METHODOLOGICAL BALANCE

[Salve]

HI all!

I'm new to the forum, so I hope you will forgive me if this question has been asked already. However, these days changes in the discipline are so fast, that perhaps information like this should be updated once in a while. 

Anyway, getting down to business. I was wondering how the top PhD programs in the US compare when it comes to their methodological allegiances. For example, I prefer qualitative methods, though I'd like to incorporate quantitative methods into my work. So when I'm going to apply for schools I want to make sure that I will not end up having to take 6 quantitative courses and only one qualitative course just because no more was offered. So if we can gather opinions on the major phd programs with respect to their methodologies, it will be very helpful for many people like me, who are wary of the increasing quantification of political science. I really am not trying to throw stones at quant people, but I feel sometimes unjustly pressured into doing much quantitative work at the expense of my own interests and epistemological choices.

So can we gather opinions on schools like: Harvard, Columbia, Cornell, Brown, UPenn, NYU, UCLA, Berkeley, UCSD, Northwestern, Emory, Wisconsin, Michigan, Princeton, Yale, Stanford, Uni of Chicago, MIT, Duke, CalTech...

Please, if I forgot to mention any other unis, just write them down as well, I just can't think of everything off the top of my head and don't mean to downplay anyone

Many thanks!

 

[PoliticalOrder]

I mean, what exactly are you asking?

Every single program you will have to take (at least) two classes in the quant sequence (unless you are doing theory). There will also be formal theory, MLE, advanced quant, experimental methods, ect. to choose from. Most programs may only have one qualitative methods class, if even that. You may be able to find a comparative-historical class at some places (Chicago for example from Dan Slater). 

More or less every single program you list will be pretty much structured the same way....around 7-10 quantitative methods courses to choose from, some required to take, while having 0-2 qualitative methods courses that are purely optional. Then also a very broad 'scope and methods' class that is required but could literally be anything under the sun.

Now, there are programs that have a more heavily qualitative faculty (places like Northwestern, Brown, Cornell), but that doesn't necessarily have any bearing on how they structure their coursework in grad programs. Although in places like this you are more likely to get a higher end of the amount of qualitative methods classes offers, like 2).

Edited August 9, 2016 by PoliticalOrder

[PizzaCat93]

Northwestern has one of the best comparative-historical scholars in the field (James Mahoney) and also is more qualitatively-inclined than most other departments. UNC has two (three soon, perhaps) game theory people. Yale also may be a good one to check out. But the above post is correct that you're going to have to take quant methods anywhere you go. Honestly, look at the course catalog and course offerings at each program and see if it's satisfactory to you. Also check out sociology course offerings, if the department allows for taking courses outside the department. However, if you're going to base your list of schools on this, it's going to be a very short list. A better plan would be to just attend IQMR during a summer and teach yourself whatever else you need for your research. 

Edited August 9, 2016 by PizzaCat93

[resDQ]

7 hours ago, PizzaCat93 said:

Northwestern has one of the best comparative-historical scholars in the field (James Mahoney) and also is more qualitatively-inclined than most other departments. UNC has two (three soon, perhaps) game theory people. Yale also may be a good one to check out. But the above post is correct that you're going to have to take quant methods anywhere you go. Honestly, look at the course catalog and course offerings at each program and see if it's satisfactory to you. Also check out sociology course offerings, if the department allows for taking courses outside the department. However, if you're going to base your list of schools on this, it's going to be a very short list. A better plan would be to just attend IQMR during a summer and teach yourself whatever else you need for your research. 

 

Agree with PizzaCat93, you should also see if the Sociology departments at the schools you are looking at lean quantitative or qualitative. Could potentially take a qualitative course in another department. 

[rising_star]

Depending on the qualitative methods you're interested in, you may also want to look at methods courses in anthropology, geography, women's and gender studies, and education, all of which are fields that tend to offer courses in qualitative methods. If you really want to have strong qual training, you could go somewhere like UGA, which has a graduate certificate in qualitative methods that any grad student can pursue regardless of their discipline.

[TakeruK]

Caltech's program is very heavily quantitative. Note also that Caltech has a very small humanities and social sciences program. There are only two PhD programs offered in these fields, one called "Social Sciences" and another called "Behaviorial and Social Neuroscience". For more information on their program and coursework, you can go here: https://www.hss.caltech.edu/content/social-sciences-phd-program or http://www.hss.caltech.edu/content/bsn-academics. And for admission requirements / course pre-requisites, see http://www.hss.caltech.edu/content/graduate-admissions

[PoliticalOrder]

I'm just gonna post something that needs to be said.

No matter how much you disagree with the quantification of political science, it is not going away and is only going to increase for the most part. And depending on your field (especially if you study American) qualitative methods may be nearly, or completely dead in the water.

The name 'Skocpol Wannabe' is alarming. Every graduate student reads States and Social Revolutions and there is a lot to learn from it. However, political science is NOT done like this anymore in basically any capacity. 

Some of the posters have advised taking extra-departmental qualitative classes, I don't think that is really a good idea (outside of something like a survey qualitative methods course or comparative-historical studies). Archival work isn't a big thing in political science. Ethnography simply does not exist in the field. Interviews, while can be a part of a comparative politics research design, are usually just reserved for theory building and getting information that isn't available otherwise. Furthermore, the critical and gender theory that is heavily used in fields like Women's studies and Anthropology have virtually no standing in political science.

If you don't have abilities in quant you probably won't get a job nor publish in top journals; it is as simple as that. You need to think about perhaps looking at other disciplines if you are dead set on qualitative methods, it's going to be a huge uphill battle throughout your career if you try to go against the grain.

[resDQ]

4 hours ago, PoliticalOrder said:

I'm just gonna post something that needs to be said.

No matter how much you disagree with the quantification of political science, it is not going away and is only going to increase for the most part. And depending on your field (especially if you study American) qualitative methods may be nearly, or completely dead in the water.

The name 'Skocpol Wannabe' is alarming. Every graduate student reads States and Social Revolutions and there is a lot to learn from it. However, political science is NOT done like this anymore in basically any capacity. 

Some of the posters have advised taking extra-departmental qualitative classes, I don't think that is really a good idea (outside of something like a survey qualitative methods course or comparative-historical studies). Archival work isn't a big thing in political science. Ethnography simply does not exist in the field. Interviews, while can be a part of a comparative politics research design, are usually just reserved for theory building and getting information that isn't available otherwise. Furthermore, the critical and gender theory that is heavily used in fields like Women's studies and Anthropology have virtually no standing in political science.

If you don't have abilities in quant you probably won't get a job nor publish in top journals; it is as simple as that. You need to think about perhaps looking at other disciplines if you are dead set on qualitative methods, it's going to be a huge uphill battle throughout your career if you try to go against the grain.

 

My undergrad profs said the same thing. They want even farther though and said that not just simple things like regressions and time series, but more advanced quant methods. Do you have any thoughts on this?

[PoliticalOrder]

Really depends on your field and/or project(s). If you are studying trade in IPE or congress in American, yes, you will need to know pretty advanced statistics and probably advanced formal/game theory. But there are other subfields that are less quantitative - for example, contentious politics/social movements in comparative politics - for the obvious reason that building datasets and quantitatively testing certain things is not really possible and/or that helpful. 

So, it depends. You do not need to be statistician to succeed in (non-theory) political science. But even if you have a qualitative bent, which is still possible these days depending on your approach, you will still need to collect data and run simple regressions/correlations for the most part. The collection of data and quantitative measures is everywhere in the field now, even the most qualitative projects - but there is still quite a bit of variance in how deep your analysis goes. 

[Salve]

Thank you all for sharing! I appreciate this information very much. 

On 09.08.2016 at 8:02 PM, PoliticalOrder said:

I mean, what exactly are you asking?

Every single program you will have to take (at least) two classes in the quant sequence (unless you are doing theory). There will also be formal theory, MLE, advanced quant, experimental methods, ect. to choose from. Most programs may only have one qualitative methods class, if even that. You may be able to find a comparative-historical class at some places (Chicago for example from Dan Slater). 

More or less every single program you list will be pretty much structured the same way....around 7-10 quantitative methods courses to choose from, some required to take, while having 0-2 qualitative methods courses that are purely optional. Then also a very broad 'scope and methods' class that is required but could literally be anything under the sun.

Now, there are programs that have a more heavily qualitative faculty (places like Northwestern, Brown, Cornell), but that doesn't necessarily have any bearing on how they structure their coursework in grad programs. Although in places like this you are more likely to get a higher end of the amount of qualitative methods classes offers, like 2).

PoliticalOrder, thank you for showing the difference between qualitative course structure and qualitative faculty. I was looking mostly at faculty, when researching programs and assumed that more qualitative faculty must mean more qualitative courses. 

Also, to your post and others above, I wasn't implying that I do not intend on taking any quant courses at all. I believe both qual and quant have something to offer. I was just concerned that because I am inclined to using a mostly qualitative methodology, this might compromise my potential success in the field. And looking at your replies, it seems highly likely (publishing, reputation, etc.).

On 10.08.2016 at 7:45 PM, PoliticalOrder said:

I'm just gonna post something that needs to be said.

No matter how much you disagree with the quantification of political science, it is not going away and is only going to increase for the most part. And depending on your field (especially if you study American) qualitative methods may be nearly, or completely dead in the water.

The name 'Skocpol Wannabe' is alarming. Every graduate student reads States and Social Revolutions and there is a lot to learn from it. However, political science is NOT done like this anymore in basically any capacity. 

Some of the posters have advised taking extra-departmental qualitative classes, I don't think that is really a good idea (outside of something like a survey qualitative methods course or comparative-historical studies). Archival work isn't a big thing in political science. Ethnography simply does not exist in the field. Interviews, while can be a part of a comparative politics research design, are usually just reserved for theory building and getting information that isn't available otherwise. Furthermore, the critical and gender theory that is heavily used in fields like Women's studies and Anthropology have virtually no standing in political science.

If you don't have abilities in quant you probably won't get a job nor publish in top journals; it is as simple as that. You need to think about perhaps looking at other disciplines if you are dead set on qualitative methods, it's going to be a huge uphill battle throughout your career if you try to go against the grain.

PoliticalOrder, I understand what you mean by my name being alarming. And, in fact, I do not have hopes of becoming the next Skocpol or even attempting to do the same work for obvious reasons that the programs in polisci right now are very quantitative. My concerns about methodology and approaches comes just from my epistemological beliefs (and maybe intuition) and not from a fear of quantitative methods. In fact, I finished my undegraduate degree in Mathematics and was among the highest ranked students (among the classes that I took were Calc III, Advanced Linear Algebra and Real Analysis). So I have no problems of doing quant for fear of "numbers". I am seeking advice on how to find a program where I will feel comfortable given my way of thinking about the world. As you mentioned in your last comments, there are  subfields which are less quantitaive, including social movements/contentious politics. Incidentally, this is what I am interested in: contentious politics, revolutions and how they relate to different types of inequalities. Comparative politics is a broader umbrella under which I would like to work. In this case, what would be your advice to me for searching appropriate schools? I come from a third world country, so cannot afford to pay for my own education or gamble with something that can prove potentially a life-defining choice. I absolutely have to get into a good school to be able to find a good job and look after my family. Nonetheless, I would want to be at least to some extent happy with what I do. Do you think this is absolutely a clash of goals?

 

Thank you!

[Determinedandnervous]

Just to weigh on on the methods debate, PoliticalOrder is absolutely right. Political science in the United States is trending towards quantitative methods, and the pendulum will probably never swing towards qualitative methods again, though it has been swinging to different types of quantitative analysis other than regressions on observational data (i.e. social network analysis and experiments).

There is still room for mixed methods. As far as I'm concerned, pretty much all work is mixed methods. Every substantive question answered with quantitative social science has to be grounded in a theoretical framework. Otherwise, the numbers don't mean anything. This framework is always qualitatively explained before it is quantitatively verified. I highly recommend reading Designing Social Inquiry: Scientific Inference in Qualitative Research by Gary King, Bob Keohane, and Sidney Verba. The authors lay out a brilliant argument as to how both methods can be used with similar thinking to answer similar questions. It certainly helped me mentally bridge the divide. Like you, I originally was more interested in mixed-methods, and I'm presenting qualitative research at APSA this fall, but I chose a quant-heavy school and after getting more comfortable with statistical methods, my work from here forward will also be quantitative.

If upon further reflection you do not wish to pursue a quantitative social science, there is always the option of doing your PhD in Canada, the UK, or the European University Institute. These places have a much more qualitative bent still and though you most likely won't have access to the US market, universities there can have great placement internationally (especially Oxford and Toronto). If you're in a third world country (and especially if you live in the British Commonwealth) then you would have no problem getting funding from a British university.

[Monody]

Extending on the previous post, I would advise you to look at universities outside the US in general. Seemingly most of the M.A. programs are taught in English and qualitative work regularly has a higher standing particularly because of the reasons you mentioned. I, of course, only speak from my experience at German universities and from what I heard from the profs there, but I think that the generalisation is not completely undeserved. You may be gambling on the future survival of qualitative research strictly speaking, but the risk is probably manageable.

[CarefreeWritingsontheWall]

I agree with what most people have said - the number of quantitative methods courses in any top American program will almost always dwarf the number of qualitative courses offered. But I would view it in a different light: quantitative methods are extremely broad and they are constantly evolving with new software developments and the incorporation of new statistical methods into the field. There are usually many offered because it's important to have a mathematical foundation before you leap into the coding, and when you're coding there's so many different types of models to learn you need multiple courses. Qualitative, on the other hand, is much more practice oriented: you can only take so many courses about how to do it and what to consider. What's important is actually going out into the field, and doing it, practicing interview skills, working with archival evidence, doing fieldwork and building up experience.

How you are able to tailor your methods training to your work is a huge variant, however. Some programs are more open to a mixed methods approach. Some are strictly positivist, stressing causal inference. Understanding your preferred epistemological and ontological approach to research before you apply is extremely helpful as it allows you to focus in on programs where you know you'll have a better fit.

Edited August 16, 2016 by CarefreeWritingsontheWall

[CarefreeWritingsontheWall]

On the other issue of methodological balance: in the US, the methodological approach suggested by King Keohane and Verba reigns supreme, which has only served to reinforce the push towards quantitative methods. I would not say, however, that it's a waste of time to do anything else. I disagree with the comment that critical theory, or ethnographic approaches are useless: they are getting published. Maybe not in APSR, but they are out there and remain influential, especially if any of your work adopts any intuition from social theory. That archival research "just isn't done" is a massive fallacy. Archival research may not require you traveling to a national archive anymore, but if you're working with polling data, survey data, or census data beyond the early 2000s chances are you're working with archived data, or at least an archivist or librarian within some institute somewhere managing that information. If you're doing anything related to institutional functions, archival data remains essential to understanding how they were developed and how they structure interaction today. If you ignore all of this, you're completely ignoring what's happening on the ground - your work automatically loses touch with reality, which means it also loses policy relevance. And while statistics can help you make a big splash, like, e.g. Rose's paper that the WTO doesn't actually increase trade, if you don't pay attention to how the institution actually works, your results can actually turn out to be crap (See the Goldstein Rivers and Tomz reply to that piece).

In Canada, programs remain mixed in their approach. It's undoubtedly a contentious issue but there is more beyond KKV and large-n analyses. It all depends on your research questions of interest. E.g. Brady and Collier - Rethinking Social Inquiry.

 

[saudiwin]

There is some good general wisdom on this thread, but I have to say that no one seems to have cited anything on qualitative methods that has been published in the 21st century. King Keohane and Verba was ground-breaking in the early 1990s, but not anymore. Qualitative methods--as an actual method, not just a culture--is an emerging and developing field in political science with a lot of work going on. It is a small community, but it is vibrant and very interesting. Where it will end up, I don't know, but for a good modern overview, read this recent book on process tracing. 

The reason that qualitative methods isn't going anywhere is because statistical, or quantitative methods, are going through what has been called the credibility revolution. It's way too complicated to get into here, but suffice it to say that statistical models are being put through a lot more scrutiny than before, and that findings are no longer taken as golden just because the authors did what the econometrics textbook told them to do. I think it's getting to the point where dissertations that rely on cross-national regressions are looked at with a dose of skepticism unless there has been some serious qualitative analysis on the back-end.

In addition, I would say that field experiments and qualitative methods are closely linked. It is difficult to pull of a field experiment without doing serious qualitative work.

As far as departments goes, as others mentioned, all top departments require a quantitative sequence, and it's generally recommended to go beyond the required courses. But there is wide variation in whether departments have qualitative methodologists or people who are well-known in that field. In particular, I would stay away from Rochester and NYU, as they tend to focus almost exclusively on statistical, experimental and formal methods.

Feel free to ping me if you want to discuss where the qualitative methods scholars are. I think my own program (UVA) is pretty good, but I am certainly biased in that regard.

[alphazeta]

On 8/12/2016 at 7:25 AM, Skocpol-Wannabe said:

Also, to your post and others above, I wasn't implying that I do not intend on taking any quant courses at all. I believe both qual and quant have something to offer. I was just concerned that because I am inclined to using a mostly qualitative methodology, this might compromise my potential success in the field. And looking at your replies, it seems highly likely (publishing, reputation, etc.).

As you mentioned in your last comments, there are  subfields which are less quantitaive, including social movements/contentious politics. Incidentally, this is what I am interested in: contentious politics, revolutions and how they relate to different types of inequalities. Comparative politics is a broader umbrella under which I would like to work. In this case, what would be your advice to me for searching appropriate schools?

I don't think that using mostly qualitative methodology compromises your potential field if you do it well and if you make a strong case that you're matching your methods appropriately to your questions.  On the other hand, it's definitely true that having strong skills in statistical methods can give you a leg up.  Just look at a few job ads, and you'll see how many jobs listed in the substantive subfields mention a preference for a candidate who can teach statistics.

There were losers in the methods war, but it wasn't the qualitative methodologists.  What is no longer possible in the discipline is the sort of sweeping, macrohistorical approach that you find in Barrington Moore or Sam Huntington - that is conclusions drawn from cross-national comparisons that sweep decades or centuries of history into a few dozen pages relying on the author's sweeping conclusions about large-scale patterns. By the same token, the broadly similar style of, say, J. David Singer (huge, time series crosssectional regressions that throw in the kitchen sink) in quantitative methods has also fallen very much out of favor.

Particularly in comparative politics, we're witnessing the triumph of detailed, context-dependent work on both the qualitative and quantitative front.  There's not much reward to quantitative scholars who just run regressions on off-the-shelf datasets and there's not much reward to qualitative scholars who just write up conclusions based on a reading of off-the-shelf historical sources.  Rewards today accrue to people who go out and gather new information directly relevant to their specific question (whether in field experiments or elite interviews; intensive efforts to gather/code quantitative data or intensive ethnography).

If it's actually the right method for your question, I can only think of a handful of places that would object to a primarily qualitative approach and they've been mentioned above (WashU, Rochester, Cal Tech, etc.).

Now, I'm assuming above that you're interested in qualitative but also broadly positivist approach.  This very much changes if you're not broadly interested in a positivist, empirical research program.

 

[Determinedandnervous]

^ I would just clarify that WUSTL, Rochester and Caltech are not hostile to quantitative methods. They train formal theorists (i.e. game theory, decision theory), who use calculus (specifically derivatives) in their work. In certain ways, it's more demanding in terms of prerequisites for quant skills than statistical methods. With statistical methods, one can drive the car without knowing what's under the hood, in a manner of speaking. With formal theory, you have to.

[waterloo715]

I am applying this cycle with my major field as CP. I am thinking of listing my intended minor as methodology, despite a relatively weak quant background. I took calculus in high school (2010 graduate), but I didn't take any quant/math courses in college beyond the required political methods course and a couple economics courses. 

How will this be perceived? Am I better off listing my minor as, say, IR instead? I have every intention of gaining a solid training in methodology, but I am afraid some adcomms may perceive the gap between my expectation and reality to be too large—hence weakening my application. Or does the minor field not matter too much from the start? 

Any thoughts?

[resDQ]

6 hours ago, waterloo715 said:

I am applying this cycle with my major field as CP. I am thinking of listing my intended minor as methodology, despite a relatively weak quant background. I took calculus in high school (2010 graduate), but I didn't take any quant/math courses in college beyond the required political methods course and a couple economics courses. 

How will this be perceived? Am I better off listing my minor as, say, IR instead? I have every intention of gaining a solid training in methodology, but I am afraid some adcomms may perceive the gap between my expectation and reality to be too large—hence weakening my application. Or does the minor field not matter too much from the start? 

Any thoughts?

 

I was told that no one cares about your second field. Not even when you are on the job market. EXCPET at liberal arts colleges where they may want you to teach IR courses in conjunction with CP or possibly methods. I doubt anyone on an admissions committee would take your 2nd field choice in consideration for admittance. I have no direct experience with this, however. 

[Determinedandnervous]

^It doesn't really matter all that much. If you're interested in methods, pick that. I had about the same level of quant experience and got into quite a few schools. Methods sequences are generally constructed with people who don't have a lot of experience with advanced math in mind. There are usually math camps at the beginning of the first year. After all, most of us majored in social sciences, which aren't usually too heavy on math.