As we have discussed in the previous two issues of this journal (Kuper et al., 2023; O’Sullivan et al., 2023), there are a surprising number of nuances to which to attend when making academic authorship decisions. Thus far we have addressed the questions of who qualifies for authorship and of the pros and cons of having co-first authors. So now you have a list of authors, and either one or two people at the start of that list. How do you decide what to do about everyone else?

The basics of author order

Although the problem of author order often seems to sneak up on novice researchers, ideally the issue should be addressed at the start of a study (albeit with the recognition that flexibility is often required as the research and writing process unfolds). The question of author order is particularly confusing in an interdisciplinary field like health professions education because author order norms are different (and sometimes contradictory) in the various disciplines and professions from which health professions education is derived. For example, humanities scholars often assume that authors should be listed from first to last in order of the importance of their contributions, while physicians generally assume that the first and last author spots are the most meaningful! Most researchers in health professions education do, however, end up following medical journal norms – which are also the norms in the health faculties in which most of them work – wherein the key positions are first and last (senior) author, then second author, and then everyone else more or less distributed equally in the middle. Usually, the first position is reserved for the principal author, who is (or should be) the person who led both the study and the writing of the manuscript. If that person is a learner, such as a PhD student, often the last author position is given to their supervisor. In situations where there are multiple authors who are peers (as with this article), the last author is generally the person who contributed the second most to the project and the manuscript. In all cases, experience leads us to advise research teams to visit the topic of author order early and to revisit it as the study progresses and it becomes clearer who is contributing what.

Unfortunately, the reality isn’t always as smooth and simple as we have described. For example, learners sometimes have more than one supervisor, and multiple supervisors may each request or instruct a learner to put them last (potentially leaving that learner in a very difficult position). Potential last authors might also argue directly amongst themselves about who has contributed how much to an article and so should have the ‘prize’ of last authorship, since it is worth more to them than middle authorship in terms of productivity measures used for things like promotion and tenure. Obviously, supervisors should not put their learners into awkward positions and should work out their issues collegially, including those related to authorship order.

Some research groups have found amicable ways of determining author order that serve their own internal purposes. For example, a team of frequent co-authors might simply rotate being first and/or last author so everyone gets some of those more highly-valued publication credits. A pair of co-authors at different career stages might deliberately give a more prestigious author role to the person nearing a promotion or tenure decision, with the intent to do the same for the other when it’s their turn. These are perhaps not what the authorial conventions would officially dictate, but they do demonstrate how authors adapt to the exigencies of the systems in which they work, and they may well be reasonable ways to make decisions if everyone in the team is happy with them.

Interestingly, although author position is important for internal purposes, it is not at all important in terms of academic footprint, often measured by the “h-index” (an author-level index which captures research output based on the total number of publications and the total number of citations to those works). There are many discussions about the limitations of this index but here we only make the point that author position has no bearing on your “h-index”. You could be first author, last author or author 425 of 900 authors (yes, there are papers with this number of authors, but not in our field!), it does not matter. On the other hand, if you are part of a larger study team, note that journals may curtail published references to a paper’s first three or four authors ‘et al.’ – while this doesn’t affect the “h-index”, people reading the paper and looking at the citations may see the names of the middle authors but not the name of the person in the coveted senior author position.

Effects of power

Given these workarounds and relational considerations, it is not surprising that access to key authorial positions (i.e., those that make a paper “count more” in metrics used for hiring or promotion) can be inequitable. Scholars are making important arguments about being more thoughtful in how we distribute these authorial positions. Running through those arguments is a concern about power and its potential misuse, whether within a single department or across continents. A simple example of this might be a team relegating a research assistant (perhaps a graduate student being paid to work on a project) who has drafted an entire article, and would normally thus be qualified to be principal author, to second author in favour of the person paying that assistant (who may have done very little of the actual work on the article).

Power dynamics often show up in international collaborations. As Wondimagegn et al. summarized in their 2020 paper, bibliometric analyses suggest that there is an over-representation of Global North first and last authors in papers arising from joint Global North-Global South projects. They discuss many potential reasons for this, ranging from colonial beliefs about research (and researchers) from the Global South to the influence of English as the primary language of research publication. They suggest potential ways to address inequities in author order (and authorship practices more generally) within such collaborations, including intentional dialogue, long-term relationality, and a conscious effort to ‘pass the microphone’ to Global South collaborators (Wondimagegn et al., 2020).

Whatever challenges you face, we advocate developing a consistent practice of working through author order considerations in a systematic way early and often during project and manuscript development, including acknowledging the role of power. Modeling this for novice researchers will provide strategies that prepare them for the potential intricacies to be encountered over the course of their careers.