Today we had a breakfast class and made bean tacos at Sherrill’s apartment. Then I took a really long nap. Afternoon classes went well; I think this was the most success I’ve had with my beginner students. We learned some basic adjectives and described some houses.
I’m currently working through an online course called Foundations and Applications of Humanities Analytics. The class is an attempt to bring text-based data analysis to traditionally-less-data-driven fields. We are essentially learning how to use big data and computers to analyze huge amounts of text and answer questions that previous generations of humanities and social science scholars could only guess at. Here is a project proposal I just submitted. This was just a proposal for an assignment and I currently have no intentions of following through on the project, but I think it’s a cool look into the kind of questions we can now answer. Feel free to do the research and publish it yourself, just credit me on the paper.
Domestic reporting on foreign affairs is far from comprehensive; newspapers simply don’t have the space to report on every story that occurs abroad. As a result, implicit in every foreign affairs piece that gets published is the idea that this particular event is more important than a lot of other events which weren’t covered. I am interested in the factors that determine which countries get reported on, and which are ignored. By examining the power of various national statistics (I would probably start with population, GDP, distance from US, and presence of US military personal) to predict media mentions, we could make an argument about which qualities make a country ‘newsworthy’. We could use the New York Times, Washington Post, and Wall Street Journal archives to track how often various countries are mentioned over time, and then see how these numbers are predicted by the other statistics. For example, I am curious whether the frequency of a country’s appearance in US media correlates more closely with its GDP or its population. If, for a significant majority of countries, GDP is a better predictor of media mentions than population is, we could reasonably conclude that economic impact is considered more significant than the number of people impacted when papers decide which stories to run. If population is a better predictor, we could conclude the opposite. We might also find that neither GDP nor population has a strong predictive effect on media mentions (which I would be very surprised by), in which case, the ‘newsworthiness’ of a story might be more closely linked with a different variable (for example, I suspect that geographic proximity also plays a role). We could also use multi-variable analysis to examine more complex questions. For example, if, as Edward Herman and Noam Chomsky put forward, the media serves primarily as a tool to manufacture consent on behalf of military and economic elites, we would anticipate more reporting is done on countries considered ‘enemies’ than countries which are considered ‘allies’. Along this line of thought, I would expect that, when media mentions are normalized for GDP, countries with low US military deployment, i.e. ‘enemies’ (the number of US military personal stationed in a country can be used as a quantifier for the strength of military alliance), would receive more media attention than counties with high deployment, i.e. ’allies’.
AdamFendos.blog 