How to use and interpret Turnitin similarity reports

Table of Contents

1 Why use it?
2 How to view a similarity report in a Moodle assignment activity?
3 Navigating the similarity report
4 Identifying matches
5 Flags
6 Matched sources
- 6.1 Other matching sources
- 6.2 What do partial highlights mean?
7 Excluding matches
8 Academic judgement

Why use it?

Turnitin is a service that matches text from student assignments against its extensive databases of previously submitted student coursework, websites and academic papers. It produces a similarity report and a score (%) of matched text.

Turnitin is a text matching tool, not a plagiarism detection tool. Therefore, to use Turnitin to investigate academic misconduct, the interpretation of Turnitin reports should be the responsibility of academic colleagues.

Turnitin provides a useful indicator to markers of the extent of matched text within a document, but it cannot determine what those matches mean. It is no replacement for academic knowledge and judgment in determining cases of academic misconduct/plagiarism.

The ‘LSE Policy on the use of Turnitin’ provides further information on how the platform is used across the School and sets out the school-wide principles of using Turnitin.

How to view a similarity report in a Moodle assignment activity?

Navigate to the relevant Assignment activity and select View all submissions.
Within the submissions table you will see a column entitled File submissions.
Locate the similarity percentage score for the corresponding submission in this column.
To access the similarity report for the submission click directly on the percentage score.

Note: If you need to view a similarity report in a Moodle Workshop activity, please refer to the Turnitin guide ‘Viewing a Similarity Report in a Moodle Workshop’.

Navigating the similarity report

Clicking on the similarity percentage score, will open Turnitin’s Feedback Studio in a new browser window/tab. This is where where the similarity report can be viewed. The first page of the particular assignment submission will be displayed with the option to scroll down to view the additional pages.

The sidebar on the right hand side is composed of two ‘Layers’ which correspond to Instructor Feedback (blue layer) and Similarity Tools (red layer). There is also a tool for submission download (grey button). For the purposes of this guide, we are only concerned with the Similarity Tools (red layer).

Note: It is important that the Similarity layer button is red – if it’s black, click it to turn it red and reveal the similarity matches.

Identifying matches

The colour of the similarity percentage score displayed in the Moodle assignment submissions table is based on the amount of matching text identified in a document.

Within the assignment, you may see that sections of text in the assignment have been highlighted, each with a corresponding number and colour. Turnitin highlights phrases which match another source within its extensive database. The number refers to the matched source identified.

Flags

Flags are used to highlight any inconsistencies in a submitted document that might set it apart from other submissions. A Flag could be referring to hidden text or replaced characters within a document.

Flags don’t necessarily indicate of a problem, but, if they appear, it is recommended to look at them in more detail.

To see if any Flags have been detected for a submission select the the red flag icon ( ) within the similarity layer. The sidebar will expand to display the number of ‘Flags for Review’ and their corresponding details.

Matched sources

The matched sources that have been determined by Turnitin can be viewed by clicking on the number (percentage similarity score) in the side bar. This opens up the Match Overview panel to show all matching sources starting with the most-used source.

As you can see from the screenshot (right), because LSE coversheets accompany many student submissions, they can inflate the similarity score (accounting for 9% of the similarity score here). This illustrates how the % score of text matches within a report can be misleading and requires careful interpretation.

Click on any of the matches in the Match Overview panel to reveal where they have been identified within the document. This will generate a pop-up window with an overview of the matched source, allowing for a side-by-side comparison with the student text.

Other matching sources

Turnitin may match a source which is not the source where the author originally encountered the text. For example, Turnitin may report a recent student essay as a match, when the student has read the words in an academic article. This is because the same text can appear within a number of different sources.

To view other matching sources, click on the arrow (>) to the right of the matching source listed in the Match Overview panel. Turnitin will display a list of other matching sources. Reading this list, you may recognise a source you put on a reading list or a source you are confident that students are likely to have read.

What do partial highlights mean?

Turnitin will often match a few phrases or words highlighted in a paragraph, as shown in the screenshot below.

Turnitin can only highlight the matched text, and cannot distinguish whether a student has:

copied words from a source into their notes, and then copied back into their work, forgetting these phrases weren’t their own words
tried to paraphrase a source, but not put it sufficiently into their own words
deliberately changed a few words in a passage to try to confuse the marker (and Turnitin)

Any of these offer an opportunity to begin a discussion about appropriate academic writing practice with the student.

Excluding matches

Using the Filter tool to exclude matches

When you click on the Filter icon in the side bar, you will be able to see which parts of the submission have already been excluded from the similarity report (this will have been set up in the submission settings in Moodle).

For example, Bibliography is excluded by default within Moodle. This has been done to make results more accurate, as bibliographies almost always match other sources, and can therefore inflate the extent to which a student’s work matches other sources.

Similarly, Quotes have been excluded by default. This includes indented blocks of text and anything in quotation marks as these are the most common ways of quoting and citing in an essay.

Therefore, if Quotes have been excluded and passages have been highlighted, it might well be that these passages have been inadequately acknowledged by the student.

Within the report, in the Filter tool, you can further exclude matches under a certain percentage or word count. However, this needs to be carefully considered. A low percentage match may seem insignificant in abstract, but if the match is fully contained with a particular section, for example the conclusion, this may require investigation.

Excluding individual sources

By excluding certain sources, you can better focus on, potentially, more revealing matches within a student’s work. There are a number of occasions on which you might want to exclude sources:

When you feel sources are falsely inflating the matching score, as we saw above with the LSE coversheet.
Where a source is referenced by a student within the body of the essay, but the way the referencing is done is problematic (i.e. in the transcription of the quotation). In this case, the issue relates to the student's ability to properly quote. Students should be offered training in academic writing. This can either be provided by the department to departments can direct students to that offered by LSE LIFE offerings.
Where you know that the student erroneously submitted the assignment to another Moodle course before submitting to your course. This is known to happen and can result in a matching of score of 100% being given to student work. However, before excluding such matches you need to be confident that student has made a genuine mistake in previously submitting to another course.

Excluding sources is a straightforward process:

In the Match Overview panel, click on the arrow to the right of the matched source. This will reveal a breakdown of all the matched sources in a new panel entitled Match Breakdown

Click on Exclude Sources at the bottom of the panel
Select the sources you wish to exclude by ticking the box next to each one

Click Exclude

You should see that the appropriate text is no longer highlighted and that the % score has been adjusted accordingly.

Restoring excluded sources

Click on the prohibited icon (excluded sources) in the side bar. This will open a new Excluded Sources panel.

Select the sources you wish to restore by ticking the box next to each one

Click Restore

You should see that the appropriate text is highlighted again and that the % score has been adjusted accordingly.

Academic judgement

A high similarity score does not always suggest that a piece of writing has been plagiarised, just as a low similarity score does not always indicate that no plagiarism has occurred. For example, multiple drafts of the same paper could have been submitted to the repository, meaning the final draft has resulted in a score of 100%.

It is important to note that Turnitin is only a text-matching tool, it cannot determine if ideas or examples are original. In addition, we saw above that there are a number of sources with which it cannot match text. Therefore, your academic judgment is still the best tool for plagiarism detection.