Is a Turnitin Score a Grade? What the Similarity Score Actually Does (and Doesn't) Do

The similarity score is the percentage of your submission's text that matches Turnitin's databases. Turnitin states plainly that it "does not determine whether plagiarism has occurred," and identical wording appears in the AI indicator FAQ for the AI percentage. Neither number is a grade. Grades come from the instructor applying the assignment's criteria and the institution's policy to the work — the report is one input in that review, not the output.

Turnitin's Similarity product page defines it precisely:

"The Similarity Report's similarity score is the percentage of text in a submission that matched against other sources."

And the same page adds a sentence that students rarely see:

"There is no 'right' or 'wrong' number to receive as a score."

Warwick's official guidance says the same thing from the instructor's side: "The score received does not automatically determine that plagiarism has been committed – there is no 'score' to aim for." The report exists so a marker can compare the submitted paper with the suspected sources and "use academic judgment" to reach a view. The number describes a comparison, not a verdict (our guide on how to read a similarity report goes deeper into the mechanism).

Turnitin says the same about the other number on the page:

"Hence, we must emphasize that the percentage on the AI writing indicator should not be used as the sole basis for action or a definitive grading measure by instructors."

"Not... a definitive grading measure" — the AI percentage is not a grading input that decides anything by itself either. Our guide on the AI report vs. the similarity report covers how the two percentages are independent of each other; the point here is that neither one is a grade.

Instructors grade against the assignment brief, the rubric, and the institution's academic-integrity policy. The similarity and AI reports feed the review — Turnitin's materials describe them as a signaling tool and "one piece of an investigative puzzle" — but the grade is written by the instructor, in the gradebook, with the assignment's criteria in front of them.

Two practical consequences for a student:

• A high score does not automatically yield a bad grade. Properly cited quotes still match; some disciplines are quote-heavy; the instructor may exclude bibliography or quoted material (our filters and exclusions guide covers those settings).
• A low score does not automatically yield a good grade. The report is one input; content, argument, and compliance with the brief still get graded on their own merits.

There is no fixed formula because there is no universal rule. The same 40% can be routine in one assignment (quote-heavy literature review) and triggered in another (math proof with no sources allowed) — the institution may even have set a threshold for that specific assignment. Turnitin's position is that interpretation is the instructor's: reports "provide a summary of matching or highly similar text" and instructors use them "as a tool within their review process to make their own determination if any academic misconduct has occurred." The grade question is always answered by the assignment's own rules, not by the report.

• Start with the assignment materials, not the report. The brief, rubric, and any stated institution threshold define what the score means for this specific assignment.
• Check your feedback before the percentage. Instructors comment on the work; our guide on instructor comments on reports covers where those comments live and how to read them together with the score.
• Ask a question that separates the two. "How does the similarity score relate to the grade for this assignment?" is a fair question; "Why is my grade lower than my score?" assumes the two are on the same scale — they aren't.
• If the AI indicator is in the conversation, the same separation applies: the AI percentage "should not be used as the sole basis for action" (official wording), so a discussion anchored on the actual flagged paragraphs is on firmer ground than one anchored on the number (our "conversation after an AI flag" guide covers that talk).

A Turnitin score is not a grade. The similarity score is a percentage of matched text — Turnitin explicitly says it does not determine whether plagiarism occurred and there is no right or wrong number; the AI percentage is explicitly not a definitive grading measure. Grades come from the instructor applying the assignment's brief, rubric, and institution policy, with the report as one input in the review. When a score and a grade don't line up, read the assignment materials, read the feedback, and ask a question that keeps the two separate.