Thursday, March 03, 2011

Out With The Old?

A question arose recently about an early career scientist* who has been slow to publish results from their PhD research. Now that this person is on the tenure track (TT), they have to make decisions about how best to spend their limited time: pursuing new research vs. finishing old projects.

(*someone completely unrelated to my research group, just in case anyone is getting paranoid)

I am not talking about unreasonable expectations by former advisors regarding post-graduation or post-postdoctoral publication; i.e., I am not referring to cases in which someone published the key papers from their previous work but their advisor would like them to publish even more. In that case, new work clearly must rule.

The tricky cases are when there are still major papers that should come out of the pre-TT years, but these have not yet been written/submitted.

Yet, if a TT professor spends time writing up old projects, there is less time for the new projects. There are only so many hours in a day and there are only so many years until the tenure evaluation. And there are an infinite number of important things to do in that time.

Factors in the decision about how to apportion time between old and new work include:

- It's important to initiate and publish results from new work that is identified specifically with the time at the TT institution and that does not involve the TT professor's PhD advisor(s) or postdoctoral mentor(s).

but:

- It's important to finish what you started, especially if your PhD and/or postdoctoral research was particularly interesting. Your visibility and reputation derive from the totality of your work, not just what you accomplish in your TT years.

Ultimately, I think that new work (research done entirely at the TT university) is more important than old work (research done during a PhD and/or postdoc), so if you have to choose one over the other, the new work is what the TT university will want to see at tenure evaluation time. Some people do get tenure based primarily on work done with their famous PhD and postdoctoral advisors, but this is not a good strategy for getting tenure and for establishing a respected research program.

As an advisor, I am not objective about this matter. Although I can write -- and even convince myself to believe -- that new work should prevail over the old in terms of publication priority -- I also feel that it's very not cool to leave advisors and other colleagues in the lurch with unpublished work, even if that was never the intention. Also, some institutions request letters from former advisors and postdoc mentors for tenure evaluations.

When I was a postdoc, I had published a few papers from my PhD, but I had some more to write. I spent most of my postdoc time on my postdoctoral research, but I systematically carved out some time for writing up the rest of my PhD research, so eventually it was all published. This turned out to be excellent preparation for being a professor and working on multiple projects at once; i.e., many of us, as professors, always have some projects in the writing-up stage and some in the data-gathering stage and some in the idea development stage and some in the glimmer-in-our-eye stage. It should therefore be possible to finish the old work without sacrificing the new work.

How's that for a mixed message?: In reality, you have to prioritize your time and probably should favor the new projects over the old projects, but, ultimately, you need to get everything done, old and new.

14 comments:

Anonymous said...

Can you offer advice on how to "systematically carve out time" for writing up old stuff while trying to get new stuff off the ground, particularly when the topics are not very related? How do you switch back and forth mentally? Do you use blocks of hours/days/weeks?

Chris said...

Another thing to think about is the funding implications. When reviewing grant applications, one of the things that gets looked at is your publications from your activities/results from the last grant. The thinking being, do you have a track record of producing published results? This may be less of an issue in the specific instance of a young researcher who had previously been a postdoc on another person's grant, of course.
In that case, any publishing inactivity "hit" is borne largely by the PI, not the ex-postdoc....

Anonymous said...

I would say that if you have all of the data necessary, you should absolutely finish the old work. Unfortunate as it is, there are many bean counters out there who review tenure decisions and often care a great deal only about the number of publications and which journals they're in. Plus if you have gathered some data which increases the knowledge of the field, it is important to share that with the community. Starting new projects is essential, but if you never publish your previous work you are getting 0 return on a (sometimes significant) investment of time.

M. S. AtKisson said...

To Anon at 02:23: The only way to carve out time and do it effectively is to put it on your schedule, and adhere to the schedule. This is true even for writing a manuscript or grant application in your current area of interest. If it's old work, and you need to be productive in the new field, I suggest you give yourself 2:00/day a couple of days a week, and be ruthless about the time. Do not make yourself available for interruptions. Try to get some solitude.

These last two points are critical in making the best use of your time. Researchers have found that it takes 15-20 minutes to get into "the zone" where you're writing productively and with focus. Any interruption requires another 15 minutes of run up to get back into that mental space. Once in the zone, most of us can write for about 90 minutes before needing a break. That 2-hour time window gives you time to get into the zone, especially If your project is the old stuff and vastly different from your new work.

My trick is to set a timer for 20 minutes, and not allow myself to check email, FSP's blog, Slashdot, whathaveyou. I turn off my email client, and sometimes I turn off my wireless connection, but that isn't always practical, as we often need to look stuff up on line while writing. Usually, by the end of that 20 minutes, I'm in groove with whatever I'm doing. 90 minutes of focused writing time are far more productive than five 30-minute bouts.

Bottom line: Schedule your writing time, and be ruthless about it.

franglais said...

In general, it is very difficult to return to old work once you have new work. This is why it is extremely important to publish PhD work during the PhD and/or very soon after. In my case, I published what I thought were the important parts of my PhD, but there was quite a bit of unfinished business. Then I was fortunate to land an academic position. I did not publish my not-so-old (PhD) work but used it as a source of ideas for proposals and projects. These ended up supporting my first two PhD students while I was a TT professor. They took the lead on these projects, and it ended up being a great situation for all of us.

Anonymous said...

So funny you wrote this today. I'm doing my postdoc and stayed home this morning to deal with some research that has maddeningly continued (>1 y) past my PhD. It's important work and will be the best thing I've done so far, but my hands are somewhat tied in how quickly I can get the results. (It's computational.) I work full-time on my postdoc projects too. Not having finished this PhD project has been SO demoralizing (as in makes-me-want-to-leave-science demoralizing--some of the delay was from bugs that I didn't catch), and I fantasize daily about getting it done. I can't imagine what my collaborators would think if I abandoned it.

Susan B. Anthony said...

Agree with franglais that unfinished work can be an excellent source of ideas and short projects for undergrad and graduate students. It definitely helps to have done some of the background legwork so that the students don't have to start from scratch. It's win-win, assuming (as FSP says) the work gets done.

(My captcha word is "wating". Ha!)

Female Science Professor said...

Perhaps I shouldn't have used the word 'systematically'; something like 'routinely' would have been more accurate because I didn't organize my schedule any particular way. I just made sure that I made steady progress on my old projects. In some ways, this wasn't difficult because I was interested in them and wanted to finish those papers. And I was fortunate to have a lot of flexibility in my schedule.

Anonymous said...

This can happen even for PhD students, particularly those in fields where you are expected to work on multiple smaller projects rather than one large dissertation. I have stuff I worked on a year or two ago that I brought to some conclusion, after which I moved on to other projects. I never got around to writing them up, and I'm scrambling to do so now that I'm in the final stages of grad school and need a publication record.

Anonymous said...

It also happens that you have a paper that you badly want to be submitted. Need to have submitted as for grants and all, but your previous supervisor is just not doing it....

alegalalien said...

Thanks for posting this! I am coming to the end of my five-year post-doc and still have a couple of PhD papers hanging over my head. It has been really hard to finish up everything and I would recommend to all grad students and new post-docs to get their PhD papers published ASAP!

I was exhausted by my dissertation and didn't want to look at that data again, and then I got distracted by the new job, a marriage and a baby (!). Now that I am hunting for a new job, I realise that my publication record is dismal and I have a lot of catching up to do. I am chipping away at that old data every evening...

Carmelo Fruciano said...

In my opinion, the ideal would be a mix of old and new (write the old while gathering new data for the new), like FSP suggests at the end of her post.
I would echo what Anon at 8:22 wrote and I would add that new stuff can hide unexpected problems (for example, what if experiments don't work as expected?) while writing up old finished stuff should be much quicker/easier in many situations.
Following M.S. Atkisson's advice of sticking to a schedule, it shouldn't take too long to write old stuff and move to the next step (working full time on new projects while waiting for the outcome of the review process)...

Anonymous said...

I do find it a bit baffling that a lot of people get 90% of the work done and then lose the steam to publish the work. I mean, don't people WANT to announce and advertise their work to the world?! After all the blood, sweat, and tears, people want to just let the work languish and not have anything to show for all that? I've even heard of a significant fraction of submitted papers to journals that are never revised after the referee's comments -- not because the work seemed fundamentally flawed, but for no apparent reason. This seems like a tremendous waste of EVERYONE's time, effort and money.

Anonymous said...

"I do find it a bit baffling that a lot of people get 90% of the work done and then lose the steam to publish the work."

I find it baffling also, but it happens to me a lot. I have several research results that I've been sitting on for years. It is not that I don't want to publish them, it's just that I have a hard time writing important stuff. Writer's block runs in my family.

I should be getting out a paper soon with a grad student who finished his PhD a year ago. And one collaboration that I started 7 years ago will finally get a paper out, now that the student involved has finished her PhD (which was mainly about the protein that started the collaboration).

I'm taking a sabbatical next year and I may decide to dedicate half the time to clearing my backlog of paper writing.