News framing effect studies usually emphasize the relevance of their results for real-life politics and political communication. But, how valuable are the results of these studies truly in terms of real-life applicability? In response to this question, a small number of studies have begun to test the duration of framing effects (e.g., Druckman & Nelson, 2003). In doing so, these studies are able to make more convincing arguments about the actual significance of their findings (Gaines et al., 2007). Based on a “classic” framing experiment (n=1,324), we empirically mimic the dynamic nature of framing effects over time. We integrate (1) multiple frame exposures as well as (2) various tests for duration of framing effects into our study design. Our results show that exposure to repetitive frames does not systematically strengthen effects on opinion formation. However, effects can get stronger when the delay between two exposures is short. Competitive news framing is characterised by recency effects; i.e. the latest frame has the strongest impact on opinion formation. Political knowledge functions as a moderator for both effect mechanisms. Participants with higher levels of political knowledge are less prone to recency effects, but show stronger signs of cumulative framing effect. The results of this study have important methodological and substantive ramifications for framing effects research, as well as for our understanding of the real-life impact of framed media messages on opinions.
