Now that applications are done, what should we be reading or studying up on to prepare for graduate school in the event that we get in?
What should one study to prepare for grad school?
(40 posts) (25 voices)-
Posted 3 years ago #
-
In general, you should be enjoying your last few moments of sanity. I would recommend reading short stories, poetry, or popular fiction -- stuff that keeps your eyes moving but that doesn't require you to make big commitments.
If you're going to insist on being Captain Spectacular, you should start working on the important skill of reading hard AND quickly, since the first few years of graduate school involve learning literatures in a hurry. Pick articles from top journals (APSR, AJPS, JOP, IO, etc.) and try and figure out the main puzzle, argument, theory, and evidence within a certain timeframe. If you plan on being a qualitative researcher, then force yourself to read quantitative or formal papers, even if you don't get the technique just yet. If you plan on going quant, read qualitative stuff to see how fast you can get what you need out of it despite thick description.
If you can master quick, thorough reading on the microlevel, then next you'll want to start thinking more macro -- the development and construction of literatures. Go to JSTOR or some other database and read articles that cited the ones you read. Which directions did they take the original article? How are they interfacing with it? Try and see the conversations that emerge.
A lot of people will tell you to start working on technique, and there's a lot of good in doing that. Many recommend basic probability, multivariate calculus, or real analysis, and brushing up on those skills is always a good idea, especially if you plan on going techy. Some recommend that you download R (it's free after all) and start tinkering. These can all help. But I really do think learning to read political science quickly, thoroughly, and critically will serve you just as well, if not better, moving forward. After all, there are methods classes.
Posted 3 years ago # -
learn python.
seriously.
Posted 3 years ago # -
I used to put together short lists of books (6-10) for former/current students of mine who were about to head to grad school. Since these were tailored to each student's interests, I won't try to produce one for you here.
But those lists were "idea" books that were accessible, but would do what Bebop recommends (get used to reading quickly) but also be more "fun" than the kind of thing you're likely to read in grad school. So, for example, reading "Clash of Civilizations" or "The Future of Freedom" would be on my list. But certainly not "Das Kapital" or anything by Habermas.
I would also read some authors who are "political" that you maybe haven't: Camus or Orwell, for example. Maybe you want to delve into a historical period you have't before. If you're going to be an Americanist, maybe read up some bestsellers that talk about periods of history or people that interest you. If you're thinking of becoming a comparativist, maybe read some bestsellers on places in Africa, East Europe, or Latin America. If you are interested in methods, try "Freakonomics" or some of the popular works of Malcolm Gladwell.
Read to stretch your mind. But also get some ideas for things you might think about or discuss in grad school, though not in a systematic or rigorous way (that's what grad school is for).
Posted 3 years ago # -
I spent a little time over the summer reading some of the work of the faculty in the relevant field in the department I was about to join. I actually didn't know all that much about them or their work, so it seemed as useful as anything.
If your department has field reading lists available, it is never too soon to start making a dent in the reading. Or check out some review essays across some relevant fields and get a feel for range of literatures and debates (to supplement reading in the original research itself).
Might also check out some sites like
http://www-personal.umich.edu/~danhorn/graduate.htmlAnd I'd recommend, http://www.phdcomics.com/ and Richard Russo's Straight Man or Kingsley Amis's Lucky Jim (focusing on a graduate student)
Posted 3 years ago # -
I am not a devotee of this board, but I check in every once in a while especially to see what kinds of questions students are asking.
You have gotten some very good advice in this case (which rare on this site). I encourage you to listen to what the above folks are recommending.
Posted 3 years ago # -
Great. Thank you all for the advice.
Posted 3 years ago # -
The above advice not to kill yourself with reading is good. My suggestion is to read things you want to read, because you won't have the chance to do that again for a long, long time.
Bebop's advice about using the PATE (puzzle, argument, theory, evidence) method to read journal articles is excellent. You should be able to summarize every journal article you read in 2-3 sentences (which you should write down in a centralized location and organize by subject so you can refer back to them when it's time for your preliminary exams). I have my senior seminar undergrads do this each week since most are headed to grad school. Learning how to identify these key points quickly will save you incredible amounts of time once you get the hang of it.
Posted 3 years ago # -
I would favor the idea of doing the fun-but-relevant readings. Learning to cram articles is necessary but you'll get that first semester, and at least you'll be with classmates in the same boat and you'll have the motivation of wanting to look good in seminars. If you want to work harder, I'd suggest taking some math classes either at your BA institution if you're still there, or at a local community college. You will need these unless both you and the department you go to are fanatically anti-quant, or you already have a math or mathematical econ major.
Posted 3 years ago # -
Cracked magazine.
Posted 3 years ago # -
Taking the spring and/or summer to take math, statistics, programming or language courses is probably a good idea. Basic skills training takes up a tremendous amount of time in early grad school, it can be very difficult, and it can be hard to have too much of it. It has probably been awhile since you've done some actual math, and brushing up on it will make life easier if you are going to a real quant or formal program. It is true that you could probably use some calm before the storm of grad school, but it is also possible to make the storm easier to ride out by preparing for it.
Posted 3 years ago # -
I would caution against going too crazy on a math-prep for grad school. From my experience, picking up stats skills (which are mostly handled by computer software) is easier than picking up research design skills. As I tell my methods students: "The computer provides the math, but YOU must provide the thinking."
Rather than taking a calc course in the summer, I'd recommend reading something "popular" like "Freakonomics." Get used to the idea of using stats to answer questions. But focus on the kinds of questions asked and HOW they're answered.
Of course, I'm not an Americanist (so my methods courses lean more 50/50 towards qualitative methods). So if you want to be the next Gary King, by all means do take some upper level calc courses.
Posted 3 years ago # -
This depends dramatically on the program (which is why I qualified my post with "real quant or formal program"). There are departments that tell you which button to push on STATA, and there are departments that expect you to do a lot of math.
Posted 3 years ago # -
Don't forget to take some philosophy courses (either logic or preferably, philo of science). Many debates within polisci are traceable to different ontological and epistemological positions that one have about the nature of social science and scientific inquiry. Most PhD programs won't offer these courses on their own, perhaps you'll get one or at most two weeks intro to philo of science in your scopes and methods course, but perhaps also none at all. I think it's better to take it when you're still in college so that by the time you enter your chosen program, you'll understand the ontological and epistemological issues surrounding the social sciences.
Posted 3 years ago # -
I agree with bianchi77 that math and stats (especially at the basic/mechanical level) are easier to pick up than research design. But that's why I think you should do it first--any community college teacher can get you through multivariate calculus and intro probability. So do that with them before you arrive. Then you'll spend less of your first 2 years in grad school over in the math department and more learning research design or other advanced/applied topics from the faculty in your department (and arguing about it with your classmates).
Also, if you don't ever learn the underlying math and you treat statistics as something that the computer software will do for you, you run a high risk of doing bad research. That would be like hiring a bunch of RAs to do field interviews for you on a topic without understanding how they are going to choose informants or how they are going to phrase their questions. You don't have to do everything by hand, but you have to know what's happening inside the black box.
Posted 3 years ago # -
How about this: Take the summer and read "Designing Social Inquiry" by King, Keohane, and Verba. Also known as "KKV" by everyone. I think it does a good job of trying to bridge the qualitative-quantitative gap. It's not "fun" reading. But it might be a good test to see whether you want to be a modern political scientist. :-)
Posted 3 years ago # -
Go out and have some fun, because you'll be doing plenty of what you are asking about for the next five to six years of your life.
Posted 3 years ago # -
"(which you should write down in a centralized location and organize by subject so you can refer back to them when it's time for your preliminary exams)."
It you are going to do this, it would be a good idea to do this organizing in Bib/LaTex if you are considering doing quant, or in some citation program that allows for notes entry. At some point in your career you will wish you had starting using citation software so you may as well start now.
Oh, and if you don't know what LaTex is, you may want to find out what it is and if your faculty use it. If they don't, it may not be worth your time to learn. But if they do, you will become more valuable as a potential co-author if you have already developed some of the skills.
Posted 3 years ago # -
I would suggest reading my books on ethics and politics. They're a bit old, but they've yet to be surpassed.
Posted 3 years ago # -
to whoever suggested learning latex - i just did the lyx training...um, awesome!
Posted 3 years ago # -
Here is another great resource for learning latex http://haptonstahl.org/latex/
Posted 3 years ago # -
"Many recommend basic probability, multivariate calculus, or real analysis, and brushing up on those skills is always a good idea, especially if you plan on going techy."
Why is real analysis important?
"Taking the spring and/or summer to take math, statistics, programming or language courses is probably a good idea."
Which programming language is most useful?
Posted 3 years ago # -
Real analysis: this is only important if you are planning to do a lot of formal mathematical proofs, and you see very little of that in political science (most "proofs" in political science are just algebra with godawful notation). Real analysis deals with the absolute fundamentals of mathematics -- how do we actually know that 1 + 1 = 2 (no, I'm not kidding, and no, the answer is not trivial) -- and that level of abstraction is used very, very rarely.
Far better to take an upper-level probability course in a mathematics or statistics department (not a social science department...): that will give you the practical experience of working with calculus that you need to derive distributions. However, this is a year (or two) project, not a summer.
Programming languages: The easiest answer -- because this is what computer science departments have generally standardized on -- is Java. Upside of Java is that it gives you experience with all of the basic control structures of structured languages, it runs everywhere, and it is graphics oriented. Also there are a lot of toolkits (e.g. for agent-based modelling) that are based on Java Downside is that it is pretty much idiot-proofed and you cannot do direct manipulation of memory (pointers). Pointers are a great way to write programs that destroy themselves, but also the only way to write programs that are very fast and efficient, and you may eventually need that if you are dealing with large data sets and/or difficult estimation problems. For that capability, the answer is C/C++ (many computer science programs will require Java as a prereq, however).
The other alternative -- and if you are going to primarily use programming for cleaning data, rather than writing estimation routines, this is preferable -- would be perl or Python. Perl is notoriously weird, but fantastically powerful if your objective is simply rearranging characters -- perl programs tend to be one tenth the size of C/C++ or Java programs that do the same thing. Python -- which I've not programmed, but a number of computer scientists I know recommend highly (and was recommended earlier in this thread) -- is apparently a good compromise between Java and perl, and might well be the best choice.
Programming is ultimately something you only pick up by writing lots and lots of programs, so the amount you will learn in a classroom is limited. One (or two) class to get you started, perhaps, but then after that it is just practice, practice, practice, preferably on problems that you really need to solve, rather than artificial textbook problems that don't really get you very far (and have been around fifty years...solve the Tower of Hanoi problem using recursion...)
Posted 3 years ago # -
What political scientists, outside of those doing agent-based modeling, use programming in their research? Other than cleaning data (which I've found to be very helpful), why else is it used?
Posted 3 years ago # -
how do we actually know that 1 + 1 = 2 (no, I'm not kidding, and no, the answer is not trivial)
Nobody actually knows how to show that correctly, except for Kurt Gödel, who read well past page 350 of Principia Mathematica.
I recommend a careful reading of two poli-sci methods books (perhaps Parsons "How to Map Arguments in Political Science" and Stinchcombe, "Constructing Social Theories" or the more recent "Logic of Social Research"), plus lots of non-work stuff, as grad school will suppress any chance to do that later on.
(e) lazy style edit
Posted 3 years ago # -
I second the motion for statistics. You'll get plenty of this once the program starts, but the pace is quick in the first year. And it really helps if you come in with a solid grasp of the basics (OLS, logit/probit, etc.). Good luck.
Posted 3 years ago # -
Work on writing style and polishing everything you write. No matter how good at stats or theory you may be, you have to be able to articulate ideas well. Worst 1st year mistake is to hand in poorly written work that appears half-assed, even if the ideas are good. Get Strunk's writing style book and other titles that float your boat, go through every page of them before grad school starts, and practice writing every day on a blog, journal, whatever.
And read something about the philosophy of science. It's important, as other posters have mentioned, and difficult to find in many poli sci departments.
Posted 3 years ago # -
I think if you're worried about stats, then by all means brush up. But I think there's plenty of time in grad school to learn & fine tune stats. What there's less time for is formulating interesting questions. That's why I highly recommend reading broadly, particularly in the area(s) you think you're interested in as possible dissertation topic. If you went to a top 50 LAC, this may be less essential. But if you didntl it may be crucial.
In another thread (on jobs) someone made metaphor of pillars & spans. Your dissertation will drill down (pillar) into a narrow area of expertise. And you'll eventually need to learn the specific methodological skills to do that (and that may or may not be stats). But to know where you want to drill your pillar, you'll want to survey the topography. That's where reading broadly comes in. Then in grad school use that knowledge to build a bridge (span) that connects your pillar to others.
Posted 3 years ago # -
Work on writing style and polishing everything you write. No matter how good at stats or theory you may be, you have to be able to articulate ideas well. Worst 1st year mistake is to hand in poorly written work that appears half-assed, even if the ideas are good. Get Strunk's writing style book and other titles that float your boat, go through every page of them before grad school starts, and practice writing every day on a blog, journal, whatever.
And get a chance to read Gary "Kissinger" King's advice on dissertations.
Posted 3 years ago # -
"Real analysis: this is only important if you are planning to do a lot of formal mathematical proofs, and you see very little of that in political science (most "proofs" in political science are just algebra with godawful notation). Real analysis deals with the absolute fundamentals of mathematics -- how do we actually know that 1 + 1 = 2 (no, I'm not kidding, and no, the answer is not trivial) -- and that level of abstraction is used very, very rarely."
If its at that level of abstraction, why is it such a big deal for econ programs then?
Posted 3 years ago #