Direct answer
Turnitin similarity is the percentage of a submitted document's text that matches other sources in Turnitin's databases, including web pages, academic publications, and previously submitted student papers [1]. It is a text-matching measure, not a plagiarism verdict: a high score can reflect properly quoted and cited material, while a low score does not by itself prove originality [1]. The score is reported alongside a detailed report that highlights each matched passage and names the source it overlaps with [2].
How Does Turnitin Calculate a Similarity Score and What Does the Percentage Actually Include?
Turnitin generates the similarity score by comparing the text of a submission against three indexed repositories: its internet archive, its publication database, and its student paper repository [2]. Every passage that overlaps with an indexed source is flagged, and the percentage shown is simply the proportion of the document's words that were matched in that comparison [2]. The report then attributes each match to a category, so an instructor can see at a glance whether the overlap comes from a journal article, a website, or a classmate's earlier essay [2].
What the number includes matters as much as the number itself. By default, quoted text and the reference list still count toward the score, which is why a heavily cited literature review can legitimately return a high figure [1]. Instructors can switch on exclusions for quotations, bibliography, and small matches, which is one reason the same draft can show different percentages in different courses [2]. Because the score is a ratio of matched words to total words, a short submission or a document with a long reference list can shift the percentage noticeably without any change in how original the argument is [1].
The score also says nothing about intent. Turnitin's own documentation is explicit that the similarity score measures matching text, not academic misconduct, and that interpretation belongs to the instructor who knows the assignment [1]. Two students can submit equally original work and receive very different percentages simply because one topic is more heavily covered online than the other [2]. Reading the highlighted matches, rather than the headline percentage, is what tells you whether the overlap is a citation, a common phrase, or genuinely unattributed borrowing [2].
What Similarity Percentage Is Considered Acceptable or Too High at University?
There is no universal cutoff that applies across institutions, because acceptable similarity depends on the discipline, the assignment type, and the marking rubric [3]. Turnitin itself advises that a similarity score is not a measure of plagiarism and that a high percentage may be entirely legitimate when sources are properly quoted and referenced [3]. A reflective essay, a lab report, and a literature review will naturally produce different baseline percentages even when all three are written from scratch [3].
That said, many departments apply informal expectations rather than hard rules. A common working range for essays is roughly 10% to 25%, with anything above that prompting a closer look at the report rather than an automatic penalty [3]. The more useful question is not "what number is safe" but "can I explain every highlighted match" — if each flagged passage is a quotation with a citation, a standard method description, or a reference entry, the score is doing its job correctly [3]. Conversely, a modest percentage driven by one long unattributed paragraph is a far bigger problem than a higher percentage spread across dozens of properly cited fragments [3].
Students should also remember that the similarity score and the AI writing score are separate outputs. A document can show low similarity and still carry a high AI detection signal, or the reverse, because the two measures examine different things [1]. Treating the similarity percentage as a proxy for originality in general — including originality of authorship — is a misreading of what the metric was built to do [3].
How Can a Student See Their Real Similarity and AI Report Before Submitting?
Most students cannot run an institutional Turnitin check themselves, because the tool sits inside the university's learning management system and is controlled by the instructor [4]. That means the first time many students see their similarity score is after the assignment has already been graded, which leaves no room to fix a genuine citation error or an over-long quoted passage [4]. Pre-submission checking services exist precisely to close that gap by returning a downloadable report a student can read and act on beforehand [4].
Turnitin0 is an independent service that provides exactly this kind of preview, and it is not affiliated with Turnitin, LLC. A student uploads a.docx,.pdf, or.txt file and receives two PDFs in a single order: a similarity report and an AI detection report, formatted the way an instructor would see them in the LMS. Files are checked without being added to the student paper repository, so the submission stays non-repository, and documents can be deleted from the account afterwards. English documents between 300 and 30,000 words and under 20 MB are supported, turnaround is under 15 minutes in 98% of cases, and more than 100,000 reports have been delivered to over 20,000 students across the US, UK, Canada, Australia, New Zealand, and Ireland. The service is pay-per-use with no subscription.
Reading a real report before submission changes how students use the similarity score. Instead of guessing whether 22% is "too high," they can open the report, see which sources were matched, and decide whether a quotation needs tightening or a paraphrase needs a citation [4]. That habit — checking the report rather than the number — is the single most useful thing a student can take from understanding what Turnitin similarity actually measures [4].
Understanding the metric is the easy part; seeing your own report before the deadline is what actually protects your grade. Turnitin0 gives you the same similarity and AI writing reports your professor will open — delivered in minutes, with no subscription and no files added to the student paper database.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Does a high Turnitin similarity score mean I plagiarised?
No. The score measures matched text, and Turnitin states plainly that it is not a plagiarism assessment [1]. A high percentage is often the result of extensive quotation, standard methodology wording, or a long reference list rather than misconduct [3].
Why did my similarity score change between two checks?
Scores shift when the comparison index changes or when exclusions for quotations, bibliography, and small matches are toggled by the instructor [2]. A revised draft with new sources will also produce a different set of matches [2].
Is there a similarity percentage that is automatically safe?
No universal threshold exists, since departments set their own expectations by discipline and assignment type [3]. The practical test is whether you can account for every highlighted match in the report [3].
Can I check my similarity score before I submit?
Institutional Turnitin is normally instructor-controlled, so students rarely run it themselves [4]. A pre-submission service such as Turnitin0 returns a downloadable similarity and AI report so you can revise before the final deadline [4].
Are the similarity score and the AI writing score the same thing?
They are separate measures. Similarity tracks text overlap with indexed sources, while the AI writing indicator estimates how much text may have been generated by AI [1]. A document can score low on one and high on the other [4].