Today we’ll be enjoying another episode of An Event in the Life. This one is really more of an experience than an event, but I already named the segment, so here we are. Semantics aside, our focus today is the paper! Specifically, we’ll learn about the purpose and the process of publishing in a scientific journal. There aren’t many objective metrics when it comes to comparing one student’s body of work to another’s, but number of papers is one of them. Publishing gets you, your group, your department, and your school fame and power (in a scientific sense, which is far removed from any sort of real sense), and it is extremely rare that a student graduates without doing it at least once (heh). Thus, I figured the process deserved a visit.
The first question, of course, is why do we publish papers? The altruistic answer would cite the advancement of science and, by extension, society. Journal articles are how scientists disseminate their findings both to one another and to the broader community. Any time you read something online or watch a news segment that opens with, “A new study suggests…”, they’re ultimately referencing a dry, boring article in a scientific journal. Every technological advancement or breakthrough has humble origins dating back to a manuscript written in the passive voice. For a recent example, here is the paper in Physical Review Letters describing the experimental observation of gravitational waves. In general, though, science is a very incremental and gradual process: each successive paper in a field offers a sliver of new insight which in turn motivates other researchers to make advances of their own, which in turn lead to another publication, and so on. Thus, a chain reaction of sweaty, nerdy fervor occurs, and in the end society moves forward and we get a bra which doubles as two face masks.
The other reason to publish (and maybe the more compelling one depending on your position) is recognition. As I mentioned earlier, publication record is one of the most objective (albeit imperfect) metrics available for judging and comparing researchers. So, when companies and institutions are trying to decide at whom they want to throw money, they’re probably going to choose the professor who farts out papers most frequently. Similarly, when companies are looking for a graduate student to hire, they’re going to look for a student with several childish blog posts publications to his or her name. In short, if someone wants a quick, digestible snapshot of your performance as a graduate student, postdoc, or professor, they’ll look at your publication record.
Of course, not all science is of equal import, and accordingly not all journals are of equal prestige. Think of it like this: most scientific articles can be boiled down to a nugget of information of the form, “When I do this, this other thing happens (and here’s why).” These nuggets vary in quality anywhere from gold to the turd variety. Those in academia refer to the nugget quality as “impact.” Nuggets about which other researchers are likely to care a lot are deemed “high-impact” and therefore merit publication in a high-impact journal. These include journals like Science and Nature. Both of these journals have a very broad reader base due to the fact that they accept only the coolest, flashiest, and most influential work. Getting published in either of these looks very good on a resume and earns you a massive science peen. Moreover, because of the broad readerships of these high-impact journals, articles here will get more exposure and are thus likely to be cited more times by other researchers. On the other hand, if your findings are more esoteric and likely only of interest to other researchers in your specific field, then you would likely submit your nugget for inspection to a more specialized journal. These journals have a narrower readership but are very helpful for researchers looking to learn more about the topic in question. Ultimately, the standard approach is to start at the most prestigious journal at which you have a nonzero possibility of getting accepted and work your way down the impact ladder with each successive rejection.
Now that we’ve talked about why we publish, we can discuss the process. Acquiring enough data to constitute a complete nugget can take anywhere from a few months to a few years. For the purposes of this post, we’ll fast-forward past the procrastination and tears to the point where it’s time to prepare a manuscript for journal submission. It varies depending on the journal, but articles are typically 2000-4000 words in length and contain 3-5 figures. I prefer to write my papers in LaTeX because user interfaces are for old people and I enjoy being confused. For figures, I first generate plots in MATLAB (R2008b edition, they don’t make ‘em like that anymore) then use Illustrator to endlessly obsess over minor aesthetic details. To summarize the typical framework, most papers will begin by explaining why the topic in question is worth researching (e.g., “Approximately 50% of people wear bras, and 100% of people are susceptible to toxic particulates”). Next, the author will usually provide a brief perspective of the field in its current state and lead into a problem or unanswered question which will be the primary subject of the paper. From there, he or she will introduce his or her hypothesis, describe the corresponding experiments, and finally spend the bulk of the paper discussing and interpreting the subsequent results and their implications. Personally, I like to wrap things up by vastly overstating the importance of the work and making questionable extensions to future research (which I claim to be “actively pursuing” to dissuade people from trying to scoop me). Add an acknowledgment to Tarun as well as to whatever funding sources Jen allocates to support my leisure-focused lifestyle, and it’s good to go!
Once you have a complete manuscript, it’s time to submit it to the journal of your choosing. Generally, the journal will have some web portal which they have deliberately made as clumsy, tedious, and user-hating as possible in an effort to dissuade all but the most resolved authors. You’ll have to upload the paper, every individual file contained within or related to the paper, receipts for every meal you consumed while writing the paper, and a video showing your face as you get increasingly angry during the submission process. After everything is uploaded, the paper will be distributed to one of the journal’s editors. The editor will skim the paper and determine whether or not it’s complete shit. If so, he or she will reject it and you’ll have to submit it to a different journal. If it’s not shit, the editor will send it out for review. Now, when you submit a paper, you’ll also typically include a cover letter in which you try to beguile the editor into choosing your friends as reviewers. If the editor decides to not immediately reject your paper, he or she will read the cover letter, examine the names of people you suggested as reviewers, and chuckle sensibly before choosing two names at random from some compilation of the biggest assholes in science (to which all editors apparently have access). These douchebags will then be asked to review your manuscript for the journal. Essentially, they’re supposed to read your paper thoroughly and advise the editor on the merit, rigor, and relevance of your work before ultimately giving a verdict on whether or not they think your manuscript should be published in the journal in question. Of course, the people who are actually at the top of the field are likely too busy to spend time doing reviews, so some dumbass graduate student will do it whenever he grows tired of writing articles for a blog which only his mom reads.
Ultimately, someone lacking the appropriate qualifications, interest, and time will read your paper and suggest that the editor do one of three things: publish the paper (with or without suggested revisions), reject it, or solicit revisions from the author before making a decision. If the reviewers express contradicting opinions, the editor will always go with the more negative of the two. As such, your paper will likely be rejected. From there, you can submit to a different journal, or, if you’re feeling particularly masochistic, you can appeal the decision (spoiler: you will fail). If the editor asks for revisions instead of giving you a full-mast middle finger, you’ll then have to update the paper and possibly even get more data to satisfy the reviewers. This is especially fun, because at this point it’s been 2-6 months since you first submitted the manuscript, so you don’t even know what it’s about anymore. Eventually, if you’re lucky, some journal will agree to begrudgingly slap their name on your work. A few days later, you’ll get proofs of your paper to make sure that everything looks as you intended. It is also your opportunity to reject any changes made by the editorial staff of the journal. For example, if it’s a British journal, this is the point at which they would go through your paper and add “u”s to words at randoum. Once you sign off on the proofs, the article will go live, and you can revel in telling the small subset of your friends and family that will actually care.
Depending on your field, your adviser, and your productivity, you’ll endure this process anywhere from once to upwards of five times before obtaining your degree. Some departments have a rigid publication requirement before you can graduate, but mine does not. If you’re planning to stay in academia, your publication record as a graduate student is hugely important. It’s less so if you want to go into industry, and if you want to leave science behind all together, it doesn’t matter at all. Therefore, the key is to decide your future as early as possible so you can spend your time in graduate school playing the appropriate amount of video games.