This past summer I took advantage of a research grant opportunity to look into something of personal interest to me. The idea of elite framing, where leaders (both political and non-political) talk about major issues in such a way as to change how the public perceives them, has fascinated me ever since freshman year. In one of the first classes I took here at W&M, we read about opponents of the estate tax proposed in the early 2000’s framed the tax as a “death tax.” By doing so, they changed public perception of the tax and got people the tax would never have effected to come out against it. I wanted to study this framing phenomenon more. I picked a policy issue of interest to me (the Affordable Care Act, or “Obamacare”), submitted a proposal (with guidance from my project advisor, Professor Jaime Settle in the Government Department, and was awarded the grant.
Armed with funding, 3 months off of school, and a kitchen fully stocked with my mom’s home-cooked food, it was time to hit the ground running.
I began my project by defining my research question. To make the endeavor feasible, I limited political elites to Congress (House of Representatives and Senate), since I am personally interested in Congress. I then narrowed framing down to social media, and specifically Facebook (because of how much data is publically available), because these media were underrepresented in the literature on elite framing. Having chosen Obamacare as my issue case study, I further condensed my research to look at the period of July 18-August 18, 2009, a month-long period with significant data available because of the August recess town halls. Finally, I established public opinion on three dominant “frames” in the Obamacare debate (cost, choice, and quality of care) as the dependent variable side of my analysis.
Data collection came next, and with it some tedium unavoidable in these kinds of projects. I first collected data on every Facebook post made by Senator/Representative on Obamacare between July 18-August 18, 2009. I used this data (as well as the partisan leanings of each state) to choose three case studies: Texas, California, and Florida. For these states, I determined how many of the Facebook posts made by the state’s Senators and Representatives were focused on each of the three frames in my independent variable (cost, choice, quality). Finally, I pulled polling data from the Kaiser Family Foundation’s Health Tracking Poll (HTP) on how people in three case study states felt about Obamacare in terms of those three frames over the course of the summer of 2009. I concluded by conducting trend analysis to see if people became more/less polarized on issue questions related to those three frames.
In the end, I did not find evidence to support the hypothesis that increased focus by elites on a particular issue frame would make the public in their state more polarized along that dimension of the issue. I did, however, find evidence suggesting that citizens in the three case study states became more polarized on all three of the frames over the course of the summer, which suggests the media storm surrounding health reform across those 3 months significantly influenced the public’s attitudes towards Obamacare.
Looking back, there are many ways I or someone else could improve this project. The relationships I found were likely muted for a number of reasons ranging from the possibility that few people care about what their Congressional delegation says on Facebook to the limits of using on Facebook to conduct my analysis. These were also numerous obvious shortcomings like hand-coding Facebook posts, the high number of Representatives/Senators with no Facebook profiles at the time, and the fact that I only used two iterations of the Health Tracking Poll. Yet in the end, I’m satisfied with how the project turned out. It taught me so much about how to design a research project, collect appropriate data, and present findings. I am very thankful I had the opportunity to complete this project.