Carini Lab in-house authorship guidelines
Our decision matrix for authorship decisions on manuscripts
Authorship on scientific papers is a form of professional currency for academics. The number and quality of papers are used as evaluation criteria throughout a scientist's career.
There are inherent ethical and power dynamics in authorship decisions. Many unscrupulous players abuse this power to control who becomes an author, and how they are listed, and to sneak onto papers they contributed minimal effort toward.
We think that’s crappy! So we outlined Carini Lab authorship guidelines for transparency, consistency, and accountability. Please consider these free for public use.
Carini Lab Authorship Guidelines
We don't make detailed authorship decisions until the manuscript is almost complete.
We're liberal with what warrants authorship. Everyone is eligible for authorship, regardless of rank. Any preparation, provision, or action that contributes to the published work warrants authorship. We also include activities that made the published work possible, but may not be in the work. A key preliminary experiment for example. Contributions may include any of the following (or others not listed):
experiments
datasets
analysis
figures
writing
infrastructure
expertise
detailed discussion
ideas
troubleshooting
organizing
coding
All authors are responsible for the accuracy and integrity of their work. They should be able to defend their contributions.
The "lead author" takes primary ownership of the work as a whole. This includes making decisions on the figures, data analysis, writing, and revising. They conduct the bulk of the organization and writing of the manuscript, even if they did not do all the work. They submit the manuscript after the final review by Dr. Carini. They take part in revisions. They may or may not have conducted the experiments. They should understand all aspects of the presented work, even if others did the work. The lead author appears first on the manuscript.
In rare situations, "co-lead" authors are appropriate. Co-lead authors contribute equal energy to "lead" activities, as defined above. The authors and Dr. Carini will discuss the listed order of the co-lead authors. Co-lead authorship is an exception, not the rule and should be agreed upon by all parties.
We list contributing authors in order of their contributions to the manuscript. Authors listed earlier contributed "more" to the manuscript. Sometimes it is tricky to compare effort or authors have equal contributions. In these situations, we list authors in alphabetical order by last name.
The last author listed is the principal investigator. Sometimes the principal investigator is the lead author. In these cases, there is no special designatory role for the last authorship spot.
The principal investigator is the corresponding author on all works. Postdocs or senior researchers are always welcome as co-corresponding authors.
All authors will have an opportunity to revise the manuscript before submission. We limit this to an "open comment" period of ~2 weeks (unless other arrangements are made).
Authors that leave the lab before a manuscript is finished retain authorship.
You can opt out of authorship for any reason.
We expect lead authors to complete manuscripts in a reasonable timeframe after leaving the group. We'll define the timeframe in our exit interview. We'll reassign the lead author of abandoned works.
We do not look kindly on "sitting" on a manuscript. Manuscripts are the culmination of a tremendous amount of work, both in terms of person-hours and financial resources. Much of the funding comes from taxpayer dollars. Thus, it's imperative to prepare manuscripts as fast as carefully possible. If there is writing to be done, do it. We’ll consider the work abandoned if progress is not made in a timely fashion, outside of extenuating circumstances.
Acknowledgements are appropriate for people who have indirectly contributed to the research. For example, undergraduates often learn experiments through watching staff perform experiments. Watching does not earn co-authorship. Preparing media for an experiment does not earn co-authorship. Learning how to run a gel does not earn authorship. These are all more appropriate as acknowledgements. If there is a question about what earns authorship please ask your direct supervisor!
Does your group have extra guidelines? Did we miss something? We'd love to see them in the comments below.
Great and important stuff… but what with our new era of robots (see eg recent tweet by @LifeAfterMyPhD), how do robots fit in the conversation? Especially as they improve on spotting seemingly incongruous findings, recommending suitable experiments, and/or interpreting/framing experimental results?