Managing the writing of a multi-author paper
Current research is very rarely done by a single investigator and thus a scientific paper almost always has multiple authors. Writing an article with multiple collaborators can be chaotic and time consuming. Adopting a few of the principles that underlie document writing operating procedures used by pharmaceutical companies can simplify the process and expedite it.
A group of writers is always comprised of people who are more deeply involved in the work and those who are more peripheral. The below described processes can help greatly if adopted by all writers, but even if they are adopted only by the close circle of authors most deeply involved.
The best practice is a process-initiation meeting (in person, by telephone or online) during which several decisions can be made:
First, the article’s "owner" must be determined. This author can be, but does not have to be, the first author. The meaning of article ownership is that the author is responsible for coordinating the writing and review activities of the various authors within the group as well as for external activities such as taking care of professional editing and proofing, submission to the journal and communication with the editorial office until the article’s publication. The scope of the owner’s independence should already be defined at the outset of the writing process. For example - in case of a rejection from the first submission target journal, is there a need for a joint discussion in order to decide on the next submission destination, or can the article owner submit sequentially to as many target journals as may be necessary to get the article published? Can the owner independently decide on further analyses at the request of reviewers?
After choosing the article owner, areas of responsibility need to be defined. Who shall be responsible for writing each section? Sometimes, these will be obvious, such as in a project that is a collaboration between a clinician, a statistician and the head of a genetic research laboratory. Here, it is clear who writes each part of the methods and results sections. But who is responsible for the survey of literature? And the discussion?
Finally, the hardest part is making the decisions pertaining to the work process. Whenever possible, it is advisable to schedule a series of joint meetings to discuss the article in advance. It is important to know what the preferred mode of communication is for the group and for each collaborator, and what is the best way to communicate with them in order to receive rapid responses. A convenient solution worth considering, is storage of data (including drafts of the article) in an online repository accessible by everyone in the group. It is very important to try to get the co-authors (or at least some of them) to commit to timelines of reading and reviewing of drafts. In this context, advanced knowledge of any prolonged absences of the co-authors is important and useful. In terms of the process, it is very useful to hold rounds of review and revision sequentially, so that each contributor is able to see the input provided by his predecessors. This prevents the article owner from having to consolidate the comments of various reviewers. For the sake of this process, it is best when notes are provided in "Track changes" mode, with the reviewer's identity and time of review.
The article owner should keep track of drafts, when and to whom they were sent and who returned their comments. “Veteran” groups (those who publish a lot together), should institute delivery of reminders to those late in providing input and group updates about the status of each article until its final publication.
Next post: a few comments on open access journals