Landing an TT job

Aug 27 2015 Published by under Uncategorized

Sergey Kryazhimskiy has a blog post up about getting a job. First off, congrats on getting a job. Not easy, and he seems to have done so.

Yet, as I read his post, my thought was, damn, I don't want anyone I know to think that This Is The Way It Is.

He starts with data. He says that explicitly. Yet, I suspect if these were ecology data and he tried to publish them Reviewer 3 would slap him down. He presents data for the two years he applied. In 2013 he put in 20 applications, got 1 interview and no jobs. In 2014 he put in 29 applications got 11 interviews and three offers (plus one that was too late).  The difference, he says, is a Glamour Pub.

Reviewer 3 would point out that his n=2 (years) and that they are not exactly independent data points. He says its the same CV, same recommendation letters. I doubt this. When trainees I know and I care about go back for a second year of job hunting, I update and freshen up the letter and emphasize what this trainee has done in the intervening year. For postdocs this is critical, because each year needs to count for growth and development. Further, if he got a Glam paper in the extra year, I can't imagine his letter writers didn't notice this.

But most of all, there is a difference in the pool of jobs. He says that of the total applications, there were 4 departments to which he applied in both years, and got interviews at 2. He doesn't say whether it was the same job re-opened or not. But if they were different jobs, maybe he was better suited for them. In my experience (which may or may not reach statistical significance), re-opened jobs can be re-opened for a number of reasons, with very different vibes attached to them. Maybe the job was offered to someone who turned it down too late for an offer to go out to number 2 on the list. Maybe the job wasn't successful because the committee, the department, the dean, somewhere, someone had issues. Maybe the job got re-defined. Which brings us to the rest of the interviews he got. Some R1 departments put out ads for "evolutionary biology" or even "invertebrate evolutionary biology" when others put out an ad for "evolutionary endocrinology in aquatic invertebrates with an emphasis on trophic adaptations to extreme environments" although they are looking to fill the same teaching /colleague niche in a department. Sometimes "evolutionary biology" means they will consider all evolutionary biologists and sometimes not.

The bottom line on the jobs is: there is a lot of missing information. Information that he just can't know, that we just can't know. This may or may not be random. One year may be an El Nino year of jobs. One year may a bumper crop of tropical fruit.

Now, the thing I found most depressing in his post (my emphasis):

First, it appears to be very important to have a major paper from your postdoc published. Not in preparation, not on bioRxiv, not in review. Published. Preferably in Cell, Science, or Nature. I guess we all know that by now.

No, actually we all don't know that. And there have been plenty of tweets and posts from folks on R1 hiring committees that say that Glamour Pubs aren't necessary. In fact, I had two friends each of whom got hired at UCSD (where Kryazhimskiy took a position)a long time ago (yes, boomer boomer, yes a different age). Neither had glamour pubs, but both were very good, both got tenure, and each has gone somewhere else (post-tenure). But that's an aside. The committees I have sat on in the last 10 years and the ones to which I have sent trainees or others I've written letters for don't really believe that Glamour Pubs are necessary.

I do agree that pubs, a major pub, a pub demonstrating accomplishment is necessary. And if he didn't have that kind of pub (not necessarily glamour) when he first applied, no he wouldn't get an interview. What is a strong pub here? One with data, with hypotheses tested, with conclusions and an interesting discussion. One that demonstrates what the primary author is capable of doing.

I checked his pub record from his lab webpage, and it looks like he has 5 pubs in 2014, and then one in 2012 and 3 in 2011 (from an earlier postdoc). If in applying in 2014 none of the 2014 papers were out, no he wouldn't be competitive. And yes, getting a bunch out changes the story. But, what I, and most committees I am on, count as important is the science. We read at least the abstracts, and talk about the significance and innovation and potential (kinda like a grant review).

A last note about his CV is that he did two postdocs lasting over 7 years. One of the calculations I've seen done for deciding on who to bring in for an interview is publications per PD-year, not counting pubs in PhD years (although you can do it that way too). When hiring committees, departments, individual colleagues-to-be care about productivity. The whole story about Maria not getting tenure was a productivity story.

So, read the post. Its interesting and in some ways encouraging along the yes-you-can lines. But don't buy into the glamour pub scam as being the basis for getting a job.

11 responses so far

  • drugmonkey says:

    Productivity for some people just isn't smooth. Whether fighting multiple rounds with journals or because a project just finally comes into its own.....sometimes there can be a drought followed by a (relative) deluge of papers.

    How is that viewed in your opinion?

    • potnia theron says:

      That happens, and not just because of pubs. Sometimes the way some data are collected, for example field seasons, or longitudinal data that require a year of animal growth mean that nothing to write up for a while and then a whole bunch of stuff submitted within a few months.

      This is not necessarily a problem. But applying for a job in a "drought" may not be the best way of demonstrating productivity. It why some committees I've been on considers "total papers published since PhD/number of years since PhD".

  • Newbie PI says:

    I'd say to postdocs out there that they should take a quick look at who has been hired recently for tenure track jobs in their field at R1 universities. (I'm talking about recruited through a job search, not internal promotions or spousal hires, etc) If the vast majority don't have a glamour pub from their postdoc, I'll eat my hat.

  • Established PI says:

    The key to getting an interview and a job is high-impact postdoc publications. To the great majority of search committees, that either means (1) C/N/S pub, (2) Nature family, Cell family journals or some other field-specific highly ranked journal pub, or (3) body of work in methods-oriented journals that may be specialized but has high impact on a specific field that is of interest to the search committee. Before his Science paper, Kryazhimskiy had 1st author pubs in Evolution and PNAS. I don't know his field, but my guess is that these did not qualify as a record of high-impact pubs by the simplistic method of looking at journal rank. Was the work in fact high-impact? I have no idea, but having Science give its stamp of approval is a major attention-getter for most search committees. Not saying that is the way it should be; that is just the way it is.

    • potnia theron says:

      Both Evolution & PNAS are relatively high impact, and well respected.

      But I disagree with your point that *impact* means glamour. Sometimes it does, but often it doesn't. The work needs to be good and it needs to be solid and it needs to be plentiful enough.

      The thing not controlled for in the original post & CV is number of pubs, and pubs/year which increased dramatically at 2014.

      • Established PI says:

        I did not say that impact=glamour, but rather that search committees by and large take that view. There are notable exceptions but they are exceptions, not the rule. Oceans of ink (and electrons) have been spilled over this topic but I don't see things changing.

        Evolution and PNAS may be relatively high impact (I know the latter much better than the former) but the reality is that they are not viewed in the same light as CNS or babyCNS papers (and perhaps eLife?). You don't need a CNS paper but it sure helps to have something the next level down - that is just the reality. His other 2014 pubs were not first author, so I don't think those had a significant impact on whether or not he got interviews or offers.

  • Dave says:

    Not this again.....

    Dude has:

    1 Science paper
    2 PNAS papers
    2 PLoS Genetics papers

    He has an awesome pub record. All of these are 'glam'. The IF alone doesn't always determine what's viewed as glam or not (i.e. PLoS Genetics, PNAS, eLife are all considered glam rags). It's not unreasonable that the Science paper put him over the edge though.

  • Anonymous says:

    "If in applying in 2014 none of the 2014 papers were out, no he wouldn't be competitive. And yes, getting a bunch out changes the story."

    Maybe you need to read the guy's post again. Because he said that the *only* difference was the addition of the Science paper to his CV. And yeah, maybe his references noted that in their letters ... not that anyone would need them to point out this obvious fact.

    But if it is true that this really was the only difference between the 2013 and 2014 apps, that does make it seem as if the committees did not, in fact, evaluate the quality of the science for themselves. He did, after all, have 2 PNAS and 2 PLoS Genetics papers, too.

    You seem to be trying really hard to come up with alternative explanations. But sometimes the most obvious explanation is, in fact, the right one.

    • A Salty Scientist says:

      I have a friend whose *only* difference was a K99. My friend could have substituted "K99" for "Science paper" and written an identical article. There is too much noise in TT searches for anecdata. Also, see Dr Becca's post "The recipe for landing a tenure track job."

      • Newbie PI says:

        The fact is, a Science paper and a K99 are both game changers in terms of getting a faculty job. Comparing yourself to those who have recently been hired is one of the clearest ways of determining if you're competitive for a job. How many people do you have to look at before anecdata becomes real data?

Leave a Reply