TrueTuesday: authorship Jun 11, 2019
Mevrouw Noemie AUBERT BONN
Being a scientist is so much more than just a title. We all share an environment, a culture and most importantly a community. Yet despite the fact that scientific findings and knowledge are covered on all possible media, the factors that affect, characterize, and influence this community and its environment are rarely discussed and often even taboo. Every six weeks, Ph.D. student Noémie Aubert Bonn targets such a topic together with 5 experts for TrueTuesday. The fourth one is Authorship.
THE MANY FACES OF AUTHORSHIP
Authorship may be an unexpected theme for our new series of TrueTuesday, but you will quickly realize that it plays an important role in research integrity and might prove to be more complicated than it seems.
There are two main concepts of authorship that are somewhat universal in academia. First, authorship works as a currency that is essential for recognition. Without authorship, researchers have little to no proof for the work they’ve done or their experience and impact within a certain field. Secondly, authorship also works as a responsibility; it shows who is willing to endorse the ideas of the published work and introduce a new landscape in the field in which they operate. Although these two concepts are rather universal, authorship traditions can differ greatly between some research disciplines.
For example, in physics, papers can have hundreds, even thousands of authors, generally ordered alphabetically (fortunately!). In other fields, papers have fewer authors and the order is essential for credit attribution. In such fields, authors are generally organized in order of their contribution, with the first author being the one who did most work (the one whose sweat, tears, and blood were sacrificed in the process). But in some fields like medicine, the last author is not the one who did the least, but the one who supervised the work (normally at least contributing in sweat).
Already, we can understand that authorship is highly field-specific. But cultures also come into play. In some countries (from which Belgium is not fully excluded), directors of laboratories or departments author all publications from their department, even if they haven’t read a single word from the paper. Researchers have also found strategies to use authorship in their favor. From gift authorship — a clever tactic which goes along the lines of ‘you scratch my back I scratch yours’ — to ghost authorship — a tactic to hide conflicting interests — some scientists certainly learnt how to use authorship to ’win the game’.
But what do the experts have to say on this matter? Let’s hear their stories that will demystify authorship or showcase its complexity.
DARREN TAICHMAN ON A NEW DEFINITION OF AUTHORSHIP
For our first interview on Authorship, we reached out to Prof. dr. Darren Taichman, Executive Editor of the Annals of Internal Medicine and Secretary of the International Committee of Medical Journal Editors (ICMJE). The ICMJE is a working group of medical editors who build recommendations for scientific publishing. Their definition of authorship has become a standard in medical and social sciences.
Q: According to ICMJE, who is an author?
A: ICMJE believes authors should meet 4 criteria. First, a substantial contribution to the design, conduct, analysis or interpretation of the work. Second, drafting or revising the manuscript. Third, approving the final version to be published. Fourth, agreeing to be accountable and ensure that concerns related to the accuracy or integrity of the work are addressed.
Q: In this definition, everyone is accountable for the integrity of the published work. So how can this work in collaborative research?
A: Coordinating a multi-author, multi-center study is certainly challenging. But, responsibility for a work’s integrity doesn’t stop with the publication. If your name is on the paper, you need to ensure important issues that arise are addressed. That doesn’t mean each author needs expertise in everything. A clinician on a clinical trial report may not know each detail of the statistical analysis but should ensure a statistician responds to statistical concerns. A statistician should similarly ensure a clinician replies to questions about clinical care. We want to avoid situations where no author responds to concerns, each saying it was not his/her part of the study. The point is this: you agree to help the community understand issues raised about the work on which you put your name. That’s good science.
Q: The definition of authorship changed a few times in the past years. Are there plans for future changes?
A: Medical science and reporting evolve. We reassess regularly if new conditions warrant adjustments to keep authorship criteria relevant and appropriate. But as Yogi Berra once said, “It’s tough to make predictions, especially about the future.” It’s what makes science so interesting and rewarding.
GEORGES AAD ON HAVING 500 FIRST AUTHOR PAPERS
In 2015, the ATLAS Collaboration published a paper which captured the attention of the media because of its record 5154 authors. The shortened reference of this paper starts as Aad, G. et al. (2015). Georges Aad, from the Center of Particle Physics of Aix-Marseille University, now has over 800 publications from this collaboration, with around 500 on which he appears as the first author.
Q: Would you say your name had an influence on the development of your career?
A: I would not completely ignore the effect of my name on the visibility advantage that it could have given me in a large collaboration such as ATLAS. It is simply easier to remember my name and this can be an advantage. However, in our field, everyone is aware of the fact that the order does not imply any particular ranking of the contributions of the authors to papers. In that sense, I believe that my name had a minimal or even insignificant influence on my career development.
Q: In 2016, a certain Aaboud also joined the ATLAS, making him the first alphabetical name of the collaboration. Did you see a difference since?
A: There was absolutely no difference. This was subject to jokes between colleagues but nothing more.
Q: Participating in the ATLAS allowed you to author over 800 papers in just 10 years! How can we know in which papers you had an active role and in which you didn’t?
A: In complex experiment such as ATLAS it is hard and subjective to define who deserves to be part of the author lists of a given paper. All papers are based on collaborative work starting from building the detector, operating it to collect data and developing the various algorithms to identify various particles and calibrate their energy and other properties before being able to use them to extract the final results that you see in a paper. All ATLAS authors are required to contribute to the various stages of this collaborative effort even if not all members contribute directly to the final steps of extracting the results for a given paper. This is the reason why all ATLAS authors sign all ATLAS papers.
LIZ ALLEN ON THE CONTRIBUTOR ROLES TAXONOMY STANDARD
Researchers collaborate but do not always contribute in the same way to a project. The Contributor Roles Taxonomy standard (CRediT) provides a detailed taxonomy, which allows authors to specify contributions. Let’s ask Liz Allen, Program Chair of CRediT, a bit more about it.
Q: Why did CRediT come to life?
A: Original research articles with one author – particularly in life sciences – are increasingly rare, and the concept of ‘authorship’ in science has become outdated. CRediT was developed through a collaboration of research funders, publishers, researchers, and institutions, following a Wellcome Trust-Harvard University workshop in 2012. The aim was to provide greater granularity, visibility, and transparency to the contributions of researchers who are listed as ‘authors’ on published scholarly research output.
Q: Are authors enthusiastic about disclosing contributions?
A: During its development CRediT was tested with authors publishing across the life and physical sciences and was well received. In 2015, PLOS rolled out the use of CRediT across all its journals. It proved to be relatively pain free for authors and provided the lead for other publishers to follow. Today, over 120 journals, publishing platforms, and system integrators implement the taxonomy, and the interest keeps growing beyond the two initial fields.
Q: CRediT was also meant to be used by evaluators such as funders and institutions. Do you see that happening?
A: One of the drivers for CRediT was to help researchers (particularly early career) get more visibility for their contributions to a research article – regardless of where they appear in an author list! We see CRediT being recommended as a tool that can help incentivize researchers to jump into ‘team science’. There are also practical uses in knowing who contributed to specific components of published work, such as helping funders identify potential grant peer reviewers (e.g. statistician; data curator) and helping researchers identify and forge collaborations with individuals with specific skills.
GREGORY PATIENCE ON SURVYING 6000 RESEARCHERS ABOUT AUTHORSHIP
The International Committee of Medical Journal Editors (ICMJE) and CRediT (previous post) guide authors to recognize contributions, but how do they do it in practice? Earlier this year, Prof. Gregory S. Patience, Canada Research Chair and professor at Polytechnique Montréal, published a survey comprising over 6000 researchers on the topic.
Q: What are the key findings from your study?
A: Research requires creativity, diligence, communication skills, resources, and time. Yet, researchers don't agree on how to recognize these activities. To reduce conflicts, principle investigators must be open from the beginning and tell people how their work will be recognized.
Q: Would you say that, in practice, authorship decisions are aligned with the definition from the ICMJE?
A: Many universities base their ethics policies around them and add criteria, which is great. However, our survey shows that these are often ignored, even within the medical field for which they were first developed, and rightly so. Penders (2016) recognized the messiness of research and Moffat asked why we should have a universal consensus of assigning authorship, which infringes on the autonomy of researchers expression.
Q: In June 2020 you will host the conference “Principles of Scientific Publication”. Why will researchers come to such a conference?
A: The PNAS organized an exclusive retreat for publishers and well established professors. They recommended tightening the ICJME criteria, which contradicts our survey. Our conference promotes open and vigorous discussion around: 1) Power imbalance: Graduate students versus professors, 2) Hyperauthorship, 3) Reviewer 2 must be stopped, 4) Does author order matter? 5) Journal selection process: Author/institute ethos, 6) Is correcting grammar a contribution? 7) Fading contributors: How to demote authors to acknowledgements, 8) PNAS authorship criteria: ICMJE on steroids, and 9) Passive aggressive reception of the active voice. A lot to look forward to.
NELE NIVELLE ON TRYING TO AVOID DISPUTES
We have seen that authorship is a complex topic and that its definition may differ between disciplines, cultures, and research teams. But even within the research team, authorship may be difficult to interpret and disputes may arise. Let’s ask Nele Nivelle, research coordinator and Ph.D. ombudsman in Hasselt University how we can avoid such disputes.
Q: The literature on research integrity suggests that authorship disputes are among the most common research integrity problems. Do you face such disputes in Hasselt University? What types of problems do you hear from?
A: It will come as no surprise that publishing and authorship are a genuine concern to our Ph.D. students. They regularly turn to me for help on the topic. When they do, they typically indicate that they have taken the lead in a publication, or extensively contributed to it, but afterwards their name turns out not to be mentioned in the list of authors, or not in the expected position. This can be a huge letdown for a Ph.D. student and it can weigh heavily on their personal motivation.
Q: How do you mediate disputes between authors? What is the procedure?
A: I always encourage Ph.D. students to communicate openly with their supervisor, to explain their perspective and concerns, in order to reach clear arrangements for future publications. If this poses a hurdle to the Ph.D. student, I can act as a mediator.
Q: Would you have any advice and tips to help researchers avoid such disputes?
A: Have a look at the charter Ph.D. supervisor – Ph.D. candidate. It describes the roles of a good supervisor and Ph.D. candidate, and it is the ideal basis to start an open talk on mutual expectations, including supervision style, regularity, mode of feedback meetings, and publication and authorship strategies. Every starting Ph.D. student should have this talk with their supervisor, and return to the principles every once in a while. This helps to avoid conflicting expectations, and it allows Ph.D. students to work in a targeted way, including towards their doctoral school publication requirements.
Be sure out to check out our social media next week Tuesday to discover what our final TrueTuesday topic will be and don’t forget to read our previous articles about ‘mental health’, ‘mistakes & failure’, and the ‘journal impact factor’.