Deciding the order of authors on a team based paper can be challenging: who should be included and in what order? What constitutes a contribution worthy of authorship rather than a simple acknowledgement? Feelings can run high and sadly I have fallen out with people over such matters, which in hindsight and the passage of time was just plain stupid. But at the end of the day does authorship order really matter? Does a reader actually care who did what?
It is a subject covered by Sebastian Frische writing in Nature this week. This whole question resonates with me at the moment. In the last year I have started to publish with a new set of collaborators who introduced me to an authorship code I had never heard of before – call me a naïve geologist if you like! My approach to authorship over the years has, with a few notable exceptions, always been based on a simple principle of inclusion if you were in the field, contributed to the debate you where an author irrespective of whether you actually pulled your weight in the analysis or write-up. He or she who does the most work and drives a paper forward goes first and the order there after reflects the level of contribution. A simple model based on simple principles. Last year I was introduced to the concept of the last author however. My new colleagues hold this position to be one of real prestige – the senior seat – and a view I now find to be widely held in some disciplines. It is something which I have to admit has past me by despite over twenty years of research. I always worked on the principle that to be lost in the ‘et al.’ was never a good thing! I have adopted this new approach in recent papers, but it leaves me intrigued to know what other conventions around authorship I am not aware off? In fact I would love to hear from my colleagues on this subject.
But to be honest to what extent does it really matter, after all an author is an author? In the context of REF it doesn’t with the Panel Criteria and Working Methods making no mention of authorship order or contribution. Sebastian Frische argues that it does to new academics trying to build a reputation and he draws the interesting analogy to the film industry where the credit list is vital to ones CV and the ability to get work in the future. In fact, he goes as far as to suggest that one vehicle is for academic networks, such as ResearchGate (which is currently sweeping through my own collaborator network), or academia.edu should allow authors to express their contribution to a given paper providing the equivalent of a credit list. I have to admit that to me this sounds like a potential for discord between authors. But in truth does it not go against the very principle that research – with the exception of the sole scholar – is by its very nature collaborative. In a football team all players receive a cup winner’s medal whether they scored a goal or saved a vital penalty. All contribute to the victory and all should be recognized equally? So in truth I am far from convinced. What do you think?
Matthew Bennett raised some interesting issues about academic authorship, although others may see it as opening a can of worms. We’d like to make two observations. First, the notion of who can be an author, who should be an author and who definitely should not be an author, is discipline specific. Secondly, the order of authors is dictated by the academic tradition from which the work comes.
1. Who should be an author?
In our field of health research there are very clear guidelines about authorship. As health research is often conducted in multi-disciplinary teams, a paper can have several authors. Guidance about co-authorship is generally based on the contributions of each author. The key questions: Who has been involved in the study (the work), the analysis, the writing of the drafts, etc.
The International Committee of Medical Journal Editors produced guidance on authorship in the form of the ‘Uniform Requirements for Manuscripts Submitted to Biomedical Journals’ which are applied globally (see: http://www.icmje.org/). This specifies that authorship cannot be ‘gifted’ since an author must have made ‘substantive intellectual contributions to a published study’. This is important as gift authorship has in the past got some quite eminent authors into trouble when the data were subsequently found to be wrongly analysed or worst fraudulent (Smith, 2006).
These guidelines also encourage teams to consider all significant contributors as authors. Junior researchers are sometimes forgotten in the authorship discussion, some may be short-term researchers who may have moved on to another university by the time the manuscript gets drafted. Furthermore short-term contracts mean that such researchers may not get the chance to develop writing skills, and it is often perceived as quicker for more senior members of the team to write up the study (Newman & Jones, 2006). This highlights the importance of identifying who will be an author early in the writing process (Simkhada et al. forthcoming).
Once you have decided who fulfils the authorship requirement, the question is ‘What will be the order of authors?’ Journal conventions differ according to the academic traditions from which they come.
2. Which order should authors be in?
We’d like to start here with a variation Matthew did not know about, simply to prove our point about the practice being discipline specific. In traditional economics authors are typically arranged in alphabetical order; so if you are lucky enough to be the economist Alexandra Aabe, your name would appear earlier in the list of co-authors than her colleague Zoë Zwyn.
The eminence of last author in not restricted to medicine (Drenth, 1998), but is also something seen in science. For example in biology the last author is often the one leading the research area and who obtained the funding for the work. However, interpreting this is anything but straightforward. A recent study of authorship position found significant difference in how readers interpreted the role of the final author, with some seeing it as eminent but others more likely to view the author as having made little or no contribution to the study (Zbar and Frank, 2011).
Group authorship has become increasingly common where large multi-centre studies have been conducted. In such circumstances the ICMJE notes that it is important that a smaller number of authors take responsibility for the paper, and the “corresponding author should clearly indicate the preferred citation and identify all individual authors as well as the group name.” (see: http://www.icmje.org/). We see a nice example of this in at the most recent edition of the the medical journal The Lancet Neurology, (Lansberg et al. 2012). The paper ‘MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study’ has twenty listed authors listed followed by the statement ‘for the DEFUSE 2 study investigators’. The latter group of even more collaborators is then listed separately at the end of paper.
Does order matter?
Yes it does, unlike Matthew’s general quote from the REF, the more detailed advice for Panel A, which covers ‘Public Health, Health Services and Primary Care’ (Unit of Assessment 2); ‘Allied Health Professions, Dentistry, Nursing and Pharmacy’ (UoA3); ‘Psychology, Psychiatry and Neuroscience’ (4) and ‘Biological Sciences’ (UoA5) includes several sections (32-37) under the heading ‘Co-authored/co-produced outputs’.
“32. Institutions may list co-authored outputs only against individual members of staff who made a substantial research contribution to the output.
Information required about the author’s contribution
34. For all sub-panels, no additional information is required in form REF2 about the author’s contribution to co-authored outputs where either:
• there are fewer than six authors or
• there are six or more authors but the submitted member of staff against whom the output is listed is identified as either lead or corresponding author (regardless of the number of authors).
36. For each submitted co-authored output where there are six or more authors and where the submitted member of staff is not identified as the lead or corresponding author, institutions are required to affirm the substantial contribution to the research by the submitted member of staff. This should ….(include) at least one element from each of a and b:
a. The author made a substantial contribution either to the conception and design of the study; or to the organisation of the conduct of the study; or to carrying out the study (including acquisition of study data); or to analysis and interpretation of study data.
b. The author helped draft the output; or critique the output for important intellectual content.”
(source: REF 01.2012 pages 23-24 /see: http://www.ref.ac.uk/media/ref/content/pub/panelcriteriaandworkingmethods/01_12_2A.pdf )
So obviously if you are the seventh or eight author of a paper (and you are not the corresponding author) in these disciplines the expectation is that your contribution to the paper has been small and not substantial.
Finally, the analogy with the football team’s glory is an interesting one. Matthew mentioned sport: “In a football team all players receive a cup winner’s medal whether they scored a goal or saved a vital penalty. All contribute to the victory and all should be recognized equally?” We have always been a little uncomfortable with the player who has not played the whole tournament, has not kicked a single ball, but sat on the bench during the final game as a substitute still getting the world cup medal.
Vanora Hundley & Edwin van Teijlingen
Health & Social Care, Bournemouth University
Bennett, M. (2012) What’s in a list?, BU Research Blog, http://blogs.bournemouth.ac.uk/research/2012/09/27/whats-in-a-list/?utm_source=digest&utm_medium=email&utm_campaign=daily
Drenth JP (1998) Multiple authorship: The Contribution of Senior Authors. JAMA. 280(3): 219-221. http://jama.jamanetwork.com/article.aspx?articleid=187761
Engers M, Gans J.S, Grant S, King S.P. (1999) First author conditions. J Pol Econ ;107:4.
Lansberg, M.G. et al. (2012) MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study, Lancet Neurol, 11(10): 860-67.
Newman A., Jones R. Authorship of research papers: ethical and professional issues for short-term researchers. J Med Ethics. 2006; 32(7): 420–423.
Simkhada, P.P., van Teijlingen, E. Hundley, V. (forthcoming) Writing an academic paper, Health Renaissance
Smith R. (2009) Research misconduct: the poisoning of the well. J R Soc Med. 99(5): 232–237.
Zbar, A., Frank, E (2011) Significance of Authorship Position: An Open-Ended International Assessment. Am J Med Sci 341(2): 106-109.