Similarity Reports

Every manuscript that is submitted to JGR-Space Physics undergoes a cross-check for identical text in previously published scientific content on the web. Specifically, we send it to a company called iThenticate to generate a “similarity report.” I have a previous post on self-plagiarism. This post is about understanding and interpreting those reports.

The iThenticate software scans the document for strings of characters that match those found in other scholarly publications. It specifically excludes the reference list, for which the entries should be identical between publications. However, scan unfortunately includes the affiliations, which are also always the same.

We (the editors) look at the similarity report for every paper that we manage in the system. This is part of our initial assessment of the manuscript: determining whether to send it out for review. The other big points we assess are whether the English usage in the text is adequate (too many errors and we will send it back to the authors for revision) and whether the paper meets the bar of an original contribution to the field.

Within the similarity report, the big issue that we are looking for is overlap of entire sentences or even paragraphs with a previous source. This is the thing that will get your paper rejected without review. Little things, like the affiliations, specific paper callouts, or standard phrases, are also identified by the software as identical overlap. These small things are discounted by us and not included in our assessment. I have never asked AGU about altering the iThenticate cross-check settings, but perhaps that could be done to omit these small/meaningless overlaps from being highlighted in the report. Until then, however, we all will just have to ignore those places and focus only on the big overlap sections.

So, when you get a manuscript rejected due to a high cross check, please look through the similarity report and find those places where entire sentences are highlighted. These are the places that we want you to rewrite in new wording. We realize that there is often significant overlap with previous papers, especially in the methodology section where the description of the instrument, model, and/or processing scheme is the same as that used in another study. Regardless, you have to rewrite it. It can (it many cases, should) say the same thing as before, but the sentences cannot be verbatim from another paper. We hope that you are able to revise these places of the text very quickly and resubmit within a week or two.

Advertisements

9 thoughts on “Similarity Reports

  1. “We realize that there is often significant overlap with previous papers, especially in the methodology section where the description of the instrument, model, and/or processing scheme is the same as that used in another study. Regardless, you have to rewrite it. It can (it many cases, should) say the same thing as before, but the sentences cannot be verbatim from another paper. We hope that you are able to revise these places of the text very quickly and resubmit within a week or two.”

    Does this not strike you as a little ridiculous? When writing a manuscript we look for the best words to describe a model/dataset/instrument and use them. Often it has gone through multiple drafts to generate the most succinct and clear form of expressing those descriptions. This rule introduces additional work for everyone involved for no real benefit. Actually the more I think about it the more I realise it is beyond ridiculous. If it can (or indeed should) say the same thing as before then why not let it say the exact same thing as before.

    I can see where ‘self plagiarism’ becomes a problem when whole chunks of a paper are reproduced or passed off as new (fraud) but this is quite some way removed from that. Fundamentally what is wrong with repeating purely descriptive text within different papers if they are referring to exactly the same thing?

    Is this essentially a copyright issue?

    • Thank you for the comment. Really quickly: a single sentence is not enough to warrant rejection. I will draft a second post on this topic with more details.

      As for methodology descriptions, if it has been described in detail elsewhere, then the authors should just refer to that other paper. There is no need to copy that other paper’s text verbatim in the new manuscript, except the few most relevant details that help the reader understand the new study. This should just be a paraphrase and, yes, rewritten fresh for every follow-on paper.

      A side note: this policy is not new. AGU has been running similarity report on all JGR-Space manuscripts since early 2011. The blog post is letting you know our editorial board’s implementation of it…i.e., we don’t sweat the small stuff, only (just like you say) whole chunks being reproduced.

      And, finally, yes, it is a copyright issue. However, it is not only that but also an ethical integrity issue.

  2. “However, it is not only that but also an ethical integrity issue.”
    But is it really at the level we are discussing? How is what we have described so far unethical?

    “As for methodology descriptions, if it has been described in detail elsewhere, then the authors should just refer to that other paper.”
    That would be great and I agree with this but have also been in the situation where this was not enough to appease referees who wanted fuller descriptions included in the paper regardless of them being described elsewhere.

    “A side note: this policy is not new. AGU has been running similarity report on all JGR-Space manuscripts since early 2011. ”
    Indeed and across JGR as a whole. A colleague ran afoul of this issue in JGR-Atmospheres last year when they were forced to rewrite several sentences (not all in a single paragraph) to get the paper to review stage. The level of nitpickery involved in that was incredible. From what you have written I think (hope) it would not have been as much of an issue in JGR-Space.

    I’m afraid that you have not convinced me this is a reasonable policy. I still feel it is misguided overkill.

  3. this is really ridiculous!
    Shouldn’t a scientific paper be self-contained?

    If I read a paper I want to find a detailed description of the methodology and the new results. On the same paper. I don’t want to jump around from paper to paper, to find the methodology description.

    What is the big issue with having the methodology copied (verbatim) in 10 different papers? What is wrong with that? Where the ethical integrity comes in?

    And why, in choosing between the quick and smart way and the dumb and tedious, are you opting for the latter?

  4. I decided to reply to the above question:
    “What is the big issue with having the methodology copied (verbatim) in 10 different papers? What is wrong with that?”

    Nothing is wrong with having the methodology copied (verbatim) in 10 different papers, just if the text is copied verbatim, quotation marks should be used with appropriate references.
    I can ask you a response question: Is there anything wrong with the requirement to use quotation marks for the verbatim text copy?

    I have been one of JGR-Space Physics editors for only about 5 months. In my experience the vast majority of rejections based on similarity reports are not because of the absence of the quotation marks but due to the original source not being clearly referenced at the end of the copied text.

    In relation to “the self-plagiarism” I did not encounter one. This is because the self plagiarism assumes that the set of authors is exactly the same in both papers. Obviously it is possible, but I did not encounter such a case during these 5 months, so it should be rather rare. If author sets differ, even if the first author is the same this is not a self plagiarism.

    I want to remind that editors are only part time editors they are also authors and understand all the concerns that you are expressing. In any particular case if you think that your paper is wrongly rejected you can always send us your concerns and we will take them into account.

  5. Pingback: Interpreting a Similarity Report | Notes from the JGR-Space Physics Editor-in-Chief

  6. Pingback: Avoiding Plagiarism in the Methodology Section | Notes from the JGR-Space Physics Editor-in-Chief

  7. Pingback: Tools For Avoiding Plagiarism | Notes from the JGR-Space Physics Editor-in-Chief

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s