Shifting Frames, Shifting Policy: How Frame Sets Influence Policy Making in Congress Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Shoub, Kelsey
    • Affiliation: College of Arts and Sciences, Department of Political Science
  • In this dissertation, I pose and answer three questions about the presence and role of policy discussion in Congress: (1) why are some issues discussed more in the House and Senate than others; (2) do the parties differ in how they discuss issues; and (3) why does that discussion influence bill outcomes? First, I posit and show that parties are reactive to each other and pursuit of electoral goals influence how much a general policy area is discussed. Additionally, I show that, while the relationships appear to be the same across the two chambers of Congress, they are conditioned by majority party status. Second, I posit and show that pursuit of the party’s electoral goal not only informs how much different issues are discussed, but also informs how they are discussed. I show that the parties differ in which frames they use to discuss policy. In doing so, I present a new measure of frames in Congress that using a set list of broad frames that facilitates comparisons across policy areas. Finally, building from the first and second questions, I question whether the policy discussion that occurs influences bill outcomes. I propose that the construction of policy discussion through the selection of frames shapes the considerations of those evaluating proposed legislation, which ultimately influences action (i.e. voting) in Congress. In doing this, I shift the focus from the use of individual frames and their characteristics to the concept of the frame set, a holistic look at how the issue is framed. Following from this, I posit and show that as the frame set changes more, policy change (i.e. bill passage) is more likely. To test these claims, I rely on large-N empirical analysis. I code the speeches in the Congressional Record for the Comparative Agendas Project general policy areas using supervised machine learning. Then, using unsupervised machine learning, I code the paragraphs within the speeches on policy for how it discusses that policy area, the frame used.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Roberts, Jason
  • Ryan, Timothy
  • Baumgartner, Frank
  • Treul, Sarah
  • Stimson, James
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2018

This work has no parents.