How Should One Design Their Dissertation Topic?

Awhile back I had a debate with some of my PhD friends over the type of dissertation project that was most likely to land you in a position at a Research I institution. We fell into 2 main camps.


Off to get some data!

The first group thought it best to pick a series of “safe projects”.  Our definition of safe project was one that had a high probability of yielding results that would meet your predictions and would build upon an existing framework.  Despite the fact that your results wouldn’t be super sexy or surprising, these experiments would show that you understand how to conduct sound science and contribute to an existing body of work.

The second group thought it best to pick a risky project that, should it be successful, would dramatically change our understanding of a particular phenomenon. Should it prove unsuccessful, however, you’re left with pretty much nothing.

At the heart of the debate was our desire to be attractive candidates for the precious few positions in academia that open up each year. Of course, an even smaller subset of these academic positions are at Research I institutions, where a number of the individuals in the debate are striving to end up.

Completing a safe project will likely land you at an institution with a heavy emphasis on teaching. After years of publishing sound science and establishing your name, then you may be able to make the jump to a Research I institution eventually.

The problem with being at a school with a less strong emphasis on teaching is that they usually can’t afford to give as much money to start up a new lab as a school with a stronger research emphasis might. Additionally, heavy teaching loads mean less time in the lab or in the field and less time to write papers. So we decided that a safe project could land you in a Research I institution eventually, but it might be a pretty long journey before you make it there.

A risky project that works out and results in a number of publications in big name journals will certainly boost your chances of getting into a good research institution early in your career. The monetary and facility support that you receive here will be much better than at a teaching institution, making it easier for you to conduct lots of great experiments and acquire high quality students to work in your lab. What researcher could ask for more?

On the other hand, if your project idea doesn’t work out, then you’re worse-off than you would have been had you done a safe project.  A series of post-doc positions can fix this mistake, but you’re once again years away from the coveted Research I position.

So what project do you choose?  Well, like so many things in science, we’ve probably set up more of a dichotomy than actually exists.  I think that the best solution is to set up a big project with one sexy, but risky experiment and a series of safe experiments.  You need 3 chapters for your dissertation anyway, right? I think that taking a risk on at least one of the chapters is worth it for the potentially big pay-off.

I’d be interested in hearing feedback from other graduate students on how they are choosing their dissertation or post-doc projects.  Additionally, I’d love to hear feedback from post-docs or professors on what they think makes a successful dissertation project (where success is defined as ending up at a Research I institution).

(NOTE: I do understand that outstanding research is conducted at non-Research I institutions.  Also, I think teaching is vitally needed and noble pursuit. I did not mean to downplay the importance of either of these points, I just personally have my heart set on a position at a Research I institution.)


16 thoughts on “How Should One Design Their Dissertation Topic?

  1. I think people get caught up on whatever is hot at their institution or to their advisers and lose track of the fact that there are actually a remarkable number of low hanging fruit out there–projects that aren’t done because the data collection is tedious but the theory is awesome, projects that aren’t done because technological limitations prevented it previously, projects that aren’t done because people haven’t got time to read all the relevant literature….

    My strategy has been to try to think of new ways to ask the questions that people in my field (primatology) have been asking for a long time. There’s so much out there in terms of interdiciplinary stuff that just so often hasn’t been integrated, that you can come up with amazing new ways to piece together. And I definitely think that a dissertation should be cool whether you prove or disprove your theory–you never want to throw all you chickens into the basket of one (really, really cool) hypothesis. It’s both unnecessary and unwise.

    What I’m saying, basically, is that we should all pick safe projects and make them sexy after the fact by integrating different perspectives. Shrug. I’m the eternal optimist though, so we’ll see…maybe I’m just talking out my ass.

  2. I didn’t even consider future career aspirations when deciding what project to pursue. My take is: pick an advisor you can work well with and pick a project that interests you enough that you are willing to invest enough time to learn the sub-field inside and out, so that you know exactly what the relevant questions are and can design experiments to test those questions. Learn how to propose new research. Finally — and this cannot be understated — learn how to demonstrate that you have done this in a 1-minute, 15-minute and 1-hour talk.

    Most departments don’t care so much exactly what you did as how you did it. People who are likely to succeed in an R1 setting will stand out no matter what their thesis topic was.

  3. I have no interest in becoming a professor, and really have no idea what I’ll be doing after I finish. I figure I can worry about that later, and in the meantime focus on just getting through grad school.

    As I was finishing up my first year, I ended up at a crossroads, research-wise. I could either take the project handed to me, that already had funding, and use that for my dissertation. It was not something I was particularly interested in, but it was important stuff, and the end result would have been used throughout the state of California. Or, I could let that go to another grad student, and plan a project of my own in an area I find more interesting.

    I ended up going with the latter. In the end, my research will not be sexy or groundbreaking, but it is very important to me to feel ownership of my dissertation project, from start to finish, which I would not have felt about the other project.

  4. Suggestion: Use science! Get a list of all of the professors at the Research I institutions in your field. Find the ones that got hired straight out of grad school (This is the hardest step, as you probably have to read their bio’s or find their CV’s) Then look up their dissertations and make a chart of whether it was based on sexy or safe (or ask them, I suppose). Then blog about your findings, with charts and graphs and methodologies.

    If nothing else, it will be helpful for others trying to make the same decision as you are, and possibly make you minorly famous in your field, helping you get that job you want.

  5. That’s a really good and original post about a semi-political aspect of science careers. I adore Anna’s answer.

    I also think it’s important to be working on something that you think is cool. if your eyes don’t light up at the idea now, how are you going to feel about it after a few weeks getting up in the middle of the night to ensure your experimental apparatus hasn’t done that thing again?

  6. I don’t know, I’m too busy trying to figure out how to eat and keep my mortgage payments with this new [lower paying] job thanks to the Microsloth layoffs. I hated that company anyway for keeping most of the hardest working geeks as perma temps. Anyway, I can’t afford grad school, and University of Washington says “we don’t have room for people like you” whatever that means. Nice huh?

    So, it’s really hard to relate to your current ideological dilemma because I’m still trying to find a grad school that accepts food stamps.

    Good luck.


  7. Currently a PhD student in solid state Physics at RMIT University (Australia), and I am much leaning towards the ‘do what you think is cool’ option. (what anna said)

    After doing a lot of safe projects back at my uni, in which we could pretty much predict the results before we even set up the apparatus, I have spent a few weeks working in the plasma science division in lbnl in Berkeley, CA…. AND ITS COOL!

    I am finding myself absorbed in this (if but slightly) risky work, developing new techniques, and finding new and better ways to get results. Often I will find myself in the lab, and think to myself ‘Geez, ive almost been here for 12hours! Where does the time go?’

    unfortunately this is my last day at lbnl, I will be doing some more work in PA (conference), and then heading back to the more boring PhD work at home. Doing ‘new’ and ‘cool’ stuff is always exciting, and if you ask me, that is reward enough in itself.

    Good luck with your dilemma!


  8. PhD, currently ABD with one chapter published (two to go).

    The most important thing to do is pick an advisor who is 1) active in your field of interest and 2) will actually help you. I broke ground in a new disciplne for my PhD (have a BS and MS) and picked a school that would fund me. However, I suffered in limbo for a while until I said, “F’this! Let me finish!” and am now making progress, but it is still a bit of a grind. If you pick an active advisor (not one who is resting on laurels and past accomplishments), you are more likely to get a ‘sexy’ project versus a ‘okay, guess I can do that’ project (mine is somewhere in between, but only because I picked it). Also, make sure you go to a school active in your field of interest with a couple different professors in the same field…numbers means support financially and begets stability of a program.

  9. I came here from SMBC but was interested in this topic. I am a professor, although in the humanities. You have to remember that the coin of the realm in academia is publications (or at least potential for publication). If sexy gets you published in Nature or Science or top journals, and you want to be at an R1, then you should go with sexy. I think that your strategy of having one sexy experiment is reasonable.

    Your dissertation will define you to potential employers and should contribute new knowledge to the discipline. Think of what you would want to be known for and head in that direction.

    Good luck.

  10. This looks cool so far, what’s up people?
    If there’s anyone else here, let me know.
    Oh, and yes I’m a real person LOL.


  11. As a postdoc on the job market I agree the sexy project amidst some safe projects is the way to go. If it fails you still have the pubs needed to get a job. As long as safe projects are still new and interesting. “We did the same thing someone else did but in a different species and saw the same thing” probably won’t get you too far unless you have some other new ideas out there. Publishing early is really the biggest thing, but I know you’re already doing that 🙂 One prof told me his search committee doesn’t even look at anyone with <8 pubs, and they count before they look at journals, topics, etc, so you've got to make that first cut first.

  12. Guy .. Excellent .. Amazing .. I will bookmark your site and take the feeds additionallyI am satisfied to seek out so many helpful information here in the publish, we’d like develop extra techniques on this regard, thank you for sharing. . . . . .

  13. Не терпится рассказать неизгладимыми впечатлениями от событий прошлого уик-энда. Примчался ко мне с сибирской столицы сотрудник в гости. основное что он мне сказал, было представить его с клубной жизнью града. Интим услуги являлись в обязательном порядке содержанием. Я не выказываюсь завсегдатаем в подобных помещениях, отчего спросил у знакомых, куда нам выехать. начальный таковой рекомендация находился одновременно присвоенный на снаряжении, и мы мгновенно отправились по подтвержденному местоположению. – У выбора версий растерялись глаза, из за того что элитные девочки имелись в том помещении.Раньше я обычно использовал уличными девками , но теперь понял, почему лучшие проститутки Москвы ютятся особенно в салонах, ну или лишь в том салоне, где мы были. Во что бы то ни стало вновь зайду туда.

Leave a Reply

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

You are commenting using your 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