Ph.D. Advice

March 28, 2009 5:21 PM

Ph.D. Advice: I'm going to be starting grad school in the fall and I wanted some suggestions on how to plan out my thesis. What should I be doing from day one to stay organized and what kind of system should I set up to keep myself productive?

I'm going to be studying for my Ph.D. in materials science and engineering. I already know which professor I'm going to be working with and I have a rough idea of what my project is going to involve. Still, I want to be sure that when I arrive I don't waste too much time. To that end, I've been reading up on what the group has done in the past and I've been trying to study papers relevant to my field of interest.

Does anyone have any suggestions about what I should do when I arrive? What kind of systems have other people used to keep their ideas organized and how did you plan out your program? I'd really appreciate any tips or advice you could give. Thanks!

posted by Aanidaani to education (13 comments total)

Don't "plan out your thesis". Your thesis should be the culmination of several years worth of work, but the nice structure of your write-up will mask months of frustration, dead-ends and U-turns. This is the nature of research. My own Ph.D. thesis was made up of a few papers, joined together with bridging material and topped and tailed with a lit. review and conclusions. The key thing for me is that I spend 18 months working on a particular topic, and then abandoned it for something else, for which I was awarded the degree. A risky strategy, but it paid off for me. The message I'm trying to put across is "don't be too focused at the start". This might be difficult if you have a very applied project, or industrial backing, but (and I hope this doesn't sound too flaky) encourage serendipity. Read widely (ie. not just the journals in your immediate field), talk to lots of people, give mini-seminars (and attend them!). I hate to use the term "network", but it is important to "getting ahead" in modern academia.
posted by gene_machine at 5:34 PM on March 28 [5 favorites has favorites]

I'm really wavering on whether to express my laughter. Ok. Hahaha!

No, seriously, relax. Planning things at this stage is probably overdoing it. I know a few people with all these time-lines for themselves and they keep getting crushed when they're forced to revise them.

Work hard, stay focused. Once your start doing the benchwork and get a better feel for the field, keep up to date with the current literature. Go to seminars and journal clubs and conferences. Talk to people, get new ideas and synthesize new ideas in these venues. Collaborate! Try to make friends with other smart grad students. Hopefully they are studying a similar system, but different subspecialty. Bounce ideas off of each other. It's like an organic backup for your ideas. It's rather common for when I'm stuck on something and I hash it out with my "lab brother" that they'll go, "Hey, weren't you talking about x last month?" where x is the solution to my current problem.

As for keeping ideas organized, that should come on it's own. Keep good records in your lab book. Write *everything* down. I'm trying to figure out an electronic system (haven't yet) that will make searching so much easier. I have to rely on general dates right now, but good notes are important! The first time I try a new technique or the first time a difficult technique works, I'll bookmark that page of my lab book with a small postit or something and label it.

Maybe have another notebook for ideas and for writing stuff down that you glean from said seminars, journal clubs, talks, and conferences. Science changes so quickly, your ideas should evolve just as quickly.

Write grants. Research proposals. Hell, submit them, maybe you'll get funded (or get additional funding). The format of research proposals sets a limitation on your verbosity and will force you to boil your ideas down. Keep copies of these (gmail them to yourself, even) and these can serve as a repository of ideas. They also keep you honest - you'll have to come up with references as well as really consider how realistic your ideas, and the implementation of those ideas, are. Unless you get lucky, most people will churn through a couple/few different projects until they find The One that'll be enough for a thesis. All these proposals will also come in extremely handy for when you do you comprehensive/qualification exam.
posted by porpoise at 5:42 PM on March 28 [5 favorites has favorites]

You'll be doing one hell of a lot of literature review, so I suggest you work out a system from the getgo that keeps your PDFs and printout copies of papers well organized. A good way for me was to rename the file as for quick reference in my files. Keeping the files arranged in folders by subject is good. Keeping a printout copy of each useful paper is also good, in a binder. And when you read a paper, create a word document on it where you can outline the major points/findings for easy use later on.

Since you'll most likely be conducting experiments, I also strongly recommend that you use a logbook and write things out in great detail (including observations, just like undergrad), save every printout in a binder (each page properly labeled), enter the data into your database (excel?) and make the appropriate charts/graphs/calculations as soon as you can. This way you can look at the data and compare it to other results quickly and know what's going on as the experiments progress. So you'll know when things aren't going well, or when they're going incredibly well and you should further explore a certain aspect of the research instead of another.

This all helps you write your reports and keep on the good side of your supervisor (so you can show him evidence you're actually working).

In all of this though, learn to be flexible and not get too frustrated - your project will likely change several times during the first year or two of your research. But by working in the above manner you can churn out a few publications as you go, which makes for super happy supervisors.
posted by lizbunny at 5:42 PM on March 28

Well, I'm not a student, but I really like the Study Hacks website anyhow. At the very least, it's worth considering his strategies...
posted by lemonade at 6:05 PM on March 28

Take a look at Getting What You Came For: The Smart Student's Guide to Earning an M.A. or a Ph.D., by Robert Peters.
posted by ads at 6:11 PM on March 28

I recommend you take a look at grad skool rulz, they are a series of posts on a blog run by some sociology and business professors, but I think they are quite applicable to different graduate programs.
posted by bove at 6:38 PM on March 28

I'll also second or third the suggestion to setup a system for cataloging related work early on. Keep copies of the papers organized somewhere. And most importanly, write up a bit about each paper as soon as you read it. Writing a lit reviews for your dissertation or for a paper will be a piece of cake if you have little blurbs ready to go on every paper you've read so far.

Personally, I use Jabref, a bibtex database management program. (Bibtex is a bibliography database system that integrates well with Latex - Latex is absolutely worth learning for writing papers if you don't know it already, and if your supervisor is cool with it.)

Any time I read a paper, I create an entry in my bibtex database for easy citation later; I save a PDF and link it to the entry in Jabref; and I write up a short review, a paragraph or two to remember the main contributions of the paper and (this is critical) how it relates to my own work, stored in the bibtex file as well, accessible from within Jabref. Jabref has a system for tagging papers, so I can easily get a view of all papers related to a particular subtopic of my research.

For organizing ideas, I have a personal wiki. I keep it simple, with one page for untried ideas saved for later, one page for meeting notes, and separate pages or groups of pages for each research direction I have. Wikis are flexible; set one up and try different ways of saving and organizing information to find out what works for you.

Also: PhDadvice.com was started by someone here on Metafilter (I think - long time ago - and it's now abandoned and full of comment spam). It has lots of posts on a wide range of topics, including some of those you're asking about. It's definitely worth browsing through to pick up an idea or two here and there.
posted by whatnotever at 7:15 PM on March 28

I once used Jabref but have since switched to Zotero and am much happier. I once kept notes with pen and paper and am now using a tool I found called KeepNote and I am happier still. I keep all my files on both my workstation at school and on a remote server I can access from home, and use a tool called Unison to keep them synchronized. Having this setup means I can work without fumbling around looking for things and that makes a big difference.
posted by PercussivePaul at 7:26 PM on March 28

ZOTERO ZOTERO ZOTERO. There, I said it three times. PercussivePaul's recommendation above makes four. You *have* to at least try it now. I've been using it for two months and have already saved an impressive acreage of old-growth forest by keeping it all digital.
posted by The White Hat at 7:51 PM on March 28

I used Writing Your Dissertation in Fifteen Minutes a Day to complete my master's thesis. It was the only thing that got me through my procrastination.
posted by orsonet at 8:51 PM on March 28

Ah, yes, lizbunny spake words of wisdom.

Do categorize your PDFs.

If you're oldschool (or just old like me) or like to have dead-tree copies of papers, get manila folders or a filing cabinet (or a tupperware cabinet if you can't scrounge a proper filing cabinet) and filing cabinet dividers to sort papers by subject. When you have down weeks (hey, sometimes they happen) spend them reorganizing your papers. You *look* busy and you run into stuff that you'd forgotten that you knew that... might just be relevant to something you're working on now. Unless you have eidetic memory, old papers like a half-forgotten scent can bring back a lot of memories.

For PDFs, I keep them organized in folders by topic. Some papers are duplicated across different folders. Of those, most have different text highlighted and different notes annotated (it's worth it to acquire the pro version of Acrobat so you can mark up and save those markups in PDFs).

I normally rename the default no-inherent-information filenames to:

Firstauthor-LastAuthor-'other important personage also on paper' (year) - keywords

I've been inconsistent with adding the journal name to the filename. It's bitten me a couple of times but not harshly.
posted by porpoise at 11:56 PM on March 28

I want to be sure that when I arrive I don't waste too much time.

Disable your metafilter account. Read all those pdf's you will have been filing away instead.

Ask your advisor for a clear, attainable goal. "Yo Prof, straight up, what do I need to do?" Then break it down from there. "Ok, what do I need to do this year, to make it happen?" Then break that down. "Ok, what do I need to this month to be on track?" Then break it down to "This week" and "This day" and finally, "Right now, what do I need to be doing?"

And then, here's the key: Write that shit down. In whatever organizational program you download from the links above. A scrap of blank paper would also suffice. It's not enough to have a goal "in mind".

Other tips/advice:

Don't discuss religion, politics, or relationships, with anyone you haven't seen naked. Don't go to any seminars, unless you're required to, or they're extremely (and I mean extremely) related to your research. Don't go more than two weeks without emailing or talking to your advisor. I know he's not technically your advisor yet, but you can still touch base with him about what progress you've been making towards that ultimate goal.

If you manage to maintain one iota of the determination you have demonstrated by asking this question, I think you'll be alright.
posted by metastability at 7:18 AM on March 29

"Yo Prof, straight up, what do I need to do?" Then break it down from there. "Ok, what do I need to do this year, to make it happen?" Then break that down. "Ok, what do I need to this month to be on track?" Then break it down to "This week" and "This day" and finally, "Right now, what do I need to be doing?"

Use this approach with caution. Some advisors are fine with this. Some will think there's something wrong with you for not being able to figure out for yourself what you need to do, at least on a day-to-day basis.

Im not saying thats how it should be. Some just think that way. Obviously, you should ask for direction when you need it, but showing initiative and foresight without needing directions all the time reflects well on you too. Anyway, I know you think you know who you want as an advisor, but I'd ask around and ask their students how they are doing. Probably the most valuable advice I heard as a new grad student, which I did not heed and I should have, is to pick who to work with based on their personality, not their research. Pick someone who seems easy to work with and whose students like them. Even if you are interested in topic X and professor Y does it, how do you know that you couldn't still do something related to topic X with nice, laid-back professor Z? You may be able to, and you may be wise to give it a fair shot. I agree with everyone else that serendipity is your friend, don't get too attached to one direction because if it doesn't work out you'll be miserable. The key here is to plan carefully to avoid total burnout/depressed misery/self-loathing, etc. Good luck.
posted by lblair at 11:21 AM on March 29