The Influence of Fiction Versus Nonfiction on Political Attitudes

Abstract

An experiment was conducted to test the influence of fictional versus real events portrayed in videos (defined by the fiction versus nonfiction labels of the videos) on political attitudes. It found that the labels interacted with individual differences in need for cognition (NC) to influence participants' support for social welfare programs such that introducing the videos as fictional films versus documentaries produced more support among low-NC participants but had little impact among high-NC participants. The interaction effect was explained by the extent to which participants argued against the videos. The findings not only shed light on the relative persuasiveness of fictional and nonfictional media messages but also suggest how people may process these messages.

Keywords:

• Counterarguing
• Fiction
• Need For Cognition
• Nonfiction
• Political Attitudes

Notes

Studies have shown that public opinion has the potential to shape policies because policy makers tend to respond to public opinion via political action (Page & Shapiro, 1983; Stimson et al., 1995).

The evidence supporting an equal impact is stronger than that supporting greater influence of fiction or nonfiction because the finding of equal influence was replicated by a number of studies (Appel & Malečkar, 2012; Green & Brock, 2000; Strange & Leung, 1999). In contrast, the one study showing that nonfiction was more influential than fiction only found the persuasive advantage of nonfiction under some circumstances (Murphy, 1998). Although a few studies have found circumstantial evidence suggesting greater influence of fiction (Green et al., 2006; Prentice & Gerrig, 1999), only one study provided direct evidence for it, and the advantage of fiction was limited to unreliable information in the messages (Prentice & Bailis, 1995 cited in Prentice & Gerrig, 1999).

Research has also shown that even when readers are informed of the inaccuracy of the information presented in fiction and are motivated to correct it, they may still be unable to shield themselves from the influence of fiction (Green & Donahue, 2011).

The two videos were from documentaries Domestic Violence and Off to War.

Although established differences exist between documentaries and fictional films in terms of visual and auditory features (Pouliot & Cowen, 2007), these differences are not hard-and-fast distinctions (LaMarre & Landreville, 2009). Some features (e.g., close shots and narration) can appear in both fictional and nonfictional films. These features are the ones contained in the videos used in this study.

Time spent watching the video was recorded. Given that each video was about two to three minutes long, subjects (N = 5) who spent less than 1.5 minutes watching the video were considered as skipping the video.

Including participants who incorrectly recalled the label in the analyses did not change the findings.

The analyses did not find a significant three-way interaction among the genre labels, video topics, and NC on support for the social welfare programs. Hence, the presentation focuses on the main and interaction effects of the labels and NC.

The veteran video induced more support than the abused woman video (β = .18; p < .01).

Additional information

Notes on contributors

Xiaoxia Cao

Xiaoxia Cao (PhD, University of Pennsylvania, 2010) is an assistant professor in the Department of Journalism, Advertising, and Media Studies at the University of Wisconsin–Milwaukee. The author thanks the editor and two anonymous reviewers for many helpful comments.