Turnitin: A staff guide to interpreting the Similarity Report

Introduction

This guide describes what a similarity report shows and how to interpret matches highlighted on a student’s assignment submitted to Turnitin.

The similarity report is an effective way to:

Check that online sources in an assignment have been properly cited and the text has not simply been copied without appropriate referencing.
Help students as a formative learning tool around referencing and that you can use with your students to improve their skills.
Identify collusion between students on their course and potentially from other institutions who use Turnitin in the UK.
Ensure a level of equality and parity when checking the similarity of students’ work against the vast range of possible online sources.
Deter students from plagiarising and encourage good academic practice.

It’s important to know that a simliarity match does not always means plagiarism. You will need to apply your academic judgment by understanding how the report works and what it shows.

The similarity report is best used in conjunction with other methods to prevent and detect plagiarism and as part of a co-ordinated approach to maintaining the academic integrity of students’ written work.

Turnitin guide: Setting reasonable expectations for the Turnitin Similarity Score

How to interpret the Similarity Report

The Similarity Index percentage

An overall percentage score (with colour code) is shown next to a student’s name under the Similarity column in the Assignment Inbox. This shows the total amount of matched text as a proportion of the assignment.

This ‘at a glance’ guide should not be used as a measure of plagiarism. Even a 1% score could potentially be plagiarised.

There is no ideal percentage to look for. Students’ work is bound to contain some words from other sources. The percentage will vary depending on the type and length of assignment and the requirements of the work involved.

Individual matches need to be investigated by opening the student’s paper and viewing the match overview and breakdown panel.

What does the Similarity Index percentage indicate?

	Blue – no matching text.	Blue indicates no text has been matched. This could mean that the work has no references at all and that there is little or no use of direct quotes. Depending on the nature of the assignment this is not necessarily an issue but a Blue score is worth checking just in case the student has simply submitted a paper with text that Turnitin cannot recognise.
	Green – one word to 24% matching text.	Green indicates matches between 1% and 24% and is the most common. While a Green score might suggest the document is OK, it is simply an indication of the amount of matched text, so potentially, up to 24% of the document could still have been copied without referencing.
	Yellow –25% – 49% matching text.	Yellow, Amber and Red denote percentage matches in bands above 24%. Higher percentage matches may indicate: • An over reliance on direct quotation as a result of poor academic writing. • Cutting and pasting from other sources.
	Orange – 50% – 74% matching text.
	Red – 75% – 100% matching text.

A 100% match means the assignment has no original work. It has most probably been submitted previously to Turnitin. This can happen if the student is making a re-submission of their work and the file had already been submitted to the Turnitin database. It could be a student error and they submitted to another assignment area by mistake. It can also indicate collusion or copying an essay from another student, either in their class, from a previous year or another institution.

Types of frequently found ‘acceptable’ matched text.

There are certain types of matched text that Turnitin will find, which can be safely excluded or ignored with discretion. These matches will be included in the overall similarity score for a similarity report and be highlighted as a matches on a student’s paper.

These include:

Quotations: Properly referenced quotations can be ignored. These can be excluded using the filter.
References and Bibliography: Other students will have used the same references at some point and these will show up.
Matching formats: e.g. the same essay title.
Tables and Charts showing shared or copied data or statistics.
Appendices may also have a large amount of matching text as other students may well have used the same sources.
Small matches that form common phrases in a sentence or subject terminology will be detected. These can be removed using the small match filter.
Paraphrasing text from a source will be highlighted even where words in the phrase have been changed. If the source has been cited, it remains the academic judgment of the tutor to decide if the text has been suitably paraphrased.


What Turnitin Similarity Reports do not detect

Reports do not pick up matches to images, drawings, diagrams or plans; print books and journals, translated foreign language works and password protected content on websites.

It is also important to state that reports do not detect plagiarism, they merely show the amount of matched text that Turnitin has found by highlighting the matched text on a student’s paper and identifying sources for the matched text. This will include correctly referenced and quoted text.

The decision as to whether this is plagiarism remains your academic judgment.

Examples of common match patterns found on assignments.

Series of small matches.

This report shows a series of 1-3% matches from different sources making up a similarity score of 9%.

It is not uncommon to see this in a long assignment where these are made up of quotes or commonly used phases. Filtering the bibliography and quotations may help to remove some of these to reveal matches of interest.

A large match to a single source.

This report shows a 14% match to a single online source.

Viewing the Match Breakdown of this source and the Full Source of the text will show how it has been used within the assignment. This will help determine if this source has been used appropriately.

Larger scores may indicate over reliance on a single source even if this is referenced correctly.

Series of matches to several sources.

This report shows a similarity score of 21%. There are a couple of larger matches to single sources. The larger percentage sources will need to be investigated to ensure they are referenced correctly.

If this is a long assignment then even 1% matches will need to be checked to see if they have been referenced properly.

A high percentage match to a single source.

This report shows a 100% match in a single assignment previously submitted to the University.

If this isn’t a match to the student’s own work submitted to another submission point (e.g. as a draft) then a request can be made to see the other student paper if you aren’t already an instructor on the area it was submitted to.

Page last updated on September 20, 2023 by adambailey