Universities need to use multiple methods in order to ascertain authenticity of students thesis work. One method is not enough. If the identification methods are few and weak, more students will try the shortcut, i e copy paste or outsource the writing for hire or using a friend.
1. Frequent communication (between supervisor and student) from initial idea to the final thesis. Oral interaction and reviewing of successive thesis drafts will provide a much better basis than receiving a final and complete thesis without interaction. If you know your student and the evolution of the trains of thought, it is easy to identify sudden changes in text which may be copied or written by someone else.
2. Peer-review. Critical students may identify plagiarism. Students don’t favour other students trying to make a shortcut.
3. Self-awareness. Students may be unaware of the definition of plagiarism, how to quote and refer correctly.
4. Text-matching software. Special software is needed in order to check if the text is copied from billions of web pages, e-journals, e-books, databases etc. This software identifies where a certain text section is found elsewhere, but this may not necessary be a case of plagiarism. Furthermore, the software only searches among digital texts and only in publishers’ databases etc to which the company has access.
5. Policy. Students may not know the rules, regulations and consequences of plagiarism. Explicit information as well as a clear university policy is needed.
6. Meaning. If students’ theses are connected to a real life R&D project with real people who are interested of learning about the results, the risk of cheating is low.
7. Oral defence of final thesis. Oral questions in a public seminar will quickly disclose if the student knows the subject and is the actual author. There is no time to search online or ask someone when receiving an oral question and asked to elaborate, explain or deepen an issue.