It takes time and effort to publish a paper. After all, if it were really easy, then publications wouldn’t be a workable (albeit flawed) currency in for success in the sciences.
I often have heard about how some labs experience a bigger or smaller MPU (minimum publishable unit) than others, as I’ve worked in biology departments with a lot of academic diversity.
For example, I once knew an immunologist in an undergraduate institution who spent five years of consistently applied effort, to generate a single paper on a smallish-scale project. This wasn’t a problem in the department, as everyone accepted the notion that the amount of work that it took to generate a paper on this topic was greater than what it would take for (say) physiology, vertebrate paleontology, or ecology.
As another example, I knew a physiologist who was one of the more productive professors in his teaching institution. This person was quick to minimize the research, claiming that it’s easier to get publications in physiology than in other disciplines. You run an experiment, you write it up, then boom, you have a paper. It’s not that much work at all, according to that person.
So, is it true that it’s harder to get a paper out in some fields?
I think it appears to be the case, especially when I think of differences among job applicants, and also that h-index values vary substantially among fields.
My lab took a molecular detour recently, and I can attest that the molecular part of the paper was not a mountain of data but required a few months of focused lab work, as well as thousands of bucks in reagents. (That said, the fieldwork for this paper took way more time and was more expensive.) And we had collaborators who smoothed out the rough spots on the molecular end, too.
What running an experiment, you need to do one thing, which takes a lot of stopping, starting, and pauses. Then do something similar again, and again. It takes months of time in the lab, even if it doesn’t take months of work in the lab. Why do scientists drink so much coffee? Because they always are waiting twenty minutes for something. (Meanwhile, genomics seems to have different constraints.)
A lot of the work in ecology can be done in parallel, while molecular work usually has to be done in serial. So if I have a small team working together really hard over a summer, we can be sure to get a paper out of it. A small team in a molecular lab? Probably not.
If you have a couple undergrads working in a molecular lab over the summer, there is a cap on productivity, no matter how hard-working the students are. You probably won’t (and probably shouldn’t) get more than 8-10 hours of benchwork out of a person per day. Moreover, in eight hours of benchwork, you’re not really getting eight hours of work, because at least from my perspective, there seems to a fair bit of waiting around for a PCR, a gel, an incubation, or something.
I manage to get a lot of hours of time – and work – per student over the summertime. I get this to happen by taking students to a distant field station, where they work for the whole summer. (This takes a few weeks of my time, plus being consistently available for email and video chat for the rest of the summer.) At the field station, students have little to do other than work. They sleep and eat right next to work, and they’re working alongside other scientists who are working their butts off. The amount of time and work that my lab might put in over one summer, counted in hours doing science, can be equivalent to several summers of work in a M-F 8-5 lab.
We can design experiments that can maximally benefit from replication and hard work in the field. The more time we put in, the quicker and better we get to a paper. That sounds normal, but I think most molecular papers can’t be cranked out in a short period of time in an undergraduate-powered lab, no matter how hard everybody works.
Here is an idea: Though it is harder to get papers out in some subfields over others, maybe it doesn’t require more work? Let me explain. Perhaps the number of person-hours required to get work done for a paper is equivalent among subfields. However, in different disciplines, it may be more or less difficult to get those person-hours accomplished. When my lab flies down to the rainforest, and we work really hard together for a few weeks, that’s a lot of time and effort. (I don’t think I’ve seen any other PI at the field station for weeks at a time for several years.) Even if one of my former deans once referred to it as “vacation.”
Okay, that idea in the preceding paragraph is probably off. I think it’s possible for a quick paper to be knocked out in evo/eco/behavior if you have an interesting idea or access to certain resources. A lot of amazing papers are now getting published with pre-existing data, by people who don’t make their own measurements and are really good at managing data and doing analyses. A good idea goes a long way.
As another example, I’m writing this post on a lazy Sunday morning in the midst of a two-week trip to the field (in Darwin, Australia). I’m setting some students up with a mentor here for their own long-term projects, but I’m able to take about eight days in the lab to do a short little project on a cool and weird behavior that nobody really understands. It won’t change the world, but it’s fascinating to me and I think it’s a nice bit of experimental natural history. If I am a writing monster, the manuscript from this project could be (should be?) finished before my plane touches down at home. It won’t be a big paper, but it’ll get published in a real journal and will “count” for those who are counting pubs. No one is going to mistake this for a big-time discovery, but I’m not going to say that this is an “easy” project. Nevertheless, my colleagues who are molecular biologists probably don’t have the option of hopping on a plane and coming home two weeks later with a brand new manuscript that could be submitted to a decent discipline-specific journal.
Do you think it’s easier to get a paper out in some fields rather than others? More importantly, what is the prevailing attitude in your department? Are these things taken into account when making hiring decisions, and in tenure criteria?
Ecologists also seem to benefit from focus; the papers often seem more discrete, about one small experiment in a system (or a few, perhaps), but they’re in isolation from a lot of other things. And a lot of it is observe and measure (though molecular ecology would take more time).
In the molecular world, the thing that takes so much time is often tool development and trouble shooting. And reviewers often seem to demand fully complete stories…how does your molecular discovery integrate with everything else going on in the cell/organism? The ‘just one more experiment’ phenomenon that seems to be less of a thing in ecology.
It’s not that it’s easy to get published anywhere, but the stories in each paper often seem less dense in ecological than molecular work. Each paper is about one thing, not trying to account for every single little thing that might be going on.
I really enjoyed this post Terry! It’s given me a lot to think about in terms of structuring research projects to try and be more efficient. You are definitely on to something when talking about the incubation times in most molecular experiments. It also takes me quite a bit of time to train each person in my lab in any particular technique to the point that they can do it well. I’ve often wondered about how quickly I could do a set of experiments, write it up, and submit a manuscript to a journal. It would definitely be longer than 2 weeks!
I’ve always felt that the amount of trouble-shooting required makes a difference. In lots of sub-organismal areas, it seems that a lot of work has to go into making procedures yield the desired outcomes (toying with reagents, doses applied to study animals, confirming the technique achieved the desired effect, etc.) In my experience, there is less of that in areas like ecology and behavior. I don’t have to worry about whether surveying vegetation or observing an animal’s behavior will ‘work’. It will work. It just does. (Well, somewhat reliably, but I still spend a lot of time trouble-shooting). I think this produces data quicker, but also perhaps raises the proportion of projects that yield ‘uninteresting’ null results.
To echo the previous comments, it’s all about troubleshooting. I’ve worked in both phys-type labs and molecular labs. Molecular biology requires way more troubleshooting than physiology. In phys, if my experiment doesn’t work after a couple of attempts and minor adjustments then I cut my losses and move on to the next project. In molecular biology, there are so many details to check before you trash a failed experiment (was my sequence correct? Were my incubations long enough? Were my ratios of reagent to DNA/RNA correct? Etc.). Of course, you can avoid some heartache by doing preliminary experiments to optimize your protocol, but that also takes time and money.
It may be a little misleading to talk about “ecology research” as if it is all one thing. Some projects will require multiple field seasons to account for inter-annual variability. And bulldozers. (Yes, one PhD project was delayed by a year because the field site was bulldozed. In a state forest reserve. Long story, but I’m glad I wasn’t relying on that one to push me over the tenure bar.)