The Quantitative Peace

Sounds About Right

courtesy: www.smbc-comics.com

Rumsfeld Interview

Given this piece from the other day, I thought this was an appropriate followup.  Daily Show fans have probably already seen his interview from last night with former Defense Secretary Donald Rumseld, but for those who have not I would recommend it.  Rumsfeld was unquestionably one of the most important architects of the Iraq war, and I think this interview fits in nicely with the discussion from the other day regarding the influence of unelected actors in the policymaking process.  I found it particularly interesting how Rumsfeld acknowledges the important role played by presidential advisors in this process, but really seems to shift accountability away from the Defense Department and onto the State Department and the non-descript "intelligence community" (of which the DOD is an incredibly important member).  

It's worth noting Rumseld's comments regarding his previous remarks on "going to war with the military you have, not the one you wish you had."  I think it's easy to dismiss this comment as another of Rumsfeld's  witticisms, but I think the question it poses is important and worth exploring.  How much do we really know about how the decisions of an administration's predecessor constrain the ability of the current administration to act in certain ways?  If we are to take Rumsfeld at his word--not much.  He seems to be implying that the capabilities of the state don't factor into decisions related to the use of force, or at least that they only matter to a certain extent.  Policy objectives come first--capabilities matter second. 

Some of his remarks also suggest some interesting things insofar as the learning process of the miltiary/DOD is concerned.  Is it preferable to invest significant amounts of money and resources up front to address possible shortfalls?  Or is it better to go to war, knowing that capabilities are less than optimal, but that investments in updating or upgrading capabilities will be more efficiently and effectively applied after more information is gained through battlefield encounters?  Rather, that it is impossible to know what the optimal allocation of funds is until more information is gained? 

Thoughts?  

Quick Post

So it's been a little while since I've posted here (very busy times these are).  Anyway, I wanted to link to this post by Phil Arena over at his blog.  Phil discusses how his views on various topics related to IR have changed over the past few years.  I agree with a lot of what he has to say on these issues--I think a lot of what IR scholars focus on, to the extent that we're interested in the domestic forces generating foreign policy outcomes, tend to place too much stock in the qualities we typically assume to characterize democracies and autocracies.  I also agree that economic forces play a huge role in shaping the incentives of political actors--particularly the elites that Arena mentions. 

Citing eBooks

I got my Amazon Kindle about a year ago and I really enjoy it, but I am  finding myself become increasingly frustrated by the apparent lack of standards when it comes to citing eBooks.  On some occassions this isn't really a problem---if you need a particular page number you can often times type a phrase into Google Books and find what you're looking for (assuming the correct edition is available).  But to some extent this defeats the purpose of having an eReader.  Furthermore, Google Books often censors the content of a variety of books, so it's not guaranteed that you will be able to find the specific page that you happen to be searching for.  

I am clearly not the first person to find this frustrating.  Upon searching for an answer I came across this page with some comments written by a certain Stephen Smith.  The location numbers that eReaders like the Kindle feature should indeed be more accurate, as I understand it.  Depending on the reader's preferences you can change the size of the font.  But, as this will change the content featured on each "page" of the eReader, page numbers are not particularly useful guides when searching for specific content. 

The page referenced above suggests a solution offered by the Chicago Manual of Style, wherein a section title or chapter be listed for eBooks in place of page numbers.  But this is really not much of a solution.  Depending on the writer's style and length of the book, particular sections or chapters could be enormous, thereby making it difficult to track down what it is that you are looking for.  True, as this page points out, even hard copies of books have some issues with page numbers varying across different editions.  But this is only a problem when tracking content across multiple readers.  And even in a situation where two or more readers are working from different editions, the content is often similar enough where the page different might only be a matter of a couple of pages. 

The problem with eReaders seems to be that the reader/author themselves cannot accurately convey, by current academic standards, where it is that they are drawing information from.  And it is more difficult to translate a location number into a page number than it is to navigate from a page number from one edition to the correct spot in another edition of the same hard copy.  I haven't yet seen an article published in a political science journal that features eReader locations for some readings, in addition to page numbers for others.  A quick Google search for political science style guides did not seem to shed any light on the issue either.  I'm not sure this problem affects that many people at present, but it does seem like an issue that would, conceivably, be fairly simple to resolve.                

Biology and Politics

A few years ago I wrote a blog post about a study that found a link between certain genes and interest in politics. Two new studies have come out since that go even further, and find a link between brain region size and genes, and political preference. I discuss them below.

Brain Region Size and Ideology

A new study by scientists at University College London has found a link between the size of certain areas of the brain and political viewpoints. In sum:

"Scientists have found that people with conservative views have brains with larger amygdalas, almond shaped areas in the centre of the brain often associated with anxiety and emotions.

On the otherhand, they have a smaller anterior cingulate, an area at the front of the brain associated with courage and looking on the bright side of life."

Although I find these findings interesting and worth exploring further, I am slightly skeptical that there is a link between fear and anxiety, and being conservative. In my opinion, fear is embedded in both the liberal and conservative viewpoints, the only difference is what scares liberals and conservatives. Conservatives tend to fear foreign enemies, terrorists, and the government overreaching its power in the economy. Liberals tend to fear strong governments that interfe with people's privacy (such as too much survelliance). They also fear big business and corruption. Hence, it is unclear whether fear and anxiety would lead anyone to be conservative.

While I do not think there is a link between anxiety and being conservative, I would not be surprised if there was a link between fear and anxiety and having extreme political positions. This is only based on ancedotal evidence, but it seems to me that those with the most extreme ideologies, both left and right, tend to be the most paranoid about government and politics (in US politics, Glenn Beck and Michael Moore come to mind). It would be interesting to see a study that examines this potential link.

Genes and Political Ideology

In another study, published in the October 2010 edition of JOP, Settle, Dawes, Christakis, and Fowler suggest that people may be genetically predisposed to being liberal or conservative. In sum:

"Scientists at the University of California San Diego and Harvard University determined that people who carry a variant of the DRD4 gene are more likely to be liberals as adults, depending on the number of friendships they had during high school....

The four authors, including UCSD's James Fowler, wanted to explore if politics were heritable by identifying a specific gene variant associated with political leaning. They hypothesized that individuals with a genetic predisposition toward seeking out new experiences would tend to be more liberal.

The 7R variant of DRD4, a dopamine receptor gene, had previously been associated with novelty seeking. The researchers theorized novelty seeking would be related to openness, a psychological trait that has been associated with political liberalism.

However, social environment was critical. The more friends gene carriers have in high school, the more likely they are to be liberals as adults."

The potential for a link between genes and ideoloy has several implications for real world outcomes. How common is this gene (and how common is it for people to have this gene and be outgoing)? How much can the existence of outgoing people with this gene in the voting age population have on electoral and legislative outcomes? Furthermore, are social movements only possible if there are a large number of outgoing people with this gene? And finally, what does this mean for how we view the leaders of social movements? Did the consciously choose to change society, or are their actions simply the result of their genes and their social environment?

CRACK

I found this link via Andrew Gelman's blog the other day:

CRACK Mapmaking software for Mac

I haven't had much of a chance to play with it, but it looks like it could be useful.  I've had a few instances in class this semester where I've wanted to make some maps of the US to illustrate various points, but alas, have not known how to do so.  Rather, I've wanted to make customizable maps, but have not known how.  It seems like this program is pretty simple to use--just drag and drop a CSV file with the relevant data onto the map.  It also appears to be able to perform some basic data manipulation as well.  When all is as you wish it to be, you can export an image file of your customized map to incorporate into documents, presentations, etc.  

As the link suggests, this appears to be only for Mac users at the moment.  I'm sure there must be some freeware out there that accomplishes the same thing for PC users as well.  I'm generally behind the curve on this kind of thing, so if anyone is aware of the existence of such software please post a link in the comments section.  Also, if anyone is aware of similar software that can be used at the global level that would also be most useful.   

Lessons From the Past? Or Justifying an Idiosyncratic Research Agenda?

I remember reading Michael Tomz's book Reputation and International Cooperation: Sovereign Debt Across Three Centuries for a seminar a couple of years ago, and someone having made a remark about how Tomz was probably just really interested in the time period or the subject matter, but had to frame the research in such a way as to justify its relevance beyond what might be the author's more narrow interests.  Since then I've had a few discussions with some of my advisers on the topic of how we frame our research--essentially how we sell a line of research as relevant to our field.  Overall though, I've found that this is a topic that seems to have stayed out of in-class discussions for the most part.

Best Practice in Political Science (or Any Field)

Over the past year or so I've been getting more and more into network analysis--Both for its theoretical and methodological components.  For the most part, this methodological approach has not seen widespread use in political science, although network approaches are steadily growing.  Emilie Hafner-Burton, Miles Kahler, and Alexander Montgomery have a relatively recent article about this issue, although it deals primarily with the application of network methods to the field of international relations (which suits me just fine as I happen to be an IR guy). 

I had started reading some texts on network analysis last spring, and subsequent to taking an introductory course through ICPSR last summer, I've also been trying more and more to get some hands on experience with network tools and data--especially when I find papers that utilize network techniques and apply them to issues or topics found in IR.  I think this is probably one of the better ways to learn new methodological techniques, and as I suspect anyone in Binghamton's political science program will tell you, diving into a good replication project can be one of the best ways to familiarize yourself with new data, methods, etc.  It can also generate some important insights into how the field, and publication process more specifically, work. 

Anywho, I am really aiming at one particular interaction I had in pursuit of some replication data.  Upon contacting one of the authors in an attempt to obtain a copy of the data (who was very prompt in his reply, so the professional courtesy should not be ignored) I was informed that it "was not the norm in [his] field" to keep replication data.  This response got me thinking. 

The point of discussing network analysis at the beginning of this post was not only to provide some background info on my own interests, but to open up the issue of inter-disciplinary exchange.  Network methods have been largely the domain of sociology for the last 60-70 years or so.  I think their migration into other fields is fantastic, and that inter-disciplinary collaboration can be a valuable way to bring new insights into fields that can be rather isolated.  But this also raises the issue of to what extent, if at all, different disciplines view/approach their work in the same way, and what impact this may/may not have on inter-disciplinary collaboration and the diffusion of new methods/data. 

The individual that I had spoken to happened to be from a business department, and the paper focused on issues typically coming from the field of IR.  Not having gone to business school, I obviously can't attest to the ways in which the profession directs graduate students regarding what it is that they should be doing, and how they should be doing it.  So it strikes me that while business schools may use the same tools that we do (regression analyses and so forth) they do not view those tools in the same way that we in political science do.  Rather, while we generally acknowledge the shortcomings of our field, data, and approaches, we attempt to approach the process as a scientific one.  This includes maintaining copies of the data so that your own work can be made more transparent to other scholars.  I also don't mean to make this sound overly Idyllic--this is clearly not how it always works.  And I would invite readers with a greater knowledge of business programs to comment on this, but it seems to me that (if this one individual's comments are in fact reflective of the broader field) business programs do not view their "mission" or process in the same way.  It also seems like inter-disciplinary work can be greatly hampered by a lack of consensus on the best practices related to the use of data.  Even if the component data sets used are all already public, the individual coding rules and decisions made by the authors can greatly affect the results.  And without more detailed coding rules or replication data, it can be more difficult to determine what it is the authors did to get their results.  So when we are trying to grapple with methods that are coming from, or have been developed, primarily outside of our own field, this could be problematic. 

So does the way we view our approach really influence the ability to effectively collaborate with scholars from other fields?  Should the spread of new methods and data also be accompanied by a discussion on the appropriate standards for data management, whatever those are?  Does it even matter?

Bayesian Approaches to Political Science

I ordered Carlin and Louis 2009 “Bayesian Methods for Data Analysis” yesterday and have been left with some questions by the first chapter that I want to raise here, because it reinforces my belief that there’s an unfortunate lacuna in this and many similar books. Specifically, is there a specific Bayesian approach to political science beyond what we think of as the Bayesian approach to statistical inference? I believe that our discipline conducts empirical testing primarily for the basis of making a convincing attempt at falsifying ourselves. If a tool is confusing to the median political scientist, can it ever be convincing? And for Bayesian statistics in particular, is an effort to fit the best model really an effort to falsify the theory?

Is our children eating margarine?

A recent medical study has gained some media attention by finding and arguing that, among other factors, consumption of margarine by infants leads to lower IQ scores.  The study, based on  590 New Zealand European children claims many other factors can contribute to the IQ of children such as consuming fish weekly and grains 4 times a day.  Also, the study suggests that women who drink moderate amounts of alcohol while pregnant have a positive effect on IQ.

The news article suffers from simplification and does not address some of the vital questions that are demanded of it such as the statistical analysis used, particular controls implemented, data collection, and everything else a statistician would want to know - but the article does go a bit further than other news pieces devoted to statistics based research.  The article does give us enough room to pause and consider whether or not the link is merely spurious - that is low IQ and consumption of margarine may be caused by poverty (and other associated factors) rather than poverty causing margarine consumption which leads to lower IQ.  This could be the case since margarine is a cheaper product than butter and other possible alternatives for consumption.

So, given the summary, the actual work becomes necessary and I believe I have found it here.  The 453 page pdf contains lots of information and various tests.  The main findings for margarine are presented in Table 64 (page 394) and come from two tests.  First, a bivariate linear regression.  Second, a multivariate regression.  The first test, while useful for initial impressions, is improper for determine causation, especially without a strong theoretical consideration driving this particular variable.  However, the multivariate regression still presents a robust statistical result (p = .03) (159).

The regression includes the following controls: "gestation, parity, gender, maternal school leaving age, parental occupation at birth, maternal marital status at birth, maternal BMI, children’s BMI, Stanford Binet examiner" (Table 64, 394).  While the result is statistically robust, it appears the magnitude is relatively small (2.81 points) at age 3.5.  By 7 years old, the correlation was significant at the .1 level only for the bivariate test and loses significance in the multivariate model.  At best, then, it appears that margarine would only be a problem for kids under 4 years of age.  While non-significance is not a safe bet for conclusion, an initial review of the evidence suggests the surprising result is not consistent but appears strongly in one of the four tests (two IQ tests, 2 age groups).

The control variables for the study contain some elements of socio-economic status (SES) but are not as complete as I would like.  The article summary suggests that only 5% of the parents in the study were of a low SES status which may factor into who is feeding their children margarine below the age of 3.5 years.  While I am not an expert in the social structure of New Zealand, this does prevent the study from being generalizable to other populations.  I would guess that part of the depression in the number of impoverished children in the study results from the target sample of New Zealanders of European descent, but that is speculation on my part.  As such, a relative measure of income should be included - occupation may proxy parts of this, but the diversity of pay may factor into this.  Additionally, marriage status may not account for single or dual-earner income households which would boost the selection and substitution choices of such a family.  Thankfully, the article also notes that exercise is not controlled for - a key component to any lifestyle choice. 

The multi-variate regression may not be enough to capture the full effects of margarine.  Sceptically, I doubt that margarine is an independent variable that meets the conditions of independence within the regression and some forms of simultaneity ought to be taken into account.  At the very least, margarine should be treated or tested as a moderator/mediator variable and see if it enhances the independent effects of poverty on IQ.

Regardless, the author is cautious in the article and has every right to be.  It is obvious more statistical refinement and further data collecting would be required before we all feed our kids butter (since it has no effect on IQ in the study) and start drinking moderate amounts of wine pregnant.  As such, "What to feed your kids to make them smarter" may be a very misleading headline for a news article.

I realize this is not political science, but it is quantitative...and sometimes news articles annoy me.