The short answer is less than 24%.
The longer answer is that it depends. Firstly you need to exclude the “Works Cited” or “Bibliography” from the test run. Then you have to look at where the plagiarism matches are coming from. Let me show you some quotes:
“Students check their assignments in Turnitin revision classes before submitting them. They attach the Turnitin report to their work when it is handed in. A generous cut-off of 25 per cent matched text was allowed in this pilot. Any essays with higher than this percentage matching must be explained by the students. Students also sign a declaration that the work is their own and that they understand what plagiarism is.”
- Sheffield University
Or this interesting point:
An originality report may have a similarity index of 24% with the similarity coming from hundreds of different sources, each making up less than 1% of the student’s work, other originality reports with an index of 24% may have the similarity coming from just one or two web pages or published sources, potentially a much more serious matter.
- University of Glasgow
Or this slightly different perspective:
Students soon become minutely concerned with what Turnitin’s percentage score means in its report. For Turnitin, zero percent means there is no overlap between a paper and any other. One hundred percent means there is a total case of plagiarism. I have found that the scores I prefer range between 20 and 35 percent. Rather than plagiarism, these scores usually confirm that this is a robust research paper with substantial use of quotations, footnotes, and bibliography. It is these two things—footnotes and bibliography—which most inflate Turnitin’s score. A very low score indicates a paper with few footnotes and references. A very high score indicates a “cut and paste” job which, even if footnoted, is unacceptable to me. The highest score I have ever seen is 85 percent in a paper in which Turnitin flagged the entire first two pages in red, indicating they came from other sources. Yet, the student had footnoted those pages faithfully and persistently throughout. For me, this was less a case of deliberate plagiarism than one of a student who had no idea how to write a research paper and use sources properly.
- California State University
And finally this from UCLA:
Therefore, the initial plagiarism percentage rating cannot be used as a trustworthy indication of the degree of plagiarism. Using an example from my course: one student posted a paper that received an initial score of 64% plagiarized, but upon inspecting the paper I saw the majority of the matches came from properly quoted material and the references cited section. After checking the accuracy of the citations and references, I excluded them from the match and the paper returned a score of two percent.
Here at Tailored Essays we look for a score of below 15% for fully original work such as the main body of a PhD. But we just want to get a score of below 24% for the Literature Review or any other work that is heavily based on existing published work. Naturally our scores are usually well below those numbers as our work is fully custom written and checked through WriteCheck (part of Turnitin).
Please email us on firstname.lastname@example.org if you have any questions about this issue. We will get back to you as soon as we can and definitely within a few hours.
If you want to know more then look at our next article on Turnitin and what the score means: Turnitin Percentages – Part II