Research Portfolio Post #7

For my small-n research design, I plan to compare two of the cases I chose for my large-n research design to examine more thoroughly: Abu Ghraib and Guantanamo Bay. I chose to keep the unit for each case the same as the large-n design, as individual prisons allow the balance I desire for more specificity than state-wide analysis yet a large enough scale to presume some degree of generalizability. I chose Abu Ghraib and Guantanamo Bay, specifically, due to the amount of similarities and control in each circumstance. They do, however, also seem to demonstrate enough variation in potential dependent variables, such as the number of years of operation as I used in the large-n design, that a comparison would be significant and meaningful.

For my dependent variable, I am interested in using news and media sources as reflections of public response to each prison in order to uncover differences in attitude by the American public as well as international community to torture at Abu Ghraib and Guantanamo Bay. This makes news articles and responses published at the time of each prison’s operation primary sources for my research, and I have started reading various articles from major national news outlets to get a sense of how I might operationalize and measure public attitude.

Comparing two Washington Post articles on Abu Ghraib and Guantanamo Bay, respectively, drew my attention to the differences in language used about each prison. In Graham’s 2004 article on the closure of Abu Ghraib, the ethics of torture in Iraq are not described as questionable or debatable, they are presumed to be wrong. The only debate in the article is regarding who to blame.¹ Conversely, the 2018 article by Ryan and Nakashima engages in the now partisan debate on the necessity of keeping Guantanamo open, which implicitly includes the continuation of torture.²

Though the differences in the two articles are apparent to me, I recognize that language and connotations are often subjective, and I am struggling with operationalization. I am considering using a software that counts how many times certain words are used in selected texts and comparing them across articles on each prison, but I fear the variation will not be apparent with this method. As I continue to read, I will be looking for more concrete trends in the articles that I may be able to measure more tangibly.

¹ Bradley Graham, “Torture and Prison Abuse,” last modified August 26, 2004, accessed October 11, 2018,
² Missy Ryan and Ellen Nakashima, “Trump, Reversing 2009 Move, Vows to Keep Guantanamo Open Indefinitely,” Washington Post, last modified January 31, 2018, accessed October 29, 2018,

2 thoughts on “Research Portfolio Post #7

  1. David says:

    Hey there Hannah,

    Perceptions of torture are definitely great material for a case study, and I think the sources you’ve outlined here are a great start to that investigation. Something that came to mind when reading your post was how governments talk about torture, and how that might be similar to or different from public reactions. If you’re interested in looking into some archives, I’m sure there are archives of treatments of POWs during World War II or Vietnam in some of the sources mentioned in the reading for today (maybe even a big source like FRUS). Archives might be more available for older sources, and so it might be worth looking into cases that could have more information. For the cases of Abu Ghraib and Guantanamo that you outline here, I would definitely recommend looking into the Senate Torture Report that was released in 2012. That whole topic area will have both government documents as well as public reactions to the release of that information, both of which might benefit your analysis. Especially since case study work is more about case selection than case analysis, broadening the scope of the cases we’re considering and looking in new places would be a good next step for most of us, it looks like. Great start and happy researching!

  2. Hannah — you are off to a good start here, but it would be good to be a bit more precise in specifying (a) exactly what the DV is, (b) what the potential values are that the variable could take in any given case, and then (c) what the values of the DV are for the cases that you have identified. There is nothing wrong with using news sources as information sources for your DV, but you will also want to think about other types of sources you can use to triangulate your data (and David’s comment has some excellent suggestions). In thinking about how to capture the various aspects of your DV it would seem that the operationalization process used by Howard in her article is a good model. What kinds of questions would you ask of the data in order to be able to establish the value of the DV in a given case?

Leave a Reply

Your email address will not be published.