Implementing CRediT: An interview with Cell Press’s Gabriel Harp

The following is an interview with Gabriel Harp, Senior Product Manager at Cell Press. Cell Press’s research journals have been leading a pilot allowing authors to optionally make use of the CRediT Contributor Roles Taxonomy, which will be integrated with Editorial Manager version 13.0, coming later this year. He describes the experience here:

Q: How did Cell Press come to be involved with the CRediT taxonomy?

The genesis of the taxonomy was a 2012 workshop jointly hosted by Harvard University and the Wellcome Trust. Emilie Marcus, the Editor-in-Chief of Cell and CEO of Cell Press, was involved in subsequent discussions that led to the creation of the taxonomy. At the time the taxonomy was rolled out last spring, the editorial team reached out and asked me to coordinate making contributor roles an option for authors. The idea has always been that this project is cross-organizational, cross-publisher – the CRediT team talked with funding bodies, publishers, authors, and others to carefully develop the taxonomy. When it came time to decide our approach to implementation, we were thinking, “Do we want to wait and see what others do? Do we want to require use of the taxonomy?” We opted for the middle ground: making the taxonomy a recommended option for authors and actively seeking their input along the way.

Q: You piloted the CRediT taxonomy on some of Cell Press’s journals. Could you explain how you went about selecting which titles to include in the pilot, and what exactly the pilot entailed?

We took a straightforward approach: we piloted the use of the taxonomy in all Cell Press research journals. The only exception was for the journals that we publish in partnership with societies; in those cases, each Society decided whether or not to join the pilot.

The pilot consisted of the following steps:
• Determine our approach to publishing the taxonomy
• Explain and promote the taxonomy
• Reach out informally to authors about it
• Formally survey authors who had used it
• Track all papers that used the taxonomy, as well as a breakdown of the rolls identified for each paper

At the outset, we decided to recommend the optional use of the CRediT taxonomy within the “Author Contributions” section of a manuscript. This section, which appears alongside the Acknowledgments, was itself optional in most cases. In other words, we had an optional taxonomy within an optional contributions section. Beginning in May 2015, we let authors know about this option via several channels: online guidelines for authors, letters sent during the revision process, a post on the CrossTalk blog, and informal communications with authors. The first paper to use the taxonomy appeared in May, and uptake has steadily increased since then. In December we decided to require the inclusion of the Author Contributions section, which has come to be seen as an important complement to the author list and the Acknowledgments. Since then, the frequency of the use of the taxonomy has doubled.

Q: Which roles did the pilot impact? The taxonomy applies to authors and other contributors. Who else within the workflow is impacted by the implementation, and how?

Thus far, the taxonomy in Cell Press papers has applied only to authors. Other contributors are mentioned in Acknowledgments sections. In terms of workflow, we needed to make sure that everyone on the Cell Press side – editors, journal associates, copyeditors, suppliers – was familiar with the existence of the taxonomy and was prepared to see Author Contributions sections that differ in format. Some would continue to take the standard, prose-based, descriptive approach, whereas those that use the CRediT approach would include author initials and the taxonomy terms.

We have a great team of copyeditors, most of them in-house, who place a lot of pride in copyediting these scientific manuscripts. We take care to ensure consistency, house style, and correct grammar and spelling – all while being extremely careful not to make any changes that could alter the meaning or voice of the author. We had to prepare the team that there could be two very different Author Contributions sections: one with the standard elements of a prose paragraph, the other with a list of terms and author initials. Because we check manuscripts carefully at various stages of the post-acceptance process, we needed to make sure that everybody – from the scientific editors handling the papers to the copyeditors and proofreaders – was familiar with the taxonomy and would not be surprised to see it in a paper. In this respect, internal communication was just as important as external communication. And implementation went much more smoothly than anticipated. We expected lots of questions from authors and internal teams, but that didn’t happen. By and large, I’ve been surprised by the rate of adoption, and the smoothness of the process.CRediT before & after

Q: What are the benefits you observed during the pilot implementation?

Thus far, authors who have used the taxonomy tend to view it favorably. To my surprise and relief, they generally report that the taxonomy is easy to use. This is really important to recognize. The publication of a research paper is often the culmination of years of work. As a publisher, we strive to make the journey as smooth as possible for our authors. Thus, whenever we introduce a change or a new step, whether big or small, we consider the potential impact on our authors. In this case, what we are hearing so far is that authors who use the taxonomy find it easy to apply, clear to understand, and important in its standardization of contributor roles. We conducted a survey of the first 100 authors who used the taxonomy at Cell Press, and I’ve discussed the results of the survey here.

Q: What questions or unforeseen issues came up during the pilot? How did you resolve them?

Not many! Throughout the course of the pilot we did encounter some questions: Should the taxonomy terms be written with capital letters or lowercase? Should we ask for clarification from authors who use the term “Writing” rather than the two options laid out in the taxonomy, “Writing – Original Draft” and “Writing – Review & Editing”? Should we allow prose descriptions in addition to taxonomy terms? For all of these cases, we decided to defer to what authors provide. And our reasoning was simple: this is a pilot, and we are keen to observe – rather than to stipulate — how the taxonomy is put to use.

Q: Will you be making the recommendation to roll the taxonomy out more widely across journals within the Cell Press portfolio?

As I mentioned, we have already rolled the taxonomy out about as far as we can at Cell Press. The next conceivable extension would be to make its use mandatory, but we are not interested in doing so at this time. From our perspective, an optional approach to the taxonomy is best. Despite the steady uptake, the majority of papers do not use the taxonomy. The last thing we would want is to force a structure upon authors who might not be comfortable with it.

Q: What suggestions or best practices would you share with other editors considering piloting or adopting the CRediT taxonomy within their workflows?

Talk to authors, talk to readers, talk to funding agencies, talk to tenure committees – talk to anyone invested in the complex systems that swirl around credit, funding, and career development. Most would agree that we need to change the ecosystem; how to do so is another matter, but this taxonomy is a clear step in the right direction.

Gabe Harp will be presenting on Cell Press’s CRediT pilot as part of an “Implementing Standards” panel discussion at the 2016 Council for Science Editors meeting in Denver, Colorado, May 14 – 17.