engines (Heberling, 2002).

Preserving Academic Integrity
Educators are faced with the task of preserving academic integrity. Although it is nearly impossible to eliminate cheating in traditional or virtual classrooms, educators can deter it by using the tools available to them. Instructors who advise their students that writing samples will be collected, term papers will be filtered through plagiarism-detecting software, pop-quizzes will be given throughout the semester, and that weekly participation in the discussion boards is a class requirement are setting up a virtual environment that will deter cheating.