Who Should Be The Authors: Guidelines for Authorship of Scientific Papers

Rob Roach, Research fellow, Copenhagen Muscle Research Center, Copenhagen, Denmark.

Sportscience Resources May 1997

Scientists "invest their lives in their research and in each manuscript" so proper credit for authorship is an important issue to scientists and research students (JAMA 1994; 271: 469-471). In fact, it's so important that a committee of medical journal editors published a guideline on authorship to be followed by several thousand biomedical journals. In many of the major biomedical journals authors now have to sign a statement that they have followed the guidelines in assigning authorship.

The Guideline


Each author should have participated sufficiently in the work to take responsibility for the content. Authorship credit should be based only on substantial contributions to (a) conception and design, or analysis and interpretation of data; and to (b) drafting the article or revising it critically for important intellectual content; and on (c) final revision of the version to be published. Conditions (a), (b), and (c) must all be met.

(From: International Committee for Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals. New England J Medicine 1991; 324: 424-428.)

Many have applauded the guidelines for attempting to reduce the number of authors and to restore value to authorship. The examples often given to support the need for a reduction in the number of authors included on a paper are multi-center trials where hundreds of names are listed. Identifying who is responsible for the integrity of the reported data then becomes problematic.

On first reading, the guidelines suggest that technical support staff are excluded from authorship. As pointed out in the following examples, if you only collect data, or run a lab, or own a gadget or have an idea you should not expect to be an author. But if in addition you make a significant intellectual contribution to study design, and interpretation and the writing, then you qualify as an author. Coincident with this discussion about authorship is a parallel suggestion to increase the use of acknowledgments. For example, some journals allow a "With technical assistance from" byline after the listing of authors. Others encourage such acknowledgments to be put at the end of the paper.
Here are a few true-to-life scenarios each with a quiz question. Try to answer these by referring to the guidelines.

Quiz: Who should be the authors?


A young post-doc works in the laboratory of a senior (HotShot) scientist with 200+ publications on his CV. Over beers at the local student pub a friend asks if the post-doc would be interesting in helping her with a project where their skills and interests overlap. The post-doc agrees. They do the study, use some resources from the HotShot's lab and also run some analyses there. At no point in the scientific process is HotShot involved. The paper gets written and post-doc shows it to HotShot. HotShot looks it over, has no comments except: Put me as senior author.
Should HotShot be an author on this paper?


HotShots can also get a bad deal. From a recent letter to the editor in Science (Science 1997; 275: 1863) comes a version of the not too common story of a senior scientist reading about an article in a major journal on which he is a co-author. He has never seen the paper prior to publication, has not reviewed the data, and is dismayed at being included on this paper in this manner.
Should HotShot have been contacted prior to publication? What should he do now?


A young scientist has a new facility, doesn't matter what exactly, could be the latest and greatest environmental chamber, nitric oxide analyzer, strength tester. Perhaps it's the latest molecular biology technique to analyze the speed gene. You name it, it is hot, and everyone wants to use it. . The young scientist decides "this is a gold mine." She knows her mentor had his name on hundreds of papers simply because he ran the lab. Young HotShot imagines a fat CV with little additional work and begins to tell collaborators that they are free to use her gadget as long as she is an author on the paper(s) resulting from the gadget's use.
Should the young HotShot be an author on these papers?

These examples illustrate the complexity of publishing scientific articles. They also point out how easy it is to get into gray areas about right and wrong. In example #1 HotShot may feel that anything that is published from work done in his lab should have his name on it. And, as I wrote the story with purposeful ambiguity, the post-doc should have asked first. This type of misunderstanding can happen once, but after that both the post-doc and the HotShot have an obligation to talk and make sure the ground rules are understood. If HotShot says that anything published from his lab have his name on it, whether he fulfills the Guidelines for Authorship or not, few students or junior colleagues will be in a position to disagree. That is reality. Only peers and supervisors of HotShot have the authority, and lack of risk in job and career security, to confront him about his unacceptable approach to authorship. The second example is a clear case of abuse of authorship. Many journals now require the signature of people that are acknowledged by name in a paper, in addition, of course, to signatures from all authors. In the third case, again only peers or supervisors can correct the behavior of this wayward young HotShot; indeed this story is just a variation on the theme of Quiz Question #1.

The leaders in the area of integrity in scientific publication argue there is a connection between unethical behavior in the drive to publish more and scientific fraud that makes the problem of defining authorship super important. For example, if HotShot #1 gets away with putting his name on everything that comes out of his lab without being intimately involved in designing the projects, analyzing the data, and writing the paper, then his name may get on a paper containing fraudulent data. Farfetched? Not really. Ask Francis Collins, Director of the multi-billion dollar Human Genome Project. Last year it was revealed that a graduate student had fabricated data that were published in papers with Collins as the senior author. Collins later admitted that he had not properly supervised the student and was not as involved in the final paper as he should have been. Remember the Guidelines! In other words, the publication system demands that everyone who is an author is deeply involved in the project so that the group of scientists claiming authorship can also share responsibility. As a group they are able to say to the scientific community this paper and the data in it are valid to the very best of their knowledge.

How to avoid problems with authorship?
Actually, deciding who should be an author can be a very simple process. First, agree with your collaborators that you will follow the international guidelines. Second, agree before starting the research who will be an author, and if necessary discuss why each person should be an author. Third, agree on the tentative order of authors and on who will be corresponding author.

Concerns for Students
The general steps outlined above apply to you, especially the advice to talk about expectations for authorship. For example, suppose you have a part-time job as a lab technician. Many publications result from the work on which you do much of the technical work, and you wonder if you qualify for authorship on those papers. Review the guidelines. If you haven't made substantial intellectual contributions to the manuscript then you deserve to be acknowledged for your technical contributions, but not to be an author.

Think carefully about technical work you are being paid to perform as distinct from thesis-related research. They may be one and the same. The point is that once you feel your contributions warrant consideration for authorship, take the initiative and bring this up with your advisor. If a situation develops where you have been thinking about the project and are anxious to make more than technical contributions but you are not given that opportunity, then you also need to talk with your advisor. It may be wrong for your supervisor to write up the work and publish it without seeking your input when the work is specifically related to your thesis research. Clear and frequent communication can help you avoid this problem.

You may feel insecure about asking your supervisor to include you in authorship. Go ahead and ask. If it doesn't work after the first paper where you think you qualified, then ask your supervisor what it will take for you to be an author. Everything will probably work out and you'll get the terrific satisfaction of seeing those long hours of work in the lab pay off with your name in print for all the world to see. What if it doesn't work out so well? Don't lose your cool. First and foremost, talk to your advisor. What happened? Is the explanation understandable? Believable? Scientists are as complicated as anyone else, and misunderstandings are common. If you cannot resolve the problem, finish your thesis and find a new lab. Nearly every graduate student I have known in the last twenty years has had to deal with problems like these. Happily I can report that almost all have had a very reasonable and productive resolution using the ideas presented in the guidelines and taking the direct approach with their advisor.

Who's on first? The order of authors in multi-author papers
After deciding who will and will not be an author it is necessary to decide the order of the authors. It is important for young authors to understand that there are two positions that count: first and last. And attached to either position is the status associated with being the author for correspondence. The best combination when you are young is to be first author and the author for correspondence. As your career progresses, being last author and author for correspondence signals that this is a paper from your lab, you are the main person responsible for its contents, and a younger colleague has made major contributions to the paper, hence they are listed first. The guidelines here are not as well defined as for authorship in general,. Here are some suggestions adapted from a longer article on the subject [JAMA 1991: 264(14): 1857]) :


The first author should be that person who contributed most to the work, including writing of the manuscript.


The sequence of authors should be determined by the relative overall contributions to the manuscript.


It is common practice to have the senior author appear last, sometimes regardless of his or her contribution. The senior author, like all other authors, should meet all criteria for authorship.


The senior author sometimes takes responsibility for writing the paper, especially when the research student has not yet learned the skills of scientific writing. The senior author then becomes the corresponding author, but should the student be the first author? Some supervisors put their students first, others put their own names first. Perhaps it should be decided on the absolute amount of time spent on the project by the student (in getting the data) and the supervisor (in providing help and in writing the paper). Or perhaps the supervisor should be satisfied with being corresponding author, regardless of time committed to the project.


A sensible policy adopted by many supervisors is to give the student a fixed period of time (say 12 months) to write the first draft of the paper. If the student does not deliver, the supervisor may then write the paper and put her or his own name first.

Keep in mind that the guidelines and suggestions presented here should not be used to condemn past practices. The rules governing authorship and the bigger issues of scientific misconduct are evolving to meet the demands of the times we live in. So employ these ideas to help prevent problems in the future.

Most scientists who have published extensively have their own policy on authorship. Whatever policy your supervisor or colleagues have, there are usually good reasons for them: tradition, experience, or just the plain old selfish genes that we all carry. Nevertheless, as long as their policies are not contrary to the above guidelines, they have a right to their policies, you have a right to query them, and you all have an obligation to negotiate a mutually acceptable outcome that meets the criteria for authorship laid out above.

Further Reading
Relman, A. S. (1984). Responsibilities for authorship: where does the buck stop? New England Journal of Medicine, 310, 1048-1049.

Huth, E. J.(1986). Guidelines on authorship of medical papers. Annals of Internal Medicine, 104, 269-274.

Edited by Will Hopkins. Webmastered by Jason Nugent. Last updated 8 May 1997.
resources=AT=sportsci.org · webmaster=AT=sportsci.org · Homepage · Copyright ©1997