Guide on the Use of Similarity Checkers (iThenticate / Turnitin) to Detect Plagiarism 

 

Plagiarism—the act of presenting someone else's work, ideas, or words as one’s own without proper acknowledgment—is a serious breach of academic and professional integrity. Whether intentional or accidental, plagiarism undermines originality, devalues genuine effort, and can lead to severe consequences, including reputational damage and disciplinary action.

Software such as iThenticate / Turnitin work as text similarity checkers, and are not actual plagiarism detectors even if they are commonly referred to as such. However, similarity checkers are essential tools that can help uphold standards of honesty and credibility by detecting similarities and identifying unoriginal content. These tools support users in ensuring proper citation, promoting responsible writing practices, and fostering a culture of ethical scholarship.

 

Requirement in NTU:

To ensure adherence to NTU's Academic Integrity policy and procedures, all theses and dissertations submitted by Master’s and Ph.D. students are to be checked for plagiarism by using iThenticate (available on NTULearn). Instructors for undergraduate courses are to use Turnitin.

Further info on Turnitin vs iThenticate can be found here.

 

What is an acceptable Similarity Score for plagiarism?

A Similarity Checker is a text-matching software that compares submitted text against a database and generates a similarity report to highlight matches for users to review. Hence, it does not check for plagiarism directly, but is used as a detector.

Therefore, there is no universal "acceptable" similarity score that automatically indicates plagiarism, as plagiarism is not solely about numbers – it’s about context, intent, and proper attribution.

The similarity report must be reviewed in context to determine whether plagiarism has occurred.

 

Limitations of Similarity Checkers:

  1. High similarity score 1 – A high score doesn't always mean plagiarism. For instance, a high similarity score is often found in the Materials and Methods section, or in properly cited literature review. Also, theses and dissertations containing own published work can have a high similarity score. It is important to review the source of the matched content to determine if there is plagiarism.

     

  2. Low similarity score 2 – A low score doesn't guarantee originality, since cleverly paraphrased but uncredited content may not be detected by similarity checkers. Also, content generated by GenAI can be highly original, but is not considered an author’s own work. Hence, theses and dissertations may have a low similarity score, but still contain rephrased ideas obtained from elsewhere.

 

Summary:

  1. The similarity score serves as a guide further investigation, rather than serve as conclusive evidence of misconduct.
       
  2. Human judgment is essential; reviewers must examine what was matched, how it was used, and whether proper citation was provided. For instance, further investigation may be required if a document has a low overall similarity score, but contains high similarity in the results or conclusion discussion. 

 

 

Footnotes:

  1. NTU allows the usage of students' own previously published materials such as book chapters, articles, reports, letters, posters, etc. in their theses/dissertations, provided they are properly cited and attributed. In this case, Turnitin typically gives a high similarity score. Supervisors should examine the Turnitin report from a qualitative and contextual basis to determine the nature of the matching. It should be reminded that reusing students' own previously published materials without appropriate citation or acknowledgment constitutes self-plagiarism.

     

  2. One student has 22% matching in the Turnitin similarity report, but all the matched material is in the bibliography, which is legitimate. Another student has 7%, but all the matching consists of direct copying, which constitutes unacceptable plagiarism. Judgments should not be based solely on the overall similarity percentage. The lecturers/instructors should examine the matched content carefully during the marking process.

 

 

Declaration: This guide was enhanced with the use of ChatGPT.