Skip to main content


This summer I spent two months at the Roper Center as a Kohut Fellow. On the first day of my fellowship, staff members taught me how to convert ASCII files to Stata and SPSS format in order to process and analyze the survey data I needed for my research fellowship. I continued to convert datasets for the next two months and while doing so I encountered several challenges. The main challenge was dropping some of the surveys I was interested in working with because the responses either conflated the military and the police or because the codebooks were incomplete. Nonetheless, I made substantial progress on my research during my visit. I converted, processed and analyzed most of the surveys I planned on working with and also found some surveys that had relevant questions for my research that I had not seen before. As part of my fellowship I also wrote a short article on public attitudes towards militarizing police in Latin America. In this article, I ask whether people are supportive of the military doing police work and why they are supportive. To answer these two questions, I used two surveys from Roper’s USIA collection: the USIA 1992 poll in Guatemala and the USIA 1997 poll in Mexico. This blog post will serve as the foundation of a larger research project on attitudes towards militarizing law enforcement which I will conduct in 2018-2019.

My time as a Kohut Fellow was extremely valuable academically and professionally. First, it contributed greatly to my academic research. By exploring and using the vast public opinion data on public attitudes towards crime and crime salience in the region, this fellowship broadened my research on crime and public opinion. Second, I plan to incorporate this summer’s research findings into my larger dissertation project on support for militarized policing and these will also inform my qualitative research in Mexico. Third, by coming to the Center I was able to find data that I would not have otherwise had access to and received constant support on how to convert and use survey data properly. Fourth, I was also able to explore the paper archives and use data that are not available online to build a dataset on responses to the most pressing problem facing Latin American countries.

As a fellow, my work will be beneficial for the Roper Center. I am contributing to the Roper Center’s goal of broadening our understanding of public opinion across the world by leaving behind 13 converted survey datasets (from ASCII to SPSS/Stata format) that I used for my research. The majority of the 13 surveys converted were sponsored by the USIA and the list includes two surveys from Colombia, three surveys from El Salvador, three surveys from Guatemala, three surveys from Honduras, three surveys from Mexico and one survey from Peru.