32

During the postdoc phase, at least in math, it's usually a very short period, like 1 or 2 years. Isn't it really important to write something and get it out there, even if it's of no real significance or interest? The reason I ask is that a postdoc could spend more time working on a more significant paper, but if you have to apply for a job in one year, if you don't do something to put something on your CV, people will look at your publication record and say, "This applicant didn't do anything!" I don't want to give people the impression that I haven't done anything if it's not true!

If a postdoc is actually doing something, but it doesn't appear on the CV, does it count? A postdoc could be working on a project that will take several years, but if people look at the application and there's no string of papers, people may automatically reject it. "Invisible work" appears to not matter.

5
  • 6
    The norm has unfortunately moved to 2 3-year postdocs, so the postdoc phase is now more like 4-7 years. Also, accepted papers are as good as published ones, and part of the job of your letter writers is to explain the impact of your papers, whether they are accepted or in review (and presumably available on the ArXiv.) Also, the papers accepted by any given journal has a wide variability in terms of quality/impact, so several experts' opinions of a paper should (but does not always) matter more. Nov 27, 2022 at 21:11
  • 20
    "I don't want to give people the impression that I haven't done anything if it's not true!" So instead of giving them the false impression that you haven't done anything (which can, as indicated in the above comment, potentially be corrected), your proposed strategy is to give them the correct impression that you have done something, but it is not of any "real significance or interest"? Your goal is not to have no-one accuse you of not doing anything (that would be a strange goal on its own), it's to convince people that you do valuable research so that they will want to hire you. Nov 27, 2022 at 22:43
  • I don't know much on your field. But this forum has several academic (and I assume, Pure) math people. If they knew the general area of your research they could offer more useful advice. Could you edit your question to elaborate ? On Buffy's idea of short but impactful papers: are you excluding any opportunities to do papers on applications of your math work in favor of doing more pure explorations ? If not, this would certainly require some degree of collaboration with the application domain experts, so start talking to postdocs in other departments.
    – Trunk
    Nov 28, 2022 at 14:27
  • 1
    Maybe I just have a different understanding of the word "impact" - but I find the premise of both the question and several answers a bit strange. The vast majority of math papers (also those that are published in decent journals and by established professors) have very little "impact" at all (and even less so after a short period of time). Nov 28, 2022 at 19:37
  • 3
    You are right: the system is just that, a system, and therefore you can play it, "game" it. However, this then comes to the question: the why the heck are you even on it? No researcher started research thinking "numbers go up", we are supposed to at least somehow believe that what we do matters. What you are proposing is more or less morally rigging the system. It does 100% work. But then, why play? Or in the real terms, why waste your talent and time in research that is unimpactful. Academia is not the only option in the world, you can do something else too. Nov 29, 2022 at 20:43

6 Answers 6

60

I disagree with one point made by Buffy. People will absolutely count your papers. Whether it is a job interview panel or a promotion, the number of publications is going to be a metric everyone will take into account. In some less research-focused places the number of papers is going to be the only measure of research performance people are going to be concerned with. In more prestigious and research-oriented Universities some people on the panel might have a better understanding of how good the journal is. And only in the top-tier places someone on the panel might actually have the expertise and time to bother reading the papers and evaluate the quality of your research.

  • Boosting the number of your research outputs (e.g. via salami slicing, proceedings, short articles with bloated list of authors, repeating the same research with minor modifications) is grumbled upon. But it will get you through the majority of doors in your career, except a few most challenging ones. People do it, and in most places they absolutely get away with doing it. This is generally a safe and reliable strategy to move your career forward -- see Mark Griffiths's Google Scholar as a notable example.
  • Focusing on quality and ending up with less publications might close many doors (as your CV will be dismissed without exploring the quality of your papers), but may potentially get you to the top-tier places. This is a high risk / high reward strategy. People who attempt it are usually perfectionists or idealists, for whom writing a sub-par paper is not an option.

This highlights a poor state of academia these days, with simplest metrics of performance being predominantly in use. As the number of people competing for place in academia increases, and the number of available positions shrinks, the change towards a more rigorous and nuanced evaluation of research efficiency seems unlikely.

17
  • 10
    If I had something like a 5-year position, then I would have a reliable source of income so I would be less pressured to just do something, write it up, and put it online for people to see. But if I have only 1 year, then I want to make sure my CV actually has changed. If a big paper takes 5 years from start to completion, then I cannot wait that long because I need to stay in the running and inform people that I am not just wasting my time. I think I have no choice because I'm doing what the situation calls for.
    – cgb5436
    Nov 27, 2022 at 23:07
  • 5
    I think to some extent this is culturally dependent. Most universities in the US, including less research-oriented ones, are perfectly willing to read your letters of recommendation to evaluate the quality of your research even when they can't evaluate it themselves. It's only cultures that are more concerned about corruption (for good reason or not) that view subjective or possibly biased information as worse than no information. Americans are usually willing to accept occasional unfairness if it leads on average to better decisions. Nov 28, 2022 at 0:37
  • 8
    200 papers a year?! Dang, he must have exposed the big lie of academia, them lazy researchers absolutely can and should work harder, let us raise the requirement for getting any funding at all to at least 20 articles a year! /s
    – Lodinn
    Nov 28, 2022 at 1:16
  • 4
    -1 for exaggerating the extent to which people just 'count papers'. Recruiters increasingly ask you to list your few best papers. This happens not just at top-tier places, but also in countrywide systems e.g. in the UK or at CNRS (France). Nov 28, 2022 at 8:20
  • 6
    On the last search we did a few years ago (when I was on the committee), we did not seriously consider the applicant who had about 5 or 6 times as many papers (all legitimate, but of negligible impact) as the next most published applicant. Nov 28, 2022 at 17:08
14

I am a professor in a mathematics department at a state research university (University of Georgia). We do not have requirements on the publication rate (or record) of our tenure track faculty, but we have some informal expectations. Two or more papers per year is pretty good. Less than one paper per year is not so good, unless these papers are exceptional in quality: in mathematics, the very top journals are so elite that the majority of lifelong research mathematicians never publish in them.

On average, a postdoc has a lower teaching load than a tenure track job, and even when the numbers are equal, postdocs mainly teach less demanding courses. (For instance, several of our postdocs are currently teaching two sections of first semester calculus. I am currently teaching undergraduate real analysis and graduate algebraic number theory.)

So the following conclusion seems reasonably inexorable: if you wish to attain a job as a tenure-track professor at a research-intensive institution, you have to show during your postdoc that you can attain the quantity as well as the quality of research that will be expected of you in that tenure-track job. Many postdocs are still finding their research footing and, in some cases, really just beginning doing their research in a largely undirected way. Such people may not be putting out multiple strong papers a year. But I can assure you that the tenure track job interviews are full of postdocs who are hitting those marks.

In 2022 an assistant professor of mathematics is not really a "journeyman" who is just starting to do business for themselves; rather these are people who already have a research program with some successes in their rear view mirror. To get to this point takes time, and as mentioned in another answer, multiple postdocs totaling 4-6 years are becoming increasingly common.

I do want to end by mentioning that we are also not looking to hire people who are in the practice of writing papers just to have publications. If half of your publications are in journals that are not especially reputable (or worse, conference proceedings or other venues where the rigor of the peer review is less than clear), you will not be an attractive candidate.

It's a competitive world out there, for sure. Good luck to everyone.

1
  • It does seem like UGA expectations in 2022 > USask expectations in 2022. (Certainly UGA expectations in 2022 > USask expectations in 2010)
    – Yemon Choi
    Nov 30, 2022 at 0:23
12

A two-edged sword, perhaps. Writing "impactful" papers in math, short or not, isn't an easy task or one that can necessarily be time limited.

Papers of low quality aren't going to help you. People won't just count your papers and assume they are good. The market is too tough for that to work.

The alternative is to have a section on your CV for Work in Progress in which you name projects that you are working on. This is worth having in any case.

If you are collaborating with people, they can help you in your job search, indicating your work and the quality of your ideas. If you're not collaborating, then consider starting.

But if you can, indeed, whip out short impactful papers, then, by all means do that. But you won't likely get published within that year's timeframe in any case, so at best you may have some submission(s) in process.

4
  • 9
    I don't know how to force my papers to be more impactful. It's like digging somewhere. You dig and then you get whatever you get.
    – cgb5436
    Nov 27, 2022 at 20:29
  • 1
    I understand. Do what you do best. Collaboration helps though. Even if it is just coffee room chatting about ideas.
    – Buffy
    Nov 27, 2022 at 20:44
  • 5
    You can't ''force'' papers to be impactful. Try to work on things that you feel are intrinsically interesting in a way others would be interested. Another way to ''cheat'' citations is to produce some workhorse result which is useful in applications, my most cited paper has a result of this kind although it's not actually really that great an article.
    – Tom
    Nov 28, 2022 at 14:05
  • 3
    @cgb5436 good analogy, only to the extent that it defines the mentioned digger as a stupid person. If you dig and don't know why you dig somewhere, you are likely to get only sweat.
    – EarlGrey
    Nov 28, 2022 at 16:24
7

The job market in pure mathematics is extremely competitive.

If you don't have an impactful paper out of your postdocs, you won't get a permanent research-oriented job, no matter how many short unimpactful papers you have.

The job market essentially is forcing everyone to gamble.

For further postdocs, what matters is convincing people that you have that impactful paper in you - in pure mathematics the value of postdoc to a hiring department is mostly that they can say 'This excellent mathematician was a postdoc here.' (Pure mathematicians do collaborate, but postdocs are usually working on their own projects in collaboration with permanent faculty, not working on the projects of the permanent faculty.) Having some moderately interesting papers does give evidence of potential for future productivity. Publishing worked out exercises does not.

5
  • 1
    My experience of the tenure-track in non-high-flying Canadian institutions is probably out of date now, but I would still prefer some clarification or hedging regarding "impactful". "Impactful" to which community, and in whose eyes?
    – Yemon Choi
    Nov 30, 2022 at 0:20
  • @YemonChoi - how about "likely to be cited outside of the circle of the collaborators of your collaborators" - though one or two papers that barely meet this standard is unlikely to be enough. (I don't think I could get hired these days.) Nov 30, 2022 at 1:20
  • @AlexanderWoo: Hmm, I'm not sure if I find the "outside the circle of collaborators of your collaborators" criterion convincing - this circle tends to be much larger than one might expect. zbMATH actually gives the number of co-co-authors for every author profile. For instance, mine is currently 345 (although I haven't been particularly long in the business, yet; started my PhD in 2013), yours is 495. Nov 30, 2022 at 21:23
  • @JochenGlueck - maybe I would have better off being more vague and say "likely to be cited by someone who did not learn of the work through a direct professional relationship with you or another of the papers' authors"? Not all collaborations are equal of course. Nov 30, 2022 at 21:43
  • @AlexanderWoo: Agreed, this seems like a better criterion. Nov 30, 2022 at 22:25
4

You are still in the beginning of your career, and writing good papers is also a skill that takes time to learn. You don't specify whether how much experience you have with getting papers published so far, but it's likely at most a couple of papers, possibly none as first author or the main writer.

Getting your first paper all the way through the publication process will teach you a lot, and will improve the chances of timely publication of your later, more important results. If you already have that basic experience, there is still more to learn from comments from new reviewers, and trying to get the papers through to higher rated journals.

Whereas if you focus on purely research for a few years, it may be that your eventual results don't get the attention they deserve due to poorly written paper.

-4

Is writing lots and lots and lots of unsubstantial papers is a career booster in academia?

Hell yes, it sure works for some.

But does this not undermine the efficacy of academia in the long run, degrade public trust in science, and create an environment in which hiring and promotion decisions are increasingly made by people who never contributed anything of substance themselves?

Yeah... what do you think?

1

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .