Scientist Sees Squirrel

How do you choose a journal when it’s time to submit a paper?


Image: Three choices – out of thousands.

Warning: long post. Grab a snack.

Having lots of options is a wonderful thing – right up until you have to pick one.  Have you ever been torn among the two dozen entrées on a restaurant menu? Blanched at the sight of 120 different sedans on a used-car lot? If you have, you might also wonder how on earth you’re going to choose a journal to grace with your latest manuscript.  There are, quite literally, thousands of scientific journals out there – probably tens of thousands – and even within a single field there will be hundreds of options.  (Scimago lists 352 journals in ecology, for example, but that list is far from comprehensive.)

What follows are some of things I think you might consider when you choose a journal.  There’s  no single “right” choice for any manuscript, because there are multiple factors at play and how you weight them will depend on things like your career stage, your budget, your coauthors, and more.  But a little systematic thought can at least narrow the field.

By the way: I’m pretty sure that parts of this post will make some people very, very angry.  I’ll be mentioning impact factor, and open-access publication, and a couple of other things, and I may not stick entirely to the sacredly anointed proper positions on them.  Failure to stick to those sacredly anointed proper positions seems to provoke incoherent rage rather than thoughtful argument in a few people.  If you’re one of those people, you might prefer to read something else.  Here’s one of my least controversial posts.

Does journal choice even matter?

 In this age of online keyword and full-text searching, some argue that it just doesn’t matter which journal your paper is in, because readers will find it anyway (hence the existence of megajournals like PLoS One).  This argument is superficially appealing, but in practice, I think journals still function to steer papers to their audiences.  Others take the argument further, holding that journals don’t matter because readers shouldn’t infer anything about your paper from where it appears (for example, during hiring or promotion decisions).  The “shouldn’t” part of this may be correct, but since we can’t enforce “shouldn’t” on those evaluating us, I think this argument fails too (except possibly for the already-tenured-and-famous).  I explore both arguments in more detail here; but in what follows, I’m going to assume that you (like me) care about journal options.

When should you choose?

I’ve had students and colleagues who wanted to write the manuscript first, and then decide where to send it.  I think that’s a mistake.  Instead, I recommend choosing a journal target very early in the writing process – certainly before writing the Introduction or Discussion, and likely before finishing the Results.  A good stage is when you’ve got data, and you’ve produced a bunch of rough graphs and tables, and you’re trying to figure out what story to tell*.

There’s a simple but important reason for this.  You can’t figure out what story to tell unless you know who you’re telling it to.  Perhaps I’m writing a paper about an invasive beetle that attacks spruce trees.  If I choose to send it to the Canadian Entomologist, I’m choosing to tell my story (mostly) to other entomologists.  If I choose Forest Ecology and Management, I’m telling my story to forest ecologists; if I choose Ecology, to ecologists in general; if I choose Biological Invasions, to invasion ecologists.  None of these decisions is wrong; but each shapes how I’ll write the paper, because these audiences will have different prior knowledge , they’ll need different context and information in what I write, and they’ll be interested in different angles on the story I have to tell.

There’s actually another reason for choosing a target journal early: you can save yourself considerable writing hassle by looking at an Instruction to Authors document early.  If your target journal has a 4,000 word limit, there’s no point polishing 8,000 words of text.  Some journals have figure or table limits, or citation limits.  Others have formatting quirks like Abstracts with numbered points.  Of course, very often your manuscript won’t succeed with the first journal you send it to, so you’ll have to make changes eventually, but why not make your first submission as easy as possible?

How do you choose?

Choosing a journal isn’t simple (unfortunately).  I can think of at least 9 factors that you might want to consider (I’m sure I’ve missed some, and I hope you’ll add to my list in the Replies).  And the choice suggested by Factor 1 might well conflict with the choice suggested by Factor 2, or 3, or 4 – which means you need to weight them to make a final decision.  I know this sounds a bit daunting; but in fact, I think a list of considerations like what follows helps give some structure to the decision-making process.

OK, here goes.

  1. Audience. The link between journal choice and audience is a 2-way street. When writing, you need to know which journal you’re writing for, so you can tell the right story.  But at the same time, the story you want to tell, and who you want to tell it to, may determine which journal you choose.  If I have data on that invasive beetle that I think are really important in answering a basic ecological question – like how southern range limits shift with climate change – then I probably won’t pick an entomology journal, because if I do, my paper won’t be obvious to the people most interested in that question.  Instead, a journal like Ecology or Global Change Biogeography would put my paper where the audience I’m after will be looking for it.

  1. Fit. Journals don’t publish papers about whatever people send in (except for megajournals like PLOS One, of course). Instead, each journal has a scope that defines the kind of paper they want to publish, and there’s no point sending your paper to a journal that will simply give it a “desk reject” for lack of fit. Usually, a statement of a journal’s scope can be found right up front on their web page, or in their Instruction to Authors.  But relying on that alone means missing more subtle cues.  Editors may have a pretty good idea what they want without having written it down in detail.  Therefore, it’s worth looking at past issues of a journal you’re considering, to find out what kinds of papers have appeared there.  Or you can go at the problem in the opposite direction, by finding half a dozen papers that resemble yours, and seeing in which journals they were published.

  1. Your CV, part 1: Impact factor**. It’s fashionable to hate the impact factor – and there are good reasons to hate it (for example, it’s easily gamed by publishers, and as a journal-level metric it has only a very loose correlation to the quality of individual papers). So many people insist that you shouldn’t infer anything about a journal from its impact factor, about a paper from the impact factor of the journal it’s in, or about a person from the impact factors of the journals they publish in.  But people still make all these inferences, and in fact they aren’t completely unreasonable.  (Again, fuller exploration of this issue here.)  Journals have reputations, and they correlate fairly well with their impact factors.  And that brings us to your CV.  If your CV can have 4 papers in high-impact journals or 4 papers in low-impact journals, even if they’re the same papers you’re better off, careerwise, with the former.  Actually, it’s not just your CV.  Impact factor does tend to correlate with visibility, and so more people are likely to see and pay attention to your paper if you publish it in a higher-impact journal.  You can overdo this, of course – nobody should care about the difference between IF = 1.0 and IF = 1.2 – but it’s entirely reasonable to think about journal reputation, and impact factor provides one signal of that.  Finally:  some granting agencies (e.g., I’m told, in Chile) etc. may have explicit requirements with respect to publication in journals with particular impact factors.  If this applies to your own situation: you can disapprove of it, but it would be foolhardy to ignore it.

  1. Your CV, part 2: Communicating interests. There’s more to the CV angle than just the (arguable) impressiveness of high-impact journals – especially if you’re early in your career. That’s because the set of journals you’ve published in will send messages about your interests to people who might be considering you as a possible hire.  Imagine that you have a pile of basic-science data, with an insect system, that have applied implications.  Where should you publish the papers?  If you’re aiming to be hired as a professor in a broad Biology department, it’s probably unwise to place all (or most of) your papers in applied entomology journals.  But if you’re heading for a career in management with a government agency, or you’d like to work for a conservation-oriented NGO, then a CV made up entirely of papers in general ecology journals may send the wrong message.

  1. Open-access vs. subscription. There are two fundamentally different ways in which journals distribute papers. If your paper is published “open-access”, it’s released free (online) to anyone at any time; but if your paper is in a subscription journal, it’s released (directly) only to those who pay for a journal subscription***.  (To make this a little more complicated, many subscription journals have open-access publishing options, although usually at substantial cost.)  It seems like an obvious proposition that open-access is better, and all else being equal, it would be.  All else is rarely equal.  Publication costs must be recouped one way or the other, so open-access publication is generally more expensive than subscription-based publication.  A decision to publish open-access, then, is a decision to forgo spending the same research dollars elsewhere.  Which brings us to…

  1. Cost. Usually, publishing will cost you; but some journals are more expensive than others. Publication costs can be substantial.  For open-access journals, the most common model is an “article processing charge”, or APC, which is a flat rate per published paper.  There are a few in the $100s, but most are US $1500 or more – and it’s not impossible to find an option costing US $5000.  Subscription journals usually levy “page charges” instead, often in the range of US $75-$100/page. Not all journals levy APCs or page charges, though; there are a few “free” ones.  (The quotation marks around “free” recognize that there’s always a cost to publishing; all that varies is who’s paying it.  “Free” journals may have foundations behind them, or may have more expensive subscriptions, but they don’t have magic.)  Many journals also have discounts on, or even waivers of, page charges that an author can apply for.  These are often available to early-career authors, authors in the developing world, authors who certify that they have no access to grant funding, or (for society journals) authors who belong to the publishing society.

  1. Society vs. for-profit. Some journals are published by scientific societies (for example, in my own field, Ecology, Oikos, and The American Naturalist). These are either non-profit or else return profit to the societies that publish them; these profits can then subsidize conferences or other society activities. Other journals are published by for-profit commercial publishers (sticking to my own field, Oecologia and Evolutionary Ecology).  These return profits only to owners/shareholders.  Most commercial journals are controlled by a few large publishers:   Reed-Elsevier, Springer, Wiley-Blackwell, Taylor & Francis and Sage.  Their profit margins are rather healthy, which revolts many people****. If you choose to avoid commercial publishers, be a little bit careful.  There are quite a few society journals that are published under contract by a for-profit publisher (Evolution is a good example, being published by Wiley-Blackwell on behalf of the Society for the Study of Evolution).  These should be considered no different from in-house-published society journals – they’re just cases of societies hiring a company to do the technical publishing work because that’s more efficient than doing it themselves.

  1. Speed. Journals love to compete on speed – especially, to boast about a short time either from submission to first decision or from submission to publication. With a few exceptions (I’ll come to them), please, please, please don’t pay attention to such claims.  There are three reasons, I think, why these claims are irrelevant at best, and actively harmful at worst.
      • Decision time statistics are easily gamed, and journals indulge in all kinds of tricks to keep those decision times short. For instance, they may favour “reject-and-resubmit” rather than “accepted with revision” because that lets them count the revision as a new submission and restart the clock.  Or they may issue a lot of two-day “desk rejects”, without careful reading, because these average out the longer time it takes for real review.  We’d all rather journals designed their procedures to allow better, fairer peer review, not to keep decision times artificially short.
      • When journals compete on speed they race to the bottom, advertising decision times that are just unreasonable.  Any journal that advertises its ability to routinely reach first decision in 3 weeks, for example, simply can’t be taking peer review seriously.  When we care about journal speed, we just encourage this behaviour.
      • For most of your papers, I just don’t think it will matter very much anyway.  I see a lot of gnashing of teeth over how slow publication is crippling the march of science, but I don’t think I’ve ever published anything that was really that urgent.  I can think of a lot of things that slow the march of science far more than a few months to get a paper published.  Besides, the easy availability of preprint servers means that if you think science really can’t wait for your results, you can post them instantly.

    Now, I’m going to allow one kind of exception here.  There may be particular times during your career when publication speed is important, not for the march of science, but for the march of that career.  You might be about to hit the job market, or a tenure deadline, or about to submit a big grant.  If one more publication could make a critical difference, then journal speed is a reasonable thing to consider.  Note, however, that data on journal speed are quite unreliable, because there tends to be high variance among papers in the same journal’s workflow.

  1. Predatory journals. Among the tens of thousands of journals in existence are thousands – at least – that are fake. They’re most often labeled “predatory journals”, which has a nice ring to it*****.  These “journals” exist only to make money, and they will publish absolutely anything for payment of an APC. Many of them pretend to have serious editorial processes, including peer review, but they don’t pretend very hard (sometimes publishing submissions within days of receiving them).  Publishing in one of these “journals” means nobody will read your paper or take it seriously – and that’s the best-case scenario.   So how to do you identify, and avoid, a predatory journal?  There are lists online, although the most well-known (Beall’s List) has become difficult to find and out of date.  Amateurish journal web sites riddled with typos can be a tip-off; an even better tip-off is the offer of a ludicrously short time to publication (often two weeks or less).  And if you get an email inviting you to submit, that’s an almost certain giveaway – unless you know the sender personally, and they’re putting together a special issue for a journal you’ve heard of.  If you’re not sure, ask colleagues; but if a journal sounds too good to be true, it probably is.

OK, that’s my list of 9 things to consider.  Exhausted?  I am.  I’ve made this sound complicated; but remember, there’s no single right choice for any manuscript.  Some of the factors I’ve mentioned may be very important to you, others may not be, and different people will have different weightings.  That’s perfectly OK, although of course it may texture discussion between coauthors.  In the end, there are almost always several suitable targets for any given manuscript.  Choose one, submit; and if you’re rejected, choose another.  Good luck!

© Stephen Heard  June 19, 2018

UPDATE: Mike Kaspari had a related post 18 months ago – it’s also worth reading.

*^Finding your story is a critical part of writing.  It just isn’t true that the data speaks for itself; you need to figure out which data, and in which order, tell a story that will grab and hold a reader’s attention.  Techniques for doing this are the subject of Chapter 7 of The Scientist’s Guide to Writing

**^Hear that?  That was the sound of a whole bunch of people clicking away in disgust.

***^The word “directly” is extremely important, because most readers don’t access papers in journals to which they subscribe personally.  A subscription-model paper (often called a “paywalled” paper) is available to anyone who pays to subscribe; to anyone working at an institution that pays to subscribe; to anyone with access to a library with a Document Delivery system (“interlibrary loan”); to anyone who’s willing to email the author (you) and ask for a copy; to anyone who’s willing to use #icanhazpdf on Twitter; to anyone who’s willing to flirt with illegality and download the paper from ResearchGate or SciHub; or to anyone who has a friend or colleague in any of the categories I’ve listed.  In other words, subscription-based papers aren’t open-access; but they’re much more widely available than critics would like you to believe.

****^Oddly, nobody seems to object to corporations making money by publishing books, by manufacturing scientific equipment or consumables, or by selling groceries.  Only publishing papers.  I don’t know why.

*****^Although I’m inclined to think it’s not a very useful designation.  The implication is that these journals are preying on unsuspecting authors by duping them into submitting their papers.  That certainly happens; but authors can also use these journals to prey on unsuspecting hiring, promotion, and tenure systems by adding CV lines quickly and cheaply.  Who’s the predator, and who’s the prey?  It depends, and so “fake” is probably a better way to describe these journals.