Stats/Math/Econ Classes

[canadianundergrad]

I'm looking to get a solid background in quant while I'm undergoing my undergrad education, but am finding it tough to balance this with my interests in political science. Just how many Stats/Math classes should I take at the undergrad level if I want to have a good shot at some of the top U.S. PhD programs? I for sure want to get Linear Algebra and Multivariable Calc under my belt, so that's a good 6 semesters of pure math, including prereqs. I figured 2 semesters of purely statistical stats would be enough, or 2 semesters of quantitave economics instead. However, would taking a 200 level intro to empirical methods in Poli Sci and then a 400 level quantitave methods/data analysis course (again, Poli Sci), totalling 4 semesters, be an adequate replacement? Or would purely statistical courses look better to an admissions committee down the road? As for Econ classes, would a couple of semesters of Micro or Macro theory suffice (aside from intro)? I would like something in Game Theory, but the prerequisites for a 300 level course on it would crowd out too many important classes.

I know this stuff is extremely subjective and that nobody will be able to make these decisions for me, but if anyone could delineate even the vaguest concept of a "solid" quantitative background (from the eyes of an ad. com.) I'd be extremely thrilled. Thanks!

[polisciphd]

Your biggest task should be deciding what courses will help you do the kind of research you eventually want to do. If you want to do game theoretic modeling, then of course you want to do as much pure stats as you can get. However, if you are more of a political economy kind of person, then by extension the econ classes would better serve you. There is no magic number of classes. I can tell you that I am going to do some pretty heavy quant stuff and that having taken both an undergrad level and grad level stats class helped (according to my ad com members I met at the open house), as well as having taken up through cal 4 (differential equations).

However, I should point out that I applied to seven schools and got accepted to one, which shows that this whole admission thing is as much of a crap shoot as going to vegas. In my opinion, it is all about spin. Show the adcom that you have done x and y to prepare you to do z kind of research as a grad student, and you should be as prepared as anyone. Then it will be all about matching your interests to a program and a facutly person (who will hopefully be serving on the adcom committee the year you happen to apply).

Scoring great on the GRE quant section will be a big help too, try getting in the 90th percentile and above...

[canadianundergrad]

Thanks, that was quite helpful. My main problem is I don't really know yet exactly what captures my interest inside the field of comparative. I'm only finishing up my first semester, so I guess that's forgivable, but the prereqs of Math classes kind of force me to plan ahead. I'm also giving as much thought to grad school as possible, because I definitely dicked around in highschool and didn't quite reach my potential. Wasn't a great feeling. Regardless, I'll have to realize the joys of specialization sooner or later. I figured I could try to get an all-around background in undergrad, but between my not doing any math this first year, my learning a third language, and my having to take some Poli Sci prereqs I'd rather not, I don't have all that much room for the Stats, Math, and Econ classes AND for taking enough poli sci to really discover my passion. I'm currently leaning towards ditching the Stats, as at least the quantitave methods Poli Sci classes sort of make up for them, whereas the (upper level) Math and Econ would simply be missing from my education. Is this a false assumption? The 400 level methods class is described as such : The use of advanced data analysis techniques in the social sciences, management and analysis of large datasets, techniques of multivariate analysis, problems of causal inference and interpretation of data. Alternatively, I could drop the third language, but I've heard languages never hurt in the app process, especially for a potential comparativist.

One final question: in what fields of poli sci is game theory usually applied?

P.S. I may seem like I'm a terrible pragmatist not making the most of the wonderful opportunities available to me to expand my mind, but I really do enjoy learning anything and everything.

[eve2008]

I'm not the most qualified person to offer advice on this topic, but I have spent a lot of time mulling over poli sci, stats, math and econ department course offerings and think about the classes I *wish* I had taken as an undergraduate. I would say, as an undergrad, you should take the math and econ you will need as prereqs for the upper-level classes in grad school, but don't worry about going crazy on upper-level classes since you will have time to get the tools you need in grad school. Now you should just worry about getting a foundation in basic math and econ, and exploring your substantive interests; I think even more important than your toolbox is your ability to frame an interesting question.

Math (2 semesters): I would take math through multivariable/linear algebra if you can stomach it (I think that's usually 2-4 semesters of math, depending on your preparation). If they have a track for economists or social scientists at your school, take that so you won't be competing with the engineers and science majors. A lot of the math you will learn in those classes you will never need, so a more tailored sequence for social scientists will probably make better use of your time.

Econ (3 semesters): I would take the intro micro/macro sequence and the intermediate micro theory class. Further classes are really at your discretion, and as far as econometrics, often times the political science department's advanced quantitative methods class is a rough equivalent. I'd also take one or two topical political economy classes (i.e., the Political Economy of X) so that you can get a sense of how economists approach political questions. I took one such seminar and found it tremendously helpful. Can you minor in political economy at your school or do a joint major between the two departments? I wouldn't worry about taking game theory now, as you will have plenty of opportunity to take both game theory and formal modeling classes in grad school, and you might find the ones offered by the political science department more relevant.

Stats (1 semester): I wouldn't necessarily take anything beyond an intro stats class, since after the first one or two semesters, classes can get highly specialized and you don't know the tools you want to get until you have a clear idea of what your dissertation would be. Really, you can take stats as a grad student, so I wouldn't take more than one or two semesters. Also, you should check the syllabi for the intro stats class and the poli sci methods sequence, as you may find a lot of overlap.

Poli Sci Methods (2 semesters): What I *would* do, if your department offers it, is take the full undergraduate research methods sequence. Just as important as getting the prereqs out of the way for grad classes is getting training in basic quantitative methods in time for you to use them in your thesis. Your thesis is ultimately one of the most important parts of your graduate school application (as is your relationship with your thesis adviser).

The question is probably not "where can I use these tools?" but what "questions interest me?" Ask the question first, then figure out what tools you need.

Game theory has tons of applications in political science, especially if you're interested in IR, institutions, collective action, legislative politics, civil war...

Also just Googled this page at Columbia. Might give you some ideas: http://www.columbia.edu/~de11/gamethry.html

This looks like a decent intro to game theory. You may find others in your school library:

http://www.amazon.com/Theory-Political- ... 0691034303

Here are some books that have been recommended to me. You might check them out from your school library and figure out what you're comfortable with already, given whatever calc you had in high school, and what might make the most sense to you in terms of math offerings:

http://www.amazon.com/Fundamental-Metho ... 610&sr=1-1

http://www.amazon.com/Essential-Mathema ... 630&sr=1-1

I'd also email the undergrad reps in the Poli Sci, Econ, Math and Statistics departments, and explain your goals. They would be the best resources for figuring out what classes you need to take to get the background you want!

Good luck. And good for you for starting so early.

[ShamPain]

if you want to impress ad-coms, then take the math, esp. the stats (at least get past multivariate regressions) and then major in Econ. ad-coms seem love that these days. or, if that's too much work, then pick up a minor. and try to tailor it to whatever your future research interests are. if it's IR, then take a bunch of trade/factor movement classes. if american, then take a bunch of macro theory stuff and labor or environmental econ classes. if it's behavior, then take a bunch of micro theory stuff. if its theory, take a bunch of econ history/philosophy courses.

get what i'm saying? the more focussed you can be, and the more you can tailor your classes accordingly, the more you can demonstrate to an ad-com that you know what you're doing and you'll be a good scholar. there isn't one or two classes that you can take that ensures you get in to a top school, but if you establish a pattern of taking and mastering tough courses in focussed direction, then that can impress.