Plagiarism detection methods include online or remotely located search tools and services, such as Turnitin; stand-alone desktop software, such as CopyCatch Gold; web search engines, such as Google and AltaVista; and subscription databases of articles, such as those found at university libraries. As of the first decade of the 21st century, Turnitin.com is the most widely used plagiarism detection program at universities.
All methods of plagiarism detection work by systematically determining the similarity of the paper submitted by the student to other, previously created papers. However, each detection program works differently. Turnitin, for example, functions by comparing the digital fingerprints created by eight-word strings in the student's paper and papers that are part of the Turnitin database---which include both previous students' papers and online articles. From this information, Turnitin produces an "Originality Report," which indicates the extent to which the text of the student's paper matches the text from papers in the database. Another electronic detection program, CopyCatch Gold, works by comparing the text of a group of papers---for instance, all the papers submitted by the members of a class.
Although the main purpose is to detect plagiarism, these programs have also been found to successfully prevent plagiarism in many universities. R. Symons says that using such programs discourages students from blatantly copying and pasting from the Internet and from using paper mills. In addition, according to administrators at Adelaide University in Australia, the use of such programs promotes greater originality and problem solving skills.
Some professors and students have misinterpreted the meaning of reports from detection programs. For instance, a 2008 study by Swapna Koshy of the University of Wollongong in Dubai found that many students strove to achieve the lowest percentage possible of matching text on their Turnitin "Originality Reports." According to Koshy, some professors based their grades on "text matching" scores. Low text matching scores can be achieved by merely rephrasing, as opposed to creatively summarizing. Such methods of cheating the system do not promote critical thinking and are thus harmful to students in the long run. In addition, sites such as Turnitin factor correctly cited references into their text matching scores, which means that these scores should not be used to measure merit. Another consideration is that checking each student's plagiarism score creates more work for professors.
If plagiarism websites such as Turnitin.com were to communicate to consumers that their reports should not be interpreted as direct measures of a paper's merit, many of the above considerations could be resolved. In the big picture, these plagiarism detection programs are effective ways of preventing plagiarism at universities.