Is it really "publish or perish" for PhD students?

Is the oft-repeated adage really true? A perspective from a survivor of low publication rate.

Go to the profile of Colleen Mills-Finnerty
Oct 10, 2019
1
0

“Publish or perish” refers to the unfortunate reality that no matter how brilliant or hard working you are, without a record of productivity you risk stalling or derailing your academic career.  Many graduate students lose sight of this as they stress over tough courses, negative feedback on qualifying exams, or disappointing results of experiments (myself included). Whether your career goals involve academia or industry, publications are the universal currency, and learning to acquire that currency is integral to graduate training. 

I was fortunate to do my PhD with a PI whose tenured status largely meant that I wasn’t under constant time pressure to produce papers.  There were distinct positives to not being pressured to crank out a manuscript every few months.  I had time to learn a diverse array of skills ranging from how to prep and teach a course to how to code.  I defended my dissertation proposal and presented my data at international conferences. I did deep dives into the literature; I took intensive seminar courses focused on my niche area; I formed collaborations. None of these activities boosted my CV like a paper would have, but they furthered my development as a scientist. 

Despite lack of PI pressure, I knew publishing was crucial for my career. While there is no hard-and-fast, field-independent rule for how many publications you should publish in grad school, some departments waive the requirement to write a dissertation if a candidate has published three peer reviewed articles.  I therefore dutifully submitted three first-author manuscripts during my PhD. Two were rejected and eventually retired to the file drawer, never to see the light of day again.  This is the story of what happened with those papers, where I think I made mistakes, and what I would advise others with the benefit of hindsight.

The first mistake I made was giving up too easily. My first submission as a second year grad student was rejected with such vigor that I never resubmitted it anywhere else.  I had no idea that plenty of respected researchers rack up 3, 5, or 10 rejections per paper. In retrospect that paper might have been publishable with a round or two of revisions, but I never found out because I let myself be easily discouraged.

My next error was not keeping up with the shifting methodological tides in my field. I had followed best practices circa 2014 when I collected the data, but was submitting papers in 2016 or later. In that time, a sea change had occurred after various papers about the reproducibility crisis, the risks of underpowered studies, and various statistical sins had gone viral. Editors were desk rejecting work based on criteria that didn’t exist when I collected my data.  My second manuscript got stuck in a cycle of review, revisions, and rejection before eventually ending up in the file drawer.

My third error was doing a lot of work without looking out for “future me.” I didn’t document my data collection meticulously because at the time I didn’t anticipate publishing on all of it.  When I returned to the data later, realizing it might make an interesting paper, I had to do a lot of extra work to make the data paper-ready. If I had maintained a code repository with version control, I would have saved myself dozens of hours of work retracing my steps.  I recommend you document well so that your lab notebook, data repository, or code comments can serve as your decentralized memory. Your future self will thank you.

In contrast to my two other papers, the one I eventually published involved a lot of preparation and a spate of luck. It took around two years to develop my initial idea into a study, collect and analyze the data, and write the manuscript. The luck  came when I received “major revisions” on my first submission. I published the paper and even won an award. However, I later learned the hard way that it is the exception, not the rule, to get major revisions on a first submission - even for work that is your best. It is much more common for papers to bounce to several journals and go through multiple rounds of review, often leading to a year or more between initial submission and publication. I ultimately graduated with only this one first author paper, which was enough to land me a postdoc position. However, not having as many publications as other postdocs has added a layer of additional pressure that I am still dealing with years later as I watch my papers go through the rinse cycle of desk rejections, major revisions followed by rejection, and other outcomes that delay publication.

There are many other reasons you should try to publish more than I did in grad school. First, you will cultivate distress tolerance for rejections - until my tenth rejection or so they all felt like the end of the world. Now they mostly feel like an administrative annoyance. You will learn when to retire a paper to the file drawer - some collaborations fizzle out, or errors in data collection render a paper more effort than it’s worth. You will learn the typical “project life cycle” of taking a scientific result from hypothesis to publication in your field, which can often stretch to several years.  Additionally, you can fine tune a sustainable writing practice instead of leaning on bad habits like binge writing or pulling all nighters.

Maybe you are reading this and think it’s too late because your defense date looms. If you can’t get the golden goose of the first-author paper in the “good” journal by graduation, consider other (faster) ways to generate a writing sample.  Can you write a blog post,  commentary paper, case report, or book chapter? A protocol paper, registered report, or publish a pre-print of a paper that is in review? A well-maintained code repository also gives other researchers a chance to evaluate your work.  You generally need some sort of writing portfolio to apply for postdocs, but don’t underestimate the importance of networking, acing interviews and job talks, and fitting into niche research areas in securing a position. A postdoctoral mentor, job search committee, or grant review committee looks for the story of who you are as a researcher. Your publication history is the most impactful way for them to glean that story.  However, extenuating factors can mitigate the impact of a low publication record when you are still considered early career. For example, I have expertise in several specialized statistical and experimental methods, and work in a high-priority research area (depression). These factors have offset my low publication record in grant review (so far - knock on wood).

I am writing this piece with full knowledge that survivorship bias likely influences my less-than-dire take on “publish or perish.” However I think it’s important to state that there are times I haven’t kept pace with publishing norms in my field, and I didn’t perish. If this gives a sliver of hope to someone feeling like they aren’t a good enough scientist or that they have no chance for a career, then I have done what I set out to do.

Go to the profile of Colleen Mills-Finnerty

Colleen Mills-Finnerty

Research Health Science Specialist, Veterans Administration Palo Alto

No comments yet.