Direct answer
A plagiarism scanner for essays compares your document against three corpora: the live and archived web, subscription journals and periodicals, and a repository of previously submitted student papers. The number it returns is the percentage of your text that matched something in that index — not a verdict on whether you plagiarised [1]. Because each scanner indexes a different corpus, the same essay can produce very different scores in different tools, which is why the only score that matters for your grade is the one produced by the system your institution actually runs [3].
What Does A Plagiarism Scanner For Essays Actually Compare Your Text Against?
Every scanner is an index plus a matching engine. The matching engine is rarely the interesting part; the index is. Turnitin's similarity report draws on the current and archived internet, a body of periodicals and scholarly journals, and the repository of papers previously submitted by students at subscribing institutions [1]. When your essay is processed, the system segments it into sentences and short phrases, searches each segment against those collections, and then reassembles the matches into a single report.
The output is a similarity score, defined as the proportion of your submission that matched sources in the database [2]. That definition matters more than the number itself. A 34% score does not mean 34% of your essay was copied from someone else — it means 34% of your text was found somewhere in the index, which may include a quotation you cited properly, a bibliography entry, a common phrase, or a passage from a lecture handout your lecturer also uploaded.
Turnitin's own documentation is explicit that a high similarity score is not automatically plagiarism, and a low score is not automatically originality [2]. A student who paraphrases a source badly can score low because the words changed, even though the ideas were lifted. A student who quotes three paragraphs of a primary source with perfect attribution can score high. The scanner measures textual overlap; your instructor supplies the interpretation.
The report is also configurable, and the configuration changes the number. Quote exclusion, bibliography exclusion, and small-match thresholds can all be switched on by the instructor, and each one alters the final percentage [2]. Two students submitting identical essays to two different modules can therefore receive different scores — not because the scanner is inconsistent, but because the settings are.
Why Do Free Plagiarism Scanners Give Different Similarity Scores Than Turnitin?
The most common reason a free tool and Turnitin disagree is that they are not looking at the same collections. A free checker typically searches the open web and, sometimes, a limited set of public documents; it usually has no licence to search paywalled journals and no access to a repository of previously submitted student papers [3]. That repository is precisely where a large share of essay-level matches come from, because essays are written from the same reading lists, the same module handouts, and the same previous cohorts' work.
Filters are the second source of divergence. Turnitin reports can exclude quotations, exclude the bibliography, and ignore matches below a word threshold set by the instructor [1]. A free tool that counts every matched string, including your reference list and your in-text citations, will inflate the score on a well-referenced essay. Conversely, a tool that strips too aggressively can hide a genuine unoriginal paragraph. Neither behaviour is wrong in the abstract; they simply answer slightly different questions.
The third difference is what the number is attached to. Turnitin separates similarity from AI writing detection, and the AI score is shown as *% rather than an exact figure when the signal falls below its 20% confidence threshold — those are low-confidence signals, not a clean bill of health. Free tools frequently blend the two into one vague "originality" percentage, which is not a metric your institution will recognise [3].
Practical consequence: treat a free scan as a rough smoke test for obvious copy-paste, not as a prediction of your grade. If you need to know what your marker will see, the scan has to come from the same engine and the same corpus your marker uses [3].
How Can You See Your Essay's Real Turnitin Similarity Report Before You Submit It?
The honest starting point is that Turnitin does not offer students a public pre-submission check; access runs through institutional accounts and assignment settings. What students can do is preview a report produced by the same engine, against the same collections, before the deadline — which is what an independent pre-submission service provides. The value is in the fidelity of the preview: same corpus, same match highlighting, same source list with per-source percentages, so the document you review is the document your instructor will review [4].
Two repository details are worth understanding before you rely on any preview. First, whether a paper is retained in the student paper repository is controlled by institutional settings, and a non-repository check does not add your file to that database — so previewing your own draft does not create a future match against yourself [4]. Second, the repository keeps growing, which means a score is a snapshot in time. A draft that scores 12% today can score higher next term when a new cohort's essays join the index. That is not a malfunction; it is an artefact of a living database.
A sensible workflow follows from this. Scan early, read the source list rather than the headline number, and fix the passages where the match is substantive rather than incidental. Prioritise matches to journal articles and to other students' papers over matches to a dictionary definition or a stock phrase. Re-scan after revisions, and keep the final report so you can show your working if a marker queries a passage.
Turnitin0 is built for exactly that pre-submission step: it returns the Turnitin AI detection report and the similarity report together, in the format instructors see, and it checks without adding your file to the student paper database.
If you want certainty rather than a guess, the next step is to run your actual draft through the same engine your marker uses and read the report yourself — source by source, before the deadline rather than after it. That is what turnitin0 is for: upload your essay, get the real similarity and AI reports back in minutes, and revise with the evidence in front of you instead of hoping the free checker was accurate.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is a plagiarism scanner the same thing as Turnitin?
No. Turnitin is one specific scanner with a specific corpus and specific report settings; a "plagiarism scanner" is a category of tool [1]. What matters for your grade is which scanner and which settings your institution uses, not the generic label.
What similarity score is safe for an essay?
There is no universal safe number. Turnitin states plainly that a high score is not automatically plagiarism and a low score is not automatically originality [2]. Read the source list and judge whether each match is properly attributed.
Why did my score change between two scans of the same essay?
Because the index changed. The student paper repository grows as new work is submitted, and instructor filters can differ between modules [4]. A score is a snapshot of a living database, not a fixed property of your text.
Do free scanners search other students' essays?
Generally not. Free tools typically search the open web and lack licensed access to paywalled journals and to a student paper repository, which is why their scores diverge from Turnitin's [3].
Will checking my essay before submission flag me later?
Not if the check is non-repository. Turnitin0 checks your file without adding it to the student paper database and does not share reports with third-party databases, so previewing your own draft does not create a future self-match [4].