(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Shortcut citations in the methods section: Frequency, problems, and strategies for responsible reuse [1]

['Kai Standvoss', 'Department Of Psychiatry', 'Charité Universitätsmedizin Berlin', 'Corporate Member Of Freie Universität Berlin', 'Humboldt-Universität Zu Berlin', 'Berlin', 'Bernstein Center For Computational Neuroscience', 'Einstein Center For Neurosciences Berlin', 'Vartan Kazezian', 'Quest Center For Responsible Research']

Date: 2024-04

This exploratory study of papers in biology, neuroscience, and psychiatry revealed several important findings. First, citations are often used in methods sections. More than 90% of papers used a shortcut citation, explaining how a method was performed. Other common reasons for using citations in the methods included “who or what” citations, that give credit to the authors of another paper or specify what was used, and “why” citations, which provide context or a justification. Different methods evolve at different rates; however, citation age assessments suggested that some methods described in shortcut citations may no longer reflect current practice. The shortcut citation chains study showed that while shortcut citations can be used effectively, they can also create problems for readers seeking detailed methods. These problems included difficulty identifying the correct citation, accessing the cited materials, finding the cited method within the cited materials, and insufficient descriptions of the cited method. Following chains of shortcut citations to find methodological details was time-consuming, and each additional step in the chain can amplify the problems described in the previous sentence. Journals typically lack policies addressing methodological reporting. Fewer than one quarter of journals had policies addressing how authors should report methods that have been described previously, or address modifications of previously described methods. While some journals (18% to 43%, depending on the field) asked authors to provide sufficient methodological details to allow others to reproduce the method, most journals (57% to 81%) had no such policy.

Using shortcut citations to foster a culture of open and reproducible methods

When using shortcut citations, authors replace a section of their methods with a citation referring to another resource. The details contained within that resource are essential to implement the method. We therefore propose that methodological shortcut citations should meet higher standards than other types of citations. Box 1 outlines 3 proposed criteria that authors can use to determine whether a resource should be cited as a shortcut. The open access criterion may be controversial for some, as it suggests that scientists who have a paywalled resource that meets the other 2 criteria should cite this resource to give credit, and create a second, open access resource (e.g., a deposited protocol) to cite as a shortcut. Nevertheless, we believe that the open access criteria is particularly important. Unlike other types of citations, readers who want to implement the study methods will need to read shortcut citations. Paywalled shortcut citations systematically deprive some scientists of information needed to reproduce experiments.

Box 1. How to use methodological shortcut citations responsibly Methodological shortcut citations replace a section of the methods. Detailed methods are essential for those seeking to implement the method. We therefore propose that resources cited as methodological shortcuts should meet higher standards than materials cited for other purposes. We propose that authors should use 3 criteria to determine whether a paper, or another resource, should be used as a shortcut citation. Resources that do not meet these criteria can be cited to give credit to those who developed the method, but should not be used as shortcuts. Detailed description: The resource should provide enough detail about the method that was used to allow others, including those who have little prior experience with the method, to implement the method. Resources with sufficient detail to be shortcut citations might include protocols, methods papers that are recent enough to reflect the current practices, or original research articles with unusually detailed methods. Similar or identical method: The method described in the resource should be similar or identical to the method used by the authors. The authors should be able to describe any modifications to the methods in the methods section of their paper. Open access: Paywalled or inaccessible shortcut citations deprive some readers of the information needed to implement the method. This creates disproportionate obstacles for researchers with limited access to publications, including researchers in countries with limited research funding and scientists who are not affiliated with a major institution. Authors using shortcut citations should: Describe modifications in detail: Deviations from the published method should be clearly described, in enough detail to allow others to implement the method. When a protocol posted on a repository is cited as a shortcut citation, the authors can version or fork the protocol to share their exact methods. Versioning is sharing an updated version of one’s own protocol, whereas forking is sharing an adapted version of a protocol shared by others. Specify the exact location of the cited methods: This might include providing page numbers where the cited method was described for books and manuals or describing the method name and location in the cited resource.

When a resource does not meet the criteria proposed in Box 1, we recommend that authors cite the resource to give credit to its authors and use other strategies to share detailed methods. Options for sharing detailed methods include supplemental files, protocol repositories, and protocol journals (S6 Table). Authors who deposit or publish protocols can cite these resources as shortcut citations.

Sharing methods in supplemental files is suboptimal for several reasons. First, readers who have access to the paper may not have access to the supplement. While completing this research, we noticed that some publishers and journals make supplemental files freely available, whereas others do not. Papers obtained through online repositories or interlibrary loan programs may not include supplements. Second, methods in supplemental files are not findable. While scientists can quickly search protocol repositories to identify relevant protocols, there is no way to identify papers that contain detailed methods in the supplemental files. Third, supplemental methods are often written as general descriptions, which are typically less useful than the detailed, step-by-step protocols shared in protocol repositories and methods journals. Fourth, supplemental methods cannot be updated after publication; hence, this format is not useful for tracking the evolution of protocols within and across research groups. This is not a problem when investigators’ goal is to describe their methods for a completed study. Methods, however, are constantly evolving and are one of the most useful outputs that researchers create. Given this, it is more efficient for research teams to use platforms that allow versioning and forking when sharing research procedures, and cite the version of the method that was used for a particular research study.

A key advantage of protocol journals is that they allow authors to obtain credit for their methods development work through a peer-reviewed publication. One disadvantage is that the publication process takes time and effort. Furthermore, the published method cannot be updated—it only shows what one research group is doing at a single point in time. In contrast, protocol repositories allow authors to create living protocols by quickly sharing updated versions [21]. Some repositories also allow forking, in which scientists can post a modified version of a protocol posted by someone else [21]. Versions and forks are linked to the original protocol. This credits the original authors for their work, while allowing researchers to see how the protocol evolved. PLOS ONE offers a new “Lab protocol” publication [22] that combines the advantages of both formats. Lab protocol publications consist of a protocol, deposited on protocols.io [21], paired with a brief peer-reviewed publication that provides context for the protocol. Authors can demonstrate that the protocol works by citing previous publications that used the protocol, or providing new data obtained using the protocol.

The decision tree shown in Fig 8 may aid scientists in using shortcut citations responsibly, while reporting methods more transparently. Achieving these goals requires a shift in incentives—the scientific community needs to value protocols as a product of scientific work, on par with publications. While depositing methods takes time, it benefits authors in the long term. Depositing protocols in a repository that allows versioning and forking allows researchers to track changes as the protocol evolves and determine what version of a protocol was used for a particular publication. Furthermore, repositories provide long-term access to protocols even when researchers have not used the protocol in years, have moved to another lab or institution, or the person responsible for the protocol has left the research group. Detailed deposited protocols can also make it easier for new team members to learn protocols. Finally, other scientists can use and cite deposited protocols. This may facilitate collaborations with others who are building on one’s methods or help to identify previously unknown factors that affect protocol outcomes.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 8. Decision tree for the responsible use of shortcut citations. This decision tree helps authors to prepare reproducible methods sections by determining when to use shortcut citations and when to share detailed methods through protocol repositories or methods articles. https://doi.org/10.1371/journal.pbio.3002562.g008

While some scientists believe that one should always cite the first paper to use a method, others prefer to cite recent papers that fully describe methods that are similar to their own. These 2 beliefs reflect different reasons for citing a paper. Scientists who cite the original paper are using a “who or what” citation to give the authors of the cited paper credit for developing the method. In contrast, those who cite recent papers that describe current protocols recognize that these citations are more useful as shortcut citations. Fortunately, authors can do both. Authors should structure the citing sentence to help readers to distinguish between the “who or what” citation and the “shortcut” citation. For example, authors might write: “Experiments were performed using an updated version [citation 1] of the protocol originally developed by Smith and colleagues [citation 2].” This phrasing clearly demonstrates that the first citation is the shortcut citation, while the second citation is intended to give credit.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002562

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.