Impacted science

The next step in writing up a manuscript is evaluating the strength of your study. Is it a novel, never seen before breakthrough, or is it simply adding on to the existing knowledge? Do you show cause and effect, or only correlations? Did you use multiple models/methods to make sure what you’re seeing is accurate? Is it descriptive (watch as I cringe at this death knell) or mechanistic?

All of these affect the quality of your work and the strength of your conclusions. You have to be honest with yourself. There's no point in formatting your research to fit a high impact journal if the reviewers are going to take one look at it and bounce it right back. There's no real harm in being rejected, but it's just not worth your time. Aim high, but be realistic. But what exactly do I mean by "high impact"?

There are good journals and there are not as good journals... and then there's flat out bad journals. But unless you've been in the field for a while and can recognize the hundreds of names that all kinda sound the same, how would you know where to submit your work? Well, by calculating its impact in the scientific commmunity using complicated equations, of course! Doesn't everybody assign value in a numerical manner?? No? Just us scientists? Ok, I'm alright with that.

Actually, to be honest, it's quite a simple calculation. To paraphrase Wikipedia, it's the average number of citations received per article published in that journal during the previous 2 years. The problem is that review articles get cited by other researchers way more often than original articles (ie: actual research), because it's easier to cite one review that summarizes the field rather than 10 individual pieces. So if a journal publishes lots of reviews, it gets an artifically high impact factor, as it's not actually showing the quality of work accepted there.

But some journals publish truly noteworthy that get deserved attention and citations. And herein lies the holy research trinity: Science, Nature and Cell. Scientists drool over these journals, they dream about their name on those papers. Tenure-track careers are built on publishing just once in these journals. They are always in the top 20 journals (most of which publish exclusively reviews), with astronomically high impact factors of 30-45. I know that doesn't sound huge, but think about the meaning of that: each paper published in Nature in 2011 and 2012 had 42 other articles cite it in the next 2 years!

The other end of the spectrum is the super low impact (0-2), not as trustworthy of journals. The science is probably still sound (it is still peer-reviewed after all), but not much is expected in terms of new ideas or details; not as many double checks need to be run to make sure everything's accurate. These articles might be mostly negative data (as in, "well, we know factor x isn't causing cancer"), or reproducing what others have seen and only adding a bit more information. This is important work, I've talked before about replicating experiments as being vital to science. But it's not flashy, it's not exciting, so it doesn't get cited much.

Most journals fall somewhere in the middle, but the numbers vary by research area. A solid journal in my field (gastroenterology) probably has an impact fact of 4-8. These tend to be more specialized and don't accept papers that are outside of their area of interest. If you're ever curious, is a great resource for looking up these numbers.

Anyway, the other factor to consider when choosing a journal is whether you have enough data! More and more often, high-impact journals are expecting in-depth studies with enormous amounts of work that simply isn’t possible for just one graduate student to complete. While that's good for encouraging collaboration, there are issues about how much data is required, how that data is presented (the majority is put into supplemental figures), and how much emphasis is put on publishing in "luxury" journals. But that's fodder for another post.

So after picking an appropriate journal, the next step is to organize your figures and write up your work in the format they want. That part is pretty straight forward in theory, but takes some work in practice. Which is what I'll discuss next time. Probably. Unless something else catches my interest!

PS: Did anyone catch the scatological humour in the title? I should have a prize for whoever sees these without me pointing it out, because I love a clever poop joke!