Direct answer
Plagiarism is the act of presenting someone else's words, ideas, data, or structure as your own, whether you did it deliberately or by accident. It covers copy-paste, close paraphrasing, uncited quotation, self-reuse of your own earlier work, and uncredited use of images or code. Institutions detect it by comparing your text against web pages, academic databases, and previously submitted papers, then reviewing the matches in context [1].
What Counts as Plagiarism?
Plagiarism is best understood as a spectrum rather than a single offence. At one end sits deliberate misconduct: buying an essay, copying a paragraph word-for-word, or inventing a citation. At the other end sits accidental plagiarism: forgetting a reference, paraphrasing so closely that the sentence structure survives, or assuming that a fact is too obvious to cite. Turnitin's own framing of the plagiarism spectrum makes the same point — the intent behind the act changes the penalty conversation, but it does not change whether the act occurred [2].
The forms that catch students most often are less dramatic than outright copying. Recycling your own earlier assignment (self-plagiarism), dropping a statistic in without a source, swapping a few synonyms and calling it paraphrase, and lifting a table, figure, or block of code without attribution all count. So does using an AI tool's output as if it were your own analysis, because the origin of the words is still not you [2].
A practical test is to ask two questions about every borrowed element: is this idea or wording someone else's, and does my reader know exactly where it came from? If the first answer is yes and the second is no, the passage is a plagiarism risk regardless of how it was produced. That test also explains why citation style errors and plagiarism overlap but are not identical — a badly formatted reference still credits the source, while a beautifully formatted one attached to the wrong source does not [2].
How Is Plagiarism Detected?
Detection works by matching, not by mind-reading. A checking system compares your document against a large index of web pages, journal and book content, and previously submitted student papers, then highlights every string of words that also appears elsewhere. The resulting figure is a similarity percentage — the proportion of your text that matched something in the index — and it is deliberately not called a plagiarism score [3].
Reading the report correctly matters more than the headline number. Turnitin's guidance on interpreting the similarity report explains that each match is listed as a numbered source with the amount of overlapping text, and that you must open the matches to judge whether they are genuine problems. A 30% match to a single paper is a very different situation from 30% spread across twelve short, properly cited quotations [3].
Two limits are worth internalising. First, correctly cited quotations are still highlighted, because the system detects matching text, not missing citations. Second, common phrases, standard methodology descriptions, and reference lists routinely match without any misconduct. That is why a human — your tutor, or you during a draft review — has to interpret the matches rather than react to the percentage alone [3].
Can You Check Your Own Work for Plagiarism Before Submitting?
You can, but usually not through your institution's own system. Most universities restrict student self-submission to the assignment dropbox, and some allow only a single submission, which means the "test run" is the real thing. Guidance on checking before you submit therefore centres on using a separate draft check, then acting on what it reveals rather than treating a clean score as permission to stop thinking [4].
The most useful habit is to review matches in priority order. Start with the longest unbroken matches and any match that runs into your analysis rather than your background section; those are the ones most likely to read as borrowed argument. Then check that every highlighted quotation has a citation and quotation marks, that paraphrases genuinely restructure the sentence, and that your reference list matches the in-text citations. Rewriting the flagged passage in your own voice is what resolves the issue — swapping synonyms is exactly the behaviour that keeps a close paraphrase too close [4].
It also helps to check the same draft for AI-generated text at the same time. A paper can score low on similarity and still be flagged for AI writing, and the two reports answer different questions: one asks whether the words came from a published source, the other whether they were machine-generated. Reviewing both before submission is the only way to see what a marker will see [4].
If you would rather see the matches yourself than guess at them, turnitin0 lets you run an independent pre-submission check on your own draft and returns the same AI and similarity report view your tutor works from — so you can fix flagged passages while there is still time.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is a high similarity percentage automatically plagiarism?
No. The percentage is the share of your text that matched something in the comparison index, and it includes properly cited quotations, reference lists, and standard phrasing. The matches have to be opened and judged one by one [3].
Does accidental plagiarism count?
It counts as plagiarism, though intent usually affects how the case is handled rather than whether it is a case. Forgetting a citation or paraphrasing too closely still presents someone else's work as your own [2].
Can I plagiarise my own work?
Yes — reusing substantial parts of a paper you submitted for another module without disclosure is self-plagiarism, and many institutions treat it the same way as copying from another author [2].
Will a plagiarism checker tell me if I cited a source correctly?
Generally no. Detection systems find matching text; they do not verify that a citation exists, points to the right source, or is formatted in your required style. That check is yours to do [1].
Can AI-generated text be flagged as plagiarism?
It can be flagged for AI writing, which is a separate signal from similarity. A draft can have a low similarity score and still be questioned over how it was produced, which is why reviewing both reports before submission is worthwhile [1][4].