iThenticate is a text-matching software tool designed specifically for HDR candidates to assist them with their academic writing. iThenticate compares documents to an extensive database of web pages and scholarly content to produce a similarity score and report.
Only HDR candidates have access to run their work through iThenticate. You, as the supervisor (primary or associate), do not have an iThenticate account. However, ANU recommends HDR candidates share their iThenticate similarity report/s with their supervisor/s (primary and/or associate) so you can guide and teach them how to improve their writing.
Note: iThenticate cannot by itself detect plagiarism; it checks a submission against the content database, and if there are instances where the submission's content is similar to, or matches against, one or more sources, it will be flagged for review to be discussed with your student.
Conversely, using it does not guarantee that writing does not contain plagiarised text. It does not check against, for example, Google Books or material that has only been published in hard copy. It is essential that using iThenticate is only part of a researcher's strategy to ensure academic and research integrity.
An iThenticate Similarity Report is now included in all final thesis submissions (go live date was 6 July 2023), as part of the eForms process. To view your role in this broader eForms process, see the User Journey Flow Chart.
You, as supervisors (primary and/or associate), will not have access to the iThenticate tool. However, you will need to know how to interpret the similarity report so that you can guide your students to improve their writing.
There are various training materials available for HDR supervisors on iThenticate including:
About the Similarity Report
iThenticate does not check for plagiarism in a submission; it checks a submission against the content database, and if there are instances where the submission's content is similar to, or matches against, one or more sources, it will be flagged for review. The database includes billions of web pages (both current and archived content), a collection of documents, which comprises of thousands of periodicals, journals, publications, and CrossRef member content. It is perfectly natural for a submission to match against sources in the database. If the submission has used quotes and has referenced correctly, there will be instances where there will be a match.
The Similarity Report will contain a similarity percentage/score that indicates the overall similarity of the candidate's submission, based on the amount of matching text that is identified. The similarity score simply makes the HDR candidate aware of any problem areas in the submission; iThenticate should be used as part of a larger review process to determine if the match is or is not acceptable.
Interpreting the Similarity Percentage
There are no standard or universal ideal maximum or minimum percentages to aim for. Even a document with a low percentage could contain plagiarised text, while a document with a high percentage may have no problems.
A document with a low percentage score (less than 20%) should still be carefully checked to ensure that citations are correct and that matching text is attributed correctly. Use the Similarity Report to do this. If the score is low researchers may wish to consider whether the document engages closely enough with existing literature in the field.
A medium to high percentage score (above 20%) requires more consideration. A higher percentage score can result from either:
significant quotation from a single source; or
a combination of many shorter quotes from multiple sources.
Both of these may be acceptable according to context and usage. An English article about a single poem or novel would likely quote extensively from the primary text, while a systematic review in medicine or the sciences might quote briefly from many sources.
The more quotations are used the more likely it is that inadvertent plagiarism will occur through errors in note-taking, typing, etc.
Even if all matching text is quoted correctly, a high percentage score may indicate that the document does not meet academic standards for originality of ideas, argument, or new discoveries. See the Self-citation and Self-plagiarism and Genres of Writing sections of this guide for more information about the levels of originality expected in different kinds of publications.
Interpreting the Similarity Report
The following video demonstrates how to interpret the similarity report using the iThenticate tool itself. However, as the supervisor, you will receive a PDF view of the report that your student can provide to you.
Supervisors should review each section of highlighted text with their HDR candidates. Below are some guiding questions that should be considered:
Is all matching text enclosed in quotation marks and correctly cited?
Are all quotations attributed to the correct source?
Are clusters of quotations in appropriate sections of the document?
Has any single source been quoted to an extent that copyright permission should be sought?
To view some more examples of common plagiarism mistakes, you can visit iThenticate's Plagiarism Spectrum
Note: The PDF will reflect the report mode and selected filters that the student has selected, so be sure to highlight any report preferences you have to your students. See "Report Modes" and "Filters and Exclusions" sections to understand more about these preferences.
Filters and Exclusions
While you as the supervisor do not have direct access to change filters or exclusions on the Similarity Report, we recommend familiarising yourselves with the possible exclusions so that you can request your student to apply these filters if required or preferred.
HDR candidates can select to exclude the following from the similarity percentage:
Quoted material: For more information, see the Quote exclusion definitions
Bibliographic material: For more information, see the Bibliography exclusion definitions
Small sources or matches: HDR candidates can select to exclude small sources and/or matches that are less than a set percentage/word limit
Sections: HDR candidates can exclude a section (e.g. abstract, method). iThenticate will exclude these by reading the document and excluding sections that have headers with the words 'abstract', 'method and materials', 'methods', 'method', 'materials', and 'materials and methods'.
Specific matches: If the HDR candidate determines that a match is not needed, you can exclude the source from the Similarity Report through the Match Breakdown or All Sources viewing modes. The similarity index will be recalculated and may change the current percentage of the Similarity Report.
Frequently Asked Questions
If you are an HDR supervisor (Primary or Associate) and have a question about iThenticate, a list of Frequently Asked Questions below: