Direct answer
A "plag report" is shorthand for a plagiarism report — most often a Turnitin similarity report — that lists the sources your text overlaps with and gives a similarity percentage. It is a matching and attribution tool, not a verdict: it shows where wording coincides with existing material, and a human (usually your instructor) decides whether that overlap is properly cited or actually misconduct [1]. A high number is a prompt to look closer, and a low number is not proof of originality [2].
What does a plag report actually show?
A plag report is a document built from text-matching, and its core component is a list of sources with the overlapping passages highlighted. Turnitin's own guidance describes the similarity report as a set of matches grouped by source — internet pages, publications, and previously submitted student papers — with each match linked to the exact strings of text that coincided [1]. The percentage at the top is simply the proportion of your document's text that matched something in those databases, which is why the same essay can score differently depending on which repositories a system searches [2].
What the report does not contain matters just as much. It does not label you a plagiarist, it does not assign a grade penalty, and it does not distinguish between a correctly quoted sentence and a copied one — that distinction is interpretive work done by a reader [1]. This is why two students with identical 24% scores can face completely different outcomes: one has a long block quote and a heavy reference list, the other has three uncited paragraphs lifted from a website.
The report also carries structural information that tells you how to read it. Matches are colour-coded and numbered, so you can click through source by source; overlapping text is shown side by side with the original; and the summary panel breaks the score down by source type [2]. Learning to read those panels is the actual skill — the headline percentage is the least informative part of the document.
One practical caveat: the terminology varies by institution. "Plag report," "similarity report," "originality report," and "text-matching report" are used interchangeably across UK, US, Canadian, Australian, New Zealand, and Irish universities, and some systems add a separate AI-writing indicator alongside the similarity figure [1][2]. If your lecturer uses a term you don't recognise, ask which report they mean before you panic about the number.
Is a high similarity score the same as plagiarism?
No — similarity and plagiarism are different concepts, and conflating them is the single most common misreading of a plag report. Similarity is a measurement of textual overlap; plagiarism is a failure of attribution, which depends on whether you credited the source and whether you intended to present someone else's work as your own [3]. Turnitin frames this as a spectrum running from accidental or unintentional copying through to deliberate misconduct, and only the human reader can place a given case on that spectrum [3].
Correctly cited work generates matches all the time. A literature review that quotes four sources per paragraph, a methodology section that reuses standard instrument wording, and a reference list full of titles will all push the percentage up while being entirely proper academic practice [3]. Conversely, a paraphrased passage with no citation can score low if the wording was changed enough to defeat matching — which is why a low score proves nothing about integrity [2].
Institutional thresholds reflect this ambiguity rather than resolve it. Many departments treat a score under roughly 20–25% as unremarkable, but those cut-offs are local conventions, not universal rules, and they are usually applied to a report a human has already reviewed [3]. If your report shows a figure above your department's guidance, the productive response is to open each match and ask a single question: is this cited?
The AI dimension adds a second layer that a similarity score cannot capture. A plag report answers "does this text match existing sources?" while an AI-writing indicator answers "does this text look machine-generated?" — two different questions with two different reports, and a document can be flagged on one and clean on the other [2][3]. Students who assume a good similarity score clears them on both counts are routinely surprised.
How can a student see their own similarity report before submitting?
In most institutional setups you cannot generate an official report yourself, because the report only exists once a submission enters the institution's own Turnitin assignment and its repository settings [4]. Whether you can see the resulting report at all depends on settings your instructor controls — some allow students to view similarity reports, others restrict visibility to teaching staff, and draft-submission checkpoints are optional features rather than defaults [4]. That is why "just check it yourself first" is often impossible inside the university system.
What students typically do instead is use an independent pre-submission checker that produces the same two artifacts — a similarity report and an AI-writing report — without adding the file to Turnitin's student paper database. The practical value is timing: you see the matches while you can still fix them, rather than after the deadline when the only remaining option is an extension request or an integrity meeting [4]. Reading a report on your own draft also trains you to recognise the difference between a benign citation match and a genuine attribution gap [3].
If you do get access to a draft report, work through it in a fixed order. Start with the highest-percentage source and confirm every matched passage is quoted and cited; then check the reference list against in-text citations; then look at whether any paraphrase has drifted too close to the original wording [1]. Fixing those three things resolves the large majority of matches that actually matter, and it takes far less time than rewriting a whole section in a panic.
A word of caution on interpretation: an independent check is a diagnostic, not a guarantee. Repository contents differ between systems, so a clean draft report does not certify that your institution's report will look identical [2][4]. Treat the number as feedback on your citation practice, not as a prediction of your grade or your integrity outcome.
If you would rather decode a real report than guess at what the percentage means, turnitin0 lets you run that check on your own draft first — the same similarity and AI-writing reports your instructor would see, delivered in minutes and without adding your file to the student paper database.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Does a plag report mean I have been accused of plagiarism?
No. A plag report is a text-matching document generated automatically for every submission, and its existence is routine rather than accusatory [1]. An accusation only follows if a human reviews the matches and concludes that attribution was genuinely missing.
What similarity percentage is considered acceptable?
There is no universal threshold; departments set their own guidance, often in the region of 20–25%, and apply it alongside human judgement about what the matched text actually is [3]. A cited quotation counts toward the score just like an uncited copy does, so the number alone cannot tell you whether you are fine.
Why does my plag report show matches to my own previous essay?
Turnitin's databases include previously submitted student papers, so resubmitting your own earlier work — or text you have submitted to another course — can appear as a match [1][4]. Whether that is treated as self-plagiarism depends on your institution's policy, and it is worth asking your tutor directly.
Can I get a plag report before I submit my final draft?
Usually not through your university, because report visibility and draft checkpoints are instructor-controlled settings [4]. An independent pre-submission check gives you the same style of similarity and AI-writing reports on your draft while you can still edit it.
Is a plag report the same as an AI detection report?
No. A similarity report measures overlap with existing sources, while an AI-writing report estimates whether text looks machine-generated — two separate indicators that can disagree [2]. Read both, because a clean similarity score says nothing about the AI figure.