C&E News on writing journal articles.

Since scientist-on-scientist communication is a longstanding topic of interest in these parts, I wanted to point out a recent (August 13, 2007) article in Chemical & Engineering News (behind a paywall, but definitely worth locating a library with a subscription) that offers tips for writing journal articles. It's quite a substantial article, drawing on advice from "dozens of scientists and engineers around the world in academia, industry, and government" -- which is to say, the people who read and write journal articles as part of their jobs.

It goes without saying that this crowd has some strong views.

One thing you might wonder is how chemists (among other scientists) learn how to write journal articles in the first place. According to the article:

Many chemists and engineers learn journal article writing do's and don'ts by example from graduate or postdoctoral advisers. That is, the students examine published articles, write drafts, discuss edits, and revise multiple times. Professors admit that a manuscript could be completed much faster if they wrote it themselves, but the exercise would lose its educational value.

I think it's a good thing that graduate and postdoctoral advisors see it as part of their job to help their advisees learn to write good papers for the peer-reviewed literature. It's less obvious, though, that advisors always have a good grasp of how to teach their advisees the relevant skills. Some scientists are more reflective about their writing, and some have probably even had good training in pedagogy. But in my experience, there's often an assumption made that if you know your subject, you can teach a trainee what s/he needs to know to get the job done. That's as risky as the assumption that if you know your subject, you can write about it clearly.

That C&E News is running this article, however, shows that someone got the memo that writing is a skill that can be learned and taught rather than a natural talent that you either have or don't.

The article gives the clearest articulation I've seen so far of the norms for authorship in the community of chemistry:

The first author is usually the person who contributes significantly to the research and assembles the manuscript. In academic settings, the first author generally is a graduate student or postdoc. In industrial or government settings, the first author may be a principal investigator (PI), staff research associate, or postdoc. Sometimes the first author and the corresponding author are the same person. The corresponding author in most cases is the PI, who probably holds the grant and oversees the work from idea through preparation of the manuscript.

Of course, there are lots of local variations in how authorship is negotiated (with some interesting shifts in where PIs show up in the author lists pre-tenure versus post-tenure), and there's no mention of the right way to work out the order of the authors listed between the first author and the corresponding author. Still, it's nice to see some awareness that the ordered list of authors communicates something to the reader. Indeed, in the list of 12 tips for writing a journal article, number 3 is:

Clarify authorship carefully. Sometimes the authorship is straightforward; sometimes it's not. Authorship on journal articles can become an ethical issue, and in certain cases, disputes have cost scientists their jobs and reputations. "Be meticulous and make sure that authorship is correct," recommends Sean B. Seymore, a professor of patent law at Northwestern University who holds a doctorate in chemistry and has written about authorship abuse (http://law.richmond.edu/jolt/v12i3/article11.pdf).

The dozen tips the articles gives are all really useful. I'll just mention two more that jumped out at me:

Tip 6. Know your audience. Nonspecialists will read your journal article. [Cornell University chemistry professor Roald] Hoffmann advises scientists to "write the manuscript for an intelligent graduate student, not a professor."

When writing up interdisciplinary work, take nothing for granted and explain everything, says University of Iowa physical chemist Vicki H. Grassian, who works on environment-related surface science and nanotechnology. For example, she has had reviewers repeatedly question particular calculations for atmospheric reactions that have been "routinely done in heterogeneous catalysis for more than 50 years," she says.

This isn't surprising advice, but it's important. The point of writing is communication, so thinking about what your audience already knows is important in getting your message across successfully.

And, a tip that's important for more reasons that you might think at first:

Tip 1. Conduct thorough literature searches and cite precedents. "Good literature searching allows you to provide a cogent paper that is well-thought-out and well-organized, and it also keeps you from embarrassing yourself," says analytical chemist W. Jeffrey Hurst at the Hershey Co. For example, it behooves you to discover earlier rather than later "that what you thought was seminal work has been reported on 12 times" already, he says.

Of course, it's important to give credit to the scientists whose work paved the way for your own. But it's also really important to be aware of what other scientists have already done in the area you're researching. Probably, it would be worth figuring this out before you sit down to design your experiments, but if this detail gets away from you, finding out whether your results are groundbreaking or instead merely confirm what others have already reported might give you more realistic expectations as to how enthusiastically the journal editor will receive your manuscript. (I don't have to tell you, do I, that one should never cite literature one has not read?)

All in all, a very useful article, and one I hope will get chemists talking with their students (and each other) about ways to make journal articles even better instruments of communication.

More like this

Dr. Isis considers a downside to having coauthors and an ethical question it raises: Imagine a hypothetical postdoc that has just left graduate school, although this could easily be an assistant professor that has just left a postdoc. She has some minor publications either published or in press.…
One of my correspondents told me about a situation that raised some interesting questions about both proper attribution of authorship in scientific papers and ethical interactions between mentor and mentee in a scientific training relationship. With my correspondent's permission, I'm sharing the…
There's a recent paper on blogs as a channel of scientific communication that has been making the rounds. Other bloggers have discussed the paper and its methodology in some detail (including but not limited to Bora and DrugMonkey and Dr. Isis), so I'm not going to do that. Rather, I want to pull…
When non-scientists think about the big ethical issues in the practice of science (beyond questions of how much freedom scientists should have with the tax-payers' money, and whether scientists ought to be "playing God"), they usually think about the three mortal sins of fabrication, falsification…

(I don't have to tell you, do I, that one should never cite literature one has not read?)
********************************************

Always amazing though when that doesn't happen. In grad school, my PI was writing a grant. He/she followed a particular statement of "fact" from the latest review on the subject back thorugh earlier reviews all the way to the original research article. What did he/she find? The "fact" everyone was not supported by the research. The findings were the opposite in fact. The first review got it wrong and no one figured out the mistake.

A very interesting article. Every scientist needs read it. Scientists can't afford to learn how to write well. Writing well entails using language easily comprehensible to one's target audience. In often times, scientists resort to scientific jargon to explain issues. This is unacceptable. Two years ago, I started a blog on agricultural biotechnology after learning that scientists were shy confronting critics of crop genetic engineering. The blog has been doing relatively well - attracts on average 800 visitors everyday. I would like to encourage scientists to engage more with the public by making use of new media like blogs, podcasts, Vlogs and forums

It has been nice to see some grad programs start to notice that they should be teaching scholarly writing, though I haven't seen a program that does it rigorously yet -- largely because, understandably, the profs never had a course like it so they have no model to draw from. The course I took was mostly gossiping about people in the field, and reading a few outdated article and two horrendous books. By doing a presentation, I did learn a lot about how my science was being conveyed to a mixed audience, which was great. But I didn't learn how to write.

In addition to teaching a skill set, I'd like to see a grad professionalization course that worked to build relationships and a safety net so that students would feel like they could really go for it in their writing. I'd rather see a really messy first draft with one or two gems than a prim and proper draft that says nothing of interest. But until grads know they won't be judged for messy drafts and making mistakes, they'll keep writing safe stuff (and doing safe science, safe dissertation proposals, etc). I just finished my PhD in May and it was a struggle to maintain the perspective over six years to really push boundaries with my ideas and my writing, because so much of it got shot down or harshly criticized.

Sorry that was so long :).

(I don't have to tell you, do I, that one should never cite literature one has not read?)

Part of the reason my Histopathology casebook dissertation for my fellowship took so long was that I did just that and read every word of each article, rather than taking the easy way and belting each of my eight cases out over twenty four hours, thirty downloaded abstracts, and a bottle of wine. Technically these are supposed to be publishable articles. In the end, much of the time wasted was time spent waiting for inter-library loans to come to my rescue, in which case I used the data in the abstract and revised it if the 'meat' of the article changed things.

The only time I did rely wholly on the abstract was when inter-library loans failed me and I wasn't willing to pay $30 for an obscure reference which was probably only going to add an insignificant amount to the article anyway.

When I was in first year in Medical School, the topic "Research design and analysis" was primarily run by a bunch of non-medicos - psychologists and statisticians for the most part - attached to the Division of General Practice. Learning how to read a paper and critique it was IIRC taught by an epidemiologist.

NOT ONE clinician or hard scientist played any significant role in that teaching, and the question of the relevance of clinicians performing research accordingly remains a contentious one for me. To this day I would put any form of laboratory based research, as well as most others, in the hands of university academics and keep it firmly out of the hands of clinicians, who already have enough to do without having to run on the academic treadmill too.

By Justin Moretti (not verified) on 05 Sep 2007 #permalink