What to do to train to apply statistical models to political science and public policy issues

Taylor Good writes:

I am a graduate of a state school with a BS in Math and a BA in Political Science, and I was wondering if you could give me some career advice.

Knowing how you got to where you are now, what path would you advise someone to take to get to where you are, presuming they started now? My interest in in applying mathematical models to political science and public policy issues, but I can’t tell if that is best served by a focus in Statistics, Political Science Methodology, or Public Policy Analysis (or some overlap of all three).

My reply: I think this is a good set of interests to have! My quick advice is that if you’re interested in policy analysis you should learn a bit about causal inference (you could start with the readings here). If you’re going to a graduate program, which you choose might depend on your strengths. If you’re good at math, then a statistics program could be a good bet. We don’t see many statistics graduates with social science interests so you could stand out a bit later on, if you can do the math necessary to excel in a good statistics graduate program. Or if you have broad interests in poli sci, that could make sense.

Perhaps the commenters have further suggestions?

12 thoughts on “What to do to train to apply statistical models to political science and public policy issues

    • I think economics (specially political economics) is what you are looking for. In political economy, they use mathemathical models (specially game theory) to explain political phenomena. There is also a lot of emprerical works in political economics to address your interest in statistics.

    • That might have been true 20 years ago, but there’s a lot of math and statistics in political science as the field is currently constituted, and quant work is very high-status. No reason to get a degree in econ if you’re interested in political phenomenon.

  1. Taylor Good seems to have broad interests and might find other diverse things of interest as time passes. I think that the advice to enter a statistics program if the math ability is good is a sound one. As Jim Berger pointed out to me once, statisticians get to play around in everyone else’s back yard, since people in many different disciplines need statistics and the advice of experienced statisticians.

  2. “Knowing how you got to where you are now, what path would you advise someone to take to get to where you are, presuming they started now?”

    Times have changed. There is presently a PhD glut…

  3. You might want to consider a program that is not well known, but is very much at the intersection of your interests; Pardee RAND. It is a PhD program in Policy Analysis connected to the RAND corporation, and probably the biggest focus at the school is applying analytic techniques to substantive policy problems.

    Check it out; http://www.prgs.edu/

  4. At big schools, math and stats programs need *a lot* of TAs to handle their intro courses. I’m not sure how it is in Poli Sci. I strongly suspect it is going to be a lot easier to get a full scholarship in math than in Poli Sci.

  5. Get a statistics Ph.D. if you can handle it–there are always jobs for statisticians. Political science–you’re looking at a very small collegium that restricts one’s behavior and thought in ways that many researchers would find difficult to tolerate. And it’s easy to move from statistics to political science (look at Andrew), but the other way around? Statisticians don’t respect political methodology, when they attend to it at all, which in general, is never–the only case I can really think of is some of the ecological analysis stuff, but that’s only because there’s money there.

Leave a Reply

Your email address will not be published. Required fields are marked *