This instructor guide explains how Turnitin’s exclusion filters refine what is highlighted in the Similarity Report, so you can focus on meaningful matches and reduce noise.
In this guide:
- Exclude bibliography
- Exclude quoted text
- Exclude cited text
- Exclude small matches
- Exclude small sources
Terminology and availability for exclusions differ between the classic and newer versions of the Similarity Report. Each section calls out availability where relevant.
Instructors: To learn how to exclude content from the report, check out our Managing filters and exclusions guide.
Students: To learn how to exclude content from the report, check out our Using exclusions and filters guide.
Exclude bibliography
This filter automatically identifies and ignores reference lists or bibliographic sections so they don’t inflate the similarity percentage. Enable this filter when you want to exclude listed references and focus only on matches in the body of a submission.
For submissions written in languages other than English, common beginning and terminating phrases are detected to exclude a section dedicated to references.
If the algorithm is temporarily unavailable when a submission is processed, Turnitin uses the beginning and terminating phrases listed below as a fallback method to detect and exclude reference sections. If bibliographic matches still appear in the report, try resubmitting later to use the advanced exclusion option.
Beginning phrases
These words and phrases will signal the start of a reference section in a submission and begin excluding the content that follows.
|
|
|
Terminating phrases
If one of the words below appears after a detected list of references, the similarity check will resume for any text that appears after.
|
|
Exclude quoted text
This setting is called Exclude Quotes in the classic version of the Similarity Report, but it operates in the same manner as described here.
This filter ignores text enclosed in quotation marks or formatted as block quotes, preventing properly quoted material from appearing as matches. Use this filter when you want to assess students’ original writing rather than direct quotations they’ve clearly marked.
The report will ignore any matches that use the following quotation marks, as well as quotations within a quotation.
| "..." | '...' | «...» | »...« | „…“ | 《...》 | 〈...〉 | 『...』 |
The report will also exclude indented blocks of text when the original file is a .doc or .docx file.
Exclude cited text
This setting is not available in the classic Similarity Report. For information about upgrading, see our New Similarity Report Resource Center.
This filter removes properly cited material, including in-text citations and their associated sentences, from the Similarity Report. Apply this filter when you want to focus on uncited or incorrectly cited material and reduce matches to legitimate citations.
The report identifies common in-line citation methods and excludes them from the list of matches. Both the citation and associated text will be excluded. For quoted text, anything inside the quotes will be filtered out of the Similarity Report. For unquoted citations, the associated sentence will be filtered.
In the rare event that citation and reference detection are unavailable at the time of submission, the report will default to header-based exclusions. You can resubmit your paper later to take advantage of our machine learning–based exclusions.
If resubmitting isn’t possible or you prefer to keep the original submission details, you can instead exclude an individual source as a backup option.
Exclude small matches
This setting is not available for the classic Similarity Report. For information about upgrading, see our New Similarity Report Resource Center.
This setting filters out short matches below a specified number of words, helping you concentrate on more substantial similarities. Turn this on to remove “noise” from brief or generic phrases, especially in assignments that use repeated wording or technical terms.
After selecting this setting, enter a value for Set match exclusion threshold. By default, the threshold is 8 words, meaning that only matches 8 words or longer appear in the Similarity Report. You can increase this number, but 8 words is the minimum allowed value.
Exclude small sources
This exclusion option only exists in the classic Similarity Report. In newer versions, the Exclude small matches setting now provides a more useful and focused way to manage similarity results. For information about upgrading, see our New Similarity Report Resource Center.
This setting, labeled as Exclude sources that are less than, eliminates entire sources whose total matched content falls below a selected threshold (by word count or percentage). Use this to remove trivial sources contributing minimal overlap so you can prioritize significant sources.
When this option is enabled, Turnitin removes sources from the Similarity Report if their total matched content falls below a chosen threshold, for example, fewer than a specified number of words or a set percentage of the total submission. This helps reduce clutter from minor or insignificant sources, making it easier to focus on more substantial matches.
This option works at the source level, not the match level. Instead of hiding individual matched highlights on the submission, it excludes all highlighted matches from a single source, like a website, journal, or another submission.