Repost: Your thesis means nothing, worry about the papers

Aug 03 2010 Published by under [Education&Careers]

Since this has topical relevance in my life right now, I thought I would repost this from October 19, 2009.

Whether it's tradition or a lack of well communicated expectations, grad students seem to be enormously focused on The Thesis. Students fixate on this document for months before writing it and then for weeks to months of actually writing it. I have to say, I've never quite understood that, because if you had to put odds on the people who will read your whole thesis it would look something like this:

2:1 Advisor
3:1 Committee/Examiners
10:1 Over-eager new student who takes on aspects of your project when you leave
1,000:1 You, after it's done.
100,000:1 Your parents
1238947692092y47nc783et687:1 Everyone else

When I got my thesis back from the binders, I opened it up and read the first sentence. In that sentence, I had a typo that made the word "three" into "tree". Seeing that, I promptly shut the thing and that was that. Never opened again.

The one caveat to this is if you never publish the papers. In that case, the community might find it and someone might crack it open, but probably not. But that's the point. Uncommunicated science might as well never have been done in the first place. Get the papers out. Don't focus on an arcane document that will gather dust for the next 50 years until the departmental office needs space and throws the old ones out. Write the papers, or at least write the chapters as papers so you can get them out quickly after the thesis. If you publish before you graduate, writing your thesis should be about as simple as slapping together a half-assed intro and conclusion (complete with typos that no one catches) and be done with it.

16 responses so far

  • Grant says:

    I agree.

    It’s one of the reasons I like the approach taken by some countries where students are asked to publish a number of papers, and the thesis is basically the papers collated with an introduction, review or whatever, giving the larger scene and rationale for the overall line of work. I believe Scandinavian countries take this approach: Bob O'H will probably be along shortly to put me right 🙂

  • Jason G. Goldman says:

    I am so glad that my advisor has the same opinion - put together a few papers, slap on an intro and conclusion and change the margins so it looks like a dissertation.

  • DrugMonkey says:

    Grrr. True enough but I don't like it. I'm a monolithic thesis type of person. That day is long gone for science dissertations though.

  • Paul Murray says:

    I had been curious for a while what happens to these theses. Surely, I thought, if getting an advanced degree means that you must actually make a contribution to your field by way of doing a thesis, then these documents would be of immense value and ought to be online.

    Seems that's not the case. Colour me disappointed, but not surprised.

  • Bob O'H says:

    Nope, Grant's right. Holland and Finland do the same thing - if you've published the paper, the proofs go straight into the paper. This makes thesis writing so much easier, only the introduction needs to be written on it's own.

  • Fat_Az says:

    This assumes that all the papers were on the same subject. The countries you're talking about treat each paper as the end point of a project. So you may have to work on 4 or 5 disparate topics (obviously within the same branch of science) to get the requirements for your PhD. This does not a thesis make in my view. I agree on the whole publishing thing, but sometimes the research just doesn't work. This would probably work well for a university that does some decent academic work and in which publishing is pushed, but I know from fighting the system in my original uni that commercial work triumphed. I have also seen far too many papers published that were obviously bit pieces of work that had been done, were not complete enough to be published in my view or were just total rubbish, but got published because of the institute or the professor on the list of authors. I have also seen some papers that were magnificent, were obviously the result of complete focus on a particular topic by a good research student over the course of their entire PhD. In that case they have a magnificent paper that summarises everything they've done. Makes for a good paper, not a good thesis. The thesis is also about the failures. The methodology of all the research, not the methodology of the research you published.

  • proflikesubstance says:

    While I agree that a student's work can end up being across a broad range of a particular field, I also think that having a central theme to the work is a good idea even if it has to be a big umbrella.

    Do you think the student gains more from writing out every little detail of their work, whether it was successful or not, than they do from getting good guidance on writing a paper and seeing it through the process? I ask this because students who concentrate on the thesis as a document often end up modifying for papers after they leave the lab, which can take much longer for a variety of reasons.

  • scicurious says:

    Even better if the intro is a review of the lit that can ALSO be published.

    But of course, this also really depends on the student having a thesis that is an overarching series of experiments on one hypothesis. Many students have these...but some students don't.

  • chall says:

    I did that (Swedish dissortation). The published papers (or submitted ones) are in the back of the thesis and then there's an intro with a short summary of the field, the interest of the research, summary of the papers and putting them in context of each other and the previously done research and then some thoughts/ideas about future studies.

    Funny enough, after looking at it - I don't know why it took my so long...(especially looking at it in hindsight) but it was harder than I thought. Good practice though.

    I wanted to add that yearlong project that didn't pan out, since I spent a lot of time on it and I didn't want to to get lost (like Journal of Negative Results) and it ended up as a paragraph after my supervisor told me it was silly to keep it since it wasn't published or publishable ... small victories 😉

    As for the thing that the papers can be in a various fields, I don't know. I've always thought that the PhD was going to have some kind of "red thread" that binds all the project together [either a technique, an approach or an organism etc]?

  • chall says:

    PLS> as for your question, I definetly learned more about writing papers from my submitting the papers and having done the whole process; "experiements, sort out which goes into the paper, make the figures, write in, revise it, send it in [did it really take 5 months to complete it?!] and then wait for revision comments... revise again, submit, get the acceptance letter, celebrate... and keep going on the new project" than I did from writing the thesis.

    I got more to benefit grant writing from the thesis, as in the thinking about the "overall picture and where does this research fit in, and what parts are missing".

    Still though, what I clearly needed to learn as a PhDstudent was "what doesn't have to go into the paper" since I write my M&Ms too detailed and it always get chopped down to at least half. Did I mention I hate writing too short and possibly unclear instructions [or comments, obviously ^^]

  • PLS we are now being made to rephrase and rewrite each paper for the thesis so that we do not plagiarise our own work. This is largely due to the rampant use of TurnItIn by our institution, which is making thesis writing more annoying than it already is. A retiring professors suggestion for writing a thesis was to write a introduction and conclusion, put your papers in the middle, staple the shit together and go start a postdoc.

  • I used to tell my students that theses were for practice and that they should save their life's work for their life. Most students try to do too much in a thesis.

    I think one reason it takes them so long is scale. Writing one big thing is much harder than writing ten short papers that add up to the same length. And for the most part, they're figuring out the science and the writing at the same time and finding out that they feed back on one another.

    One issue is whether we need to train people to write longer docs or not, since they rarely ever have to write something thesis-like again (unless you're like me and prefer writing research monographs to papers).

  • Beth says:

    The University of British Columbia allows students to write a "manuscript-based thesis" (i.e., chapters are published (or soon to be submitted) papers, plus a lit review chapter and a discussion/conclusion chapter to explain how it all goes together) for exactly the reason you are talking about here - no one actually cares about the thesis - only the papers you publish matter. One thing to be aware of though - since it's not that common in North America, not all external examiners understand that this is a legitimate format! (My defense was a few years ago, so it may be that things have changed now).

  • I have never understood the reluctance to write papers. For me writing has always been a reward for all those tedious experiments I had to do.

    I agree that students needlessly stress about the dissertation, but for me it was a labor or love . I could have just slapped a couple of papers together and wrote a short introduction. However, I felt I damn well deserved the paid time away from the bench to indulge myself in the process of writing a document that was truly my own. I took great pride in writing a review of the field, and my methods sections are used in that lab and shared with anyone who requests detailed protocols.

  • proflikesubstance says:

    But of course, this also really depends on the student having a thesis that is an overarching series of experiments on one hypothesis. Many students have these…but some students don’t.

    I don't think it is a requirement to have all of your thesis work fall under one hypothesis. In fact, I would think this would be very hard to do. What Chall mentioned seems more plausible.

    One issue is whether we need to train people to write longer docs or not, since they rarely ever have to write something thesis-like again (unless you’re like me and prefer writing research monographs to papers).

    I'm not sure we can really call writing one document that is really long as training, per se. I guess it is practice (for something one may never do again), but for all the classes and workshops on how to write a good paper, I have never seen one on how to write a good thesis. I think the format is antiquated, personally.

  • Dele says:

    Makes a lot of sense.

Leave a Reply