We worry so much about measurement in epidemiology. (Or at least we should.) We teach about the bias and lack of precision that can occur due to measurement error. We ask participants in our studies to be as accurate as possible when filling out questionnaires to avoid measurement error to the extent possible. So where does this concern about measurement go when it comes to authorship?
Most fields have come up with ways of indirectly communicating to the reader the relative contribution of each author. In epidemiology, the first author is the lead author of the paper, the last author is the senior author (although this can mean different things) and sometimes the second author or second to last author can have some additional prestige (depends on who you talk to). Everyone else is just in the middle author soup.
Why don’t we make clearer what each authors’ contributions are? There are a whole host of ways you can contribute to a paper. CRediT (Contributor Roles Taxonomy) identifies 14 and there are likely many more. There are many roles on that list such as data curation, funding acquisition and reviewing and editing, that, by themselves do not qualify for authorship but are nonetheless important roles. Understandably, we want to compensate people who fulfill these important roles and the only way we have to do so is authorship.
Imagine if, instead, we could directly acknowledge how each author contributed. It would give so much additional information for people who need to evaluate researchers. If someone has a lot of publications because they created a popular dataset (a crucially important part of research) they would be recognized for their extraordinary contribution to data generation rather than just being a middle author. The same for people who perform critical review of a manuscript towards the end of the project. The more eyes that see a manuscript, the better and we should recognize directly the service they have performed rather than mix them up in the middle author soup with people who maybe contributed data or performed analyses or any other of a whole range of roles.
We should also think about the consequences of this information not being generally available. It’s possible that everyone benefits and is hurt equally by this authorship measurement error but it’s also possible that some people benefit while are people are more hurt by it. I suspect the latter is more likely true and I’m not sure it would even be easy to identify who is benefitting or who is being hurt. Making the roles of authors clear will eliminate this inequity.
Some have mentioned that it’s not comfortable to have these conversations with co-authors because it’s hard to agree what everyone contributed. I agree. But we ask participants in our studies to answer sensitive questions all the time and so I think it’s only fair that we be prepared to experiences a little discomfort ourselves if it leads to better science.
I don’t think this will solve all our problems. People will still be listed as having contributed to a category they didn’t contribute to or listed as a manuscript reviewer when they only gave the manuscript a cursory look. But I think it would be a step in the right direction. And I believe that thinking about this as a measurement error issue is the right way to think about this. Giving the people who evaluate us more information about what our contributions are will help eliminate at least a small part of this measurement error.
