Framing responsibility for political issues: The preference for dispositional attributions and the effects of news frames

This study takes an individual differences perspective in explaining framing effects. In an experiment manipulating episodic and thematic framing, the trait preference for dispositional attributions was measured and treated as a quasi-experimental factor. Beside a significant framing effect, the general attributional preference of an individual determined responsibility attributions. However, the trait exerted an impact only in the thematic condition, not in the episodic condition. Framing Responsibility Running head: FRAMING RESPONSIBILITY Framing Responsibility for Political Issues. The Preference for Dispositional Attributions and the Effects of News Frames. Jörg Matthes University of Zurich Jörg Matthes (PhD, University of Zurich, 2007) is a Post-doctoral researcher at the Institute of Mass Communication and Media Research, University of Zurich, Switzerland. Address correspondence to Jörg Matthes, Institute of Mass Communication and Media Research, University of Zurich, Andreasstrasse 15, 8050 Zurich, Switzerland, [email protected] Framing Responsibility 1 Abstract This study takes an individual differences perspective in explaining framing effects. In an experiment manipulating episodic and thematic framing, the trait preference for dispositional attributions was measured and treated as a quasi-experimental factor. Beside a significant framing effect, the general attributional preference of an individual determined responsibility attributions. However, the trait exerted an impact only in the thematic condition, not in the episodic condition.This study takes an individual differences perspective in explaining framing effects. In an experiment manipulating episodic and thematic framing, the trait preference for dispositional attributions was measured and treated as a quasi-experimental factor. Beside a significant framing effect, the general attributional preference of an individual determined responsibility attributions. However, the trait exerted an impact only in the thematic condition, not in the episodic condition. Framing Responsibility 2 Framing Responsibility for Political Issues. The Preference for Dispositional Attributions and the Effects of News Frames. Iyengar’s (1990, 1991, 1996) seminal studies have demonstrated that media frames can shape attributions of responsibility for political issues. However, the exact process of attribution has not been fully understood by framing scholars. Most importantly, attributions of responsibility are necessarily subjective and, as a result, may differ from individual to individual. The present study tests the idea that cognitive personality traits such as general attributional preferences can influence the susceptibility to framing effects. News Frames and Attributions of Responsibility Iyengar (1991) distinguishes between episodic and thematic framing. When news is framed episodically, a story focuses on individuals that illustrate and exemplify an issue (e.g., the depiction of a victim for the issue terrorism). In contrast, thematic framing emphasizes broader trends or backgrounds, providing more “in-depth, interpretive analysis” (Iyengar, 1991, p. 14), e.g., discussing the issue terrorism in historical and religious terms. In a series of experiments, Iyengar (1991) found evidence that subjects shown episodic TV reports were less likely to consider society responsible than subjects exposed to thematic stories, and subjects shown thematic reports were less likely to consider individuals responsible compared to subjects that watched episodic news stories. Iyengar (1991) states that explanations of issues depend on the reference points furnished in media presentations. However, seminal work on individual differences has shown that some people are more inclined to perform dispositional attributions while other people tend to prefer external attributions (Kelley & Michela, 1980; Rotter, 1966). Persons with a preference for external attributions tend to see the reasons for the behavior of others in concrete, contemporary events that directly impinge on the individual. In contrast, people with a dispositional Framing Responsibility 3 attributional style see the causes for others’ behaviors in terms of general dispositions such as personality or abilities. Taken together, this work suggests that attributions of responsibility depend not only on the mass media’s framing, but also on general attributional preferences. Put differently, while episodic framing leads to dispositional attributions and thematic framing leads to less dispositional, external attributions, these effects may be hampered or reinforced by the general attributional preference of an individual. Thus, the following two hypotheses were formulated: Hypothesis 1: Episodic framing of a news item leads to stronger dispositional attributions of responsibility compared to thematic framing. Hypothesis 2: When reading a news article, individuals with a general dispositional attributional style tend to prefer individual causes over external causes for the problem or issue at hand. Method Participants, Design, and Procedure One hundred and six students in mass communication classes voluntarily participated in the study (67% women) in groups of five to eight individuals. They were told the investigation was about evaluating newspaper stories. Participants were randomly exposed to a newspaper article about poverty (either framed episodically or thematically), completed a web-questionnaire, were debriefed, and thanked. The trait was measured after the experiment. Cognitive distraction tasks were meant to prevent context or priming effects. Before the trait measurement, participants had to evaluate journalistic features of the article; they were also asked to write down and explain ten adjectives to describe themselves and their personality in detail. Twenty-five items about various personality aspects were applied as a further distraction. No effect of the dependent variable on these answers was observed. Stimulus Material, Measures, and Data Analysis Framing Responsibility 4 Based on real newspaper stories, the thematic article approached the issue in general terms describing statistics and definitions. A brief description of a young mother was given. This episodic component in the thematic frame is in line with Iyengar’s (1990) view of a dominant frame. The story then covered possibilities of how poverty can be diminished, and a sociologist gave background information and statistics about poverty in the country. The episodic story reported about the young mother without giving background information. Article evaluation of both stories was equal (Mthem = 3.82, SD = .46; Mepis = 3.78, SD = .55; F =.21, n.s.) As the independent variable, four items from the dimension tendency to perform internal attributions (α = .71) from the trait questionnaire by Fletcher et al. (1986) were applied (e.g., “I think a lot about the influence that society has on my own behaviour and personality” (reverse coded); index M = 2.24, SD = .79). As the dependent variable, dispositional attributions of responsibility for poverty were measured (α = .80) with four items (e.g., “I think the person in the news story can be blamed for her poverty”; index M = 3.25, SD = .95). Article evaluation was assessed with four semantic differential items (e.g., “credible”, “realistic”, α = .71). Both measures were developed for use in the present study. All items were administered on a fivepoint scale. The data were analyzed with hierarchical moderated multiple regression. Framing was coded with 1 = episodic and 0 = thematic. Both independent variables were entered in the first step, and the interaction term was entered in the second step. The interaction was probed with the MODPROBE macro (Hayes & Matthes, 2008) that applies the Johnson–Neyman technique (Johnson & Neyman, 1936). This technique identifies regions in the range of the moderator variable where the effect of the focal predictor is significant and not significant. Results Framing Responsibility 5 In the first step of the regression, news framing was a significant predictor of attributions (B = 1.10, β = .59, p<.001). Moreover, a significant effect for attributional preference was observed (B = .24, β = .20, p<.01). The explained variance for both effects was R2 = .38. Adding the second block of the regression, the interaction term was marginally significant (B = -.32, β = .14, p = .08). The explained variance after including the interaction term was R2 = .40. In order to interpret this interaction, the effect of trait attributional preference was examined for the thematic and episodic conditions using the modprobe macro. In the thematic (t(103) = 3.08, p < .01) but not in the episodic (t(103) = .56, n.s.) condition, trait attributional preference was found to be a significant predictor. In contrast, the Johnson–Neyman technique revealed that news framing was a significant predictor across all levels of attributional preferences. However, this effect was much larger for low values of the trait compared to high values (see figure 1). Thus, Hypothesis 1 was supported, but Hypothesis 2—as it predicted an unmoderated relationship— was not. The tendency to perform dispositional attributions only had an impact on responsibility judgments when individuals were exposed to a thematic frame. Discussion Attributions of responsibility are not only shaped by the mass media’s framing but also by general attributional preferences. However, the effects of attributional preferences were only found in the thematic condition, not in the episodic condition. The thematic frame gave less information about poor individuals than the episodic frame. Thus, subjects had less information on which to base their responsibility attributions and, consequently, general attributional preferences exerted a larger impact. This suggests, the more judgment-relevant information a news frame provides, the more will subjects base their attributional judgments on the news frame, and the weaker are the influences by general personality traits. Framing Responsibility 6 The results are important in three aspects. First, surprisingly, no studies so far have explicitly replicated the Iyengar (1991) results. The present investigation has done so with respect to newspaper frames. Second, previous studies of framing effects have focused on the influence of media content, entirely neglecting the role of individual differences. This study demonstrated that individuals are not only influenced by media frames, but also possess general traits that influence their opinions and attitudes. In order to fully understand framing effects, this perspective is necessary and overdue. Third, the marginally significant interaction is a hint that the predictive power of traits may vary depending on the amount of judgment relevant information provided in a frame. There are limitations to this study. Only a single topic was used and the study relied on quite simple measures for the dependent variable. An extensive distraction task was applied before the trait measure, but an independent trait measure one week before or after the experiment would increase internal validity. Furthermore, no conclusions about long-term effects can be drawn. A control group without any frame exposure would allow greater insights. Finally, the concept of episodic and thematic framing is only one way of describing the mass media’s framing of issues. For issue-specific frames (e.g., Matthes & Kohring, 2008, Shen, 2004), other individual difference variables might be at work. Exploring the impact of these variables is a fertile area for further study. Framing Responsibility 7 ReferencesFletcher, G. J., Danilovics, O. P., Fernandez, G., Peterson, D., & Reeder, G. D. (1986).Attributional complexity: An individual differences measure. Journal of Personality andSocial Psychology, 51, 875-84. Hayes, A. F., & Matthes, J. (2008). MODPROBE: An SPSS and SAS procedure for probingsingle-degree-of-freedom interactions in linear regression models. Manuscript submitted for publication.Iyengar, S. (1990). The accessibility bias in politics: Television news and public opinion.International Journal of Public Opinion Research, 2, 1-15.Iyengar, S. (1991). Is anyone responsible? How television frames political issues. Chicago:University of Chicago Press.Iyengar, S. (1996). Framing responsibility for political issues. The Annals of the American Academy of Political and Social Science, 546, 59-70.Johnson, P. O., & Neyman, J. (1936). Test of certain linear hypotheses and the applications tosome educational problems. Statistical Research Memoirs, 1, 57-93.Kelley, H. H., & Michela J. (1980). Attribution theory and research. Annual Review ofPsychology, 31, 457-501.Matthes, J. & Kohring, M. (2008). The content analysis of media frames: Toward improvingreliability and validity. Journal of Communication, 58, 258-279.Rotter, J. B. (1966). Generalized expectancies for internal versus external control ofreinforcement. Psychological Monographs, 80, 1-28.Shen, F. (2004). Chronic accessibility and individual cognitions: Examining the effects ofmessage frames in political advertisements. Journal of Communication, 54, 123-137. Framing Responsibility 8 FiguresFigure 1Mean Attribution of Individual Responsibility as a Factor of Framing and TraitAttributional Preference 11,522,533,544,55 1,00 1,16 1,33 1,49 1,65 1,81 1,98 2,14 2,30 2,46 2,63 2,79 2,95 3,11 3,28 3,44 3,60 3,76 3,93 4,09 4,25Attributionsof Individual Responsibility Trait Preference for Dispositional AttributionsEpisodic FrameThematic Frame