The Nature of Framing Effects: Connecting Theory and Evidence

• In Event: Poster Session: Current and Classical Topics in Political Psychology & State Politics and Policy
  In Poster Session: Poster Session: Current and Classical Topics in Political Psychology

Abstract

The expectancy value (EV) model is ubiquitous in framing research, but its utility has been circumscribed by its treatment as merely a conceptual tool. This article posits that the EV model is most useful insofar as it directly corresponds to observable quantities in the framing process, but these quantities go unmeasured in standard designs. In particular, no studies to date have tested whether within-subject changes in opinions and importance weighting before and after treatment assignment correspond to the model’s predictions. This article develops a theory of issue framing based on the EV model and suggests multiple improvements to experimental designs. Two studies confirm the model’s success: a novel repeated-measures design validated by replicating a canonical framing effect, and an extension of this design analyzing competing frames of protests for racial justice during the Trump era. These results encourage renewed scholarship in the framing tradition, and explicitly challenge the notion of an “end of framing.”

Author

• Joseph Essig, University of Rochester