Direct answer
Checking text for plagiarism means comparing your draft against a large database of existing sources — student papers, web pages, and academic publications — and reviewing the matches before anyone else does. The goal is not a single "safe" number but a report you can read, understand, and act on: which passages matched, where they came from, and whether your citations cover them. Turnitin's similarity check and its AI writing indicator are two separate signals, and a draft can score low on one while carrying a flag on the other [1].
How Can I Check My Text for Plagiarism Against a Real Source Database Before I Submit It?
A useful plagiarism check does three things at once: it finds matched strings of text, it tells you which source each match came from, and it separates genuine copying from properly attributed quotation. Turnitin's similarity report is built around exactly that logic — every highlighted passage links back to the source it matched, so you can open the source and judge it yourself rather than guessing [2]. Matches are grouped by origin, so a hit against a journal article in your reference list reads very differently from an unattributed paragraph that matches a student paper [2].
The practical workflow is to run the check on the version of the file you actually intend to submit, then read the report top to bottom instead of looking only at the headline percentage. Turnitin lets you exclude quoted material and bibliography from the similarity index, which gives a much more honest picture of how original your body text is [2]. If a paragraph still lights up after those exclusions, that is the paragraph to rewrite, paraphrase properly, or cite explicitly.
Two limits are worth knowing before you rely on any single tool. First, the database matters more than the interface: a checker that only crawls the open web will miss paywalled journals and previously submitted student papers, which is precisely where a lot of accidental overlap comes from [2]. Second, the report is a review instrument, not a verdict — Turnitin itself frames matches as material for a human to interpret in context, which is why instructors are told to look at attribution and placement rather than the raw score [2].
What Similarity Percentage Is Considered Plagiarism, and When Should I Worry About a Turnitin Score?
There is no universal plagiarism threshold, and any service that quotes you one is oversimplifying. Turnitin does not set a cut-off; individual institutions and departments set their own acceptable limits, and those limits vary by discipline, assignment type, and level of study [3]. A 15% score in a heavily referenced literature review may be entirely normal, while a 15% score in a reflective personal essay with no sources might deserve a closer look [3].
A high similarity score is not automatically plagiarism, because the index counts correctly cited quotations and standard methodological phrasing exactly the same way it counts copied prose [3]. The question that matters is where the matches sit and how they are attributed: a block of matched text inside quotation marks with a citation is scholarship, whereas the same block with no attribution is a problem. This is why reading the report matters more than reacting to the number.
The reverse is equally true — a low score does not prove originality. A narrow source base, a topic with little published material, or paraphrasing that the repository does not happen to cover can all produce a reassuringly small percentage over text that is not genuinely your own [3]. Treat the score as a signal that tells you where to look, and treat the highlighted matches as the actual evidence.
Can a Turnitin Plagiarism and AI Report Be Generated on My Own Draft Before Submission?
In most institutions, students cannot run a full repository check on themselves, because a Turnitin assignment has to be created by an instructor before a submission can be processed [4]. That is a structural limitation, not an oversight: the repository comparison only happens inside an assignment that an instructor has set up. As a result, many students only see their similarity and AI reports after the deadline has already passed.
Pre-submission draft checks exist to close that gap. A non-repository check compares your file against the same categories of sources — student papers, web content, and publications — and returns a similarity report plus an AI writing indicator, without adding your file to the student paper database [4]. That last detail matters: a draft that gets indexed can inflate your own official similarity score later, so a non-repository preview keeps your final submission clean.
What you get back is the same kind of artifact your instructor would see: matched passages linked to their sources, a similarity summary, and an AI score. If the AI indicator shows an asterisk instead of a number, that reflects Turnitin's low-confidence display below its 20% threshold rather than a clean bill of health [1]. Reading that report before you submit converts an anxious guess into a specific, fixable list of passages.
If you would rather see the actual report than guess at your odds, turnitin0 lets you run a non-repository check on your own draft and download the Turnitin AI and similarity PDFs together — the same documents your professor sees in the LMS, usually back in under 15 minutes.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Does Turnitin check for plagiarism and AI writing at the same time?
Yes, but they are separate outputs from the same submission. The similarity report lists matched sources, while the AI writing indicator estimates how much of the text appears machine-generated — a draft can be low on one and flagged on the other [1].
Will a pre-submission check add my paper to the student database?
Not if you use a non-repository check. A repository submission can raise your own similarity score on a later official run, so a draft preview that stays out of the database is the safer option for a final version [4].
What does an asterisk instead of a number mean on the AI score?
Turnitin shows *% rather than an exact figure when its AI detection confidence falls below the 20% threshold, so the result is a low-confidence signal rather than a confirmed zero [1].
Can I lower my similarity score by deleting my reference list?
You can exclude quotes and the bibliography from the index to see your body text more clearly, but removing citations from the submitted file itself would misrepresent your sources. Use the exclusion view to diagnose, not to disguise [2].
Is a 0% similarity score always a good sign?
No. A very low score can simply mean the checker's database did not cover your topic, or that the overlap was paraphrased beyond string matching. Read the matched passages, not just the percentage [3].