Educated Woman - The Grad-School Adventures of Micella Phoenix DeWhyse, Chapter 35 - Thesis Proposal Time


Well folks, for the benefit of those of you who are just joining me on this journey, and those who might have forgotten what I'm up to: I'm trying my hardest to slide into 3rd base and preparing to steal home at the first opportunity. I've accomplished enough research on the project that will become my thesis, I've gathered my committee and now it's time to propose my official thesis work. Once I've been counted safe on third--had my proposal accepted by my committee--I can start the sprint (or maybe the slow crawl) for home. Or, if not for home, for anyplace other than here.

In my department, most students propose their thesis work as an extension of a project that is already well under way. This is especially true if you happen to be the first or only student working on a particular project, as you are proving concepts that haven't been worked out by your research group. Your committee sees that you have been able to perform the research satisfactorily, and together you formulate what will constitute the rest of your Ph.D. work.

As one of my elders told me, you can often work 6 months to a year off a few good pieces of data. My work has recently taken a turn for the better, and lately I've been doing just that: living off scraps that are a little less meager than the ones I'm used to. Things have improved greatly, actually. Large chunks of progress have been made. I am no longer the only student working on this project; other people are producing data that can help me tell my committee a coherent story so they'll let me leave.

Yet, like all the other steps in the Ph.D. -- qualifying exams, possible failure, group hell, committee choosing, research anxiety, project hating, advisor issues, and so on -- the stress level involved in proposing your thesis work is high. There is a dance to perform--an oral defense of the work you have already done along with a proposal of the work you plan to do to finish your Ph.D.--and the usual fears creep in. Fears that you aren't prepared enough for this performance, that the critics--you're advisor and committee--will rake you over the hot coals excessively, and that your committee will decide you need to do more, or something different, before granting you your degree, is enough to give you nightmares and keep you working hard.

Several things need to be accomplished before you present your proposal to your committee. I've broken it down to four main points: searching the literature, reading, collecting and culling data, and (finally) writing. Searching the literature and reading will be covered in this chapter; collecting/culling data and writing will be covered in the next, chapter 36.

Searching the Literature: Who, What, When, Where, Why, and How

Remember those 5Ws and an H? They can be a big help in your literature search. Actually, an exhaustive literature search is not what needs to be done at this stage. Hopefully you've done that already, more than once. The goal at this stage is to make sure that you're up to date, and that you haven't missed anything that your committee or advisor might find and call you on. You should be able answer the following questions:

  • What are the seminal papers in your field?

  • Who are the bigwigs and the hangers-on in your area?

  • What is the state of the art in your field, and how is that research being accomplished?

  • When and where did your field evolve over time? Historical background information will help you put your own work into perspective. It will also look great in the introduction to your dissertation.

  • Why is your area of research important? What larger questions does it address? Why should we, who are not in your field, care? Will it cure diseases? Rid of us world hunger? Help us make big-screen TV's cheaper? Look really cool on the cover of Science? (What wouldn't?)

The first bullet point--that list of seminal papers--contains the answers (or at least other references) to all the other bullet points. But in most areas of science, the literature is a vast and shapeless thing. Narrowing down this vast list to a manageable number of seminal papers can be hard. The prestige of the journal and software/indexes that allow you determine the number of citations of each paper can help lower and stabilize that mountain of papers on your desk. Reviewing articles, if they exist, can also help quell the madness and keep a few pennies on that print/copy card. If you've already written a paper this experience, too, will help to cut down on the information overload, since you will have done some culling already.

And remember: papers are being published all the time (assuming you're working in a field that someone else is interested in, which you hopefully are), so before you defend, make sure you know what has happened in your area since your last major literature search and have thought about how this impacts your own work.

As you complete your literature search consider this: Citation software is an absolute must for at least 3 reasons:

  • You need a mechanism to organize all of the papers you have acquired, read, and not read. Folders and piles are a great way to keep everything you do have together, but if you haven't listed it somewhere you might forget all about it.

  • Writing out bibliographies is a waste of your time, and you will always get something wrong. Wouldn't you rather spend your time making sure all of your actual proposal writing is in order? Or do you want to spend hours formatting your bibliography by hand? Stop using your hand-written bibliography as an excuse to put off actually writing. It's not allowed.

  • Many search engines allow you to download information directly into citation software. I love technology, don't you? Remember: Efficiency = Graduation. Graduation = A Life You Just Might Like.

  • If your advisor doesn't own a group copy of citation software, convince him or her to purchase one. If your PI's a cheapskate, invest in your own copy. It will save you time and agony ... and when you consider that time is money--especially given the level of typical grad-school stipends--it'll save you money as well.


    There are effective ways to read, and there are ineffective ways to read. You should choose the effective ways. Consider the following examples:


    Reading with purpose. You obtained the article for a reason: an experimental technique, a particular finding, a historical perspective, or an interesting innovation. Make sure the paper serves its purpose; don't just glaze through it, or run your eyes across all the words without noticing what they say. Chew on it. Digest the part you can use and spit out the rest.

    Reading in a journal group. Two or more brains are better than one. Two or more brains with different backgrounds are even better than two or more (similar) brains. Find friends to read with and ask questions as you immerse yourself in your topic.

    Writing while you read. What questions does this paper lead you to ask? Write them in the margins. Do the authors answer these questions? Do you agree with the conclusions that they draw? Is their methodology sound? What experimental traps might they have fallen into that would call their data into question? What other experiments would you perform? Are the conclusions they draw supported by the data? Are the data/conclusions in line with other literature? Is there something new and profound here? Or were they just sloppy?

    Understanding the figures. Some scientists/engineers, once they've seen and done it all, look only at the figures. While I don't recommend you do this until you've become fabulous, you should be able to explain a figure, the technique used to obtain that figure, and the significance of the data to the rest of the paper. If you can't, someone on your committee will call you on it. Count on it.


    Limiting the scope of what you read or reading without access to sources that can put items in context. Yes, you might not understand an article the first time through, or even the second. That's okay. But if you don't have resources/knowledge to understand it eventually, you're just wasting your time. Often the things that trip us up initially when we are reading a new topic are vocabulary and technique. Do you have at least a basic understanding of the techniques and terms that are being used in the paper? Do you have a basic book that can begin to explain these things? Is this paper essential, or something optional, for later? Can you afford the time you would have to spend on it? How important is it likely to be, really? Here's an important trick that can save you loads of time: if reading the paper seems hopeless, call up one of the authors and ask him or her to explain it. It takes a little courage the first time, but you get used to it.

    Highlighting every sentence. Are you doing this just to keep yourself awake? Or do you just like the way that highlighter smell makes you feel? Stop wasting ink! Highlight sparsely. Main ideas, points the authors are trying to prove/disprove. Everything is not important; learn how to distinguish. Go get a cup of coffee.

    Assuming you got it all the first time. Technical papers need to be read more than once or twice. Take the time to understand the new knowledge you've acquired and what it might mean. Do you know how it will be applicable in your own research?

    Reading in bed. We know it's a good tranquilizer, but you might actually need to know some of this stuff. Try to keep your bed free from papers and work, you might sleep better. I hate nightmares with equations and scientific instruments, don't you?

    Space and memory constraints prevent me from providing even more tips, tricks, and secrets for literature searching and reading. Lots of them are out there. Slightly senior colleagues can probably provide even more. If you have any that you'd like to offer our readers, please send them to and I will disseminate them to the world. Good luck!

    Follow Science Careers

    Search Jobs

    Enter keywords, locations or job types to start searching for your new science career.

    Top articles in Careers