Posts Tagged 'papers'

How to contribute to a paper

This one is rather off-centre, but my guess is that it might well be useful for some students. I think most people get their first papers published either working alone, or by collaborating with their supervisors or perhaps another PhD student, and thus you are in a familiar environment with people you know well. It can be intimidating to contribute to a paper with a couple of senior researchers or even to try and handle a big research group on your own as a senior author on a paper. While this micro-guide is aimed at the former situation, it should also serve to help with the latter as well.

Continue reading ‘How to contribute to a paper’

Analysing palaeobehaviour

With the recent publication of a new paper on pterosaur ecology it seems like a good time to strike up on the vexed topic of determining (or trying to) the palaeobehaviour of many animals. Actually there is not much to say about it that is not covered in some detail elsewhere (there are a few links here, but obviously check out Tet Zoo and Mark Witton’s site) but I can use the opportunity for another wild stab at people who come up with wildly inappropriate statements about palaeobehaviour. One point that gets repeatedly hammered by the authors is the endless statements (well speculations really) made in the past about pterosaur behaviour when not only were they not backed up by the slightest bit of evidence, but also were obviously false and could be shown to be, as it flat contradicted the basic morphology of the animals under study, or the principles being invoked. This happens alot, and not just with pterosaurs.

I understand that when you have described a new taxon or have been working on some functional aspect of an animal, you will want to try and interpret it and put it into context. But that is not a license to write whatever you want, without giving it proper thought and research. It might be a very minor part of the paper, but it should still be treated in a scientific manner. If you have an interesting idea about how an organisms might be acting or what a certain functional model might allow behaviour wise, its relatively simple (especially in the Google Scholar age) to hunt for a few papers on that area in extant taxa for comparison. If the basic principles are funamentally violated by your concept, or the extant organisms practicing it show none of the adaptations your taxon has, it is time to rethink the idea. Sadly, this does not seem to happen as often as you might think. The paper demonstrates this admirably (and Darren comments on it on his blog too) that some ideas just will not die, no matter how much evidence you can provide that rebutts it, or how obviously that concepts flat contradicts our understanding of the funamental biology.

The real problem for me, is that these ideas then get picked up and perpetuated in the literature until they become some kind of common consensus, without any back-up at all. Skimming by pterosaurs is a classic example that it taken apart rather ruthlessly by Witton & Naish. Even before the recent paper refuting skimming in pterosaurs on energetic and mechanical grounds, it should have been pretty obvious that no pterosaur really bears even the slightest resemblance even in gross morphology to extant skimmers. That does not rule it out of course, animals can evolve to do similar things in very different ways on occasion (though the rampant convergence of many groups shows that there is often is only one really good way of doing somehing – look at formicivores) but one would expect at least some of the highly specialised adaptations to appear in pterosaurs as they are going to be under similar pressures and experienceing similar problems, and they are easy enough to look for. So why did no-one check, at any point, and instead just recycled the idea again? And then why did other people not notice that this had not been checked and check it themselves, rather than just repeating the idea? It does all get a bit tiring when you keep seeing these unjustified or supported speculations attached to papers.

There is nothing wrong with concluding a piece of work with ‘I don’t know’, ‘We can’t tell’, ‘There is no obvious analogy for this’ or ‘This unusual feature is hard to interpret’, and it is certainly better than coming up with some unjustified specualtion that is not only transparently incorrect, but will then be recycled itself. It might seem frustrating that 20 or 30 pages of excellent work could end with ‘Don’t know’, but is that really worse than something incorrect? At the very least researchers (and once more, referees and editors) need to tighten up on this and not let unjustified speculation creep through into print. If you have a good idea, check it. The literature is available and some basic comparisons (like skull shape, foot size, neck length, whatever) can form a very simple basis for comparison and allow the elemination of at least a few concepts and the support of some others. It need not be in depth, that is not the point, but it should at least be supported and currently the opposite is far more common that it should be.

This is a modified Mk.1 post, to see the original with comments etc., go here.

Fodonyx – a new genus of rhynchosaur

Yes my new paper is out (co-authored with my PhD supervisor, colleague and all-round-rhynchosaur expert Mike Benton) and as a result I want to talk about rhynchosaurs. OK, so they are not really dinosaurs, in fact they are not actually even archosaurs, but they are reptilian, lived alongside the dinos and are dead so they more or less count. And given what has passed for ‘dinosaur’ posts by me of late something on some other archosaurs seems appropriate. Besides, it’s not like you lot are going to stop me is it?

So rhynchosaurs then – the chisel lizards – so called because of their bizarre ‘tusks’ at the front of the jaws. These look rather like giant incisors and as a result give them a bit of a rabbit-like appearance, but these things aren’t teeth. In fact they are just modified jaw bones (the premaxilla above and dentary below) that in life would have stuck out through the flesh of the animal to form these pseudo teeth. In fact you can see marks on the bones where the skin would have stopped and the bony protrusion started.
Continue reading ‘Fodonyx – a new genus of rhynchosaur’

Publications – more effort than you might realise

What exactly do we do as scientists? Well, I know the DinoBase readers are a bit more ‘science aware’ than average (well they seem to be), but I suspect that more than are few are not really aware of the work that goes in to getting a paper published. I’m not here talking about writing a paper – checking up on previous research, collecting the data, doing the analyses, collating the references and writing it in a coherent form is work enough. Getting the damn thing into print is entirely another matter.

My first paper!
Once you have written the paper and you bung it off to a suitable journal the editor will read it. If he thinks it is interesting and is suitable for the journal then it will be sent out for review (and if you are a decent researcher this should happen pretty much every time – the obvious exceptions being Science and Nature). The review process is one of the most import features of scientific study, something often overlooked by the public who assume that we can happily publish any old tosh that gets sent in. It is there to make sure that no one published false or misleading results as well as to check the suitability of the analyses used and the conclusions drawn by the researchers.

The reviews are carried out by at least two experts in the filed that the researchers have written about. The reviewers are also active researchers and so are familiar with the existing work and methods used in this area. They will read the paper very carefully and check everything about it – really. Are the any major mistakes in the introduction about the current state of the subject, has the data been collected properly from the right sources and with appropriate methods, are the statistics correct, is their discussion justified? Reviewers will check all of this in detail and report any mistakes to the journal’s editor – right down to spelling mistakes and missed full stops.

If there are too many mistakes, or some serious problems with the methods and results then the editor will reject the paper and ask the researchers to start again. That is generally not the death of the manuscript, but it is going to take some serious work and lots of time to correct and most often you will have to find a new journal who is interested. However, if the mistakes are not too serious, then he will send the reviewers’ comments back to the researchers and ask them to do some more work to bring it up to scratch. They might need to collect some more data, or look at some other papers on the subject, or just use some extra statistics to back up their arguments. Even this ‘little’ work can take days or weeks. If you missed one critical data point you will have to go and get it, and then redo all the stats, redraw all the graphs and check that it doesn’t affect your conclusions, and if it does start writing again.

Once you have got this extra work has been done the revised paper can go back to the editor. If he and the reviewers are satisfied with the changes then it can be made ready for publication and will eventually appear in the journal. All of this takes time though. Reviews often take several months if the reviewers are busy, and it is not uncommon for a paper to appear in print 2 years after it was first sent to the editor and some can take even longer. One of mine took closer to three and I know of odd one that have taken 3 or 4 years from submission to publication even though there was little wrong with them. Everyone is busy these days, and doing reviews on your own time when you have exams to mark or grant deadlines to meet is rarely anyone’s priority. So, although it might appear easy, it is a mammoth task to get even one paper published. Really. It’s just easier with time and experience on your side, but it is always a challenge to get it right and make it interesting and accurate.

One thing that often annoys me (here he goes again) is often a piece of work is dismissed by others who are unaware of the work that has gone into it (yes I am looking at *you* DML). Yes bad papers get published, and yes, referees and editors make some terrible mistakes, but they are the exception, not the rule. Pretty much by definition anything that gets published has been reviewed by two (and generally more) people in the field of work the paper is about, plus a journal’s handling editor and probably the main editor too. If it is a paper with multiple authors they will have read it (well, you hope so), and generally you will have asked a few friends or colleagues for comments too. To then have someone turn around on the internet the day its published to suggest that your work is wrong and you don’t use the right analysis is pretty galling. Yep we make mistakes. Often. But the idea that about a dozen of us (authors, referees, editors, colleagues) are all completely wrong and failed to spot an enormous flaw is frankly, pretty unlikely. Far more likely is that you have missed the exact nature of the analysis or the reason that it was done, not that so many experts are wrong (and have been checking the work over months if not years) and you are right. Read it again!

Well, apart from a relatively small rant on the side, that’s the process of publication. It is slow, often frustrating, time consuming and apparently unnecessary. But it is also essential, and can’t be ignored so for now we just have to stick with it. Sadly, they won’t just let you publish whatever you want with no justification or data, though I could mention a more than a couple of journals where that does not actually seem to be the case…..

This is a revised Mk.1 post, to see the original with comments etc., go here.


@Dave_Hone on Twitter

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 503 other followers