Math requirements in American politics

[Kendra Smith]

I graduated from a top liberal arts college with a double-major in economics and political science.  Upon graduation, I went to work for an economic consulting firm, deciding that I wanted to move away from the private sector and submitted applications to study economics in graduate school.  I achieved admission to a top program on the strength of my recommendations and grades.  However, upon arriving at school, I was shocked by the level of mathematical training and ability that is required in graduate level economics (the burden of which grows heavier every year, I understand), and that despite my excellent performance in the undergraduate-level courses and on the GRE (I graduated in the top 2% of my class, and scored 800 on the math section and 700 on the verbal), there was no way that I was going to get through what actually felt like a Ph.D. in physics or mathematics, especially without an undergraduate major in math and perhaps a couple of graduate courses as well (and frankly, I have to admit that I am not exceptional at writing mathematical proofs nor do I enjoy it particularly much, a disposition that I discovered the hard way is fundamentally incompatible with graduate school in economics).  Confident that I was going to fail out, I voluntarily withdrew during the first semester.

 

The reasons, then, that I am considering studying political science instead are as follows: 1. It was my other undergraduate major, precisely because it was one of the two things that I found most interesting, 2. In contrast to what I saw in grad-level economics, I hope that there would be more of an opportunity to focus on substantive issues in political science, and 3. While I understand it is a difficult and grueling path, I would like to have an academic career, with a think tank/DC career being an acceptable fallback if academia did not work out.  I certainly understand that I would need to take some methods courses, but I expressly am not interested in, say, developing new statistical methods. Overall I wish to combine qualitative work with a regression or a little time series or probit here or there--if all of American politics is actually about who can do the fanciest math or the trendiest new Bayesian analysis or the most innovative formal models, then I think that this would be the wrong discipline for me.

 

My main concern, then, is that with a substantive interest in American politics but not theory, the mathematical demands of today's political science Ph.D. programs would also be excessively onerous.  At econ grad school, for example, there were poli sci Ph.D. students (who I believe were Americanists) in my Ph.D. microeconomic theory course with me.  I am not totally incapable of math--I got As in Calculus I and II and Linear Algebra, and I am willing and able to take courses in applied statistics--I got As in each of intermediate and upper-level econometrics in undergrad.  However, if ever again I had to take anything approaching the econ Ph.D. courses, it would just not be workable.  In seeking advice out about applying to economics programs, Ph.D. after Ph.D. I talked to vastly underestimated what I would be up against and even seemed sanguine about the fact that I hadn't taken calculus-based statistics or real analysis (the latter of which I did eventually do, albeit a not so rigorous summer course at a local public university), so I am almost shell-shocked and terrified of another seeming bait-and-switch--I don't want to head off somewhere intending to study politics only to find myself struggling to prove that the eigenvalue decomposition of something is some other unintelligible thing.

 

I am wondering, then, if I am right even to be considering applying for such programs, or whether, even deep into the poli sci rankings, American politics is just mathematical proofs and hardcore econometric theory all the way.  Then, assuming that I should apply, what should I say to convince graduate schools that I am not going to abandon them or fail required methods courses, given my previously abandoned Ph.D.?

[catchermiscount]

It varies, and it's the rare Americanist that has to have the all-around technical training that an economist does.  You may need to know one of the tools in the kit really well, but it's unlikely that it will be the core of your training the way it is in economics. 

 

Some of it varies on what it is that you want to study within American politics.  If you're interested in behavior (voting, public opinion, political psychology, etc.), then you probably won't need very much theory in your life.  But people that study opinion have to be well trained in the tools for analyzing survey data.  People that study political psychology generally know something about psychometrics (much of which leverages your beloved singular value decomposition).  And so on.  Generally, folks that apply statistical tools to substantive problems don't have to prove anything about their models, because they use off-the-shelf tools readily accessible in Stata (which is still probably the most used package for applied folks) or R with well-known properties.

 

Now, if you're interested in institutions, you might need to pick up a bit more theory.  Obviously there are plenty of empirical folks in the study of just about any institution (though straight-up empirical bureaucracy folks are somewhat rarer than others), but so too are there theory folks.  Consider the study of the American congress:  there are all kinds of empirical people (those ideal points are coming from somewhere and being used for something), and there are all kinds of theory people (everybody Riker and Fenno trained).  All else equal, folks that apply theory probably have to act a little more rigorously than folks that apply stats. 

 

Of course, there's always the qualitative route.  It seems like saying that you study "American Political Development" is insider jargon for "I'm a qualitative, history-based Americanist."  There are plenty of those folks.  They use remarkably few proofs ;-).

 

As for application tactics...well, let's jump off that bridge when we get to it.

[Kendra Smith]

Thanks for your reply--to clear up something that I was ambiguous about, when I said that I was not interested in theory, I meant normative political theory/philosophy (although it is clear from my post that I am not a fan of positive political theory either).

Edited July 31, 2013 by jayalthouse

[polisci12345]

American isn't as math intensive as as econ, but still requires some amount of math. I don't think more than one or two people had taken any sort of analysis and I would not imagine most would be comfortable with mathematical proofs. 

 

While there is some APD work going on in my department, I think most of the Americanists who have taken quals took either methods or formal theory as their secondary field and almost all have gone outside of the department in search of additional methods training (mostly in the econ, stats and CS departments).

 

American politics is generally fairly quant heavy. We know a lot of the fundamentals of how the US government functions so there are a lot of people who spend their time trying to refine that knowledge by building better mousetraps 

[Kendra Smith]

polisci12345--thanks for your input.  I am a little confused though.  If these Americanists had not taken analysis and were not comfortable with mathematical proofs, how would they do formal theory or take courses in the economics department?

[Penelope Higgins]

By the standards of where you're coming from, even most quantitative and formal work in American politics won't compare. There are departments (like the one coach is attending) that are exceptions, but in general most grad students in American aren't developing new methods or models. Look at the American politics syllabi and at the work of faculty and you'll get a sense of this.

This statement, to be honest, isn't based on direct experience: I am not an Americanist. But I have served on hiring committees for American politics and in economics, and the difference in methods expectations is dramatic.

[polisci12345]

kendra: no one is taking the full sequences, just the particular course or two that they need. by the time they get to needing that particular method, they have the requisite knowledge to take the class. If there are a few topics that are too far...grad school grades don't matter anyways

[bees]

 

I graduated from a top liberal arts college with a double-major in economics and political science.  Upon graduation, I went to work for an economic consulting firm, deciding that I wanted to move away from the private sector and submitted applications to study economics in graduate school.  I achieved admission to a top program on the strength of my recommendations and grades.  However, upon arriving at school, I was shocked by the level of mathematical training and ability that is required in graduate level economics (the burden of which grows heavier every year, I understand), and that despite my excellent performance in the undergraduate-level courses and on the GRE (I graduated in the top 2% of my class, and scored 800 on the math section and 700 on the verbal), there was no way that I was going to get through what actually felt like a Ph.D. in physics or mathematics, especially without an undergraduate major in math and perhaps a couple of graduate courses as well (and frankly, I have to admit that I am not exceptional at writing mathematical proofs nor do I enjoy it particularly much, a disposition that I discovered the hard way is fundamentally incompatible with graduate school in economics).  Confident that I was going to fail out, I voluntarily withdrew during the first semester.

 

The reasons, then, that I am considering studying political science instead are as follows: 1. It was my other undergraduate major, precisely because it was one of the two things that I found most interesting, 2. In contrast to what I saw in grad-level economics, I hope that there would be more of an opportunity to focus on substantive issues in political science, and 3. While I understand it is a difficult and grueling path, I would like to have an academic career, with a think tank/DC career being an acceptable fallback if academia did not work out.  I certainly understand that I would need to take some methods courses, but I expressly am not interested in, say, developing new statistical methods. Overall I wish to combine qualitative work with a regression or a little time series or probit here or there--if all of American politics is actually about who can do the fanciest math or the trendiest new Bayesian analysis or the most innovative formal models, then I think that this would be the wrong discipline for me.

 

My main concern, then, is that with a substantive interest in American politics but not theory, the mathematical demands of today's political science Ph.D. programs would also be excessively onerous.  At econ grad school, for example, there were poli sci Ph.D. students (who I believe were Americanists) in my Ph.D. microeconomic theory course with me.  I am not totally incapable of math--I got As in Calculus I and II and Linear Algebra, and I am willing and able to take courses in applied statistics--I got As in each of intermediate and upper-level econometrics in undergrad.  However, if ever again I had to take anything approaching the econ Ph.D. courses, it would just not be workable.  In seeking advice out about applying to economics programs, Ph.D. after Ph.D. I talked to vastly underestimated what I would be up against and even seemed sanguine about the fact that I hadn't taken calculus-based statistics or real analysis (the latter of which I did eventually do, albeit a not so rigorous summer course at a local public university), so I am almost shell-shocked and terrified of another seeming bait-and-switch--I don't want to head off somewhere intending to study politics only to find myself struggling to prove that the eigenvalue decomposition of something is some other unintelligible thing.

 

I am wondering, then, if I am right even to be considering applying for such programs, or whether, even deep into the poli sci rankings, American politics is just mathematical proofs and hardcore econometric theory all the way.  Then, assuming that I should apply, what should I say to convince graduate schools that I am not going to abandon them or fail required methods courses, given my previously abandoned Ph.D.?

 

 

Kendra, if I may, I think you may be over-thinking this. You're clearly very smart and would be a very competitive applicant. Which Americanist political scientist do you most admire? Which sub-field? Go to the Annual Review of Political Science, pick out a few of the articles that sound most interesting to you in terms of your substantive interests and see if you could see yourself doing that kind of work in the future. My guess is that you'll find stuff that suits both your interests and your abilities.