Matt Winters points us to this paper by Brian Rathbun. Matt writes: “I came across this article today that reminded me of last week’s discussion in the playroom regarding isolationists and internationalists. I haven’t read it, but he appears to use principal components analysis of survey questions to identify people as caring about community or hierarchy at the international level or else being isolationists, and then he looks at how people’s attitudes affect their responses with regard to proposed actions in hypothetical scenarios.”
Here’s the abstract to Rathbun’s paper:
Although there is increasing evidence of a relationship between domestic and foreign policy attitudes among American elites, we have less of an idea about why these sets of attitudes cohere. The answer lies in a better understanding of what we mean when we talk about “left” and “right” or “liberal” and “conservative.” Drawing on the literature on rights theory, partisan cleavages, and ideological continua, I posit the existence of two core values, hierarchy and community, that should manifest themselves both at home and abroad. I perform a principal components analysis on data capturing both the domestic and foreign policy attitudes of American elites. The results indicate an almost identical structure of attitudes in both domains, indicating that it is generally inappropriate to distinguish between the two. Using factor scores in a series of logistic regressions, I demonstrate that support for community is most important for predicting support for humanitarian military operations, while hierarchy and community both help determine positions on strategic missions.
The topic came up because we were discussing ways to distinguish among (in statistical jargon, “to identify”) various foreign-policy-related attitudes, including militarism, liberal interventionism, internationalist ideology (for example, anti-fascism or anti-communism), and domestic political preferences. The idea is that any of these could be rationales for supporting a war, but under different circumstances, support could change. For example, a Republican might have opposed the Vietnam war under Johnson and supported it under Nixon, or a Democrat might have supported Clinton in Kosovo but opposed Bush in Iraq. But as these examples illustrate, other factors are changing at the same time, so it’s tricky to try to separate these different attitudes. Especially because, at all four levels (policy rationales, elite discourse, arguments in the media, and public opinion), the different motivations get mixed.
Also, I haven’t read the paper in detail either, but I can say that Tables 2 and 3 should be graphs. Figure 1 is ok but could be improved, maybe with three columns.