By Kirsten A. Johnson
WJMCR 28 (January 2011)
Introduction | Agenda-Setting | Hypotheses | Results | Discussion | Conclusion
AbstractThis study examined 329 posts made by citizens to CNN’s iReport.com about the 2008 presidential election. A content analysis showed a correlation between stories posted on iReport and stories covered by traditional media. The top stories covered by traditional media were the same stories written about by citizen journalists. Story tone was also explored, and John McCain received the most negative coverage on both the iReport site and by traditional media, followed by Sarah Palin, and Barack Obama. This study is an important step in exploring and beginning to understand the agenda-setting relationship between mainstream media and citizen journalism web sites.
Introduction
If the campaigning process across the twentieth century moved, if even roughly, from a candidate-centered to a media-centered and now to a citizen-centered communication process, we are at a point of a possible paradigm shift in American politicking.1
Citizen created content played a large role in the 2008 presidential election, as people shifted from passive consumers of media to active content creators. People went online to post stories using text, pictures, video and audio. According to Bimber2 we are in the midst of a fourth information revolution that is changing American politics. Bimber argues that this revolution has been brought about by new media. Evidence of this shift could be seen during the 2008 presidential election. Americans did not just passively consume campaign information, they also created it. Five percent of all adults reported posting their own political commentary to web sites, blogs, or online news groups.3
Even though citizen journalists posted stories about the election, traditional media still played an important role in informing the citizenry about the campaign. This study seeks to examine the intersection of citizen reporting of stories, reporting of stories by traditional media, and agenda-setting theory. The types of stories and the tone of the stories covered by both traditional media and citizen reporters on CNN’s iReport were examined to see if an agenda-setting effect exists. The relationship between traditional media and citizen journalism is an important one to understand as citizen journalism web sites become an increasingly popular way for citizens to express themselves. This study is the first one to look specifically at citizen journalism and agenda-setting during a political campaign.
Agenda-Setting
According to the agenda-setting theory, there is a correlation between how much the media covers certain issues and whether people perceive the issues as being important.4 Much research has been conducted in the area of agenda- setting and its relationship to politics and political campaigns.5
Agenda-setting among different news outlets (intermedia agenda-setting) has also been studied often. Intermedia agenda-setting posits that media outlets can influence the pattern of coverage by other media outlets.6 Intermedia agenda setting has not just been studied among traditional media, it has also been studied, although to a lesser extent, in the online news environment. Lim7 found that a prominent online newspaper had a strong impact on the agenda of another online newspaper and an online wire service. In another study of agenda-setting online, Roberts, Wanta and Dzwo8 studied agenda-setting theory in the context of online users and posts to an electronic bulletin board. They found that the media influenced people to not just take notice of issues, but also to post about them on electronic bulletin boards. In many cases the posts were made 1 to 7 days from the time that traditional media covered a story. In a study of the 2004 presidential election Sweetser, Golan and Wanta9 found that candidate blog posts were influenced by reports from mainstream media, reinforcing findings from previous studies that show traditional media continue to set the agenda. A study by Lanosga10 showed an agenda-setting effect between traditional media at a local level and local bloggers. There was a strong correlation in this study between what traditional media covered and then the posting of a reaction and/or discussion about the report on the blogs studied. In a study that explored the perceptions of bloggers who post political stories online Tomaszeski, Proffitt and McClung11 found that those surveyed saw their role not as generators of original content, but as those who comment and provide insight on stories reported by traditional media outlets. In a similar study of political bloggers Wallsten found they use their blogs to “express their political beliefs, to interact with like-minded people, to inform their readers, and to influence the political world around them.”12 Political blogs are often a mix of fact and opinion, links to material produced by others, and attempts to sway readers’ views.13
It is not just agenda-setting, but also the tone of media coverage that is thought to be important in shaping public perception of political issues.14 The tone of media coverage has become more negative.15 Negative coverage of a particular politician by the media can negatively impact peoples’ perceptions of that politician. In a study of the impact of media story tone on a senate race Druckman and Parkin16 found that the tone of media coverage can influence voting decisions. Studies show that traditional media coverage of candidates is getting more negative because the stories tend to focus on candidate character as opposed to issues.17
iReport.com
As Americans’ appetite for posting content online grows, sites are being created to support this citizen storytelling. One such site is CNN’s iReport.com, launched in 2008. This site offers a unique open environment where anyone can be a journalist and post a story.
The current iReport web site evolved from a previous attempt by CNN in 2006 at generating user created content for the station. Citizens were allowed to submit photos, video clips and text in hopes of having their work used by the news source on television or its web site, CNN.com. While this was certainly an attempt at becoming open to citizen journalism, it was still a selective process. Content had to be approved and chosen by CNN reporters, leaving other citizens’ work unheard. Despite the odds of having their contributions chosen by the news team, CNN’s iReport received close to 100,000 photos and video entries of news content in its first two years, with almost 10,000 of those sent in January 2008, just before iReport.com was fully launched in March 2008. Currently users can add any combination of hyperlinks, audio, video and up to ten still images with each post.18 About one million people a month visit the site according to Quantcast.com. It became especially popular following the controversial 2009 election in Iran. From June 13-18 iReport received 4,919 submissions from citizens. Posts on Iran helped the site to generate more than 11 million page views in 5 days.19 Stories posted by users on iReport don’t just stay on this site, they are also shared with other users. One of the all-time most shared stories on the site is about the typhoon that hit southern Taiwan. The story, posted in September 2008, was shared more than 5,000 times, and was viewed more than 9,000 times. The most viewed story of all-time on iReport is about thousands of Christmas lights on a house in Pittsburgh, Pa. This story was viewed more than 400,000 times and was shared more than 1,000 times according to iReport.com.
In many cases citizen journalists have been the first on the scene of breaking news events. For example, the 2007 shootings at Virginia Tech yielded 420 user generated video clips, and more than 11,000 stories on the California wildfires were submitted to CNN.20 Reporting the news is no longer left solely up to professional reporters; in other words, “freedom of the press now belongs not just to those who own printing presses, but also to those who use cell phones, video cameras, blogging software and other technology to deliver news and views to the world.”21
Civic and Citizen Journalism
Allowing citizens to have a voice in news coverage is not new. The civic, or public journalism movement, allows citizens’ concerns to help shape the news agenda.22 The presidential election in 1988 is often cited as the time civic journalism emerged. During this time some journalists were concerned that the news being covered was not of interest to citizens.23 Citizens also became unhappy with media coverage of the election because it lacked perspectives from citizens, seemed to pander to political consultants, and failed to seek a wide variety of perspectives about issues.24 The 1988 election also represented a blurring of the lines between political advertising and news coverage, a failure by mainstream news media to check facts, and a preponderance of news stories that focused on candidates as performers, as opposed to the issues that candidates supported or opposed.25 The practice of soundbite politics and politicians focused only on the image they portrayed through the media was also of concern.26 The growing discontent among citizens that sparked the civic journalism movement, along with the development of new technologies that allowed people to easily publish information online, led to the emergence of the citizen journalism movement. Citizen journalism is the idea that news content is produced by ordinary citizens with no formal journalism training.27 As more and more citizens go online and engage in content creation the number of citizen journalism web sites continues to grow. While it is difficult to track how many citizen journalism web sites are in existence, according to j-lab.org, there are now more than 800 of these sites.
Hypotheses
H1: There is an agenda-setting effect between the issues covered by traditional news media sources and stories posted on a citizen journalism web site.
H2: The tone of stories covered by traditional media influences the tone of stories posted on a citizen journalism web site.
This study aims to extend the role of agenda-setting into the realm of citizen created content, specifically posts on CNN’s iReport.com, a citizen journalism web site. It also seeks to build upon previous studies28 about agenda-setting in online news environments. In this study traditional media is defined as network and cable television, newspapers, and radio. CNN’s iReport.com is used in this study as an exemplar of a citizen journalism web site. This study also focuses on tone, which is important to examine because it can play an important role in shaping public perceptions of the news.29 Tone in this study is defined as the number of positive and negative comments made in each story.
Methodology
In order to assess the agenda-setting effect, and compare the tone of citizen journalism posts to traditional media, content analysis data collected by the researcher from iReport.com were compared to data collected by the Pew Center’s Excellence in Journalism Project Winning the Media Campaign: How the Press Reported the 2008 Presidential General Election.
A content analysis was conducted on 329 stories from iReport.com, CNN’s user-generated content site. The stories were all posted to the site from September 8 through October 16, 2008, these dates represent the day following the end of the Republican National Convention through the day after the final presidential debate between Barack Obama and John McCain. This time period was chosen because it represents a critical time in the presidential campaign and it coincides with the time period examined in the Pew report Winning the Media Campaign: How the Press Reported the 2008 Presidential General Election to which the data were compared. All the stories examined on the iReport site were tagged and grouped into the category 2008 election on the site. A total of 2,262 stories was posted during this time, and of those, 329 stories were randomly selected in order to achieve a confidence level of 95% and a margin of error of 5% . Using Cohen’s Kappa, intercoder reliability was calculated at .86 or above for the two coders on each of the variables.
In order to make the comparison of citizen stories on iReport and stories reported by traditional media, data were examined from a Pew Research Center’s Project for Excellence in Journalism report called Winning the Media Campaign: How the Press Reported the 2008 Presidential General Election. Data were collected from September 8 through October 16, 2008 and included 2,412 stories from 48 news outlets. Fourteen people coded the stories and reached a level of agreement of 80% or higher.30 In the Pew study the tone of the stories was determined by looking at the number of positive and negative comments made in each story. In order for a story to be coded as positive or negative it had to have 1.5 times the number of negative to positive comments to be coded negative, or vice-versa to be coded positive. This same method was also followed while coding the citizens’ stories on iReport.com. This coding procedure follows an established coding methodology used by the Pew Center.31
Results
This study compared stories posted on CNN’s iReport to stories covered by traditional media fromSeptember 8 through October 16, 2008. The section begins with an examination of whether there is a significant correlation between the top stories covered by iReporters and traditional journalists using Spearman’s rho. Next, differences in story tone between iReport and traditional media are assessed using a Chi-square analysis. The section concludes with an analysis of six distinct issues/events during the time period examined in this study that served to drive mainstream media coverage, and how that coverage related to reporting by citizen journalists.
Story Content
Stories on iReport were coded for story content into a number of categories that coincided with the categories used in the Pew research study.32 One category was added, an “undecided voter” category, to accommodate stories that did not fit into the Pew study’s classification of stories. See Table 1 for a list of categories and percentages. A Spearman’s rho computed for the top 10 stories covered by both traditional media and iReporters, excluding the “undecided voters” category, showed a significant correlation between the two lists, rs = .70, p < .03.
Table 1: Frequency of Story Topics on
Traditional Media & CNN’s iReport
Campaign Stories
|
Percentage of Coverage/Traditional Media
|
Percentage
of Coverage/iReport
|
Economy & Financial
Crisis
|
18%
|
13%
|
Presidential Debates
|
17%
|
16%
|
Palin-related Stories
|
14%
|
19%
|
McCain vs. Obama Polls
|
6%
|
9%
|
Candidate Attacks
|
6%
|
36%
|
McCain Suspends Campaign
|
4%
|
1%
|
Swing State Strategy
|
4%
|
0%
|
Voting
Issues/Irregularities
|
3%
|
.3%
|
Profiles of McCain
|
2%
|
0%
|
Iraq War as an Issue
|
1%
|
4%
|
Note: The traditional media data are from the Pew Research Center Study “Winning the Media Campaign: How the Press Reported the 2008 General Election”
On the iReport site Republican candidates were featured most often in stories. Palin was featured in 34% of stories, followed by McCain (30%), and Obama (26%). In terms of overall story tone, 68% of the stories were negative, 22% were positive, and 10% were neutral. When broken down by individual candidate, Obama had the largest percentage of positive stories written about him, followed by Palin, and McCain.
How prominently candidates were featured in traditional media stories was also explored. Obama was a significant presence in 62% of stories, McCain in 62% of stories, Palin in 28%, and Biden in 9%. A candidate was considered a significant presence if he or she was featured in 25% or more of the story content. Early in the 2008 campaign Obama received much more media attention than McCain. However, from the end of August through the months leading up to the election the two candidates received more equal coverage. In fact, when stories about Palin are considered, the Republican candidates received more coverage. However, for the Republicans that coverage came in the form of negative stories. In terms of traditional media coverage, McCain received much more negative than positive coverage between the GOP convention and the final debate, while Obama received more positive than negative press coverage. Table 2 shows the stories broken down into positive, negative and neutral stories about each candidate.
Table 2: Stories
Posted About the Candidates Based on Story Tone
|
|
iReport
|
|
|
Traditional Media
|
|
|
Obama
|
McCain
|
Palin
|
Obama
|
McCain
|
Palin
|
Positive
|
50
|
10
|
19
|
36
|
14
|
28
|
Neutral
|
0
|
0
|
2
|
35
|
29
|
33
|
Negative
|
50
|
90
|
79
|
30
|
57
|
39
|
Note: The traditional media data are from the Pew Research Center Study “Winning the Media Campaign: How the Press Reported the 2008 General Election”
A Chi-square analysis was computed to examine whether there were differences between the tone of stories reported about each candidate on iReport and traditional media. There were significant differences found for each candidate. When examining stories about Obama there was a significant difference found, c2(2, N = 200) = 42.6, p <.001. Cramer’s v was calculated to assess the strength of the relationship and was found to be .46, which indicates a strong relationship. A significant difference was also found when examining stories about McCain, c2(2, N = 200) = 37.1, p < .001. Again, a Cramer’s v was calculated and showed a strong relationship, .43. Stories about Palin were also examined, and a significant difference was found in terms of story tone for iReport stories and traditional media stories, c2(2, N = 200) = 42.7, p < .001. Cramer’s v was found to equal .46, which indicates a strong relationship.
Overall, traditional media paid an equal amount of attention to the Republican and Democratic candidates, but Obama received more positive press coverage than McCain. Even with the severe economic downturn during the campaign, the character of the candidates was an important part of the press coverage leading up to election day. As Obama’s poll numbers rose, so did the amount, and positive tone of his press coverage. As McCain’s poll numbers dropped so did the amount of media coverage he received, and the tone of that coverage was decidedly negative.33
Events Covered By Traditional Media and CNN’s iReport
There were six distinct issues/events during the time period examined in this study that served to drive mainstream media coverage. They included: the week following Palin’s nomination, the beginning of the economic crisis, McCain suspending his campaign and the first debate, the vice-presidential debate, the second presidential debate, and the third presidential debate.34
During the week following Palin’s nomination (Sept. 8-14) she was a significant presence in 45% of the stories posted on iReport. When compared to traditional media she was a significant presence in 52% of the stories. Also, 30% of the stories posted on iReport during this period were specifically about her.
The week of September 15-23 the economic crisis became a topic covered extensively by the mainstream media. The topic was also written about by iReporters. During this week 23% of the stories on iReport were about the economic crisis, and 40% of the stories covered by mainstream media were about the economy.
During the time period of September 24-28 McCain suspended his campaign, and the first debate took place. During this time period 42% of the stories on iReport were about the presidential debate and McCain was a significant presence in 45% of the stories. When examining mainstream media, McCain was a significant presence in 74% of the stories, however the economy continued to be the most covered story that week, with 23% of all stories focusing on the economy.
From September 29 to October 5 the time of the vice-presidential debate, 41% of the stories posted on iReport were about the debate, and Palin was a significant presence in 47% of the stories. Palin was a significant presence in 51% of the stories covered by traditional media.
During the time period from October 6-October 10 (the time of the 2nd presidential debate), the debate was covered in 12% of iReport posts. During this time period 61% of the stories on iReport were about candidate attacks. Obama was a significant presence in 57% of the stories, followed by McCain at 33%. In terms of mainstream media during this time period, candidate attacks were covered most often, and Obama was a significant presence in 80% of the stories, followed by McCain at 75%.
During the final time period in the study, from October 12-16 (the time of the final presidential debate), 26% of the stories on iReport were about the presidential debate. McCain was a significant presence in 48% of the stories, followed by Obama with 31%. When examining traditional media, the debate was also the story most often covered, and McCain was featured most often in stories at 72%.
Discussion
Research Question 1 focused on whether traditional news media sources have an agenda-setting impact on stories posted on a citizen journalism web site. It does appear that this is the case. A Spearman’s rho showed a correlation between the top 10 stories on the iReport site and those covered by traditional media. While there was a significant correlation, the stories were not in the exact same order. The top story covered by traditional media was the economy/financial crisis, while the top story covered by iReporters were stories about candidate attacks. This story distribution (candidate attacks) at the top, reflects iReporters’ propensity to write stories that largely drew upon their opinions and reactions to stories. The following post To McCain: Stop Attacks, Tell Us Your Plan written by “censorednews” is one example:
I am increasingly upset with John McCain and his campaign. America is spiraling into its worst financial crisis in decades, and instead of concentrating on telling the American people specifics of how he is going to help, all he has chosen to do is attempt to run smear ads and spread misinformation and lies about his opponent. This is a time for Americans to come together, not try to divide us through fear mongering. That is exactly what the Bush administration has tried to do for the pas several years, and people have "wisened" up.
When examining the timeframe for the study broken down by week, there was a clear pattern exhibited between when the mainstream media covered an issue and when the issue appeared on iReport. For example, during the week of September 15 the economic crisis was the story covered most often by mainstream media and iReporters. This post on September 15 by “gvscmr” called Lehman Brothers is typical of posts on the iReport site during this time period:
Lehman Brothers to file for bankruptcy – Wall Street received a huge jolt today when Lehman Brothers, a 158-year-old investment bank undermined by bad bets on real estate, said it was filing for bankruptcy.
- Lehman Brothers Made it through World War I
- Lehman Brothers Made it through World War II
- Lehman Brothers Made it through the Great Depression
- 150 Years of a powerful financial instruction weathering huge world changing events and remaining strong. It took only the remaining 8 years under George Bush to bring Lehman Brothers to Financial Ruin!
- Think about it! Eight is enough! We the People of the United States of America CANNOT Survive with 4 more years of George Bush under John McCain.
The above post begins with a statement of the facts about Lehman Brothers and at the end the story clearly attacks McCain. Many of the stories on the iReport site followed this pattern—a statement of fact followed by some sort of endorsement or condemnation of one of the candidates.
Another indication that agenda-setting may be occurring is that 55% of the stories on iReport specifically cited a mainstream media source. These mainstream media sources varied, and included The New York Times, The Wall Street Journal, CNN, and Fox News. It is important to note that there were many iReport posts that clearly used information from mainstream media sources, but did not cite them specifically.
In terms of how prominently the candidates were featured in the stories there was a difference between iReport and traditional media coverage. Palin was the candidate featured most often in iReport stories, whereas McCain and Obama were featured most often in traditional media stories. On the iReport site the citizen journalists seemed to enjoy waiting for the mainstream media to report something about Palin and then post something negative about her. The following story Poor Sarah, posted on Oct. 1 by “blueken”, is an example:
You have to feel sorry for Palin. Why was she picked? Why did McCain do this to her, the poor thing? Once again stumbling through an interview. It’s so sad. Here is a direct quote:
Is that coherent? Is it just me? Does that make any sense at all?Asked what newspapers and magazines she reads, Palin – a journalism major in college – could not name one publication
"I’ve read most of them, again with a great appreciation for the press, for the media," she said at first.
Couric responded, "What, specifically?"
"Um, all of them, any of them that have been in front of me all these years.""Can you name a few?"
"I have a vast variety of source where we get our news," Palin said. "Alaska isn’t a foreign country, where it’s kind of suggested, ‘wow, how could you keep in touch with what the rest of Washington, D.C., may be thinking when you live up there in Alaska?’ Believe me, Alaska is like a microcosm of America."
The second research question focused on whether the tone of stories covered by traditional media influenced the tone of stories posted on iReport. This study confirmed what previous studies have found, namely that mainstream media coverage tends to be more negative than positive.35 This negativity in coverage was also found on the iReport site. On both the iReport site and in traditional media reports McCain received the largest percentage of negative coverage, followed by Palin, and then Obama. A Chi-square analysis showed a significant difference in terms of the tone of stories covered on iReport and those covered by traditional media. Stories posted on iReport were significantly more negative and less neutral in tone than stories covered by traditional media journalists. This is not surprising, as traditional journalists are trained to report in an objective manner (although this may not always be achieved), whereas citizen journalists are not. On the iReport site many of the negative stories posted were very harsh and based on little fact. The stories often expressed the writers’ political viewpoints. The notion of “objective” reporting on issues existed in very few of the posts.
Conclusion
In conclusion, as was found in previous studies36 about online media content, traditional media still play an important role in setting the agenda. In this study, similarities were found between traditional media stories about the 2008 presidential election and stories posted by citizens on CNN’s iReport. The top stories covered by traditional media were the same top stories covered by iReporters. There was no indication of reverse agenda-setting in this study. In other words, it did not appear that stories posted on iReport had any impact on stories covered in mainstream media, in fact, just the opposite was found. In many cases when a story was covered by mainstream media citizen reporters used the facts from the stories reported, analyzed the information, added their opinion to the story, and then posted the story on iReport.
What became very apparent in this study was that citizen journalists on the iReport site were not generating original content based on research or information they had collected themselves, instead they acted as analyzers of information reported by mainstream media, or saw their role as making sure stories in the mainstream media that they felt were important were posted for iReport readers. This is very similar to what Tomaszeski, Profitt and McClung37 found in their study of political bloggers.
This study also confirms findings in the Roberts, Wanta and Dzwo38 study of those who post stories to electronic bulletin boards. They found that there was a very short lag time between when traditional media reported stories and when people posted those stories on electronic bulletin boards. On the iReport site, after major stories or events were reported by traditional media, either an analysis or reaction to the story was usually posted very quickly (sometimes in a matter of minutes as was the case when the financial crisis began).
As has been shown in previous studies39 tone plays a role in shaping public perception of political issues. A content analysis of stories on the iReport site showed that there were more positive stories posted about Obama as compared to the other candidates. When examining traditional media coverage there were also more positive stories about Obama than the other candidates. This pattern also continued in the other direction for McCain, with a majority of stories that were negative in tone appearing in both traditional media outlets and on the iReport site. It is interesting to note that when looking at the number of neutral stories, very few were posted about the candidates on iReport, whereas in traditional media around 30% of the stories for each of the candidates were neutral40. This could indicate that traditional media is doing a better job of trying to remain objective in their reporting, whereas the notion of objectivity is not something that many citizens journalists may understand, know, or care about.
One limitation of the study was that not all stories posted to the iReport site about the 2008 presidential election were analyzed—only those that had the 2008_election tag were included in the sample. If a user did not include this tag on his or her story, it was not aggregated into this category, and was not part of this study. A second limitation was that only stories posted on one citizen journalism web site, CNN’s iReport, were analyzed. Future researchers may want to consider including more citizen journalism sites in the content analysis to give a more complete picture of what citizens are posting. A third limitation was that the users’ political affiliations and real identities were not available. Only usernames were available so no concrete conclusions could be drawn specifically about party affiliations. Future researchers may wish to explore what motivates citizen journalists to post on citizen journalism web sites. These underlying motivations could be interesting in gaining a deeper understanding into what they write and why they choose to post what they write to an online audience.
Kirsten Johnson is an assistant professor of communicates at Elizabethtown College in Pennsylvania. She would like to thank all of those who have taken the time to read and comment on this article– your feedback was invaluable. This article is based on a paper presented at the 2009 AEJMC conference in Boston