Jae Kook Lee* & Jaekwan Jeong
Visit for more related articles at Global Media Journal
This study analyzes coverage of Iranian and North Korean weapons of mass destruction (WMD) by blogs and mainstream media, examining how the two types of media deal with international news, based on the theoretical framework of second-level intermedia agenda-setting. Attribute agendas of blogs are found to be positively correlated with the mainstream attribute agendas with regard to the issue of WMD coverage. The results indicate that despite many distinct characteristics, blogs cover international news in ways very similar to mainstream media.
In the contemporary cyberworld, blogging has become a prominent way to communicate. Blogs arguably function as news media providing information to significant numbers of audiences, although it is questionable whether the contents of blogs meet professional criteria of mainstream media journalism. The A-list blogs,1 commanding great attention online, have made their voices increasingly louder, as the size of blog audiences has exploded (Rainie, 2005). Blogs have gained extensive readership as a result of several important incidents: the 9/11 terrorist attack, the U.S. invasion of Iraq, and the 2004 U.S. Presidential election. Blogs function as news media that attract large audiences with the world of blogs—the blogosphere—being composed of constellations of individual blogs. With millions of different blogs, it is possible that many blogs replicate contents from other blogs as well as from mainstream media.
With regard to blogs, especially political or public affairs blogs, operating as news media with a significant number of readers, it is natural to question whether blogs’ coverage of news events is similar to, or different from, that of mainstream media. To answer that question, this study investigates coverage of weapons of mass destruction (WMD) with second level agenda-setting as a theoretical framework. Weapons of mass destruction in Iran and North Korea have been heavily covered both by mainstream media and by blogs since President Bush identified those two nations as members of the “axis of evil.” Because intensive coverage has produced large amounts of news reports in both blogs and mainstream media, it is possible to compare how the two media types cover a specific subject.
The goal of this study is to examine coverage of WMD in two different media from two different angles. First, this study compares the blog agenda and the mainstream media agenda at the attribute level for purposes of investigating the relationship of blog coverage with that of mainstream media. Second, the coverage of the Iranian WMD is compared with that of North Korean to find differences or similarities between the two. Based on comparisons from those two perspectives, the study seeks to contribute to an understanding of blog reporting with regard to international news. Further, intermedia agenda-setting at the second level is tested against global issues in the context of international communication.
Agenda-setting research has traced how people think about a variety of social issues and how likely their perceived importance of different issues corresponds with media coverage. Since the inception of the agenda-setting concept in the Chapel Hill study (McCombs & Shaw, 1972), researchers have refined the theory in fields such as contingent conditions and have further expanded applications to new arenas including attribute agenda-setting and consequences (McCombs, 2004).
With the advent of the Internet, the agenda-setting theory faces both opportunities and challenges in its application to new communication phenomena introduced by new technologies (Chaffee & Metzger, 2001; Takeshita, 2006), a prominent example being that of blogging. Because blogs are distinguished from traditional media in many different ways, the study of blogs’ news coverage offers yet another approach to comprehend the blog phenomenon. A combination of intermedia agenda-setting and attribute agenda-setting factors are viewed as effective tools for this investigation in that intermedia agenda-setting is useful to compare contents across media type while attribute agenda-setting offers advantages for comparing substantive information in news stories in greater detail. For those reasons, a brief discussion of intermedia agenda-setting and attribute agenda-setting is necessary.
Studies of the processes of agenda-setting have led researchers to examine origins of the media agenda that, in turn, has spawned studies of intermedia agenda-setting. The literature illustrates that the intermedia agenda-setting concept is useful to compare the effects of different media acting upon each other in terms of coverage.
A primary question in this line of research is “Who sets the media’s agenda?” (Turk, 1986). Studies on origins of the media agenda have revealed two main sources of influence: governments and other media. Major news media, especially The New York Times and the wire services, are found to influence other media in shaping their agendas. In a study of 52 Ohio newspaper and television journalists who were responsible for regularly selecting wire service stories to run as broadcast and print news, the decisions made by a small number of wire-service editors were found to greatly influence the news agendas of local media (Whitney & Becker, 1982). One U.S. leading newspaper has also been identified as leading the media agenda nationwide. In their study on the coverage of the drug issue, Reese and Danielian (1989) found that The New York Times exerts a significant impact on agendas of other print and broadcasting media. After examining four newspapers and three TV networks (ABC, CBS, and NBC), they concluded that The New York Times has great impact on shaping agendas of other newspapers and television networks (Reese & Danielian, 1989).
The original hypothesis of agenda-setting posits that salience of issues in the media is transferred to the public (McCombs and Shaw, 1972). During the process of the news coverage, media selectively emphasize certain issues rather than others which lead to increased salience of the issues in the public mind. Attribute agenda-setting, another aspect of agenda-setting that operates at the individual story level, employs a mechanism similar to that of the original agenda-setting. The hypothesis of attribute agenda-setting posits that salience of attributes of an issue is transferred from the media to the public (Ghanem, 1997; McCombs & Bell, 1996; McCombs, Llamas, Lopez-Escobar, & Rey, 1997).
Empirical findings are consistent in supporting the hypothesis of attribute agenda-setting. In a study of the 1995 Spanish regional and municipal election, McCombs and his colleagues (1997) found that attribute salience in news stories of candidates’ images were significantly correlated with the priority of attributes in people’s minds. In a follow-up study of the 1996 Spanish general election, McCombs and his colleagues (2000) reaffirmed the hypothesis of attribute agenda-setting. They found that attribute agendas of seven different media highly corresponded to voters’ attribute agendas for each of the three candidates (McCombs, Lopez-Escobar, & Llamas, 2000).
Blogs and News Coverage
As popular communication channels, blogs occupy an inter-connected space, the blogosphere, where people can express and discuss their opinions at minimal expense. The blogs’ potential for individuation has drawn scholarly attention to the blogosphere, in that each blog may seem to present a collection of idiosyncratic information and opinions. The uniqueness of blog contents, therefore, makes the blogosphere a testing ground for comparing news coverage by blogs and by mainstream media.
The blogosphere has expanded amazingly fast, with blogs, estimated at fewer than 50 in 1999, growing exponentially into thousands by 2000 (Mead, 2000). Around 2003, the estimated number of blogs skyrocketed again—to a range between 2.4 and 4.1 million (Henning, 2003; Wolff, 2003). A blog-tracking company, Technorati, Inc. reported almost 57.4 million blogs worldwide, as of October 2006. According to a January 2005 Pew Internet Survey, out of 120 million American Internet users, 27 percent or 32 million people say they read blogs (Rainie, 2005) and 7 percent of Internet users report owning a blog or web-based diary. The percentage translates into more than 8 million people. According to Truthlaidbear.com, a blog-tracking website, top blogs have almost 1 million hits a day. Thus, although blog audiences still constitute a minority, blogs collectively have a sizable audience.
Numerous blogs are connected to one another by hyperlinks whereby bloggers make references to other blogs. Hyperlinks provide blogs with a specific characteristic—connectivity. With hyperlinks, users are able to consume news material in multiple ways (Rich, 1999). By facilitating interactivity, hyperlinked texts enable the users to choose from an existing selection of pre-packaged information in a bilateral system (Jensen, 1998; McMillan, 2002). Collectively, blogs thus provide to the public a wide spectrum of information not available on an individual blog, an observation that supports the claim that, with regard to public affairs, blogs function as news media for much of the public.
Of particular interest in this study are the substantive characteristics of posts on blogs that function as news media. Based on recent empirical findings, we speculate that agendas are likely to become more homogeneous among public affairs blogs as well as between blogs and mainstream media. Studies of the blogosphere reveal that blogs have strong tendencies to refer to mainstream media news stories as sources of original information. The outgoing links of blogs to mainstream media news stories account for 38.6 percent of all outgoing links, the single largest percentage of sources that blogs mention followed by links to other blogs, official government sources, and so forth (Reese, Rutigliano, Hyun, & Jeong, 2007). Reese and his colleagues conclude that blogs depend heavily on news reports of mainstream news media in producing their posts. Iraq war blogs are found to have even stronger outgoing linkages to mainstream media. About 60 percent of all outgoing links to Iraq war blogs were linked to mainstream media news and editorials (Tremayne, Zheng, Lee, & Jeong, 2006). Given evidence of intermedia agenda-setting and the high tendency of blogs to hyperlink to mainstream media, it is highly probable that blogs and mainstream media share common attribute agendas. An investigation of blog agendas found strong correlations between political blogs and the mainstream media in the 2004 presidential election (Lee, 2006). Given those findings relate to object-level agendas, the results naturally lead to questions about the relationship between attribute agendas of blogs and of mainstream media.
The nature of blogging suggests a rationale for agenda-setting effects between mainstream media and blogs that operate in two ways. First, bloggers’ heavy dependence on mainstream news media as sources of information for their posts suggests that blogs rely upon limited resources for news coverage, having only one or a handful of contributors at best, with few exceptions. Resource limitations, in turn, lead to the second observation that the agenda-setting function of mass media can be extended also to blogs so that bloggers, who represent a subset of the general public, are not free from the psychological blanket effect of agenda-setting. Based on these observations, it is therefore appropriate to consider the intermedia agenda-setting effects that occur between the traditional media and blogs at the level of attributes.
The Issue of Weapons of Mass Destruction
Terrorism has obviously become a main issue that news media have dealt with since the September 11, 2001 terrorist attacks, logically prompting the news media to cover more intensively the issue of weapons of mass destruction, especially relative to Iraq’s alleged nuclear program. Since the fall of Baghdad, the focus of WMD news coverage has moved to other two countries, Iran and North Korea, and has resulted in the production of a large number of news stories available for investigation. Further, the issue of WMD offers many different aspects that can be studied with regard to the conceptualization of attributes in second level agenda-setting. In this sense, an investigation of the coverage of Iranian and North Korean WMD issues offers an interesting testing ground to compare coverage of different countries about the same issue, in that both nations are similarly situated to confront the U.S., but in different geopolitical contexts.
After the fall of Iraq’s President Sadam Hussein, North Korea and Iran have become the two most prominent nations in the eyes of U.S. foreign policy relative to WMD. Both countries, suspected to have ambitions to possess WMD, have come under heavy international scrutiny. Since the terrorist attacks on September 11, 2001, U.S. policymakers have increasingly turned their attention to North Korea and Iran. The U.S. government’s foreign policy has become focused on (1) the successful development of WMD and (2) potential transactions of WMD between the countries and terrorist groups, including Al Qaeda. Recently, the issue of WMD has taken a new twist relative to North Korea’s declaration of nuclear arms possession and Iran’s uranium enrichment program.
The first nuclear crisis in Korea began to emerge in the early 1990s when North Korea declared its intention to withdraw from the Nuclear Non-Proliferation Treaty. As a result of various interventions, including former U.S. President Jimmy Carter’s visit to Pyongyang, an agreement to freeze WMD development was reached in 1994. After North Korea admitted to a secret nuclear weapons program and then withdrew from the international Treaty on the Nonproliferation of Nuclear Weapons (NPT) in 2002, the situation became rapidly transformed into a second crisis (Bleiker, 2003) when, in early 2005, North Korea claimed it had successfully built and possessed nuclear weapons.
Iran’s nuclear program also entered the media agenda during this time period. Under the 2003 nuclear safeguards agreement, signed with the United Nations International Atomic Energy Agency (IAEA), Iran is required to provide the IAEA with extensive information about its nuclear activities and better access to sites. Iran has subsequently come under escalating international pressure to prove economic motives for its proposed nuclear energy program. Disclosures have been announced about Iran’s past failures to reveal work on uranium enrichment and plutonium separation and international suspicions have intensified that Iran is secretly developing nuclear weapons. Nevertheless, Iranian officials have emphasized that the country’s interest in nuclear energy stems from purely civilian goals.
The two countries are currently marching along similar paths with regard to international relations with the United States. Historically, both countries have experienced violent confrontations with the U.S. government: North Korea in the Korean War from 1950 to 1953 and Iran in the American hostage crisis in 1979. Both countries have been under strict economic sanctions imposed by the U.S. and have suffered severe economic failures. Both are believed to have ambitions to possess WMD. Yet, situations between the two countries differ significantly, especially with regard to geopolitics. Iran is deeply involved with the complicated Middle East political climate whereas North Korea is surrounded by four international powers—China, Japan, Russia, and U.S. forces in South Korea.
Against this background, an investigation of WMD news media coverage with the help of the attribute agenda-setting framework may effectively reveal the similarities and differences about media agendas regarding Iran and North Korea. Most new stories about WMD are found to be based on seven major attributes that may reveal the importance of the issue. (1) Conflict heads the list as the most frequently mentioned feature of WMD, a topic that Goldstein (2003) explored as the most crucial feature of a likely future conflict between states possessing WMD. (2) Security is another prominent aspect of WMD, regarded as a threat to international stability as well as national security (Harnisch, 2002). (3) The issue of WMD inevitably involves the process of negotiation between parties and their strategies, a topic that Smith (2004) identified as the problem of strategy in U.S foreign policy at the start of the Iraqi war. (4) Economy is another conspicuous item related to the WMD issue with recent moves by the Libyan government exemplifying the economic component of WMD development. Decade-long United Nations sanctions were lifted in September 2003 to reward Tripoli for its dramatic transformation from a “pariah state” to a partner in the war on terrorism and adherent to the international norms against the proliferation of WMD (Bahgat, 2005). (5) Since the terrorist attacks of September 11, 2001, terrorism has become one of the most prominent attributes of the media agenda related to WMD. Rademaker (2004) warned of the additional challenge of worrying about WMD technology and materials falling into the hands of terrorists who may be prepared to use such capabilities to wreak destruction and devastation. (6) There also are explanations that connect WMD with moral issues, such as democracy. It is indicated that, concerning Iraq, four stand out as particularly important: specifically, internal U.S. pressures, oil, WMD, and democracy (Rockmore, 2004). Brooks (2005) also dealt with democracy relative to WMD. (7) Science and technology are found as other, yet rare, components in the issue of WMD.
Hypotheses and Research Questions
Guided by understandings of blogs and agenda-setting, this study posits two hypotheses and four research questions.
H1: Attribute agenda in mainstream media coverage of WMD in Iran will be positively associated with blog coverage of WMD in Iran.
H2: Attribute agenda in mainstream media coverage of WMD in North Korea will be positively associated with blog coverage of WMD in North Korea.
RQ1: What is the relationship between attribute agenda about WMD in Iran and WMD in North Korea - in mainstream media coverage?
RQ2: What is the relationship between attribute agenda about WMD in Iran and WMD in North Korea - in blog coverage?
RQ3: Is the attribute agenda about WMD in Iran in blog coverage positively correlated with that about WMD in North Korea in mainstream media coverage?
RQ4: Is the attribute agenda about WMD in North Korea in blog coverage positively correlated with that about WMD in Iran in mainstream media coverage?
The present study compares distributions of WMD attributes from mainstream media coverage and blog coverage over a four-month period before and after the declaration of nuclear development in Iran and North Korea. A content analysis is applied to determine mainstream media and blog agendas about WMD news and to look for relationships.
This study performed content analysis of two network TV channels - CNN and ABC - and two nationally circulated newspapers - The New York Times and The Washington Post. The print media were chosen due to their strong effects on national audiences, as documented in previous content analysis studies. The TV channels were selected as representative agendas of television news, cables and networks. Justification for the broadcast media selection is based on literature that gives evidence of relatively homogeneous agendas across different broadcasting channels (Reese and Danielian, 1989). All news stories were coded for the period of November 1, 2004 to March 1, 2005. This four-month period included important timelines related to nuclear issues in Iran and North Korea. News stories of these four media were retrieved from the Lexis-Nexis database. The key words: “Iran and Nuclear,” “North Korea and Nuclear,” “Iran and Nuke,” “North Korea and Nuke,” “Iran and WMD,” and “North Korea and WMD.” The search was restricted to headlines and lead paragraphs. These search commands produced 153 stories about Iran and 167 stories about North Korea were found on CNN; 25 stories about Iran and 23 stories about North Korea were found on ABC; 143 stories about Iran and 170 stories were collected from The New York Times; and 81 stories about Iran and 97 stories were found on The Washington Post. The unit of analysis was an individual news story. The stories were coded for their issue attributes. A maximum of three attributes per story were coded and identified in case of multiple attributes contained in one story. The numbers for coded attributes were totaled to determine media agenda of WMD attributes.
A content analysis on the Internet is problematic because it is almost impossible to capture the true nature of an audience population (Rossler, 2002; Schafer, 2002). There is no comprehensive directory of complete samples since the contents of the Internet are in constant flux. Therefore, researchers sometimes rely on alternative methods that provide the most extensive available directory of websites, such as search engines. This study chose a blog-specific web directory. The web directory (www.truthlaidbear.com) provides a list of individual blogs taken from a list of 59,000 blogs searched by the website. The web directory also provides a daily-updated list of rankings among those 59,000 blogs, based on their citation from other websites and blogs. The top 150 blogs, from the list of rankings, were selected and content analyzed. Each blog provided a search function from its electronic archives of posts. The same key words used for finding media news stories were used to identify and search relevant blog posts in an effort to guarantee compatibility with the data from news media coverage.
The coders eliminated some of the stories that were not relevant to this study. For instance, although some of the stories included the key words, such as “Iran and Nuke,” the stories dealt with different issues. After the elimination process, a total of 689 stories were analyzed from the four news media and 150 blog sites. Table 1 summarizes the number of attributes analyzed from selected media. The unit of analysis for the blog agenda was a blog post. The same coding procedures used for news media content analysis were also applied to blog posts.
Table 1 about here
Ten percent of mainstream news and blog stories coded by two authors were selected and compared to calculate intercoder reliability. Scott’s pi was used for this study because the formula makes it possible to adequately control coder agreements by chance (Riffe, Lacy, & Fico, 1998). The reliability was .91.
A comparison of selected stories by countries showed that both news media and blogs covered the two countries during different time periods. As Figure 1 illustrates, the coverage of WMD stories about Iran reached a peak during mid-November while coverage about the North Korea peaked during mid-February. At the same time, when coverage of one country was at its peak, coverage of the other country also increased. This pattern suggests that WMD issues of both countries are distinctive, if related, concerns for the U.S.
Figure 1 about here
Before testing hypotheses and answering research questions, it is worth paying attention to distribution of WMD attributes by country. As summarized in Appendices 2 and 3, distributions of WMD attributes by country and media types are generally similar to each other in rank order. While the number of attributes analyzed between Iran and North Korea were fairly equal (685 of Iran and 511 of North Korea), those between mainstream news media and blogs were skewed in favor of the news media due to differences in the number of news reports per day (869 for traditional news media and 327 for blog sites).
Table 2 about here
H1 predicted attribute agenda in mainstream media coverage of WMD in Iran would be positively associated with that of blogs’ coverage of WMD in Iran. Table 2 summarized distribution of issue attributes and their rank order between mainstream media and blogs. The results indicate that mainstream media and blogs were similar to each other in the coverage of the given issue, with blogs putting more emphasis on the terrorism attribute and less emphasis on moral issues. Therefore, H1 is supported (Spearman’s rho = .75, p < .05).
Table 3 about here
Table 3 summarizes the results of H2. The second hypothesis predicted a positive correlation between the mainstream media agenda and the blog agenda about WMD in North Korea. Similar to the comparison of coverage of Iran, strategy and security attributes were the most important concerns of both mainstream media and blogs, although blogs covered more terrorism attributes. Compared to Table 2, blog coverage tended to pay more attention to personally-related issues, such as terrorism, and less attention to policy related attributes. In general, the correlation between the WMD attributes of North Korea among media and blogs was significant (Spearman’s rho = .91, p < .01). H2 was supported.
The first two research questions were set up to investigate the relationship of attributes between the two countries by the media. RQ1 asked the relationship between mainstream media agendas about WMD in Iran and in North Korea. Table 4 summarizes the results of the analysis. The rank order as well as the percentage distribution of attributes tracks a near perfect correlation between the variables (Spearman’s rho = .96, p < .01). The result suggests that mainstream media have consistent attribute agendas about WMD for both countries.
Table 4 about here
RQ2 asked the relationship between the blog agendas about WMD in Iran and in North Korea. Again, a significant relationship was found between two variables. As summarized in Table 5, the rank orders of issue attributes between the two countries were highly correlated with each other (Spearman’s rho = .94, p < .01). In addition, more terrorism attributes were found in blog stories about the North Korea, while conflict attributes were higher in blog stories about Iran.
Table 5 about here
RQ3 asked the correlation between the blog agendas about North Korean WMD and the mainstream media agenda about Iranian WMD. An analysis of Spearman’s correlation showed a considerably high level of correlation between two variables (spearman’s rho = .88, p <.01). Refer to Table 6.
Table 6 about here
Table 7 summarizes the last question, which posits a correlation between the blog agendas about Iranian WMD and the mainstream media agenda about North Korean WMD. The rank order correlation showed a relatively high level of association between the two variables (Spearman’s rho= .79, p <.05).
Table 7 about here
This study examined how two different types of news media —political or public affairs blogs and mainstream media—covered the issue of WMD by applying intermedia attribute agenda-setting to the coverage of WMD in Iran and in North Korea. The tests of hypotheses and answers to research questions contribute to our understanding of blogs that have emerged as a fresh format of news communication.
Findings in this study indicate that, in terms of WMD coverage, the mainstream media agenda is positively correlated with that of the blog agenda. The six correlations designed to examine association between the blog attribute agenda and the mainstream media agenda were found to be all statistically significant and very strong.
Two relationships between mainstream media and blogs (H1 & H2) showed particularly strong correlations. The values of these correlations were +.75 and +.91 suggesting that blogs cover WMD issues in a manner very similar to the mainstream media, despite the unique characteristics of blogs. This result appears to stem from the fact that blogs seldom gather news independently but instead depend heavily on mainstream media for news. Blogs may have different points of view in processing raw materials of news or in expressing opinions about them. However, as the present study suggests, blogs think mostly about the same attribute agenda as the mainstream media, with regard to foreign affairs issues.
We also found that coverage of Iran is correlated with that of North Korea, both in blogs and in mainstream media. The two relationships between coverage of Iran and North Korea in blogs and media (RQ1 & RQ2) showed strong correlations, the values of which were +.96 and +.94, respectively. This finding suggests that, despite geopolitical differences between the two countries, blogs and mainstream media appear to cover the issue with very similar agendas because of the two countries’ similar situations in confronting the U.S. government. In further investigations of association between WMD coverage of blogs and mainstream media (RQ3 & RQ4), the blog agenda and the mainstream media agenda once again showed strong correlations. The values of these coefficients were +.88 and +.79.
However, the results of this study are not sufficient to claim an agenda-setting effect from traditional media to blogs. Because this study showed only correlations between the mainstream media and blog agendas, future studies are needed to address issues of time order and functional relationships for the argument of causality. In search of time order, many studies have made numerous efforts to measure the optimal time lag for agenda-setting. As Roberts and colleagues (2002) suggested, the time lags in online and offline environments are likely to be different. Therefore, measuring the optimal time lag for agenda-setting from mainstream media to blogs presents another challenge for future studies.
The interpretation of the present study should be limited to international news coverage of the two media types. The association between the blog and the mainstream media attribute agendas relative to international news does not refer to associations about news in general for reasons that may stem from the characteristics of international news. Blogs are most likely to depend on mainstream media for the coverage of international news, because the news is one of the most difficult genres to produce without extensive sources of production.
Despite these limitations, the current study finds that blogs cover WMD news in ways that are very similar to those of mainstream media. Further studies in different settings, such as election news, may illustrate more diverse aspects of blogs, as vehicle for the dissemination of news.
1 A-list blog is a casual term that usually refers to the most popular blogs in terms of the number of incoming links (See Herring et al., 2005).