Who can haz job?

Dr. Isis asked a question about how many pubs you need to get a job. This varies by field, of course, where you publish, etc. I’m a bit skeptical that publications are the key determinant of getting shortlisted or an offer relative to pedigree/connections and maybe unstated goals of the search committee regarding subfield/organism/methods. I’ve been told that publications are what matters most by people who, when actually on search committees, shortlisted several well-connected applicants with no or mediocre postdoc publications. These same individuals have assured me that my single CNS paper will be enough to at least get me some interviews, but I know this is far from a sure thing.

I’ve developed some theories about why the advice I get seems out of step with my observations. The main one, I think, is that some of the faculty I’ve spoken to “grew up” in pedigree bubbles. The influence of their prominent mentors on every stage of their career – fellowships, publications, job opportunities – is essentially invisible to them. Contests always seem meritocratic from the winner’s circle: as far as they can tell, all the hard bench work just kinda paid off. So when I’m trying to get them to help me strategize about my search, I get responses like “You know who has a great neuroscience program? Columbia. You should email someone there.”  Thanks a fucking lot.

Anyway, I was thinking of how to represent the multivariate nature of trainee publications graphically. Because I hate bubble charts most of all, I went with that. I have taken data of a fictional postdoctoral fellow who is going on the job market. x = time, y = IF (not an endorsement), area = citations. For scale, the biggest bubble = 55 cites, smallest = 1. Obviously IFs and years are estimates for the unpublished ones, and they are represented with a single fictive cite.



9 Comments on “Who can haz job?”

  1. toto@club-med.so says:

    1- Lovely graph (despite the bubbleness). What are the scales on these axes?

    2- Spare a thought for the poor souls who started in a field where conference papers actually count (like CS-AI), and then transitioned to a field where they don’t (like neuroscience).

    Imagine having 10 papers on your CV (sorry, on the CV of “a fictional postdoctoral fellow”) that potential employers won’t even look at!

  2. rxnm says:

    x-axis is 2005 – 2013. y-axis is “respectable niche journal” – CNS.

    That is harsh… but I think in neuroscience you get a boost if you have demonstrable quantitative/computational chops. Also, you can probably look in biophysics programs, where they tend to be less fuckwitted about impact factors and other meaningless crap.

  3. neuromusic says:

    maybe mendeley or researchgate should integrate such a visualization into their profiles?

  4. rxnm says:

    That’d be cool. Kind of a Rorschach to put at the top of everyone’s CV. The blob pattern + PI name should be all the search committee needs to know.

  5. jbashir says:

    There are a variety of ways to get the attention of the search committee. Having a lot of pubs is one way to do that. So is having a famous advisor, or working in whatever area is super hot this year. Of course these three things may correlate with each other.

  6. rxnm says:

    Yes, all this thinking about it is just hand-wringing. Ultimately I think each search/shortlist/offer is a highly idiosyncratic mix of department needs and wants, connections, general sniff test of ability, impressions from visit, etc. But this leaves you nothing to lie awake worrying about or to try to rationally analyze… I want a regression.

  7. Dr Becca says:

    There’s not much strategizing to be done, really. Apply anywhere you would go. I sat on a search committee last year, and we passed over applicants with CNS papers and K99s for those who told a better story in their research statement, or whose plan seemed to jive better with the department. Of course, you must have >0 1st author papers from both grad school and post-doc, and probably some demonstration of fundability.

    Good luck!

  8. rxnm says:

    Thanks, Dr. B. Yes, I really need to get a draft research statement out for feedback before this first round of applications is due. I am very often wrong about what other people will think is interesting about what I do.

  9. Arseny says:

    Dear rxnm,
    this summer I’ve created a different, but related kind of plot for publications history benchmarking. I figured you could find it interesting:

    Your type of plot better “represents a person”, in terms of generating a unique image that one can quickly discern. My type of plot would probably work better for personal benchmarking and goal setting. But still in spirit they are quite similar.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s