Reducing the inadvertent spread of retracted science: Shaping a research and implementation agenda

This is the Working Draft (v3) of the recommendations for Reducing the Inadvertent Spread of Retracted Science (RISRS), a stakeholder collaboration funded by the Alfred P. Sloan Foundation. We would appreciate feedback in the Google Docs linked from the project website: https://infoqualitylab.org/projects/risrs2020/ or by email to jodi@illinois.edu. Our goal is to develop an agenda for reducing the inadvertent spread of retracted science. This includes identifying how the gatekeepers of scientific publications can monitor and disseminate the retraction status of a paper, and determining what other actions are feasible and relevant. EXECUTIVE SUMMARY Retraction is a mechanism for alerting readers to unreliable material, effectively removing from the published scientific and scholarly record articles that are deemed to be unreliable or seriously flawed whether due to misconduct or honest error. As noted in the Committee on Publication Ethics (COPE) Retraction Guidelines, retractions may also be used to address, “redundant publication, plagiarism, peer review manipulation, reuse of material or data without authorisation, copyright infringement or some other legal issue (e.g., libel, privacy, illegality), unethical research, and/or a failure to disclose a major competing interest that would have unduly influenced interpretations or recommendations” (COPE Council, 2019). Retracted papers insinuate themselves into the scientific publication network via citations both before and after retraction, which inadvertently propagates potentially faked data, fundamental errors, and unreproducible results, or can lead to misattribution of results or ideas (e.g., in cases of retraction due to dual publication, plagiarism, or ownership). Research over the past decade has identified a number of factors contributing to the unintentional spread of retracted research. Many retracted papers are not marked as retracted on publisher and aggregator sites, and retracted articles may still be found in readers’ PDF libraries, including in reference management systems such as Zotero, EndNote, and Mendeley. Most publishers do not systematically surveil bibliographies of submitted manuscripts, and most editors do not query whether a citation to a retracted paper is justified. When citing retracted papers, authors frequently do not indicate retraction status in bibliographies or in-text citations. Collaboration across diverse stakeholders in the academic publishing ecosystem is needed to reduce the inadvertent spread of retracted science. This is a critical moment for stakeholder dialogue: There is growing concern about the quality and reliability of scientific and scholarly information both within the research enterprise and in the broader public discourse; and the data needed to identify retracted research has become available, particularly from the Retraction Watch Database. The goal of the Reducing the Inadvertent Spread of Retracted Science: Shaping a Research and Implementation Agenda (RISRS) project is to develop an actionable agenda for reducing the inadvertent spread of retracted science. This includes identifying how the gatekeepers of scientific publications can monitor and disseminate retraction status and determining what other actions are feasible and relevant. Herein the term, paper, is used for the published item that is retracted. We focus on whole, published items such as journal articles, conference papers, book chapters, and monographs. Except where otherwise stated, we exclude items that are posted as opposed to published such as preprints or data deposits as well as specific objects within a publication such as figures and images. The RISRS process included an exploratory environment scan, a scoping review of empirical literature, and successive rounds of stakeholder consultation, culminating in a 3-part online workshop that brought together a diverse body of 70 stakeholders to engage in collaborative problem solving and dialogue. Workshop discussions were seeded by materials derived from stakeholder interviews and short original discussion pieces contributed by stakeholders. The online workshop resulted in a set of recommendations to address the complexities of retracted research throughout the scholarly communications ecosystem. The recommendations below are being circulated for further refinement (e.g., through the draft you are now reading) with the aim of producing final recommendations in June 2021. The RISRS team will continue to solicit feedback from across the scholarly communications ecosystem, through discussions about a NISO Work Item and presentations this spring to the Society for Scholarly Publishing and other groups. We welcome your feedback to refine the recommendations and the implementation agenda. For instance, you might help form a professional working group to further develop or refine these recommendations; present about retraction and related issues at professional and academic meetings; take on an implementation or policy project; or outline further research to be conducted. Recommendations Develop a systematic cross-industry approach to ensure the public availability of high-quality, consistent and timely information about retractions. Recommend a taxonomy of retraction categories/classifications and corresponding retraction metadata that can be adopted by all stakeholders. Develop best practices for coordinating the retraction process to enable timely, high-quality outcomes. Educate stakeholders about publication correction processes including retraction and about pre- and post-publication stewardship of the scholarly record. Feedback requested Your feedback will help shape this document. The proposed schedule for ongoing feedback is: Apr 27 Version 3 circulated publicly for last look  May 28 Deadline for comments, suggestions  June 15 Recommendations completed We invite you to provide additional feedback to help shape the scope of the recommendations. This version is being circulated publicly as a preprint to maximize the feedback we receive. We would appreciate feedback in the Google Docs linked from the project website https://infoqualitylab.org/projects/risrs2020/ or by email to jodi@illinois.edu. An updated version can be viewed here .