« Nassim Taleb's "The Black Swan" | Main | Strategic extremism: why Republicans and Democrats divide on religious values »
April 10, 2007
Brendan Nyhan's political science links
Brendan Nyhan (who arranged my fun visit to Duke's quantitative social science center in Feb) sent a bunch of references. I'm commenting on them here for convenience (easier than storing in my inbox!).
1. Cooperative game theory (looks at combinations of coalitions): a paper of Brandenburger and a syllabus of a course of Gilboa and Scarf.
2. NetLogo, a popular automaton simulation environment. This looks cool. I want to use something like this to do simulations to extend the ideas of this paper: Forming voting blocs and coalitions as a prisoner's dilemma: a possible theoretical explanation for political instability. (This is why I'm interested in item 1 above also.)
3. Computational and Mathematical Modeling in the Social Sciences, by Scott De Marchi: I ordered it, will report back. Brendan said I should read it because Scott's views on statistics are completely different from mine.
4. Fearon and Laitin's paper on civil wars, which is a controversial example of political methodology because they try to interpret zillions of regression coefficients at once. Also these supplementary tables.
5. Arthur Brooks's survey data on civic engagement and inequality, and Brendan's comment on Brooks's writings in this area. (I'd earlier noticed some of Brooks's interesting work on fertility differences between Democrats and Republicans and charitable giving.)
Posted by Andrew at April 10, 2007 12:56 AM
Trackback Pings
TrackBack URL for this entry:
http://www.stat.columbia.edu/~cook/movabletype/mt-tb.cgi/860
Comments
I wonder how much of the first linked paper was borrowed from Stuart, whose office is probably a stone's throw from yours.
Posted by: Bob Dobalina at April 9, 2007 11:21 PM.
I'd be very interested in running some simulations on that myself, and if I can come up with a way to work it into my work in network science and scientometrics, I'm going to take a stab at it.
Just perusing the paper, I'm thinking that to start with agents will have some preferences on a variety of issues and some way of empirically and naively measuring if they're satisfied with their voting power -- say, votes that went their way divided by total votes. Decreases in that value should make the agent more likely to create a coalition or join an existing coalition.
This is very interesting to think about.
Posted by: Russell Duhon at April 10, 2007 12:01 PM.
Am interested in your critique of the Fearon-Laitin study on civil wars.
Please email me relevant studies.
K
Posted by: K at April 26, 2007 11:18 PM.