First Monday

Agenda–setting, opinion leadership, and the world of Web logs by by Aaron Delwiche

More than 350 studies have explored the agenda setting hypothesis, but most of this research assumes a clear distinction between reporters and their readers. Web logs erode this distinction, facilitating participatory media behavior on the part of audiences (Blood, 2003). The activities of journalistically focused Web log authors give us new ways to understand and measure the agenda–setting process. While previous researchers have explored issue salience by focusing on audience recall and public opinion, Web logs invite us to consider hyperlinks as behavioral indicators of an issue’s perceived importance. This paper tracks news stories most often linked to by Web log authors in 2003, comparing the results to stories favored by traditional media. Arguing that Web log authors construct an alternative agenda within the admittedly limited realm of the blogosphere, I note that their focus has shifted from technology to broader political issues. My findings support Chaffee and Metzger’s (2001) prediction that “the key problem for agenda–setting theory will change from what issues the media tell people to think about to what issues people tell the media they want to think about”.


Literature review
Web logs and agenda–setting
Directions for future research




According to the Pew Internet and American Life Project, fewer than seven percent of Internet users maintain their own Web log, and only one in ten read blogs on a regular basis (Jesdanun, 2004). Nevertheless, online journals continue to make their mark on our cultural landscape. Many analysts cite the attacks of September 11 th as a watershed moment when Web logs emerged as a vehicle for personal journalism (Halavais, 2002a). With the fall of Trent Lott, Web log authors showed that they could also influence the agenda of traditional media (Shachtman, 2002). A new wing has been constructed in the Fourth Estate. Armed with little more than reason, wit and hyperlinks, certain members of the Web log community regularly present their own view of “the news that matters.” This sphere, situated between the mass audience and traditional journalists, raises important questions about the role of media coverage in constructing public discourse.

Forty years ago, Bernard Cohen (1963) famously observed that “the press may not be successful much of the time in telling people what to think, but it is strongly successful in telling its readers what to think about” (p. 13). Is this still true in an era when communicative power has been radically decentralized? Does the media agenda still influence the public agenda? In an attempt to answer these questions, this paper examines Web logs from the standpoint of research on agenda–setting. After briefly discussing theoretical foundations, I compare issues that dominated the Web log agenda in 2003 to those deemed most important by journalists and the public during the same period.



Literature review

When Cohen first suggested that media coverage affects the public’s perception of issue importance, communication researchers had grudgingly concluded that the effects of media content on public opinion were quite limited (Lowerey and DeFleur, 1988). Inspired by Cohen’s work, McCombs and Shaw (1972) asked Chapel Hill voters to identify the most important issues in the 1968 Presidential election. Comparing the results to news coverage during the same period, they found a strong relationship between media coverage and issues that respondents considered to be important. They were the first to refer to this process as “agenda–setting.”

The agenda setting paradigm was enormously attractive to researchers who had failed to identify direct effects at the individual level (McCombs, 1981). Since the phrase first entered the lexicon, close to 400 studies have been published on this topic. Despite the usual quibbling over definitions and methods, the majority of empirical studies support the claim that the media agenda affects the public agenda. As Dearing and Rogers (1996) noted after exhaustively reviewing the empirical work, “this relationship of the media agenda to the public agenda seems to hold under a wide variety of conditions, for a diversity of issues, and when explored with diverse research methods” (p. 92).

Agenda–setting research methods have changed over time. In the earliest studies, the media agenda was operationalized as the amount of coverage (measured in air time or column inches) devoted to a specific issue. The public agenda was measured by aggregate opinion polls that presumably expressed issue salience (McCombs and Shaw, 1972). If the two agendas could be significantly correlated, researchers inferred the existence of agenda setting effects.

Of course correlation does not constitute causality, and subsequent researchers identified a host of intervening variables that might affect agenda setting. In a recent essay, McCombs (2003) noted that personal experience, information–seeking behavior, and style of coverage are all contingent factors in the agenda–setting process. Attention to such factors has been accompanied by disputes over research methods. Is issue salience best measured by public opinion polls or audience recall? Are disaggregated data sets preferable to the larger data clumps used in classic agenda setting studies? Each of these issues is discussed in turn.

Issue salience. In an article that distinguishes agenda–setting from work on priming and framing, Scheufele (2000) claims that a memory–based model of information processing is the theoretical foundation of agenda setting research. “Mass media can influence the salience of certain issues perceived by the audience,” writes Scheufele. “That is, the ease with which these issues can be retrieved from memory” (p. 300).

In actuality, much work on agenda–setting is not premised on a memory–based model. Nevertheless, the appeal of this approach to a certain school of communication scholars is obvious. The ability to recall an issue is a cognitive behavior that can be measured at the individual level, and it is a more concrete mental operation than the vaguely defined concept of “an attitude.” Yet, there are reasons to question an approach that conflates memory with salience. We often remember important things, but not everything that we remember is important. Furthermore, we often forget or repress important details (Freud, 1951).

A more compelling measure of issue salience is found in Roberts, et al.’s (2002) exploration of agenda–setting effects among chat room participants. In one of the few studies to explore agenda–setting online, the authors operationalized issue salience as the decision of Internet users to discuss certain topics in an AOL chat room. Comparing media coverage to the topics discussed online, the authors identified agenda setting effects for the New York Times, Reuters and the Associated Press. Online conversation is a more reasonable indicator of issue importance than the ability to recall a topic, but the link to media exposure is tenuous. The authors provide no evidence that chat room participants even read the newspapers in question.

In an ideal world, researchers would measure salience by tracking specific behaviors that that articulate the importance of an issue while making an explicit connection to a journalistic source. For example, consider the following comments that might emerge in a conversation between ordinary citizens.

Each of these statements explicitly links a salient issue to an instance of media coverage. When such a connection arises in conversation or print, we refer to it as a reference or a citation. When it happens on the Web, we refer to it as a hyperlink. In blogs, a hyperlink is a measurable behavior that expresses issue salience by making a link to media content.

Disaggregation. A second dispute revolves around the shortcomings of aggregate data sets. In the earliest studies, researchers compared public opinion polls to content analyses of mainstream media messages (McCombs and Shaw, 1972). This approach made sense in the 1970’s when media options were limited. At the time, it was still possible to speak of a “mass audience.” More recent approaches have exposed individual audience members to carefully controlled media inputs (Iyengar and Kinder, 1987). As Dearing and Rogers (1996) observe, this trend toward disaggregation is facilitated by the development of computer–based tools for data collection and analysis.

A fine example of this approach is Althaus and Tewksbury’s (2002) investigation of individual–level agendas in persons exposed to print and digital versions of the New York Times. Disaggregating measures of the public agenda while also manipulating media content, the authors found clear agenda–setting effects. They also found significant differences between online and off–line versions of the newspaper, suggesting that print media were more effective in orienting the audience to a particular agenda. Though remarkably well–conceptualized, the study was limited by the artificial nature of its experimental design. The “disaggregated public” turned out to be little more than isolated individuals who responded to questionnaires in a laboratory. The findings are useful, but this approach extinguishes the sense of vibrant conversation that is vital to the public sphere.

The problem is not that agenda–setting researchers have relied too heavily on aggregate data; the problem is the type of aggregate data that they have used. A public opinion poll tells us what people think about a particular issue, but attempts to correlate this with patterns of media coverage are vulnerable to arguments about intervening variables. A stronger approach would be to aggregate all of those moments when individuals explicitly express a link between a political issue and media content. That is the approach taken in this study.



Web logs and agenda–setting

The past decade has been an exciting time for communication researchers. When Dearing and Rogers (1996) published their comprehensive review of agenda–setting research, Internet Explorer was less than a year old. Since then, the growth of the Internet has dramatically decentralized communicative power. Web logs are the most recent manifestation of this trend. At the same time, traditional media have embraced the potential of the Web. Newspaper stories and audiovisual broadcasts are no longer ephemeral. Today, they are archived, indexed, and fully capable of receiving links.

Situated between traditional journalists and the public, blogs disrupt the assumptions of agenda–setting research. For one thing, blogs undermine the distinction between the audience and the journalist. Commentators have variously termed this “pamphleteer journalism” (Gillmor in Lasica, 2002b), “amateur journalism” (Hiler, 2002), and “do–it–yourself journalism” (Halavais, 2002a). Blood (2003) describes blogs as participatory media forms that engage in something slightly different than journalism. Whatever term is used, it is clear that many blog authors ask questions about current events, read stories against one another, fact–check, and synthesize multiple accounts of reported events. Although the author’s personal voice is the foundation of every blog (Lennon, 2003), hyperlinks to external information are also vital to this form of journalism. As Hiler (2002) notes, the majority of the top 50 links on Blogdex are pointed toward traditional news stories.

Herring, et al. (2003) are properly skeptical of theorists who suggest that all blog authors are journalists. Yet, though small in numbers, journalistically oriented blogs play an increasingly important role as mediators between traditional journalists and those who read blogs.




In many ways, the world of blogs is a researcher’s dream. Time–stamped, searchable, and coded according to structural properties, blogs lend themselves to a range of research questions and analytical approaches. For example, many researchers are interested in locating the most influential nodes in the blogosphere. BlogRunner, BlogStreet, and Technorati rank blogs on a daily basis, though Gill (2004) notes that their algorithms are weighted toward those who post most frequently. An alternative technique (iRank) developed by the HP Information Dynamics Lab uses cluster analysis and time–stamps to identify blogs that function as “patient zero” in information epidemics (Adar, et al., 2004).

Such methods are useful in identifying opinion leaders, but agenda–setting is more concerned with ideas circulated through the network than with the individuals who circulate them. One approach is to track information propagated through the network. Using a customized Web crawler, Halavais (2002b) analyzed word bursts in 125 blogs over a two–week period to develop a list of leading topics on a weekly basis. Working in the same vein, developers of BlogPulse track the dissemination of URL’s, proper nouns, word bursts, and “blog bites.” (Glance, et al., 2004) Unfortunately, word bursts and “blog bites” are not helpful for this study. Computers can find patterns in miniscule chunks of data, but human beings require narrative context to make sense of information. When investigating possible agenda–setting effects, the most useful unit of analysis is the individual news story.

For this reason, Daypop was used to track the most frequently linked news stories in 2003. Characterized as “primitive but effective” for its reliance on link counting (Glance, et al., 2004), it is one of the only indexing tools that clearly delineates news stories from other types of content. This distinction makes it possible to examine the relationship between web logs and traditional media.

Using the Dapop archives, I extracted the top ten news links from every fifth day of 2003. This produced a sample of 781 news stories. The archives included the hyperlink, number of blog citations, the article headline, the date of publication, and the first sentence of the story (see Figure 1). When the original article was not available — or hidden behind a subscription gate — adjacency searching for words in the first sentence usually led to a usable copy of the article elsewhere on the Internet.


Figure 1: A typical top news page in the Daypop archive
Figure 1: A typical “top news” page in the Daypop archive.


In this study, all content was coded by the author. Reliability issues are a concern, but the use of a single author is an accepted approach in content analysis (Stempel, 1981). Articles were analyzed with standard content analysis techniques (Krippendorf, 1980). Because no pre–existing classification scheme could accommodate the unique range of topics, a recursive inductive process was used to develop a list of coding categories (Thomas, 2003). Stories were coded according to their primary theme. For example, a story about the efficiency of solar power was categorized “Environment — Alternative Energy.” An article about the discovery of a new planet was categorized “Science news — Astronomy.”




Understanding the blog agenda

The most frequently cited news categories of the past two years are depicted in Table 1. For each year, the column on the left identifies the most popular topics, while the column on the right shows the total number of related citations. Attentive readers will notice that a different method was used to calculate the rankings for each year. In 2003, Daypop began archiving the number of citations that pointed to each of the top–ranked stories. This makes it possible to more accurately gauge the popularity of linked items. Since this data was not available in 2002, the previous rankings were based on the number of times that a topic appeared in the daily rankings.


Table 1: Most linked news topics of 2002 and 2003

2003 (n=781)



1. Iraq war

2. Web logs

3. Copyright issues

4. Presidential campaign

5. News of the weird

6. Bush administration

7. Technology frontiers

8. Civil liberties

9. Deteriorating U.S.–Europe relations

10. Peace movement











2002 (n=451)



1. Copyright issues

2. Israel/Palestine

3. News of the weird

4. Technology frontiers

5. September 11

6. Bush Administration

7. File sharing

8. Computer security

9. Civil liberties

10. Blogs














Politicization of the blogosphere. When blogs first emerged, they were accurately perceived as the exclusive domain of computer experts. In 2002, technology topics still dominated the rankings, but a gradual shift toward political concerns was discernible. By the end of 2003, the ratio of political and technological news had shifted dramatically. Two–thirds of the top links pointed to political news, while approximately one–fourth covered technology trends. The expanded focus of blogs may be attributable to the advent of services such as Blogger and LiveJournal that lower the technical barriers to entry. Of course, greater attention to public affairs is also to be expected during the run–up to an election year and when the nation is at war.

Indeed, more than a fifth of all citations related to the Iraq war. Bloggers regularly linked to news about diplomatic initiatives, car bombs, pitched battles, and the pursuit of Iraqi leaders. Along with attention to the war itself, bloggers continued to express concern about the erosion of civil liberties. The sequel to the Patriot Act received extensive attention, as did loosened restrictions on search warrants, DARPA’s proposal for Total Information Awareness (TIA), and ongoing attempts to monitor the reading habits of library patrons. The deteriorating relationship between the United States and Europe ranked ninth, followed closely by articles about the nascent anti–war movement.

Technology. Reinforcing the impression that bloggers are narcissistic navel gazers, blogs were the second most popular topic. Copyright issues ranked third, which will not surprise students of Internet culture. This category included stories related to RIAA attempts to prosecute file swappers, congressional legislation, and digital rights management. Articles about innovative technological developments ranked seventh.

News of the weird. As in the previous year, a significant number of blogs pointed to articles about strange animal behavior, foolish criminals, and UFO sightings. Identifying a similar trend among the top one hundred forwarded news stories of 2003, the editors of the New York Times noted that there is an “abiding relationship between the Web and bizarre ephemera.” This seemingly innocuous finding reminds us that people share information because it is interesting, not just because it is important.

Global events. Excluding war coverage, news about other countries was conspicuously absent from the top rankings. Foreign events accounted for just three percent of total links. Significant world events during this period included upheaval in Liberia, the collapse of the Aristide government in Haiti, destabilization of the Chavez government in Venezuela, and the outbreak of SARS and the bird flu. Some of these stories charted on a daily basis, but they received little sustained attention. World news was only ranked consistently when it was linked to the war on terrorism. This is consistent with patterns of news coverage during the Cold War; events beyond America’s borders were deemed newsworthy only when they related to the conflict between the U.S. and the Russia.

Comparison to the media and public agendas

Having identified salient issues in blogs, the next step is to compare these findings to the agenda articulated by traditional media. In classic agenda–setting studies, the media agenda has been operationalized as the amount of coverage devoted to key issues (Funkhauser, 1973). Yet, the explosion of media inputs erodes the notion of a unified media agenda (Chaffee and Metzger, 2001). In this study, it was not practical to comprehensively sample a year’s worth of coverage. Instead, a rough view of the media agenda was obtained by analyzing “year in review” stories that listed the top events of 2003.

Table 2 compares the blog agenda with stories identified as “most important” by 287 editors polled by the Associated Press. Although the Iraq war ranked at the top of all three lists, there was little correspondence on other issues. The discrepancy could be explained by the fact that the units of analysis are not comparable. The blog agenda lists issues, while the media agenda ranks the top stories of the year. From an editorial standpoint, an important story might be one that sells papers or wins awards. From the standpoint of a blog author, a link–worthy story might be one that provides a basis for secondary analysis and commentary. The crash of the space shuttle Columbia was significant, but it did not have much staying power in the blogosphere. Although wildfires, the blackout in the Northeastern United States, and Elizabeth Smart yielded big headlines, there simply was not much for bloggers to say about these events.


Table 2: Comparison of 2003 blog agenda to indicators of the media agenda

Blog agenda

Associated Press — Editor survey

1. Iraq war

2. Blogs

3. Copyright

4. 2004 campaign

5. Weird news

6. Bush Administration

7. Technology

8. Civil liberties

9. U.S.–Europe relations

10. Peace movement

1. Iraq war

2. Columbia

3. California recall elections


5. Northeastern blackout

6. Economic recovery

7. Wildfires

8. Bush tax cut

9. Elizabeth Smart

10. 2004 campaign


Table 3 compares the blog agenda to issues identified as significant by respondents to a Gallup poll. The obvious divergence between the two lists may be partially explained by the fact that the lists were developed in very different ways. The blog agenda was induced from an aggregate of hyperlinks, while the survey respondents were forced to choose from a closed set of categories.


Table 3: Comparison of 2003 blog agenda to indicators of the public agenda

Blog agenda

Gallup — “Issue ranking” *

1. Iraq war

2. Blogs

3. Copyright

4. 2004 campaign

5. Weird news

6. Bush Administration

7. Technology

8. Civil liberties

9. U.S.–Europe relations

10. Peace movement

1. Education

2. Economy

3. Terrorism

4. Health care

5. Iraq war

6. Taxes

7. Budget deficit

8. Foreign affairs

9. Environment

10. Corporate corruption

* Gallup Poll. February 6–8, 2004. (“Now I am going to read a list of some of the issues that will probably be discussed in next year’s Presidential election campaigns. As I read each one, please tell me how important the candidates’ positions on that issue will be in influencing your vote for president – extremely important, very important, somewhat important, or not important. How about [see below]?”)





Situating blog authors in the physical world

Taken as a whole, blog authors were relatively unconcerned with world events unless they were explicitly linked to the war on terrorism. Some might argue that this is proof of agenda setting in a very broad sense. The lack of attention to foreign events could be perceived as a myopic register of news frames that dominate traditional media in the United States.

This finding challenges more optimistic predictions about the possibilities of global media. After all, extensive international coverage is available for free in the online versions of newspapers such as the New York Times and Christian Science Monitor. Web surfers have access to hundreds of overseas publications that cover world events in English. Meanwhile, in some countries, citizens are starved for access to news about the outside world. In nations such as China and Myanmar, access to overseas information is blocked with a combination of proxy servers and extremely harsh penalties. Such technological and political barriers do not exist in the United States; perhaps they would be redundant.

Blogs, the media agenda, and the public sphere

In 2002, when technology issues dominated blogs, deviation from the media agenda made sense. As the attention of bloggers shifted to political issues, one might expect that they would focus on issues foregrounded by traditional media. Clearly, this was not the case in 2003. Bloggers continued to be more concerned with invasions of civil liberties, they drew attention to the breakdown in relations between the U.S. and Europe, and they pointed to stories about anti-war activity. Articles about the Bush Administration tended to be negative, while analytical pieces about the war were more critical than supportive of the war effort [1].

It is intriguing to note that these themes have become more pronounced in media coverage during the past several months, even before the Abu Grahib scandal. It is possible that trends in the blogosphere were an early indicator of eroding support for the war in Iraq. This idea is not as far–fetched as it might seem. We have seen other instances in which web logs anticipate the media and public agendas (Regan, 2003), and some members of the intelligence community now scan web logs for insight into emerging trends (Tsuroka, 2004).

The intellectual foundations of our field may provide some clues about the role of blogs. In a classic study, Katz and Lazarsfeld (1955) suggested that media influence could be conceptualized as a two–step flow in which opinion leaders mediated between journalists and the publics. More recently, Brosius and Weimman (1996) have linked the two–step flow to agenda–setting research, arguing that agenda setting is a process in which influential individuals “collect, diffuse, filter, and promote the flow of information” (p. 564). This is a perfect description of journalistically focused blog authors (Branum, 2001).



Directions for future research

Agenda–setting research explores the flow of information from media outlets to the public. This paper has modified the traditional approach by inserting blog authors between the public and the press. Although the data does not support firm conclusions about agenda–setting effects or about the role of bloggers as opinion leaders, more investigation is warranted. A crucial next step is to find more comparable data sets. Narrowing the time frame of the study would make it possible to sample the media agenda and the blog agenda in identical ways [2].

We also need to know more about blog readers? Who are they? How many are they? How have blogs affected their consumption of traditional media? As Huesca and Dervin (2003) observe, we have little empirical data about how audiences consume hypertext media. Creative data collection methods will be required to answer these questions. Daypop and Blogdex report the relative popularity of sites among other blog authors, but there is no equivalent tool for tracking blog audiences.

It is possible that data collection methods for this side of the equation may not involve computers at all. Although new media channels lend themselves to an array of automated approaches, we should not limit ourselves to questions that can be answered with digital tools. If we rely too heavily on data that can be easily imported into software packages such as SPSS and InFlow [3], we risk lapsing into structuralism. Such an approach would reduce communication to the simple transmission of data while overlooking the vital role of human beings in constructing meaning (Carey, 1989).




The novelty of blogs has diminished during the past three years. Far from an intrinsically revolutionary technology, they are best understood as a part of a broader trend toward the decentralization of communicative power. Attempts at amateur journalism constitute only a small part of the overall blogosphere, but they have demonstrated their ability to affect the flow of information between traditional journalists and audiences. From the standpoint of agenda–setting, the most important thing about blogs is the way that they bridge these components of our public sphere.

In an article published shortly after his death, Steve Chaffee (writing with Miriam Metzger in 2001) argued that new media transform the assumptions of traditional communications theory. Anticipating the developments we have seen with blogs, he predicted that “the key problem for agenda–setting theory will change from what issues the media tell people to think about to what issues people tell the media they want to think about” (p. 375). This study suggests that he was correct. End of article


About the author

Aaron Delwiche is an Assistant Professor in the Department of Communication at Trinity University. He has worked as a technology consultant in both the private and public sectors and directed a team of interface specialists at one of Hong Kong’s leading Web design firms. Aaron maintains an award–winning site on propaganda analysis. His research interests include new media, youth culture, and global civil society. He teaches courses on media messages, multimedia design and criticism, film studies and video game theory.



1. Of course, a preponderance of pointers to a news story — whether positive or negative — does not prove that those linking actually agree with the content of that story.

2. Observing that unequal distributions characterize all social systems of sufficient size, Shirky (2003) argues that a small percentage of blog authors wield disproportionate influence. Some might suggest that such distributions make attempts to measure the blog agenda analytically useless. Yet, so–called power laws do not invalidate the methodological approach used in this study. As Ito (cited in Rosen, 2004) points out, “the tail of the blog distribution is much more spiky and interesting than the A–list.” If anything, link aggregation mitigates the issue by giving a citation in InstaPundit the same weight as a citation in a blog maintained by a high school student in Erie County, Ohio.

3. InFlow is one of many applications used to visualize and analyze social networks.



E. Adar, L. Zhang, L. A. Adamic, and R. M. Lukose, 2004. “Implicit Structure and the Dynamics of Blogspace,” In: WWW 2004 Workshop on the Weblogging Ecosystem: Aggregation, Analysis, and Dynamics, New York (18 May).

S.L. Althaus and D. Tewksbury, 2002. “Agenda Setting and the ‘New’ News: Patterns of Issue Importance Among Readers of the Paper and Online Versions of the New York Times,” Communication Research, volume 29, number 2, pp. 160–207.

B. Berelson, 1952. Content Analysis in Communication Research. New York: Hafner.

R. Blood, 2003. “Weblogs and Journalism: Do they Connect?” Nieman Reports, volume 57, number 3, pp. 61–63.

J.M. Branum, 2001. “The Blogging Phenomenon: An Overview and Theoretical Consideration,” at, accessed 20 September 2003.

H.B. Brosius and G. Weimann, 1996. “Who Sets the Agenda? Agenda–setting as a Two—step Flow,” Communication Research, volume 23, number 5, pp. 561–580.

J. Carey, 1989. “A Cultural Approach to Communication,” In: Communication as Culture: Essays on Media and Society. London: Routledge, pp. 13–35.

S.H. Chaffee and M.J. Metzger, 2001. “The End of Mass Communication?” Mass Communication & Society, volume 4, number 4, pp. 365–379.

M. Ceglowski and C. Marlow, 2004. “Upflux: An Open Index for Web Logs,” WWW 2004 Workshop on the weblogging ecosystem: Aggregation, analysis, and dynamics. New York (18 May).

J.W. Dearing and E.M. Rogers, 1996. Communication Concepts 6: Agenda–Setting. Thousand Oaks, Calif.: Sage.

A. Delwiche, 2003. “Reconstructing the Agenda in the World of Do–It–Yourself Journalism,” paper presented at 4th Annual Conference of the Association of Internet Researchers. Toronto (16–19 October).

K. Foot and S. Schneider, 2002. “Online Structure for Action in the September 11 Web,” paper presented at the 3 rd Annual Conference of the Association of Internet Researchers. Maastricht (13–16 October).

S. Freud, 1951. Psychopathology of Everyday Life. Chicago: Mentor.

K.E. Gill, 2004. “How can we Measure the Influence of the Blogosphere?” WWW 2004 Workshop on the Weblogging Ecosystem: Aggregation, Analysis, and Dynamics. New York (18 May).

D. Gilmor, 2003. “Moving Toward Participatory Journalism,” Nieman Reports, volume 57, number 3, pp. 79–80.

N.S. Glance, M. Hurst, and T. Tomokiyo, 2004. “BlogPulse: Automated Trend Discovery for Weblogs,” WWW 2004 Workshop on the Weblogging Ecosystem: Aggregation, Analysis, and Dynamics. New York (18 May).

A. Halavais, 2002a. “The Web as News: September 11 and New Sources of Web News,” paper presented at the 3rd Annual Conference of the Association of Internet Researchers. Maastricht (13–16 October).

A. Halavais, 2002b. “Blogs and the Social Weather,” paper presented at the 3rd Annual Conference of the Association of Internet Researchers. Maastricht (13–16 October).

A. Halavais, 2003. “Urban Sociology and a Research Agenda for the Blogosphere,” paper presented at the 4th Annual Conference of the Association of Internet Researchers. Toronto (16–19 October).

S.C. Herring, L.A. Scheidt, E. Wright, and S. Bonus, 2003. “Beyond the Unusual: Weblogs as Genre,” paper presented at the 4th Annual Conference of the Association of Internet Researchers. Toronto (16–19 October).

J. Hiler, 2002. “Blogosphere: The Emerging Media Ecosystem,” at, accessed 15 September 2003.

R. Huesca and B. Dervin, 2003. “Hypertext and Journalism: Audiences Respond to Competing News Narratives,” In: H. Jenkins and D. Thorburn (editors). Democracy and New Media. Cambridge, Mass.: MIT Press, pp. 281–307.

S. Iyengar and D.R. Kinder, 1987. News that Matters: Television and American Opinion. Chicago: University of Chicago Press.

A. Jesdanun, 2004. “Study: 2 to 7% of Web Users Keep Blogs” [Electronic version], Editor and Publisher (1 March).

B.M. Jenkins, 1981. The Psychological Implications of Media–Covered Terrorism. Santa Monica, Calif.: Rand Corporation.

E. Katz and P.F. Lazarsfeld, 1955. Personal Influence. New York: Free Press.

S.H. Kim, D.A. Scheufele and J. Shanahan, 2002. “Think About it This Way: Attribute Agenda–setting Function of the Press and the Public’s Evaluation of a Local Issue,” Journalism and Mass Communication Quarterly, volume 79, number 1, pp. 7–25.

K. Krippendorf, 1980. Content Analysis: An Introduction to its Methodology. Beverly Hills, Calif.: Sage.

P. Krugman, 2002. “The Other Face,” New York Times (December 13), p. 39.

J.D. Lasica, 2002a. “Blogging as a Form of Journalism,” USC Annenberg Online Journalism Review, at, accessed 12 October 2005.

J.D. Lasica, 2002b. “Web Logs: A New Source of News,” USC Annenberg Online Journalism Review, at, accessed 12 October 2005.

S. Lennon, 2003. “Blogging Journalist Invites Outsiders’ Reporting In,” Nieman Reports, volume 57, number 3, pp. 76-79.

M. Lombard, J. Snyder–Duch, and C.C. Bracken, 2002. “Content Analysis in Mass Communication: Assessment and Reporting of Intercoder Reliability,” Human Communication Research, volume 28, number 4, pp. 587–604.

S.A. Lowery and M.L. DeFleur, 1988. Milestones in Mass Communication Research: Media Effects. New York: Longman.

M.E. McCombs, 1981. “The Agenda Setting Approach,” In: D.D. Sidney and K.R. Sidney (editors). Handbook of Political Communication. Beverly Hills, Calif.: Sage.

M.E. McCombs, 2003. “Myth and Reality in Scientific Discovery: The Case of Agenda Setting Theory.” In: B. Dervin and S.H. Chaffee (editors). Communication, A Different Kind of Horserace: Essays Honoring Richard Carter. Cresskill, N.J.: Hampton Press, pp. 25–37.

M.E. McCombs and D.L. Shaw, 1972. “The Agenda–setting Function of the Mass Media,”Public Opinion Quarterly, volume 36, number 17, pp. 176–187.

L. Morano, 2002. “Propaganda: Remember the Kuwaiti Babies?” United Press International (February 26).

T. Regan, 2003. “Weblogs Threaten and Inform Traditional Journalism,” Nieman Reports, volume 57, number 3, pp. 68–70.

M. Roberts, W. Wanta, and T. H. Dzwo, 2002. “Agenda Setting and Issue Salience Online,” Communication Research, volume 29, number 4, pp. 452–465.

J. Rosen, 2004. “Will Mainstream Media Co-opt Blogs and the Internet?” paper presented at the World Economic Forum — Annual Meeting, Davos, Switzerland (22 January).

D. Scheufele, 2000. “Agenda–setting, Priming, and Framing Revisited: Another Look at Cognitive Effects of Political Communication,” Mass Communication & Society, volume 3, numbers 2/3, pp. 297–316.

N. Schactman, 2002. “Blogs Make the Headlines,” Wired News (23 December), at,1284,56978,00.html, accessed 12 October 2005.

C. Shirky, 2003. “Power Laws, Weblogs, and Inequality,” at, accessed 19 May 2004.

P.J. Shoemaker, W. Wanta, and D. Leggett, 1989. “Drug Coverage and Public Opinion,” In: P.J. Shoemaker (editor), Communication Campaigns about Drugs: Government, Media and the Public. Hillsdale, N.J.: L. Erlbaum, pp. 67–80.

G. Stempel, 1981. “Content Analysis.” In: B.H. Westley and G.H. Stempel (editors). Research Methods in Mass Communication. Englewood Cliffs, N.J.: Prentice–Hall, pp. 124–136.

D.R. Thomas, 2003. “A General Inductive Approach for Qualitative Data Analysis,” at, accessed 12 October 2005.

M. Wendland, 2003. “Blogging Connects a Columnist to New Story Ideas,” Nieman Reports, volume 57, number 3, pp. 94–95.

J.P. Winter and C.H. Eyal, 1981. “Agenda–setting for the Civil Rights Issue,” Public Opinion Quarterly, volume 45, number 3, pp. 376–383.


Editorial history

Paper received 14 October 2005; accepted 10 November 2005.

Creative Commons License
This paper is licensed under a Creative Commons Attribution–NonCommercial–NoDerivs 2.5 License.

Agenda–setting, opinion leadership, and the world of Web logs
by Aaron Delwiche
First Monday, Volume 10, Number 12 - 5 December 2005