Need more math background: what would you suggest?

[wannabee]

Need some help and advice. My quant background is minimal. I am in comparative with an undergrad degree in area studies. My gpa, language preparation, and references are strong, and I am fortunate to be admitted to two good programs. As far as I know,neither program has a "math camp" but I will check on that. My own future work and close advisors will definitely have a qualitative orientation, but I know I need to sharpen my background in terms of math to be able to read and understand the literature in the field. However, I only took basic college statistics and before that AP calculus. (Who knows how much I even remember?) My verbal GRE is much higher (99%) than my quant (66%), and I have not taken any formal econ.

I am thinking that I should make time before grad school to strengthen my skills to make my path a little easier. I am uncertain which way to go and, since I have taken no real poli sci courses outside of those focusing on my chosen region, I don't even have an undergrad faculty member to approach. Once I determine where I am going, I will correspond with the Director of Graduate Studies to get his/her advice. Meanwhile I'd like to start thinking about my options.

Should I take calculus at the local community college? Or a refresher in statistics? Or maybe linear algebra...if I can even qualify for that?

What about MIT's online graduate level course?-- Quantitative Research in Political Science and Public Policy taught by Prof. Stephen Ansolabehere.

http://ocw.mit.edu/courses/political-science/17-872-quantitative-research-in-political-science-and-public-policy-spring-2004/

I believe one poster said he worked his way through Simon's Mathematics for Economists. http://www.amazon.com/Mathematics-Economists-Carl-P-Simon/dp/0393957330

If anyone--prospective or current grad student, or a faculty member--has any ideas on this, I would appreciate some suggestions. I should be home at the end of May for the summer and will have time then.

Edited February 28, 2011 by wannabee

[maicondouglas]

I think it really depends on how you learn best. Some people will have a hard time learning math directly from videos and textbooks. Others will have no problem with that. Whatever you do, remember that with Math you do most of your learning by practice. So it's important that you do a lot of problems and not just follow lectures.

Another online resource is http://www.khanacademy.org/. He has a ton of short lectures on Calculus, Differential Equations and Linear Algebra that are quite helpful. I never really used it much for Calc or DE, but his videos came in really handy for Linear Algebra.

[Penelope Higgins]

Take a stats sequence when you get to grad school. You'll be fine. There are better things to do with the summer before you start. So long as you understand basic statistics (significance, control variables, size of effect) before you start, and so long as you work at it once you start grad school, you'll be able to get up to speed quickly.

And your plan of talking to the grad director is a good one. I would also have a conversation with the faculty you expect to work with to pick their brains about this issue.

Need some help and advice. My quant background is minimal. I am in comparative with an undergrad degree in area studies. My gpa, language preparation, and references are strong, and I am fortunate to be admitted to two good programs. As far as I know,neither program has a "math camp" but I will check on that. My own future work and close advisors will definitely have a qualitative orientation, but I know I need to sharpen my background in terms of math to be able to read and understand the literature in the field. However, I only took basic college statistics and before that AP calculus. (Who knows how much I even remember?) My verbal GRE is much higher (99%) than my quant (66%), and I have not taken any formal econ.

I am thinking that I should make time before grad school to strengthen my skills to make my path a little easier. I am uncertain which way to go and, since I have taken no real poli sci courses outside of those focusing on my chosen region, I don't even have an undergrad faculty member to approach. Once I determine where I am going, I will correspond with the Director of Graduate Studies to get his/her advice. Meanwhile I'd like to start thinking about my options.

Should I take calculus at the local community college? Or a refresher in statistics? Or maybe linear algebra...if I can even qualify for that?

What about MIT's online graduate level course?-- Quantitative Research in Political Science and Public Policy taught by Prof. Stephen Ansolabehere.

http://ocw.mit.edu/courses/political-science/17-872-quantitative-research-in-political-science-and-public-policy-spring-2004/

I believe one poster said he worked his way through Simon's Mathematics for Economists. http://www.amazon.com/Mathematics-Economists-Carl-P-Simon/dp/0393957330

If anyone--prospective or current grad student, or a faculty member--has any ideas on this, I would appreciate some suggestions. I should be home at the end of May for the summer and will have time then.

[wannabee]

Many thanks for the suggestions and advice, especially Realist and Penelope. I have been pushing for so long to get into a program that it is hard to put "pushing" on hold! But I can definitely think of more pleasant things to do with my summer.

[UFGator]

Free online real analysis course- http://beta.learnstream.org/course/6/

[Charlie2010]

This is the correct advice. I frankly don't even think that you need to understand basic statistics before starting graduate school; if your school has a research design or first-semester statistics course you will learn everything that you need.

Spend your summer doing something interesting or relaxing. Travel. Be a lifeguard. Learn how to cook. Once you hit grad school that stuff is over.

I disagree with both the Realist and Penelope Higgins on this. Most of the top ten programs do have a "math camp" or even a math semester because to learn methods at the level required to intelligently read (read, mind you, not produce) a large part of the research on almost any substantive area, you need more math than most political science undergrads take. You can focus on qualitative work and not learn the methods if you don't want to really understand those branches of research, just as someone could rely entirely on English-language sources to study EU politics, but I don't think this will result in the best research.

[The Lorax]

Speaking as a current master's student about to go into a PhD program...I agree with Charlie2010. If you feel your quantitative skills are weak, then spend some time brushing up BEFORE grad school. I know mine are, and I wish I had. If you're entering a decent program, the competition will be hot, and you will busy. So Penolope Higgins and The Realist are right - don't kill yourself or anything, but maybe take your algebra/stats book with you to the coffee shop, or on the train as you backpack through Europe, or whatever it is you decide to do with your summer.

[firefly28]

I think it really depends on how you learn best. Some people will have a hard time learning math directly from videos and textbooks. Others will have no problem with that. Whatever you do, remember that with Math you do most of your learning by practice. So it's important that you do a lot of problems and not just follow lectures.

Another online resource is http://www.khanacademy.org/. He has a ton of short lectures on Calculus, Differential Equations and Linear Algebra that are quite helpful. I never really used it much for Calc or DE, but his videos came in really handy for Linear Algebra.

I enthusiastically second the Khan Academy recommendation. I never took calculus in school, but I now can do what would be expected after a year of taking calc, thanks to him.

Two more very impressive math resources:

http://www.patrickjmt.com

http://www.integralcalc.com

Both of these are very good, and I would use them in conjunction with one another: I've found that Khan is very good at explaining some things, but on a few things, I get more out of PatrickJMT. I liked Integralcalc's limits videos, and I just covered cross product through the videos.

In terms of material to learn:

* Statistics. Try to learn everything through ANOVA. Generally schools will go over this in the first methods course, but you want to make sure that this isn't new to you, since for many students it will and thus the professor may move at a speed that would be uncomfortable for you in that instance. Make sure you also learn fundamental probability concepts.

* Calculus. Learn derivatives and integrals. This leads to...

* Differential Equations. Very useful for modeling. Cover this AFTER you have the fundamentals of calculus down.

* Linear Algebra. This can be very important, as some of the more sophisticated quantitative techniques make use of linear algebra. Learn this after your basic stats and calculus are good, because some of the linear algebra topics coincide with multivariate calc topics.

If you get a chance, cover predicate logic too. If you take a class on formal methodology, this will be important. As will game theory. I recommend this website:

http://www.gametheory101.com/

Some programs require a game theory course, so having a head start will certainly help you

Edited March 1, 2011 by firefly28

[Penelope Higgins]

I 100% agree that the skills are needed. I just know that my current department which has no math camps does a lot to train students to read quant work in the first semester and year of grad school. We do more in scope and methods, research design, and quant 1 courses than my grad department, which had a math camp, did. That was the reason for my suggestion that if there was no math camp, the department likely is aware that students are not prepared in this way, and does something to prepare them. So I think we are all on the same page as far as the set of skills you need; I'm just suggesting that you check in with the department about how to acquire them before you dive in to summer preparation that may or may not be useful.

I disagree with both the Realist and Penelope Higgins on this. Most of the top ten programs do have a "math camp" or even a math semester because to learn methods at the level required to intelligently read (read, mind you, not produce) a large part of the research on almost any substantive area, you need more math than most political science undergrads take. You can focus on qualitative work and not learn the methods if you don't want to really understand those branches of research, just as someone could rely entirely on English-language sources to study EU politics, but I don't think this will result in the best research.

[Charlie2010]

At this point I wouldn't worry about trying to do stats, methods, or game theory. Just focus on improving your math basics--single and multivariate calculus, probability, linear algebra. But it's true that most incoming students even at top programs won't have much more than a couple of college calculus classes from freshman year, and departments will compensate as best they can.