Methodology as a primary / substantive interest?

[northstar22]

I notice a few (poli. sci.) PhD programs offer primary focus areas in research methodology or related fields, like formal theory and empirical theory. Does anyone here have methodology as a primary focus area or substantive research interest? I'm pursuing my master's degree now, and as strange as it sounds, I've found that I really enjoy statistical / quantitative analysis in a poli. sci. context (yes, I'm a nerd). Does anyone know of some good programs for methods? Would getting into such a program be realistic (I didn't take any calculus courses as an undergrad)?

Edited October 6, 2011 by northstar22

[Penelope Higgins]

Methodology as a field goes far beyond enjoying statistical analysis. If you do methods as a field, you'll be trying to develop new statistical techniques to address the problems with existing ones in studying certain questions. Look at the work of King (Harvard), Imai (Princeton), Mebane (Michigan) or Brady (Berkeley) to get a sense of what this implies. And if you want to study methods as a field, that's not a bad list of departments to look at, along with Stanford, Stanford GSB, Caltech, and Rochester. Without calculus, your chances of getting admitted to focus on methods are going to be slim - you'll need calculus, multivariable calculus, and linear algebra at a bare minimum to even understand the math these folks do.

[mv0027]

I disagree that you need calculus, etc for admission. You could probably just say you want to take those courses during your first two years (assuming they do not list them as required...which I think caltech and GSB do).

I discovered my love of methods during my master's as well

[Penelope Higgins]

I stand corrected - my impression from the few methods files we get (I am not at one of the schools listed above) was that everyone has a lot of math and/or econ under their belt already, but perhaps the broader pattern is different. Certainly if you want to do methods, you should plan to get the math skills as soon as possible.

[RWBG]

I think it would be extremely difficult to signal to an admissions committee that you're going to be a strong applicant for a specialization in political methodology without some background in mathematics and statistics. Mv0027 (and maybe the original poster) appear to be conflating applied methods - i.e. using regression and other statistical methods to deal with other substantive questions - with political methodology work, which as Penelope noted, often involves developing new statistical techniques, demonstrating how particular techniques should be used when dealing with certain kinds of political questions, developing new methods of measurement, etc. This requires more than a working understanding of statistical programming; you need to thoroughly understand the mathematical assumptions being made in the model you're estimating. By all means, indicate an interest in learning statistics and applying it in your research, but don't paint yourself as a pure methods applicant if you haven't taken calculus. If you find you're really excelling at it during your Ph.D, then maybe you can shift your focus after gaining admission. Anecdotally, most of the people I know who specialize in methods had an undergraduate math major.

As an aside, when thinking about whose advice to take seriously on this forum, Penelope's will be amongst the best. She's served on admissions committees, and her advice is usually sounder than the speculation of a lot of others on this board (myself included.)

Edited October 6, 2011 by RWBG

[northstar22]

Thank you so much for the responses! I should clarify that I have taken statistics, algebra, and economics courses at the undergrad level, just not calculus, and I'm also taking (and doing quite well in) two graduate-level classes right now which involves regression analysis, etc.-- one course is pure statistics, and the other is a substantive course that heavily emphasizes statistical analysis. I definitely have the option of taking calculus courses during my master's program (I still have a few semesters left).

Does anyone know of any lower-ranked methods programs? I think Stanford, Princeton, Harvard, etc. are a little (WAY!!!) out of my league, and I'd prefer to stay in the Midwest if possible. I know Nebraska, Purdue, and Missouri have methods focuses -- any insight on those programs?

Edited October 6, 2011 by northstar22

[mv0027]

well, I would certainly listen to some with a PhD that has been on committee's over little ole me. I just wanted to encourage a fellow applicant to pursue their interests. Plus, since they are in a professional master's program, I knew what type of research they were talking about. I've heard the "you have to take all these classes before they will even consider you file" arguments a lot, so I wanted to avoid going down that road.

I'm also from the "write the application that gets you in" school of thought. Which lends itself to taking all those classes after you get in, if you need them for what you are interested in.

[northstar22]

Why do you think it's better to avoid taking such classes before applying? I don't think it would hurt my application to come in with previous relevant coursework.

Edited October 6, 2011 by northstar22

[mv0027]

Oh yea, take them if you can. I just assumed you didn't have the (1) time (2) money or (3) energy to take them while doing a masters. But if you can, go for it!

[RWBG]

Mv0027, sorry if my last post sounded a bit pointed; it wasn't intended to sound as such. My main points are this; (1) Your best bet is not going to be to paint yourself as a pure methods applicant if you are without a background in calculus (or a formal theorist applicant, for that matter.) This will lessen your likelihood of admission. (2) You should be sure you have a good conception of what work in political methodology entails before deciding it's what you want to do, and how it differs from applied methods. If you do, that's great. It's just hard for me to imagine having a clear idea that you want to do methods proofs, etc. without a solid background in the calculus required to understand the work being done in the field. I also think it would also be extremely difficult to identify pure polmeth research questions you want to address without that background, which could create difficulties when writing your SOP. Northstar; you mentioned taking algebra - if that's a linear or complex algebra course that involves proofs, that might give you a better sense of whether pure methods work is for you, and might be a better signal for admissions committees.

Once again, this shouldn't prevent you from pursuing training in econometrics, and looking to apply it in your research, even if your main substantive focus is not political methodology. However, if you do decide polmeth is what you want to focus on, you should also look at University of Georgia; their current strength in the subfield is not yet reflected in broader rankings, and is probably similarly not reflected in admissions competitiveness.

Edited October 7, 2011 by RWBG

[Charlie2010]

I'll echo Penelope's and RWBG's comments, and also say that they both seem to really know what's going on in all their responses.

To be crystal clear (and a bit rude), if you tell any top twenty (thirty?) program that you want to be a methodologist, but have not taken calculus, you will be laughed out of town. Stats courses that do not require calculus are not considered to be particularly rigorous or demanding by the standards of this group. Better than nothing, but that puts you way behind where any methodology specialist should be starting grad school.

That said, I think what you really mean is that you like doing applied work using statistical methods. That's great! Lots of people are doing that, in every subfield, in every department, and many of them start out without the necessary calculus background (although this is rarer and rarer). So really any department would be fine. If what you mean is that you want to be in a program that will let you take more stats and calculus, and will trade off by requiring fewer book-intensive seminars, then you could try one of boutique departments, Rochester, Stanford GSB, or Caltech Social Science, with a second tier being NYU or WUSTL.