There’s No I in ‘Research’ (But There is in ‘Science’)
by
A wise man once said “Stop, collaborate and listen.” Few people knew he was making a profound statement regarding the nature of scientific research worldwide. Probably because they couldn’t get past his outfit. Or maybe his point wasn’t that deep afterall…
In the lab, scientists face a choice when tackling a project – to work alone or to collaborate. How we decide which approach to take may depend on a number of factors including funding (can we afford to do all of the experiments ourselves?), expertise (are we capable of performing all of the required techniques?), timing (do we need to get this finished before our competitors?), and personal preference (do we like working with others?).
Of course, other issues may also come into play such as Do we want to share the spotlight with others? or Do we trust our collaborators? Some departments and PIs are notorious for their collegial atmosphere in which collaborations may be born out of a meeting in the hallway. At other places, it feels more like a fist fight may break out in the hallway before a collaboration would.
There are those who would argue that additional authors on the biline will dilute the recognition given to the first author. On the other hand, a good collaboration will generally speed up a project, increasing the amount of work – and possibly publications – completed in a given timeframe.
As a result of the many factors affecting the decision, it may be less than obvious to students and postdocs whether there is a general consensus about whether to collaborate on their main project or not. We’ve selected two variables that are important to most researchers – authorship concerns and time to publication – and have created two likely scenarios to find out which one most scientists prefer.
.
.
[poll id=”68″]
.
.
Natalie Sashkin Goldberg
wrote on March 8, 2011 at 4:15 pm
Who would have foretold such bearing from the teachings of Vanilla Ice?
I wouldn't have thought that the difference between 2 and 6 authors would be very great in terms of diluting credit, and that maybe 8 or over would have defined that line. This is my naivete in the world of publishing, however. Because if 2 more people make the story in the manuscript significantly more powerful, I would be inclined to allow 8 authors. So where is the line?
alan@benchfly
wrote on March 9, 2011 at 10:00 am
Yeah, who knew a true visionary could have hair like that…
I don't think there's a specific line that's drawn in how many authors is "too many" on the biline. If there are two authors (first author and PI), then it's pretty clear that the first author did nearly 100% of the work and thus, they'll get credit for it. However, if there are 6 authors (5 in lab and 1 PI), then it's not clear whether the first author did 25% or 95% of the work – and that's a big range. Yes, some journals require author justification, but that won't matter when people look at your publications on a CV or job/grant application.
At the extreme, consider large genome publications with tens, or hundreds of authors. If the same paper had only two authors, you'd have a completely different impression of the amount of work performed by the first author, just based on the number of authors on the biline. So in between "2" and "tons" of authors exists a large grey area…
Natalie Sashkin Goldberg
wrote on March 9, 2011 at 1:45 pm
excellent points. thanks Alan :)
enervate
wrote on January 11, 2012 at 9:24 pm
I have to agree with you. I feel that science is a conjunction of labors. Nothing is discovered in a vacuum, and no one owns knowledge. It's unfortunate how funding drives this competition. :/
Jack H. Pincus
wrote on March 8, 2011 at 4:19 pm
Alan-
Intersting post. Your poll accurately captures the authorhship dilemma of collaboration. It would also be interesting to know how much weight responants give the benefits fo collaboration you describe when deciding to collaborate or work alone.
Jack
alan@benchfly
wrote on March 8, 2011 at 5:10 pm
Hi Jack- thanks for the feedback! We're planning a follow-up on that exact issue especially given that it appears most people lean toward the collaborative effort as evidenced by the early poll results. Would be interesting to know which elements of collaboration people are drawn to most.
Thanks!
Alan
Chris B
wrote on March 16, 2011 at 2:07 pm
Alan, from my perspective it's very easy to make this decision as a former biomedical student/postdoc. #1 of 6 authors in the current environment in 1 year is an easy decision to make. Any grad student or postdoc looking at that will feel this is a win-win for their CV, and when I read this poll it was pretty easy for me to select the first option.
Not to say that the second option is not worthy; but if we assume an average time span of 5 years for a grad student plus another 5 years as a postdoc, and you are looking at the academic career track – if you can add even two pubs of that type to your CV in that time, in my experience this is no mean feat.
However, if over the same time frame, you could choose adding 6-10 pubs of the first type instead, I think most people would instinctively select that option. 6 authors is not so many as to imply that you (as first author) didn't do the bulk of the work (unlike the genomic studies you give as an example above), and I would argue in most cases of this type many of the lower/middle authors are PIs in collaborating labs anyway. So I think that there is a bit of an imbalance between the two choices here – but that is my own personal observation based on experience.
A good CV is going to be a balance of both publication types, as well as other projects you have been involved in, and I am sure this will vary by field, but my impression from looking at this is that if I had had the choice, over my '10-year training' period, to have 10 publications of the first type, versus 4 of the second type, I would pick the first option without too much though going into the decision. So I think that I would be interested in having your follow-up give some decision options that will be tougher to weigh against each other – this will be very informative
alan@benchfly
wrote on March 17, 2011 at 12:31 pm
Chris, I agree. I think a good CV should have a mix of publication types. In fact, I think exclusively publishing papers where you're the lone author may even raise flags about your ability to share, collaborate and work with others. Based on your grad school/postdoc math, I'd much rather have 10 collaborative first author papers than 5 of the second type- but that's just my preference.
We'll follow-up with a broader range of trickier options- if you'd like to add any scenarios just let me know ([email protected] or in comments).
12 Reasons to Establish a Collaboration (or Not) | BenchFly Blog
wrote on March 21, 2011 at 11:36 am
[…] out, we scientists are a very collaborative bunch! Well at least according to our recent article, There is No ‘I’ in Research, where researchers overwhelmingly opted for sharing the biline with others in order to complete […]