Abstract
Identifying narratives within the news media is a challenging task. In this research note, we utilize the Narrative Policy Framework (NPF) to identify news media narratives and characters (heroes, victims, villains) on hydraulic fracturing in two local Colorado newspapers. We find the victim is the public and the villain is most often the oil and gas actors. Furthermore, the two newspapers present heroes differently: the conservative newspaper presents oil and gas industry actors as heroes while the liberal newspaper presents environmental organizations as heroes. This research note demonstrates the potential of the NPF to evaluate media reporting of contentious issues.
Introduction
The media have a responsibility as a source of scientific information for citizens as well as framing stories that depict different interpretations of events (Carvalho, 2007; Shanahan et al., 2011). The media can be understood as a conduit to transmit their issue in order to mobilize groups and citizens to achieve policy success (Shanahan et al., 2008) and the narratives in the news media have the potential to reinforce or convert an individual’s belief surrounding a contentious topic (Shanahan, et al., 2011). The standards intended to prevent bias reporting by journalists and editorial staff can create conditions that lead to distortion of the news (Antilla, 2005). These issues have led to the news media being accused of biased or selective coverage of events based on ideology and trust in the media has decreased.
Policy actors—those that are actively involved with the politics or governance of a given topic or issue—can strategically use narratives to influence policy debates and to expand or contain a coalition (Heikkila et al., 2014; Jones and McBeth, 2010). These narratives can be represented in the news media as policy actors seek to communicate to the general public or decision-makers. At this time, the use of characters within policy narratives presented by the news media is an understudied area of research. The Narrative Policy Framework (NPF) provides a methodological approach for using a content analysis to identify the elements of policy narratives (Jones and McBeth, 2010). However, theoretical and empirical contributions are currently limited regarding the elements of the NPF presented by the news media, along with the potential reporting biases associated with these elements.
The primary elements of a narrative as defined by the NPF include a setting or context, plot, characters, and moral of the story. 1 The character element of narratives includes identifying the role of heroes, villains, and victims. An issue with identifying the narratives within the news media is that actions are not clearly attributed to characters. As with most media reporting, an action is often accredited to an actor, but determining whether the action makes the actor a hero, villain, or victim is difficult. For example, when unconventional oil and gas development using hydraulic fracturing commences in an area, determining whether those involved are heroes, victims, or villains cannot be accomplished without introducing an opinion on hydraulic fracturing. In other words, when unconventional oil and gas development using hydraulic fracturing begins, does that make industry a hero or villain?
Given the central role of the news media in relating information to mass publics, exploring the media coverage of unconventional oil and gas development that utilizes hydraulic fracturing is of critical importance. Furthermore, empirical and theoretical advances are needed to further understand the narratives presented by the news media surrounding contentious environmental issues. To overcome the ambiguity and explore the use of characters presented within the news media, we completed a content analysis focusing on three issues involving hydraulic fracturing: the economy, environment, and public health. When an issue was found, the character could be identified as part of the problem being presented. Thus, in this research note, our goal is to examine local Colorado newspapers to determine whether coding characters only when an issue is identified as present can overcome this challenge in coding narratives and characters.
Methods
The setting for this case study is the hydraulic fracturing debate within the news media in the State of Colorado in the United States. To evaluate how the news media covers the different issues associated with hydraulic fracturing and the characters with each issue, we performed a content analysis of articles in two Colorado newspapers: the Boulder Daily Camera and the Colorado Springs Gazette. These newspapers were selected due to the varying political endorsements and their close proximity to active hydraulic fracturing sites. The Boulder Daily Camera is considered liberal having endorsed the Democratic candidate for the past three Presidential elections. The Colorado Springs Gazette is considered conservative where they endorsed the Republican Mitt Romney in 2012 and did not endorse a candidate in 2008 and 2004. In general, conservatives are more in favor of hydraulic fracturing while liberals are less supportive of the practice (Boudet et al., 2014).
We obtained the Boulder Daily Camera articles through LexisNexis and the Colorado Springs Gazette articles through Newspaper Archive. We used key search terms of “hydraulic fracturing”, “fracking”, “fracing”, and “shale gas”. The search dates spanned from 1 January, 2008 to 31 December, 2013. For the Boulder Daily Camera, 462 articles met the search criteria and 127 articles met the search criteria for the Colorado Springs Gazette. Our content analysis included articles selected through a random number generator and 66 articles were selected per newspaper for a total of 132 articles. The articles were briefly reviewed to ensure the articles included a substantial section regarding hydraulic fracturing. In addition, articles that were not news stories (e.g. editorials, commentaries) were omitted and replaced with another randomly selected article.
The documents were coded by one of two coders using a codebook similar to Heikkila et al. (2014), with the unit of analysis at the article level (versus the sentence or paragraph level). The codebook includes basic identification information about each article, such as the newspaper, title, author, and date. With respect to evaluating the risks and benefits of hydraulic fracturing, the codebook included variables to identify three issues associated with framing hydraulic fracturing: (1) “harms” or “benefits” to the economy; (2) “harms” or is “safe” for the environment; and (3) “harms” or is “safe” for public health.
When one of the three issues was found to be present in a document, the character linked with the issue was coded. To identify characters, coders looked for quotes or direct accreditation of a hero, victim, or villain action to one of the actor categories. The codebook included key actor categories involved with or concerned with hydraulic fracturing in Colorado: (1) government employees and elected officials; (2) oil and gas industry organizations and associations; (3) environmental advocacy groups or organizations; (4) general public/individual citizens; (5) none or the author of the article; and (6) other. The actors selected for this work are similar to the previous research conducted on interest groups involved with hydraulic fracturing in the State of New York (Heikkila et al., 2014). The Cohen’s Kappa was between 0.55 and 0.85 for all codes presented and the values are considered moderate to substantial agreement for the codes (Viera & Garrett, 2005).
Results and discussion
Table 1 shows the use of the NPF in terms of character by the Boulder Daily Camera (leans liberal) and Colorado Springs Gazette (leans conservative). The framing of characters was similar for the two newspapers for the victims and villains. In terms of victimizing characters, the prominent victim in both of the newspapers was the public. Boulder presented the public as a victim at a similar rate as Colorado Springs, 19 versus 15 times respectively. The prominent interest group that was vilified by both newspapers was the oil and gas industry.
Percent and count of articles with narrative policy character construction used by the Boulder and Colorado Springs newspapers.
Bold signifies statistically significant difference in presence of character class per article between Boulder and Colorado Springs newspapers, Chi-square p < 0.05.
The primary difference between the two newspapers was in the use of heroes. The Boulder newspaper used the narrative strategy of presenting environmental organizations as heroes while the Colorado Springs newspaper used the narrative strategy of presenting the oil and gas industry characters as heroes. This would align with the general political views of the topics that those with more liberal leaning beliefs are opposed to hydraulic fracturing and those with more conservative beliefs are in favor of the practice.
While this analysis focused on the State of Colorado, the results may be generalizable beyond this state or this issue. The NPF may be utilized as a tool to understand framing and narrative biases to evaluate other issues where narratives are important (immigration, welfare, etc.). We posit that the coding of issues and the actors affiliated with these issues may offer insight into how narratives—including the presentation of various coalitions as villains, victims, or heroes—can be presented by the news media.
Conclusion
This study finds differences in the news media presentation of heroes within narratives as defined by the NPF by liberal and conservative leaning newspapers. By coding the character associated with an issue, we found differences in conservative and liberal leaning newspapers in Colorado regarding the portrayal of heroes. In the conservative newspaper, we see a greater portrayal of the oil and gas industry actors as heroes, and in the liberal newspaper, we see a greater portrayal of environmental organizations as heroes. The future research objectives based on this research will be to conduct similar coding on a different political subsystem and to complete a comparative analysis between different environmental issues. This research note offers insight into the use of characters when a frame is presented within the news media and finds different patterns based on the political leaning of the newspaper.
Footnotes
Declaration of conflicting interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Alfred P. Sloan Foundation (grant number 2012-3-24). The research design and results are the authors’ alone.
