Retracted references

It is vital for IPBES products to be based on solid evidence in order to safeguard their quality and reputation. Using retracted publications in the assessment can jeopardise the credibility and reputation of the platform, the assessment reports and the authors. It is therefore important that the assessment authors and technical support units for the assessments are aware of the recommendations regarding the handling of retracted publications.

What are retracted publications?

Retracted publications are any type of material that has since their publication been withdrawn. This concerns not only publications in scientific journals, but also data deposits in e.g. Zenodo or other repositories. Here, mainly scientific journals will be addressed, but similar caution needs to be applied to any source which could be retracted.

Publications can be retracted for a number of reasons which can be broadly placed into one of two categories: honest error, for example when there are errors in the data or analyses, and misconduct, for instance when data was falsified or there are undisclosed conflicts of interest. When publications are retracted, a retraction note is published in the original source of the reference (i.e., academic journal), which includes the date and reason for the retraction. The retraction of publications can be unknown to its readers and users, for instance when publications were downloaded on a computer.

How to identify retracted publications?

Since 2019, Zotero integrated the Retraction Watch database plugin, which flags detected retracted publications. When a retracted publication occurs in a Zotero library, a ‘Retracted Items’ folder is automatically added to the library. In this folder, the retracted publications can be selected, which shows a red bar at the top with ‘This work has been retracted’, followed by the reasons for retraction and the link to the retraction notice. When a retracted reference is cited using the Zotero plugin in Word for an in-text citation, a warning that the reference is retracted pops up. This last warning only occurs at the time of the citation, there is no warning pop up when a reference has been cited already before the retraction.

The handling of retracted publications

IPBES strongly advises against the use of retracted publications, regardless of the reason for retraction: if retracted, it should not be considered to be a reliable, ethical or legal source. However, the ultimate decision is to be made by the coordinating lead authors in agreement with the technical support unit for the assessment.

Therefore, when a retracted reference is identified it should be communicated to the technical support unit for the assessment as soon as possible. If a retracted reference is identified in the assessment period up to the delivery of the third order draft, authors should replace the citation wherever possible, or delete the section for which the retracted reference was cited. If a cited publication is retracted after the delivery of the third order draft or the authors decide to cite the retracted publication, the doi of the retraction note should be added to the bibliography, as well as the date and source where the retraction note was published and preferably a justification explaining why the publication was used when its retraction was known.