Your PlagAware report What the percentage really says – and what it doesn’t.
First, breathe.
You upload your thesis, wait, and then there’s a number. 12%. Or 23%. And straight away your head starts racing: is that too much? Will my supervisor notice? Am I in trouble?
The percentage is not a verdict. That isn’t us being nice – PlagAware says so itself: the software shows matches, and whether any of them is plagiarism is always decided by a person who looks at the passages.
So you can read those passages the way your examiner will, let’s walk through the report step by step.
What the number measures.
The percentage result is one simple ratio: words matching relevant, not-declared sources ÷ all words in your text.
Which also means: a correctly cited sentence still counts until someone marks its source as “declared”. PlagAware reports everything it finds first – your clean quotations included.
What the number does not know:
- whether a passage is in quotation marks with a reference,
- whether it is a stock phrase (“In summary, it can be said that …”),
- whether it is a mandatory part, such as the declaration of authorship from a template,
- which text is the original – PlagAware cannot tell original from copy.
According to PlagAware, the result should be 0% or close to it once every source is classified correctly. A high number before that step is normal.
The four source types in the report
-
🔴 Relevant
Sources that may be uncited or cited incorrectly. Only these words make up the percentage. Red in the overview bar.
-
🟢 Declared
Classified as correctly cited – automatically by URL or by hand. Still highlighted in green, but left out of the percentage.
-
⚪ Excluded
Matches that were never cited or got in by mistake, like a template declaration or an earlier version of your own work. Hidden, but still listed.
-
🫥 Covered
Sources whose matches are already fully contained in another source. They don’t count twice – if a blog copies Wikipedia, the sentence isn’t added to your number again.
How to work through your report
-
1. Open the red sources
Click each relevant source and open the side-by-side view. Bold means copied word for word, regular type means reworded. Source colours mean nothing – they only tell sources apart.
-
2. Classify every match
Correctly cited? Then it’s declared. A stock phrase, a technical term, a mandatory text or your own earlier work? Then it isn’t a suspicion. What’s left is what was really taken without a reference.
-
3. Close the real gaps
Reference missing? Add it. A whole passage too close to the source? Rewrite it with your own line of thought – and cite anyway. Rewording never replaces a reference.
-
4. Check again
Run the check again after revising. Note that your own classifying changes nothing in your university’s report: there, the examiner classifies the sources.
High isn’t bad. Low isn’t safe.
Many universities deliberately work without a fixed threshold. FH JOANNEUM, for example, states explicitly that there is no “everything under 15% is fine” – because the percentage alone says little.
What examiners look at instead:
- a few small similarities without marking → usually harmless,
- many copied or lightly changed sentences without marking → critical,
- longer copied sections or core ideas without marking → attempted deception.
And the other way round: PlagAware only knows paywalled articles and books if they’ve been added to the library, and by its own FAQ it can’t detect fully reworded text. A low score isn’t a free pass – a properly referenced thesis is.
And the AI detection? The style analysis shows how “AI-typical” the wording is – and people write like that too. It’s a reason for a conversation, not proof.
