The short answer
Griffith University runs submissions through Learning@Griffith (Blackboard), and work submitted there is checked by Turnitin. The similarity score it returns is a measure of matching text, not a plagiarism verdict: your quotations, your reference list and standard phrasing in your subject all count towards it. Griffith does not publish one percentage that applies to every assignment, so the figure in your module handbook is the one that matters. A marker reads which passages matched before deciding anything.
What the number actually measures
Turnitin compares your submission against web pages, published work and other students’ submissions, then reports how much of your text matched something else. That is all it does. It has no view on whether a match is legitimate, because a correctly quoted and cited passage matches its source exactly, which is the point of quoting it.
This is why the percentage is a starting point for a conversation rather than a result. At Griffith University, as everywhere, a human opens the report and looks at which passages matched and whether each one is referenced.
What counts towards your score
| Part of your work | Counts? |
|---|---|
| Direct quotations, even correctly cited | Yes, unless excluded |
| Your reference list or bibliography | Yes, unless excluded |
| Standard phrasing in your discipline | Yes |
| The assignment question, if you restate it | Yes |
| Tables, headings and templates you were given | Yes |
| Your own earlier submitted work | Yes, and it has its own name: self-plagiarism |
Whether quotes and the bibliography are excluded is a setting your module leader chooses per assignment, which is the main reason two students on the same course can see very different numbers for comparable work.
The AI indicator is a different number
Similarity and AI detection are two separate systems producing two separate figures, and they are routinely confused. The AI indicator estimates how much of the writing looks machine-generated. It is not enabled at every institution, and it is known to return false positives on genuinely human writing, most often for students writing in a second language or in a very plain style.
If you are flagged and you wrote the work, your defence is your process: drafts, version history, notes, search history, the messy first attempt. Keep them. They are worth far more than any argument about the number itself.
Before you submit at Griffith
- Check the assignment page in Learning@Griffith (Blackboard) for a draft or self-check submission point. Where one exists, use it days early: reports on resubmissions are often delayed.
- Confirm your referencing is in Griffith Harvard, APA 7th, AGLC4 in law. Consistent, correct citation is what turns a match into an obviously legitimate one.
- Check whether quotations and the bibliography are excluded on this assignment. It changes the number substantially.
- Read which sources matched, not just the total. One 8% match to a single source matters more than 30% spread across forty.
- Never paraphrase with a spinning tool to lower a score. Institutions treat that as misconduct in its own right, and it reads as mangled English to a marker.
What Griffith says
Griffith law runs AGLC4 footnoting while the rest of the university runs Harvard or APA, and criminology sits across both faculties so students often get the style wrong. Academic integrity regulations are revised annually and are published by the university itself. This page explains how the tool works; your institution’s current regulations and your module handbook are the authority on what is permitted, and they are the documents to read if you are facing a specific allegation.
Where we come in
Everything we write is written from scratch by a person, and every delivery ships with a free similarity report and a free AI report so you can check it yourself rather than take our word for it. You read the work, check the reports, and pay only after that.