You are not a bad person if you do the things you need to do to get/keep a job in academia

Aug 28 2013 Published by under Uncategorized

While the desk-drawer bourbon is still coursing through my bloodstream, I feel compelled to weigh in on a discussion that has been on and off on the tweets and blogs for oh, a long time now. The question is re: whether you are a morally bankrupt individual if you publish in non-open access journals. TL;DR, the answer is a big fat NO. I'm too lazy to storify or whatever, but you should definitely read Dr. Isis's predictably excellent post, which is framed in the context of the choice to go open access or glam mag, the former perhaps/likely crushing her chances of adding to the despicably low numbers of TT Latina women in science.

But I am here to take this mentality a step further. It doesn't matter WHO you are. If you are a person at any pre-tenure stage of an academic career (incl grad students & post-docs), the reality is that you are judged by a finite number of things: 1) where you did your PhD; 2) who you do your post-doc work with; and 3) the IF of the journals you publish in. It's not rocket science/brain surgery, people. Now, there are most definitely arguments to be made that these are not the be-all end-all of ACTUAL merit , but this is the world we live in. If you think these are terrible metrics and wish to push forward with no regard for such customs, I wish you all the best.

Open Access is an awesome thing, and I hope one day that all people everywhere have access to every scientific paper ever written. This might actually happen. But in the meantime, we need to get/keep our jobs, mkay? And this means doing the things that impress people according to the 3 criteria laid out above. You do not need to do ALL of them at the level of like, Harvard/Nobel Laureate/CNS, but if you have the OPTION of publishing in a glam mag as a pre-tenure person, do not fall on your sword for the sake of the general public, FFS! Make yourself the most impressive scientist you can possibly be, because nobody else is going to do it for you, and a whole lot of people are probably going to do it instead of you.

34 responses so far

  • DrugMonkey says:

    Well put. But tenure doesn't magically make all concerns disappear either. Fair warning on that one.

  • Dr24hours says:

    Ding. Someone is right on the Internet.

  • jmcin9 says:

    Thanks for saying it. Agree with you wholeheartedly.

  • Everybody has to make their own choices about what's important, but I think you do yourself and your readers a disservice with your list about what matters to careers in science.

    You say you are judged by a finite set of things - pedigree and publications. I see some version of this list all the time. But it's something: whether you do good science or not.

    The reason why I get so frustrated every time this issue flares up is not because I don't understand the challenges people fact in navigating their careers. It's because there is a complete lack of proportion in how these things are discussed. I've sat on lots and lots of hiring and tenure committees. Do pedigree and publications come up? Yes. Are they the first thing that comes up, or the most important thing, or even the deciding factor? NEVER. Not once.

    It's not to say that these things don't matter at all. They do. They help people get noticed, which is not an insignificant thing. And I'm not even saying this because I want you to all publish in open access journals. I'm saying it because I think telling people to focus on getting their work into the highest impact journal and that everything will be ok is manifestly bad advice. For every person we've hired with a Nature paper, there are dozens who we've not hired because we didn't think their work was that good or interesting.

    So, by all means, do what you think you have to do to succeed in your chosen profession. But base your decisions on a rational assessment of the costs and benefits of different decisions. And remember that by far and away the most important this is to do good science.

  • dr24hours says:

    The reason I can't get behind your vision, Michael, is in large part because you seem to conflate "rational" and "agrees with Michael Eisen".

  • That's completely irrational.

  • drugmonkey says:

    Do pedigree and publications come up? Yes. Are they the first thing that comes up, or the most important thing, or even the deciding factor? NEVER. Not once.

    People on the job market have a very hard time believing such protestations when they look at the people being short listed and hired for the jobs that are open that round. Confirmation bias? Excuse making for own shortcomings? perhaps. But maybe there is some crowdsourced truth to it.

    For every person we've hired with a Nature paper, there are dozens who we've not hired because we didn't think their work was that good or interesting.

    All that sort of analysis (and I believe you) does is say that for your department, CNS is the *minimum standard even to be considered*. Why would it not be? It is a way to gate the no doubt scores of applications you receive for every job. The idea that you go through scores of applications with a fine tooth comb, ignoring any of the more bean-countery measures, seeking the [subjective, amorphous, ill defined] essence of quality is not credible. And even if your department is weird (which would not be unthinkable, you are an influential part of it, after all), many many folks in other departments report things about their hiring that suggest bean counting measures are indeed highly critical.

    whether you do good science or not

    There should not be *anyone* with half a brain that trusts that some version of this will see them through. Why? Because the notion of what is good science is so subjective, nobody agrees, everyone thinks that what makes themselves look like "best" is of course the only objective standard, etc. And the standard moves- so the minimum or mean performance of the existing tenured faculty when *they* were hired is no longer sufficient.

    It's the same at tenure, its the same at promotion to Full. Probably the selection for HHMI as well. Nothing ever changes because the contingencies don't change. There *is* no way to objectively compare scientists who work in different subsubsubfields, using slightly different models, addressing slightly different questions.

    Ultimately, the best we can do is to attempt to grasp at measures that appear to have the patina of objectivity. JIF, h-index, citations.... these are what we fall back on. As institutions trying, trying to get some objectivity in to the judging. (Because we know full well if we rely exclusively on subjective evaluation that bad things happen. Like departments full of clonal white men all trained in the Ivy league).

    far and away the most important this is to do good science.

    Disagree. One should strive to do the science that they themselves find important and interesting in the best way that they can do it.

    • me: far and away the most important this is to do good science.

      you: Disagree. One should strive to do the science that they themselves find important and interesting in the best way that they can do it.

      How do we disagree?

      • DrugMonkey says:

        Every kind of science has limitations but some kinds tend to violate the more common and/or canonical views of high quality science. Think, for example, of the huge limits of doing human subjects science. In a Science 101 sense it can be a poorly controlled, under powered, never to be repeated mess. There will be no "mechanism". Etc. Some might recognize it isn't the "best science" but still think it hugely important to do.

    • Dave says:

      One of your best ever comments DM.

  • I have long forged and strongly help beliefs on this issue, and we happen to disagree on them. But I don't think I dismiss people's views on this matter out of hand. Quite the contrary, I spend a lot of time trying to explain why I feel this way and why I don't agree with them.

    • Cate says:

      "I don't think I dismiss people's views on this matter out of hand. Quite the contrary, I spend a lot of time trying to explain why I feel this way and why I don't agree with them."

      How is it not dismissing other people's views if your only response to disagreement is to try to convince them why you are right? Not dismissing them, to me, means listening to what they have to say and considering that they have a valid point of view, even though it is different than yours.

  • drugmonkey says:

    I would really like to hear more about how, in a department as broad and large as yours, you come to agreement on new hires without reference to the JIF of the articles on their CVs and without reference to pedigree, focusing only on "the best science" or even "the best person for us who does good science".

    I mean, maybe there is something in the air or water around Berkeley that alters normally contentious department members into a unified mind. But I doubt it. I bet you have just the same differences of opinion on what is the "best" science that any other department has.

    And I'd like to hear more about what possible techniques are deployed to harmonize decision making so that most of the department agrees that it is the "best science" that is being brought in for the short list interviews.

    It would also be of interest to hear how your contentious tenure cases are decided, again without reference to the usual pseudo-objective bean countery.

  • Dr Becca says:

    telling people to focus on getting their work into the highest impact journal and that everything will be ok

    I most definitely did not say this--no one on this planet has the answer to what will make everything ok.

    We seem to be dealing with some sufficient/necessary confusion here. As I've written previously, there is no magic formula to landing the job of your dreams. Nor do I think people should strive only to publish in Glamour Mags - I certainly haven't, and I've done OK. But as DM points out, having a at least one high IF publication is a pretty damn good start for getting yourself onto short lists. What happens after that is up to your performance at the interview.

    Finally, remember that today's conversation started with Dr Isis's situation in which she had a manuscript that was already "glamour-worthy". Would you actually advise someone pre-tenure who could potentially successfully submit a complete body of work to a glamour mag not to do so?

  • drugmonkey says:

    cross posting from Isis' place

    ok, I took a look at all the assistant professors listed in Eisen's department.

    I counted 16 that list CNS publications on their splash page, or in a couple of cases I had to hit pubmed. This was on the assumption that if there was a splash page and they *did* have CNS pubs they would list them.

    4 assistant profs listed no CNS pubs but of these 2 had Nat Neuro/Neuron level baby glams and one was from an OA wackaloon lab.

    One of the no-CNS trained with Greengard. ( and actually, trying to retrace my steps I may have mis-scored one of these b/c the CNS pubs were way down the list. )

    The only two that I could make out as the very most recent hires (i.e., they aren't there quite yet) both had CNS pubs.

    So clearly CNS pubs are important for Eisen's department. In hiring or as tenure approaches. The very latest hires both had glamour pubs....perhaps a coincidence of small N stats, perhaps a feature of increased recent pressure in the job market. Two of the nonCNS asst profs had subGlamours and we can assume that means they are expected/expecting to be duking it out for a CNS level soon. Of the other two I may have miscounted one and one is from what I think is an trubelieverz open-access lab.

    I'm sorry but I'm having trouble crediting Eisen's attempts to characterize his Department as anything other than the worst sort of GlamourDouche environment.

    • I stand by what I said. I looked a while back and 5/15 of our assistant profs didn't have CNS publications when they were hired. Again, I'm not trying to hold up the department as some kind of paragon of virtue - my colleagues are as bad on the OA front as anyone. My point was that these things are not necessary, and, at least in my experience, don't define success.

      I would also point out, as I have before, that there's a logical fallacy in assuming that because so many faculty in top-tier departments have CNS publications that there is a causal relationship. An alternative explanation is that the same basic criteria that are used in hiring are used by these journals, which induces a correlation. I'm not defending this - it's a pathology of the clubbiness of science - but it's one of the reasons I don't buy the argument you're making.

      • RB says:

        A quick question for you, Mike: of the people invited to interview for the open TT position in your department last year, how many did not have CNS papers?

        The answer is zero. Every candidate (of >5 that interviewed) had at least one published CNS paper. Do you still think this is not a necessary condition?

  • Kaymtye says:

    Great post @doc_becca and great debate in the commentary.
    I'm glad there are idealists out there, because I agree that it *should* be about the best quality science. And in the very long run, this will be true for a given individual who has been publishing and having their findings vetted across decades. One sexy paper doesn't make a career any more than 30 seconds of twerking.

    That said, publishing glam-worthy science in an open access journal is like sending your fashion masterpiece to a thrift store... Does it make it more accessible? Only sort of... It will not attract the same attention and careful reading because only a subset of readers have the time and energy to look through all the lower quality stuff in the hope of finding a gem. @drugmonkey is right on point, because in the world we live in, search committees and tenure committees and other decision makers are human beings. They are busy and they need way to filter through the masses. Are college applicants only as good as their SAT score? Does a good one ensure your future? Of course not. But is looking at a number faster than reading pages and pages of text? You betcha.

    So is it sad the system is the way it is? Sure. The parameters we use don't always perfectly match our ideals. But it would simply be poor mentorship to tell trainees that the world works the way it should instead of the way it does.

  • You are all way to smart to be making the elementary mistake of confusing correlation for causation.

    Having a high impact paper is one thing.
    Publishing in a journal with a high IF is another thing
    Publishing in a open access journal is yet another thing.

    All the above are orthogonal to one another. I know from readership stats on Mendeley that there are tons of CNS papers that no one reads. I know there are tons of papers in low IF journals that are widely read and widely cited, some of which are in open access papers and some of which aren't. Furthermore, there are some high IF open access journals, like PLOS Biology.

    I know you may have missed some of the stuff going on because you're heads-down working on grants and research (and bless you for that!), but your view that a high IF paper is necessary for success just isn't true.

    I don't think you should feel bad for not publishing open access, but I will point out that the NIH and every other federal funding organization wants you to publish open access. Publish whereever you want, but don't conflate open access and low impact. How about just listing the number of citations you have for each paper? Impact Story (ex. http://impactstory.org/WilliamGunn) is a free service that tracks all that for you, and keeps it up to date.

    The practice encouraged by many scientific associations and most publishers is to not use IF for evaluations. See http://am.ascb.org/dora/ and one reason for this is specifically to prevent the phenomenon where someone gets a Nature paper just because they came from Joe Famous's lab. Supporting this system supports the old boy network, so people like Dr. Isis should particularly be in favor of it, if their rhetoric matches their motivations.

  • ecologist says:

    Some of the comments have mixed hiring and tenure as if those operated on the same principles. They don't. At hiring (and at any pre-tenure promotions that exist; my institution has one such) one of the big considerations is potential. Potential for doing good science and potential for developing a successful research program. After all, if you are hiring someone to do something (be a professor) that they have never done before, you must be judging potential.

    At tenure time, potential no longer counts, the question is impact. Has the candidate had an impact in their field? I have sat on many tenure decisions, and by far the biggest factor in those decisions is a set of external evaluation letters responding to exactly that question.

    I'm surprised that no one in the discussions has mentioned these letters. There will be 5-10 of them. The writers will have been chosen by the department, unbeknownst to the candidate, as international experts in the field. They will be asked a detailed set of questions about the accomplishments of the candidate and the impact of her work.

    The letters tend to be thoughtful and often in depth analyses of the candidates impact. They will often cite specific parts of a research program, in terms of their impact ("her experiments have completely changed the way that we view X, and have since become the standard approach ..."). They will pay little or no attention to where they were published.

    They will also discuss other ways of having an impact. Has the candidate participated on national or international panels? Organized big events?

    These letters are by far the most important thing in a tenure file, precisely because they are written by the world's experts in the field who will be able to speak first hand about the impact of the candidate's science, and they provide an alternative to IF-obsessed publication counting.

    So, if your institution uses this external letter step in its tenure review process (and I get asked to write such letters for lots of other places), pay attention to it. Doing science that has an impact in your field is far more important than where you publish it. If publishing in Glamour Mags helps have an impact in your field (in mine, it doesn't much), go for it. Just don't think it's a guarantee of anything.

  • ecologist says:

    the shorter version: you want them to know your name.

  • Comradde PhysioProffe says:

    The writers will have been chosen by the department, unbeknownst to the candidate, as international experts in the field.

    As an official matter, whether the candidate provides some suggested letter writers varies by institution. As a practical matter, any chair who is supportive of the tenure case is committing malpractice if she fails to discuss at least some of the letter writers with the candidate.

    They will pay little or no attention to where they were published.

    This is delusional. Even open access enthusiasts know the difference between PLOS Biology and PLOS One.

    • ecologist says:

      "As an official matter, whether the candidate provides some suggested letter writers varies by institution. As a practical matter, any chair who is supportive of the tenure case is committing malpractice if she fails to discuss at least some of the letter writers with the candidate."

      In all cases I know of, the candidate provides a list of suggested names. Some of those are used, and others are added by the department, but the identities of the final selected pool of letter writers is confidential and is not revealed to the candidate.

      "This is delusional. Even open access enthusiasts know the difference between PLOS Biology and PLOS One."

      No, it is not delusional; it is a fact. Yes, reviewers may pay a little attention to it (as I said), but it is by far a secondary consideration. The question du jour is, how much impact have you had? Aim for that.

  • Ola says:

    In addition to the 3 criteria you list, I would add a fourth which trumps all the others - demonstrated ability to attract external funding. Sure, at the AP-->tenure stage it's NIH or bust, but even at the earlier stages you should never underestimate the ability of dollar signs to offset deficiencies in the other areas.

    Old Scrooge McDean really could not give a flying fuck about OA/glamor/IF, so long as the candidate keeps those indirect dollars flowing.

  • […] that researchers like Michael Eisen are so passioned about this topic, but I have to agree with Dr. Becca and Proflikesubstance when they say that this must be a task for senior scientists, not for […]

  • else says:

    Ahaa, its pleasant conversation regarding this article at this place at this weblog,
    I have read all that, so at this time me also commenting at this place.

  • […] You are not a bad person if you do the things you need to do to get/keep a job in academia The Morality of Open Access vs Increasing Diversity Which came first, hermits or kings? Genes Aren’t Just Architects; They’re Actors A Question Remains […]

  • We had a big discussion on the usefulness of the journal impact factor (JIF) on the american biotechnologist blog and may commented that the JIF has been abused by academic institutions and is being improperly used to determine staff eligibility for promotion. If people hate JIF and what it represents, then why not publish in open access journals?

    In fact, "peer review" can often mean that people who have "friends" on the reveiw board get published. Those that don't will be more successful publishing open access. To be honest, at least in theory, open access allows for many more qualified (and not so qualified) peers to add their 2 cents about your publication. What could be more robust than that approach?

    If you're interested visit The ugly side of the journal impact factor

  • […] split into options that involve staying in academia and research vs. options that do not. Much has been said recently about the general difficulties facing PhD graduates in terms of leaving […]

  • Anders says:

    All of you seem to be missing an important aspect. With parallel publishing you can publish your paper in any journal and then publish the author's version in a university repository. By doing so, you get to publish in your fancy journal and anyone can read the copy in the repository.

  • […] It seems to follow that plans for the wide adoption of pre-prints would need buy-in from junior researchers. Yet, it is not clear that the awareness of pre-prints among new scientists is anywhere close to a necessary critical mass. The subset of junior scientists that are aware of the existence of pre-prints, are likely to carefully weigh the benefits and costs of any new publishing innovation with their career advancement prospects. […]

Leave a Reply