In any case, Monroe's talk was on "Digital Political Science," which I'm not sure I can explain in a short post. His Penn State website that I linked above describes his work this way:
He is currently director of a multidisciplinary NSF-funded project with political scientists, computational linguists, and statisticians on The Dynamics of Political Rhetoric and Political Representation, developing methods for the statistical analysis of political speech. This team was awarded the 2006 Gosnell Prize for Excellence in Political Methodology.Many of the applications he demonstrated used Wordle, which many readers have probably seen in the New York Times or on various websites. Monroe says he can explain a lot about political behavior by examining word patterns -- perhaps more meaningfully than other scholars who use voting records, for example.
When I returned to my office, I compared a Wordle result from my latest blog entries to my blogging from 5 years ago. It's clear that I'm writing a great deal more about my interests in film and baseball -- and much less about Iraq, the war on terror, etc. Of course, you knew that, since most of my IR blogging is now at Duck of Minerva.
Visit this blog's homepage.