Skip links
erudit.org

FAQ on Post-Dissemination Corrections

Érudit’s Policy on Post-Dissemination Corrections has been in effect since October 1, 2024. Here, we respond to the most frequently asked questions about that policy, but please contact us at edition@erudit.org for further information. Also, kindly note that we exclusively accept requests from journal editorial teams; we invite authors to directly contact the journal that published their article.

What is the nature of the problem or the error you are trying to correct?

  • Minor typo or error: see question A.1
  • Major error: see question A.2
  • An author has changed their name: see question A.3
  • Plagiarism and fraud: refer to Section C.

A. Correction in an Article

The best practice is to publish an erratum to inform readers of the typo. You can consult the section B. on errata in this FAQ to learn more about the procedure. Érudit does not make corrections in the text body of articles (HTML or PDF) or their metadata, unless the typo affects metadata key to the visibility of the article or its ability to be cited. Therefore, only corrections to the following metadata will be accepted (fees apply, and we will send an estimate for your approval before making the correction):

  • volume(s)
  • issue(s)
  • year(s)
  • title
  • subtitle
  • parallel title(s)
  • parallel subtitle(s)
  • references(s) to listed work(s)
  • author(s)
  • page number(s)

For transparency purposes, we add a note to the article to highlight any modifications made. If your journal has a printed version, an erratum would ideally be added in a future issue to inform your readership. In such an erratum, you can state that the digital version on Érudit has also been corrected.

To assist us in making the correction on the platform, please provide the following information to production@erudit.org (or ULavalErudit@bibl.ulaval.ca if you usually communicate with the production centre at Université Laval):

  • URL of the article on Érudit
  • the erroneous metadata to correct
  • the correction to be made
  • the article’s PDF file with the correction, if relevant

Depending on the correction to be made, several scenarios might apply:

a. If the scientific validity of the article is at stake, a retraction is preferable. You can consult the section C. on retractions in this FAQ to learn more about the procedure.

b. However, if the article’s conclusions are not compromised, meaning that the error does not affect the conclusions of the research, publishing an erratum to inform the readers of the error is recommended. You can consult the section B. on errata in this FAQ to learn more about the procedure.

c. Finally, if the error affects a metadata element that is fundamental to the ability of the article to be cited, a direct correction on the platform is possible, with a note to highlight it. You can find more detail in the answer to question A.1 in the FAQ to learn more about the procedure. Publishing an erratum is still worthwhile in this context, especially for journals with a printed version.

In line with the recommendations put out by COPE in 2021 for a more inclusive vision of publishing, Érudit is committed to reflecting name changes on its platform, no matter the metadata being changed or the reason for the change. The modifications will be made with discretion and within reasonable lead time, at the author’s request.

To assist us in making the correction on the platform, please provide the following information to us at production@erudit.org (or ULavalErudit@bibl.ulaval.ca if you usually communicate with the production centre at Université Laval):

  • URL of the article on Érudit
  • the name of the author to modify
  • the new name of the author
  • the metadata and paragraphs to modify
  • the article’s PDF file with the modification, if relevant

To help match authors with their article, we recommend that journals integrate ORCID Identifiers into their publishing policies.

Érudit cannot guarantee that every indexing tool will integrate modifications, as they each manage post-dissemination corrections differently. Please contact us at edition@erudit.org for more information. Here are two examples:

If Érudit is responsible for creating your DOI within Crossref, we make sure to update the data at the DOI level. However, if your library or your publisher generates the DOI, you will need to make your request with them.

Based on the terms of our agreement with Google, the only situation that can lead to a quick update on Google Scholar is a correction tied to an author’s name change. In that case, we submit an update request for the notice. For any other modification to an article, Google updates the notices independently. Please note that the update can take several months.

B. Adding an Erratum to an Issue

It can seem easy and inconsequential to quickly and discreetly correct an error in an article disseminated online. However, since an article can be read, cited and redistributed as soon as it is published, we recommend avoiding making such corrections, either in the HTML or the PDF versions. When an article is modified, several versions of that article are created, leading to the possibility of non-matching citations from the same article and of confusion for its readers. We implemented our Policy on Post-Dissemination Corrections in October 2024 to prevent such situations.

Moreover, the directives to make a post-dissemination correction can vary from platform to platform depending on the features offered (for example, Érudit does not manage version histories). As it is impossible to guarantee that the correction will be reflected elsewhere, the best practice is to publish an erratum in a future issue to highlight the correction to the readers and improve the chances that indexing tools will integrate it.

On the erudit.org platform, the erratum and the articles to which it applies are linked to inform the readers of the correction (here’s an erratum on the platform as an example). The hyperlinked mention “See the erratum for this article” is added to the HTML and PDF versions of the article. A note with a hyperlink to the article is then added to the HTML version of the erratum.

There is no need to inform us when adding an erratum to your journal’s next issue. We will process the erratum as part of the production process for that issue.

An erratum must include the following elements:

  • the complete reference to the article to which the erratum applies
  • the errors
  • the correction

If the error was made in a table, the entire corrected table must appear in the erratum along with its title, its number, its legend and its source, if applicable. This also applies to any graphic or figure.

Here is a fictional example:

Please note that an error appears in the article “Cognitive Biases” by Dominique Malton, which appeared in Volume 10, Issue 2, 2023, of our journal on pages 7-20.

At the end of the fifth paragraph of page 8, “Erroneous sentence” should read as “Corrected sentence.”

Thank you for your understanding.

We would like to bring to your attention two errors made in the article “The Notion of Atychiphobia: Rise and Social Dissemination” by Lucille Beverly Lavoie-Gagnon, which appeared in Volume 2, Issue 1, 2005, of our journal on pages 32-42.

  • The name of the author reads “Lucile Beverlly,” when it should read “Lucille Beverly.” The correction has been made to the article published on Érudit.
  • Note 13 should have been removed.

Thank you for your understanding.

We are conscious that certain corrections need to be communicated without delay to your readers. It is therefore possible to add an erratum to the same issue of your journal on Érudit. Please note that additional fees apply and that you will receive an estimate for approval before we undertake the process.

To assist us in disseminating the erratum, please provide the following information to us at production@erudit.org (or ULavalErudit@bibl.ulaval.ca if you usually communicate with the production centre at Université Laval):

  • URL of the article to which the erratum applies on Érudit
  • a brief explanation of the problem
  • the PDF file of the erratum

On the erudit.org platform, the erratum and the articles to which it applies are linked to inform the readers of the correction (here’s an erratum on the platform as an example). A hyperlinked mention “See the erratum for this article” is added to the HTML and PDF versions of the article. A note with a hyperlink to the article is then added to the HTML version of the erratum.

There is no need to inform us when adding an erratum to a future issue. We will process the erratum as part of the production process for that issue.

C. Retracting an Article

A retraction identifies a published article as scientifically invalid, whether due to scientific fraud or an honest error. It is a healthy and normal mechanism to maintain the accuracy, credibility and integrity of scholarly literature. To respect current norms and best practices, it is important to retract an article, rather than delete it completely. This prevents confusion for those who might have read or cited the article since its publication.

COPE highlights the reasons that can lead to the retraction of an article, as well as reasons not to do so. Here are the scenarios requiring a retraction:

  • solid proof of inaccurate results (major error, fabricated or falsified data)
  • proof of plagiarism or non-authorized republication
  • copyright infringement or other legal and ethical problems
  • compromised or manipulated peer-review process
  • non-disclosure of a major conflict of interest

For more detailed retraction scenarios, Appendix B of the Retraction Watch Database User Guide is a good resource.

A retraction is not appropriate in the following cases:

  • The attribution of the article, the contribution of an author or the order of the authors is contested, but the article’s results are not.
  • The results remain trustworthy despite errors that can be corrected via a correction or an erratum (see section B. on errata of this FAQ to learn more about that process).
  • Proof is insufficient to justify a retraction, or the investigation is still ongoing. An expression of concern can be appropriate in that situation (see question C.5 about expressions of concern in this FAQ to learn more about this procedure).
  • Minor conflicts of interest that do not affect the article’s conclusions.

While it is preferable that the journal and the authors agree on the contents of a retraction notice, if an agreement between the parties cannot be quickly reached, a retraction notice can be published without the authors’ approval. The final decision to retract an article ultimately belongs to the Editor-in-Chief, who is responsible for guaranteeing the integrity of the scientific content published in their journal. This is one reason why it is recommended to establish a policy on retractions and post-dissemination corrections for your journal (for example, see the Canadian Journal of Bioethics’ policy and that of PLOS One.)

The retraction notice can be written by the author of the retracted article, by the Editor-in-Chief, or by someone chosen by the journal’s editorial committee.

The title of the retraction notice must contain the title, the parallel titles (if applicable) and the name of the author of the article being retracted. Here is the format: “RETRACTION OF: [title of the retracted article]/[parallel title(s)], by [author]”

The retraction notice must contain the following elements:

  • the complete reference to the article with the DOI as a hyperlink
  • the name of the entity that initiated the retraction process (author, journal or institution)
  • the reason for the retraction, in clear, objective and unambiguous language
  • the complete chronology of events (date of publication of the article, start date of the retraction process, date of publication of the retraction notice, etc.)
  • the concerned part of the article, the erroneous data, the invalid conclusions
  • whether an institutional investigation or legal action is underway

Here is a fictional example:

RETRACTION OF: Les biais cognitifs / Cognitive Biases / Los sesgos cognitivos, by Dominique Malton

In conformity with the recommendations on retractions from the Committee on Publication Ethics (COPE), the article “Cognitive Biases” by Dominique Malton, which appeared in Volume 10 Issue 2, 2023, pp 7-20, is being retracted (https://doi.org/10.7202/0000000ar/).

A reader has brought to the attention of the editorial team at Apprentissage et neurosciences that a large part of the Analysis section of the article, mainly paragraphs 18, 19 and 22, is the result of plagiarism from another article on the subject (TREMBLAY, J. Pour comprendre les biais cognitifs : une revue de littérature. Revue d’études philosophiques et cognitives, vol. 8 issue 1, 2019, pp 108-132).

We first contacted the author in March 2024. We deemed the explanations provided by the author unsatisfactory, and we contacted her institution in April 2024. An institutional investigation, which had begun in May 2024, concluded that the article was indeed the result of plagiarism. We published the retraction notice on July 8, 2024, in Volume 11 Issue 2 of our journal.

Alix Magloire, Editor-in-Chief of Apprentissages et neurosciences

RETRACTION OF: Émergence et diffusion sociale de la notion d’atychiphobie / The notion of atychiphobia: Rise and social dissemination, by Lucille Beverly Lavoie-Gagnon

Upon request from the author and with the support of the editorial team of the journal Approches sociologiques du numérique, the article “The notion of atychiphobia” by Lucille B. Lavoie-Gagnon, which appeared in Volume 2 Issue 1, 2005, pp 32-42, was retracted on May 22, 2024 (https://doi.org/10.7202/0000000ar/).

A few months after the publication of the article in January 2024, the author identified major errors in her research data, mainly presented in Table 3, that invalidate most of the article’s conclusions. Consequently, we must retract the article.

To learn more about our retraction policy: Journal Policy

Suzanne Bélanger, Journal Director of Approches sociologiques du numérique

If an article in your journal disseminated on Érudit is retracted, you must notify us as soon as possible. Please provide the following information to us at production@erudit.org (or ULavalErudit@bibl.ulaval.ca if you usually communicate with the production centre at Université Laval):

  • URL of the article being retracted on Érudit
  • a brief explanation of the situation
  • the PDF file of the retraction notice

The digital version of the retraction notice should be published as soon as possible. It will therefore be added to the summary of your journal’s latest issue on the platform.

NISO has provided rules on the correct way of identifying retracted articles in a database. The following modifications will be made by Érudit’s team to the retracted article:

  • the mention “RÉTRACTÉ / RETRACTED” is added to the article’s title.
  • the mention “RÉTRACTÉ / RETRACTED” is watermarked on all pages of the PDF version.
  • The date of the retraction is added to the page footer of every page of the PDF version and as a note in the HTML version.
  • if a complete HTML version of the article is available, it will be deleted and only the metadata will remain available.

Moreover, the retraction notice and its related article are linked (see this article). The hyperlinked mention “See the retraction notice for this article” is added to the header of the HTML version of the article. A note with a hyperlink to the article is then added to the HTML version of the retraction notice.

To make the retraction effective, these changes should be reflected in every digital version of the article. It is the journal’s responsibility to ensure that the retracted article is similarly modified along these guidelines on the journal’s website, in OJS and on every other platform.

Note that it is possible, in exceptional cases, to completely delete the HTML content and the PDF version of an article for legal or public safety reasons. However, the notice remains present in the minimal metadata, if available: title, subtitle, parallel title(s), parallel subtitle(s), author(s), affiliation(s), ORCID, DOI, page numbers, license and publishing date.

An expression of concern is a notice that informs the readership that the conclusions of an article could be invalid, but that there is currently insufficient proof to proceed with a retraction. For example, if an article is in the process of being retracted, but an institutional investigation is delaying such a measure, it might be appropriate to quickly publish an expression of concern while waiting for the completion of the investigation.

This is a rather recent practice. The procedure to publish an expression of concern is consequently less standardized than for a retraction.

The expression of concern contains essentially the same elements as a notice of retraction. You can consult question C.3 on how to write a retraction notice in this FAQ to learn more about the procedure.

Here is a fictional example:

Expression of concern: Les biais cognitifs / Cognitive Biases / Los sesgos cognitivos, by Dominique Malton

May 2, 2024

This is an expression of concern for the article “Cognitive Biases” by Dominique Malton, which appeared in Volume 10 Issue 2, 2023, pp 7-20 (https://doi.org/10.7202/0000099ar/).

Following a request by our editorial committee last November, an investigation pertaining to a potential problem with the research methodology of the article was initiated by the author’s institution. We are therefore publishing this expression of concern to inform readers that the conclusions of this study might prove invalid. Based on the information shared with us, the committee responsible for the investigation is mainly considering a potential flaw in the consent form sent to the study’s participants. We would like to highlight that the author is not necessarily at fault.

This notice will be updated upon completion of the investigation.

Alix Magloire, Editor in Chief of Apprentissages et neurosciences

To assist us in disseminating the expression of concern, please provide the following information to us at production@erudit.org (or ULavalErudit@bibl.ulaval.ca if you usually communicate with the production centre at Université Laval):

  • URL of the article to which the expression of concern applies on Érudit
  • a brief explanation of the situation
  • the PDF file of the expression of concern