Departments

Authorship

By: Iain Taylor, Bruce Dancik
May 31, 2014
Authorship Best Practices Ethics

The author byline tells the editor, peer reviewers, and readers who did the work. Simple, isn’t it? Except that throughout the scholarly community, “who did the work” may include the originator of the idea, the writer of the funded grant proposal, the critical readers who honed the proposal into a “doable” piece of research (interested colleagues, interested students, or visitors who provided opinions and improvements), the person who designed the particular experiment, the student or postdoctoral associate who executed a major or critical part of the study, the paid technicians and statisticians (who also may have analyzed and interpreted the data), the professional writer, or an institutional editor. You may have struggled looking for the end of that sentence, but we could probably have added a few more people.

The list goes on and may eventually include one or more of the peer reviewers and even the “first-decision” journal editorial-board member. The long-standing tradition in science is that the person who has been designated as the principal investigator is one of the authors, commonly the first person, the last person, or the person designated as the corresponding author. In one of the recent preliminary papers reporting the discovery of the Higgs Boson,1 more than 2,400 people are listed in a supplementary statement, whereas the byline states simply “The ATLAS Collaboration”. A recent paper in Science2 had as the byline that “All authors with their affiliations appear at the end of the paper.”

Credits for scholarly advances are recognized largely through the lists of authors on the bylines of published research reports, so it is perhaps not surprising that professional advancement is determined largely on the basis of authorship of published (usually peer-reviewed) papers. However, it seems that members of appointment and promotion committees occasionally superimpose their own understanding of how author sequence reflects the importance of a candidate’s contribution to publications. Such cases may be avoided if there is a clear explanation from the originating department head or chair. Peer review is a process that provides the journal editor and the reader with an assessment of the originality and quality of submitted research and some sense that the work is perceived by informed colleagues as being relatively reliable and thus likely to add to the base of knowledge and understanding in the field. The paper is of course not “the truth” but is a report of the results of research undertaken with attention to acceptable standards in a discipline.

It is increasingly apparent that the requirements of authorship can be flexible. Authorship has also been used in attempts to gain favor with mentors, senior colleagues, eminent peers, and others. Some scholarly subcultures add the names of distinguished scholars to the author list (gift or guest authorship) in the belief that peer reviewers and even editors will be swayed to recommend acceptance. The level of trust that prevails in the scholarly community generally makes it unlikely that the appearance of these so-called guest authors will be questioned, but several leading journals increasingly require a statement of what each named author contributed to the paper. Such practices are more common, and apparently continue, in countries and communities where outside financing seems to depend on the established or perceived excellence of a research group, such as a department or a major research organization. One may be left to wonder how a person managed to publish more than 3000 papers in his or her career that spanned 40 years!

The single-author paper may be the norm in some fields and identifies the person who did all the work, but there are still cases in which a graduate student or research assistant may be unlisted because the identified author perceives that the student has simply done as directed. Many years ago, one of us was admonished for including a research technician as first author. The concern was that the technician was simply following instructions (he was really a major contributor to all the work) and was not considered to have the stature to be an author. Donald Kennedy, the former editor-inchief of Science (2000–2008) and a former president of Stanford University, wrote in his book Academic Duty,3

The most common abuse in authorship is the addition of the names of academic supervisors to largely independent work done by students. The fact that this has become a custom in a number of fields does not make it less pernicious. In some institutions, co-authored articles are even allowed in a student’s Ph.D. thesis. This practice not only confuses later evaluators about the degree of independence involved in the work of a student, but also constitutes a challenge to the very notion of a doctoral dissertation. The idea behind the thesis is that it is a piece of original work, done to demonstrate that the student is ready for an independent scholarly career. But if as part of it, there appears a work jointly authored by the student and his or her academic advisor, one of two things is amiss: either the work isn’t independent, in which case it should not be part of the dissertation, or it is (independent), in which case it is a plain case of complimentary (guest) authorship.

Ed Huth4 suggested some years ago that an author is a participant in the research and “can take public responsibility” for the work. However, although that description opens a door for several participants, such a test may also raise the question of the sequence of names.

The authorship issue can be broken into the following:

Who has earned authorship?

What shall the sequence of names indicate?

Who shall be acknowledged?

An increasingly common approach requires full agreement among all participants as to who should be an author after completion of the work and the written report. Indeed, the writing author needs to know who the participating authors are and be assured that the list of authors reflects the reality that the original work could not have been achieved without specific contributions by each named person. There may be useful measures—time spent, expertise to obtain or analyze and interpret the new data or ideas, critical evaluation of the presentation, and limits of the written report. Supervision of students or successful application for research funding may not merit authorship if a research supervisor has done little to direct the work or if the work has little or no relation to the funding.

Interdisciplinary research requires a particularly thoughtful approach to considering the authorship criteria of the participating disciplines. The purpose remains the same: to ensure that recognition goes to the doers of the work. Expectations will vary when the research brings together people who have widely different traditions and practices, as perhaps in education, political science, clinical psychology, and clinical and laboratory medicine.

Work that requires the use of surveys and work that involves interpretation of both qualitative and quantitative studies may require amendment of starting operational details as new information is obtained, particularly when survey subjects drop out and a statistical assessment changes or is even invalidated. Design of a survey may warrant authorship especially when the people who plan the work rely on expert details for subject recruitment and the continuing support of recruited participants. Clinical studies may rely heavily on expert knowledge to recruit subjects, and evidence of researcher activity may be measured by the number of survey subjects recruited. People may have only one function—to recruit subjects who meet stated criteria—and such activity may or may not require intellectual understanding of the work. Although acknowledgment clearly is essential, authorship criteria may not be met.

The problems become more complex if planning is determined by the nature of a question posed in theoretical or contract work in which research is undertaken to find support for a theory or to satisfy the contractor’s desire for a specific answer. Regulators of medical or industrial product safety may prefer arms’-length studies, but under best practices of research it may well be that a study conducted by a contractor’s research employees yields high-quality information. However, it is apparently common for staff of contracting companies to prepare a manuscript (written by a “ghost author”) to present the company’s research in the most positive light and then invite an expert “independent researcher” to provide a presubmission opinion for a monetary reward in return for placing the independent’s name on the byline (as a “guest author”). Such ghost authorship is much frowned on and may be declared gross misconduct by journals, but university authorities seem unwilling to take disciplinary actions against what the journal-editing community sees as a major breach of scholarly integrity. Conflicting interests are common throughout academe, government, and industry. “Named” guest authors, from the earliest-career researcher to the most senior academics and research managers, especially where scholarly integrity may be compromised, must be eliminated from research reporting in order for publications to earn and retain the highest levels of public and professional trust. Every author has a duty to take full responsibility, as Ed Huth so clearly and elegantly described. It is still a problem when earlycareer researchers name a distinguished researcher (perhaps a mentor) as an author with or without the agreement of the senior person, who possibly has little or no knowledge of the work. The reverse may also occur when a senior researcher adds a junior’s name “because you really need another paper to increase your publication list”. In short, although “ghosts” may serve a commercial purpose to steer results to increase chances of regulatory approval or direct advertising, “guests” have no place in research reporting. The obligation remains: Authors must be fully responsible when their names are on a byline.

“Big science” projects often require much diverse expertise and long-term research commitment to the content of a paper. Particle physics and astronomy are two fields in which research is no longer possible without numerous collaborators in many sites around the world. Whose names should be on the resulting papers? The information magazine Science Watch5 noted that the Fermilab had captured the top quark, and hundreds of researchers had shared in the hunt. The paper, published in Physical Review Letters, had nearly 400 authors. A paper titled “Correlation of the Highest-Energy Cosmic Rays with Nearby Extragalactic Objects”, published in Science6 by the Pierre Auger Collaboration has 450- plus authors listed under the Collaboration byline. It also has an authors’ summary of contributions. A short piece titled “Really Big Science: Multiauthor Papers Multiplying in the 1990s”7 reported that Science Watch published a graph showing a dramatic rise in papers that had more than 50 authors during the years 1981–1994. The greatest numbers at that time were in physics, but there was also a steady rise in numbers of authors in the medical literature. Individual contributions to the papers may seem fundamentally different from those to two-, three-, or four-author papers, but it is apparent that research could not be completed without so many contributing researchers, so each was clearly deserving of recognition on the byline. There are occasional papers in which equal author contributions are stated; see, for example, reference8, in which the four named authors are explicitly stated to have contributed equally to the work.

(Although rewards of authorship are related largely to career advancement or funding, one case of election to the US National Academy of Sciences resulted in a strange predicament. A news item9 in Science reported that Nancy Jenkins was elected to membership in the academy but refused because husband Neal Copeland was an absolutely equal partner in their research and they alternated as last author on 750 papers. Dr Jenkins argued that competing with him would mean not staying married.)

Contributorship and guarantorship policies obviously remove much of the ambiguity surrounding contributions, but they leave unresolved the question of the quantity and quality of contributions that qualify for authorship. The International Committee of Medical Journal Editors (www.icmje.org/) has recommended the following summarized criteria for authorship, which are appropriate for journals that distinguish authors from other contributors10:

  • Authorship credit should be based on 1) substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND 2) drafting the work or revising it critically for important intellectual content; AND 3) final approval of the version to be published; AND 4) agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
  • Authors should meet conditions 1, 2, 3, and 4.
  • If a large multicenter group has conducted the work, the group should identify the persons who accept direct responsibility for the manuscript. Those persons should fully meet the criteria for authorship, and editors will ask them to complete journal-specific author and conflict-of-interest disclosure forms.
  • Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content and identify which of the co-authors are responsible for other portions of the content.

Several examples of presenting the order of authors in a byline exist, and they seem to be generally accepted by all named authors. The published list of authors may include a footnoted statement, such as “the first two authors contributed equally to this work.” It is often presumed by readers and by research administrators that the first author is the primary doer of the research and the last author is the professionally senior participant, or in some cases it may be the reverse. Some research groups determine who has earned authorship and then list the authors alphabetically. In such cases as the Higgs Boson work, the authors are identified by the name adopted by all the participants, hence “the CMS Collaboration” and “the ATLAS Collaboration”. Each of those articles included an acknowledgment that recognized the outstanding performance of the Large Hydron Collider at CERN and the Partners in the WorldWide LHC Computing Grid for crucial computing support.1 Partnerships and contributions to the research were presented without the need to spell out all details of individual contributions.

Smaller research groups seem generally accepting of giving credit where credit is due. In a roundtable conversation, it can become clear that each participant can explain and defend his or her contribution. Principal investigators who wrote the funding proposal may want their names on the byline regardless of their not contributing to the research, but a principal investigator may not require recognition as an author when a junior researcher has undertaken a project that is of peripheral interest to the group and did all the work. Every researcher has a professional responsibility to ensure that recognition is afforded appropriately, for example, as an author assistant or contributor in some material way. One of us, on a couple of occasions while editor of a journal, was contacted by authors who asked that the referee of a manuscript or the journal associate editor for a manuscript be added to the list of authors because their remarks were helpful in analyzing results and improving the manuscript. In each case, the referee and associate editor declined the offer of authorship, we believe appropriately, on the grounds that they were just doing their job and trying to help an author to communicate his or her work more effectively to readers.

But does willingness to share experimental materials or data that others may wish to reinterpret or to share information and ideas provided during open peer review merit recognition as authorship? In the new world of “open access” through open peer review before official publication, do we need to think more freely about authorship or even find some other way in which appropriate credit can be assigned for new work? Should open reviewers who contribute substantially to a manuscript be added to the author list? Within a specific research community, members learn or come to understand the ways and means of their colleagues (competitors or non-competitors) so that credit is fairly understood. Problems may arise, however, when professional advancement is granted by administrators (members of appointment and promotion committees) who are not part of this research community.

References

  1. The ATLAS collaboration. A particle consistent with the Higgs Boson observed with the ATLAS detector at the large Hydron Collider. Science 2012;338:1576–1582.
  2. GWAS of 126,559 individuals identifies genetic variants associated with educational attainment. Science 2013;340:1467–1471.
  3. Kennedy, Donald. Academic Duty. Harvard University Press 1997 P 97.
  4. Huth, E. Guidelines on authorship of medical papers. Ann Intern Med 1986;104:269–274.
  5. ScienceWatch Newsletter. July 2012. http://archive.sciencewatch.com/newsletter/2012/201207/multiauthor_papers/http://archive.sciencewatch.com/newsletter/2012/201207/multiauthor_papers/. Accessed 28 March 2014.
  6. The Pierre Auger Collaboration. Science 2007; 318:938–943.
  7. Really big science: multiauthor papers multiplying in the 1990s. Curr Content. 1995.
  8. Edwards EJ, Osborne CP, Stromberg CAE, Smith SA. The origins of C4 grasslands: integrating evolutionary and ecosystem science. Science 2010;328:587–591.
  9. Bhattacharjee Y. The cost of genuine collaboration. Science 2008;320:859. 10. Defining the role of authors and contributors. The International Committee of International Journal Editors. www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-roleof-authors-and-contributors.html. Accessed 28 March 2014.

IAIN TAYLOR is professor emeritus, Department of Botany, University of British Columbia, Canada, and BRUCE DANCIK is editor-in-chief at NRC Research Press/Canadian Science Publishing and professor emeritus, Department of Renewable Resources, University of Alberta, Canada.