You are here
Graduate students and publishing
I just stumbled on this very interesting discussion on why science graduate students should publish, regardless of their later career intentions. I agree with the author on most points, but believe it really comes down to two things: (1) if you aren't going to communicate your results (both good and bad!) then you might as well have not bothered to do the work, and (2) becoming a good writer is a skill that every technical person will need in any career.
»
- MichelleLOyen's blog
- Log in or register to post comments
- 14702 reads
Comments
Publishing
Michelle,
My experience is that null results (bad results?) are not publishable. And results that contradict established methods are extremely difficult to publish. Here are two more things I'd like to add to the list:
1) Sometime reviews will take a year to come back to the author asking for you to do a couple of months worth of additional work (that has happened in my case). By that time you will probably have moved on to other things and its very difficult to get back, finish, and resubmit the paper. I would strongly recommend that authors in that situation immediately take some time off their current work and finish the additional work requested by the reviewers and resubmit. Otherwise the initial excitement of doing the work will fade and you will never have the motivation to finish the paper.
2) I have learnt, after much denial on my part, that papers are the currency of the academic world (though they don't have as much value as patents in industry). Every project that you work on should have associated papers as the final goal. You should start writing the papers from day one of the project and keep adding/polishing until the project is complete. You will have a paper ready to submit as soon as you move on to something else. Otherwise the new project will take all your time and you will never have the time to write up a paper on your work.
Biswajit
Whitesides on writing a paper
I particularly appreciate your point 2). For each beginning student in my group, I ask them to read an essay by George Whitesides. He advocates an approach that is quite similar to yours. That is, one should begin to outline a paper at the very beginning of research. I stronly recommend his essay.
That was really
That was really useful...Thank you :)
Publication issues
Certainly now that people have switched over to electronic publishing it can't be that common for reviews to come back after a year?
It's an interesting question/point about publication of null results or things that did not work. I guess to a certain extent it's difficult to generalize: it will depend on the nature of the poor or null result. Certainly in biomedical circles it's quite common to publish a null result, although the issue almost universally arises of the statistical power required to say that something had no effect (and lacking statistical power is a frequent limitation of such studies).
However, that issue is more interesting in the context of your other point, on the difficulty of publishing a result that goes against current thinking, or even worse of challenging large players in the field. Again, isn't the boom in electronic publishing, and the associated proliferation of journals, somewhat of a help here? I have heard it argued both ways, on the one hand that with more journals, only results published in high-impact venues get any attention (thus the proliferation of Nature journals), while on the other hand good work published in any venue, regardless of impact factor or journal reputation/history, can find an audience and make a significant impact. I can't decide which of these extremes is more likely to be the "winning" scenario in the end.
re:publication issues
Just a couple of thoughts on this topic.
To some extent, iMechanica appears to be the perfect forum for discussing "null" results. The experts can then chime in on whether or not they've seen the same thing, or even hypothesize why something doesn't work.
I typically find that reviews come back within six months. The bottleneck is not the submission process, but rather the human side of things. Reviewers don't always respond in the time requested, and editors are hesitant to bother them. Some of the journals have begun to use incentives (such as book points) for reviewers to provide timely responses.
I think you're correct Michelle that the proliferation of journals should help with the publication of results that cut against the grain. However, I think the problem may be that many libraries only subscribe to the top journals. When something interesting turns up in a database search and my library doesn't have an electronic subscription, I use interlibrary loan or simply contact the authors. I don't have the impression that many others take the time to do that.
review times
MIchelle and John: Certainly review times in the mechanics community are on the longer side. There are a some exceptions e.g. JMPS (based on my limited personal experience). In an extreme case, my wife recently recieved a review on her paper submitted after nearly two years! ---the paper was submitted to an ASME journal. On the other hand, the physics community for the most part appears to have resolved this issue. The Phy. Rev. journals as well as APL/JAP tend to do very well in returning reviews quickly. I always wonder why this is so?
John: You are right that the delay in review is due to the "human factor" but some journals and communities are handling it better and perhaps we can try to figure out they do it.
Length of papers
Pradeep,
Most Phy. Rev. papers are short (moreso the Letters) and hence it might be rather easy for the reviewer to gauge if it will be accepted/rejected. Then,given that the reviewer has to read only a few pages and if he/she is well-versed with the topic, a critique of the paper can be done in relatively short period of time. Also, given the space-constraints, I presume, on average, the papers are more focused and well-written. Acceptance rates of these journals might also be a factor. Probably `short papers' need to be encouraged in some of the mechanics journals, especially when a new theory/approach is being proposed.