Turnitin Similarity Report, also called the Turnitin Originality Report, is a report that detects accidental or on-purpose plagiarism. iThenticate is a product of Turnitin that is basically designed for academic authors who want to submit manuscripts to scientific journals and is not used for classroom use.

Comparison of Turnitin to iThenticate
Comparison of Turnitin to iThenticate

iThenticate report, also known as CrossCheck or Similarity Check, is used before publishing a study in a scientific journal, submitting a thesis or dissertation, or submitting a grant proposal.

In this task, I used iThenticate as part of Turnitin. This article applies to Turnitin and iThenticate because they use the same database and are under the same ownership. However, their interpretations are different. Therefore, please refer to a wide range of frequently asked questions to get a deeper understanding of the Turnitin and iThenticate similarity reports.

Even though the Turnitin website tells authors that the score might normally be up to 20%, most Journals do not accept any iThenticate similarity rate above 10%!

Turnitin Report as well as the iThenticate Report can be understood as “plagiarism detectors” for non-academic people.

Task

In March 2023, a Turkish academician applied for our “manuscript translation services.” The client also wanted us to submit a report about the Journal, to which they would submit their research article.

Price of the Translation and Journal Adaptation

This was a service provided by our Turkish company, and we applied Turkish agency prices, which were $225 for the manuscript translation to English (6600 words) and $125 for the adaptation of the manuscript to a specific scientific journal, including the three revisions that I am explaining below.

Turnitin Similarity Report Score from 32 to 9
Turnitin Similarity Report Score from 32 to 9

Preparations

I first read and summarized the concerned journal’s guidelines and detected that the manuscript had missing titles, was not anonymous, and exceeded the limits for word count and the number of references. Besides, the article was not compatible with any of the SEO rules (wrong location of keywords, very long paragraphs, etc.). I have known this client since 2012; therefore, I accepted the task despite my busy workflow.

It took a whole week to prepare the manuscript’s Turkish version. We had high phone traffic during this phase. After three days, I sent the client an email explaining what to change. The client worked on her article for 2 days and sent it back to me, and I sent it again because of the missing titles, etc.

I am a real perfectionist, and I didn’t start translating until she adapted her manuscript to the Journal’s guidelines. In addition to these revisions, she had to delete 2000 words from the main article and remove 12 references.

To be honest, preparing the article for the Journal was much more difficult than translating the Turkish article into English.

The Academic Translation Phase

I translated the document in three days. Meanwhile, I used QuillBot to check grammar and paraphrased some sentences when I saw a necessity. There was no problem with the translation phase. The original Turkish text was now 4100 words, and the Tables and Figures were around 2500 words. The price of the translation was calculated based on the original article’s 6600 words, including tables and figures, excluding references. I also edited the references properly.

Revisions: Adjusting to the Turnitin Similarity Report

Adjusting to the Turnitin Similarity Report

Ok, now it is the phase where the translator waits for the revisions. Revisions?? Weird, right? Revisions are not typical revisions. In the academic translation world, revisions mean “adjusting to the Turnitin similarity report.” In our case, it was adjusting to the iThenticate similarity report.

Initial originality report check

The first Turnitin similarity rate was 32%, excluding the references, tables, and figures. When we add those, the similarity increased by +9 points! This originality report check was a complete disaster.

I asked and learned the reason (mostly self-quotations). At first, it looked impossible to decrease this rate because the article contained lots of quotations, and the scale items in between quotation marks should be written as they are in the original study. Even though Turnitin does not take the Table items, it detected similarity because the article was using an already published scale’s items in the current study. It may look confusing to you, but actually, it is not. This happens when you adapt a published study’s scale to another language.

First Revision: Turnitin Similarity Report 22%

I changed the sentences one by one without changing the basic meaning of the original Turkish article. Since almost half of the article had red or orange colors, it took me two more days to send my first revision. Meanwhile, I intensively used the Quillbot paraphrasing tool.

Excluding references, tables, and figures, I could manage to decrease the score from 36 to 22. However, I asked some academicians on Quora and learned that it is still a high score.

Second Revision: Turnitin Similarity Rate Decreased to 14%

Because I was aware that the Turnitin similarity report also considered sentence structure and that academicians used Quillbot more frequently than I had anticipated, the second revision was a complete waste of time for me. The task here was not only to replace words with their synonyms. The Turnitin software was detecting any similarity that could easily be used in other articles, such as “the study conducted by xxx et al., etc. I mean, even the phrase “et al.” or “the” was still red.

Third Revision: Turnitin Similarity Report Decreased to 9%

Until this revision, I was using a synonym dictionary to make the changes, and I was also using the Quillbot paraphrasing tool. In this revision, I completely made the changes manually. I informed the client about the problematic sentences and references. She worked on those sentences and deleted some of them. She also deleted a Table that was creating big trouble.

The final score for the Turnitin originality report was 9%, including tables and figures and excluding references. This was the hardest phase, and it took me two more days to complete it.

Last Revision: The Turnitin Similarity Issue is Over

I don’t want to write the whole story here.

At first glance, the translation was an easy one for me. And I thought I would submit it in one week. But I ended up working on that manuscript for three weeks.

The client offered me almost double the price because of the effort she witnessed.

However, at some stage, money loses importance to me, and I took this task as a challenge. I didn’t charge any extra for all of these unanticipated tasks, and I didn’t know the dangers of qualitative research papers in terms of Turnitin similarity reports.

We submitted the paper.

Turnitin originality report
Turnitin originality report

The answer arrived in one week. The research article was not accepted by the Journal the client selected. However, the platform offered easy submission to seven more journals.

Since the double-blind peer review is over and the reason for the Journal not to accept the paper is NOT the similarity issue, I can say that a 9% rate is acceptable by the majority of reputable journals. 

A Q2 journal with an impact factor of 2.11 published the article. Is this a success? To be honest, this is not my business anymore. Nevertheless, even though it took three months, I’m glad that an international journal published the manuscript. The client had applied to another Q1 journal, and when the second Q1 journal did not publish, she tried a Q2 journal, and they accepted the article.

Please note that I wrote this text to explain my own experiences, and the ideas presented here belong to me and do not bind Turnitin, iThenticate, or any official Journal’s guidelines.

My Final Words

To decrease the rate of Turnitin similarity reports, I used no specific techniques. In other words, I tried some techniques, but they didn’t work. At last, I made all the changes manually. I used only a synonym dictionary when necessary.

Paraphrasing with Quillbot did not decrease the score significantly. I also did two small revisions and tested the basic two keywords by selecting these two keywords from the whole document and changing them with another word. These efforts decreased the score substantially. However, we could not touch or change the keywords, so we could not use this option. Despite this, I could manage the decrease in Turnitin score from 32% to 9%, including tables and figures but excluding the references list.

If you wish to read Pelotek’s real experiences, follow our “Case Studies” category. You may also want to read USCIS-approved translators from Turkey.

Frequently Asked Questions on Turnitin

The optimum similarity rate for Turnitin is below 15%.

As I tried to explain above, a good Turnitin score depends on the type of your article. As a general rule of thumb, try not to exceed 15% similarity.

As stated on Turnitin’s website, any score above 25% may imply high plagiarism. However, this also means that one-fourth of your article is already published. You should give yourself a break and honestly think about the originality of your article.

You should read each highlighted similarity on your Turnitin report. Make changes manually, one by one. Insert proper quotations. Change the row of listed items or the sentence structure. Make sure you are commenting in your own words.

No. It means that you didn’t utilize any of the scientific publications in your field. Although your article is totally unique, communication in science is another factor to consider when writing a manuscript. To me, a 0% Turnitin similarity rate implies that the authors might not check all the relevant articles in their fields.

No, the Turnitin report does not detect paraphrasing as long as you are trying to cheat on the system (like stealing someone else’s study and changing it by paraphrasing). You can comfortably use any paraphrasing software as long as you do not change the original meaning.

If you are planning not to submit the article in a short time, or, in other words, if you will work on the article and plan to publish it later, I advise you to delete your article from the Turnitin database. Turnitin does not care who the author is. It will detect similarity as long as you upload your manuscript to its database.

In addition, if your article was not accepted and you plan to resubmit it after making major changes, you still have to delete your manuscript from Turnitin’s database.

Turnitin keeps papers forever. You should consider deleting your paper if it is a draft or if you consider submitting it at a later date or under a different title.

Frequently Asked Questions on iThenticate

Based on my experience, the optimum similarity rate should be between 5% and 10%. However, if you have many quotations or a manuscript that relies heavily on other studies, 15-20% similarity is acceptable. Try not to exceed 5% similarity with one individual reference.

Based on my experience, any score between 5% and 10% is a good score.

You should read each highlighted similarity on your Turnitin report. Get rid of unnecessary references and quotes. Avoid replications within the text. Use synonyms for the highlighted words and change the sentence structure one by one.

No, you don’t need to delete your article from the iThenticate database because the systems are different. On Turnitin, there is no way back unless you make an effort to delete your article. On the other hand, iThenticate allows you to keep your article offline.

No, iThenticate is paid software. However, universities may get a license, and generally, faculty and graduate students use it for free. If you don’t have this option, you can buy credits to use iThenticate on their website. The pricing changes based on the number of documents and the word count of each document. For example, 100 credits cost $100 and can be used to check up to 25 documents of 400 words each.

Authors, publishers, and organizations use iThenticate to check submitted materials for originality and proper citation. The iThenticate Similarity Report will highlight matched content in a document. Focus on the red-painted ones because they indicate a high level of duplicate content. You can employ some controls, such as excluding quotes, excluding references, or excluding small matches. I kindly advise you to exclude only the bibliography or references sections.

The answer really depends on the various policies of various journals or whoever is publishing your article. In my experience, Q1 journals generally accept lower similarity rates, such as those below 5%. However, I believe that the optimum similarity rate should be a maximum of 15%. Exclude your references but include tables and figures.

Sule Kalmis

Sourcing Expert

Experienced business consultant, translator, and co-owner of a 14-year-old chemical company.

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *