A seven-step authorship plan for scientific papers

N.B. This is a do-as-I-say, not a do-as-I-have-done plan. This is the plan I intend to implement explicitly from here on out. I’d love to hear your thoughts and feelings on this! 

1. At the start of the project, have a list of things that feel like substantive contributions for your project (idea generation, data collection, statistical analysis, context and interpretation, writing, funding, etc). Allow this list to evolve as the project evolves.
2. Any time you invite anyone to work on any aspect of the project, explain this method of authorship assignation to them.
3. At the point when you are ready to start WRITING THE PAPER, make a list of everyone who has contributed to the project by seeing which substantive contribution boxes they check.
4. Anyone who checks two boxes is an author. Offer them the chance to contribute to the writing because they might enjoy it, might learn, and may make your paper better.
5. To anyone who checks one box, OFFER THEM A CHANCE TO CONTRIBUTE TO THE WRITING. If they take it and do any non-zero amount of work, they’ll have checked two boxes, so make them an author. If they don’t take it or agree but contribute zero work, they’ll understand that they had this opportunity and chose not to take it. You can tailor what it means to “contribute to writing” as appropriate for your constraints but be as clear and as generous as you can.
6. Get really good at wrangling contradictory comments :p.
7. Ultimately you are first author, you get to decide what goes in the paper. Embrace that responsibility, deferring to the expert authors for sections where you lack expertise.

On the flip side, sometimes we find ourselves being offered authorship as collaborators and not being entirely sure of whether to accept or not, of whether we’ve done enough work to warrant authorship. A rule-of-thumb that I learnt from Ben de Bivort that works for me also is to ask myself the question: have I done one full day’s worth of work on this project on my own and beyond that which my role asks of me? (e.g. in Ben’s case, he was on my dissertation committee, and chose not to count the time he spent advising me in one-on-one or committee meetings, deduced that he had not spent a full day working on the project outside of those meetings, and therefore declined my offer of authorship).