[An early version of this paper was presented at the Media in Transition Conference at MIT on October 8, 1999. This revised version was completed in 2001.]

Media decisions as to what newspaper stories should be published, and therefore reach the public, and, of those stories, which should receive prominent display have often been described as elements of the editorial "gatekeeping" process. This term figuratively refers to the editor(s)' content selection process, opening and closing the news gates that determine the flow of information to the audience.

The sociological concept of gatekeeping dates back to Kurt Lewin's 1947 ideas on social influence. He asserted that creating widespread social change could be achieved by focusing on people in key positions of influence who functioned as "gatekeepers" of the flow of ideas.[1]

Paul F. Lazarsfeld had earlier noted the intervening factor of interpersonal interaction in the development of public opinion in his "The People's Choice" study of the 1940 election. Lazarsfeld et. al. reported that "molecular leaders" held influence among their acquaintances "even though they were not necessarily prominent within the overall community."[2] Charles Berger and Steven Chaffee note that Katz and Lazarsfeld's seminal work, Personal Influence (1955), crystallized the social-mediation effects framework in their "description of the movement of information and influence from media to recipient as a 'two-step flow' mediated by group networks and opinion leaders."[3]

However, before societal group opinion leaders exert influence on general public perception, they, themselves, may be limited by the topics that are put before them, an agenda of issues that is controlled to a degree by the media. Sociologists and media scholars trace the concept of "agenda setting" to Bernard Cohen's comments on the role of the media: "The media may not be successful much of the time in telling people what to think, but they are 'stunningly successful in telling people what to think about.'"[4] Studies of the 1960s and 1970s tested the hypothesis that topics the press emphasized would become those that the public found most important, beginning with McCombs and Shaw's study on the 1968 presidential election.[5]

More recently, Roberts and Bantimourodis considered the evolution of agenda setting and gatekeeping concepts as they researched the flow of news in the international setting. They also explored the phenomenon of intermedia agenda setting, in which they suggested some news media, including wire services and global broadcasters (such as CNN and BBC's World Service Television), establish the diet of news and issues that other media will disseminate.[6] Of course, this news food chain is partly determined by budgets and the necessity for smaller news outlets to purchase and/or rely on news services and syndicates for international coverage.

Shepard suggests that gatekeeping was a role not just attributed to the media by analytical, academic observers, but one that members of the press had assumed as an obligatory mantle: "For years the mainstream media considered themselves the gatekeepers, the wise ones with the good sense to decide what the public needed to know and what it would be better off without." She also argues that the press corps, pre-Wilbur Mills and the Tidal Basin bombshell, had behaved in a paternalistic fashion, deciding what to print and what to withhold and sometimes ignoring politicians' private and sexual peccadilloes.[7]

Media scholars have examined not only the gatekeeping phenomenon and its effects but also the process through which it occurs. Clayman and Reisner conducted a study of the gatekeeping function evident in contemporary editorial conferences, focusing on the social process in addition to internal news value criteria used by decision makers. The authors explained that in the group editorial meeting two main factors influenced what the leading editor (often the managing editor or front page editor) would select for publication. The two factors were the story presentation process by each editor, including strong and favorable evaluations of newsworthiness, and the sequence of stories within that presentation. They observed that these influences were factors in article placement and prominence (such as on the front page, above or below the fold, and as the "lead" story). Clayman and Reisner asserted, "Gatekeeping itself is fundamentally a social and collaborative process." They suggested that researchers analyzing these crucial news content decisions review not only what an editor might internally deem newsworthy but also how the social and public process of the editorial conference functions.[8]

The Changing Nature of Gatekeeping in the New Media

Although many have accepted the gatekeeping metaphor as an accurate representation of this decision-making process, journalists, among others, have noted the gradual changes in their role created by the explosion of audience-seeking news sites. Newsweek columnist Jonathan Alter recently lamented the slow fading of the authoritative gatekeepers' voices that used to rise above the din: "But for most of this century there was a greater order to the babble than exists today. Journalists and those who influenced them served as gatekeepers, deciding what the people could know."[9]

Alter reflected on a "new news" that "now comes from a thousand mouths." And, he warned, "If everything's for entertainment, everything shrinks in significance, and we risk losing ourselves in the new American wilderness of noise."[10]

Alter is probably correct that media divergence and popular access have created a "wilderness of noise," an exponentially expanding marketplace of news and opinion, albeit sometimes hard to distinguish from each other. Yet, many of the daily newspaper Web sites that have burgeoned over the last half decade have emerged from within a traditional daily newsroom structure, still carrying its trappings of reportorial style and editorial gatekeeping. As with any new medium, or, in this case, hybrid medium, there are differences. This upstart cyber offspring is more sociable, that is, more interactive, than its paper counterpart; it is more attractive than its graying parents; and it is more independent, often cut loose from ritualistic, formal meetings about its content and display. As a child of the late twentieth century, the newspaper Web site is electronically sophisticated and keenly aware of its earning potential, tracking daily who reads it, what they read, and how often they read or click through.

Studying the News Decision Making Process

In order to investigate how newspaper Web sites make content decisions, I have examined a range of daily newspaper content and display decisions through selective content analysis and probed the article linking and placement process through an informal survey of online editors. When discussing this developing hypernews format, we can overlay the terminology of print-side news production on the gatekeeping process as it plays out in the electronic sphere. Visually, of course, paper pages become computer screens; "above the fold" transfers to pixel and screen size. Yet, electronic newspaper pages retain some print medium features. On those screens, one can still see a combination of type, photographs, and graphics. Stories are still stories, and writers are still writers.

A noticeable difference in online vernacular is that Web page viewers more often are referred to as the "audience" than the "readers." Reading or showing interest in an article is defined as clicking on the story link or a "hit." Thus, a successful article may have a thousand or more hits in a day (for a medium-sized daily newspaper). Circulation has been transformed to the number of page views or page impressions, that is, how often any reader opens the Web page. Editors and advertisers both seek golden "click through" rates, or the number of times a reader actually double-clicks to see more of an article or advertisement. Most advertising rates on the Web continue to be measured in CPM (cost per thousand) units, but with the thousand representing numbers of people opening to the Web page not putting their quarters into a news box.

Herein lies one of the strategic differences between print and electronic media in acquiring demographic and circulation data. The quarter-dropper is nameless, faceless, and enigmatic in her or his reading habits. However, the clicker, although not known by name (in most cases), can be tracked more closely. Most Web page nerve centers are able not only to count each page viewer but also to record at what points the reader/viewer enters and exits the Web site - at what link. So, if Reader A opens the main Web page, links to an article on a stranded pet lizard and exits without further perusal of the nation's political and economic news, the top sports scores, or what's happening in county government, the wizards behind the magic monitor are alerted. Maybe the prominently displayed county government story looked too dull. Perhaps the tax cut was an old issue. Or, could it be that Reader A really only cares about reptiles in distress? The individual case may not jump out at the bean counters, but the aggregate data pile up, designating "what's hot" and "what's not," by article, section, and time of day, for editors to contemplate.

The news media always have run surveys and reader polls to ascertain what readers read, what they enjoy, and what will draw them to peer closely through those gated news box windows. (The old joke is that hardworking young reporters are often horrified to learn their readers are most faithful to the obituaries and the comics.) With the electronic delivery of news, however, the decision-making powers that be have greater and more specific information about what finds favor with their audience. And, as they begin to sift through reams of statistical printouts, what impact does this knowledge have on content selection and display?

Methodology

To examine the content selection process, I began with traditional newsroom criteria for selecting top stories of the day: a significant national issue; a significant international issue; breaking news; an article of local interest or about a community issue; reader interest (a story that draws a reader in and that the reader will likely read); a strong article package (including good photos, graphs, or illustrations). Of those criteria, local news and breaking news inevitably rise to the top.

If all politics is local, then surely all coverage of politics, and other issues close to the heart, is local, too. Newspaper Web sites have been touted for their ability to bring local news to the reader electronically, at all times of day, not just in the driveway at dawn.

J. D. Lasica remarked, "The irony is, as online papers acquire a global reach, they need to become even more local."[11] Local stories and local angles on national stories are surfacing as popular on newspaper Web sites and, in the traditional gatekeeping sense, important for community education and awareness.

Breaking news, naturally, is the area in which online news sites have the greatest one-upmanship over their in-print rivals and, perhaps, their long-time broadcasting adversaries, as well. Felicity Barringer, in an article in the New York Times, quotes Rich Jaroslavsky, managing editor of The Wall Street Journal Interactive Edition, praising the renewed competitiveness that online news sites offer: "The Web allows newspapers to do both - to be competitive on a time basis with television, and to provide perspective as well." Barringer also offers the example of Web site staff members at the Chicago Tribune who believe "what they can do best is cover breaking Chicago news because that is where their resources are concentrated."[12]

Survey of Online Newspaper Staff

"Breaking news" and "local news or community issue" were ranked as the most important factors in a pilot survey of Knight Ridder, Inc., newspaper Web staff that I conducted in September 1999. The Knight Ridder constellation (then comprised of 28 newspaper Web sites) provided an intriguing cross-section of electronic journalism, including a few large dailies (Detroit, Philadelphia, Miami) and many medium-sized dailies from a range of locations across the country. The survey, with a 36 percent response rate, was directed to the online editor or other staff member most responsible for making article content selection and display decisions.

Respondents ranked the traditional content selection criteria (discussed earlier) in terms of most important factor (1) to least important factor (6) when "selecting stories given major play (receive most prominent placement) in the online edition." Fifty percent of respondents ranked "breaking news" as the most important factor; 40 percent listed "local interest or community issue" as the most important factor. Ten percent ranked "reader interest" as the most important factor. In open-ended comments, one respondent noted that the emphasis of the Web site was almost entirely local. Another simply asserted, "Breaking news is king!"

One survey section provided a breakdown of reader interest factors. Respondents were asked to rank the most important to least important factors based on four categories of reader interest: "Readers will be curious about the subject of the article; readers will be educated by the article; readers will be entertained by the article; readers will likely respond to the article" (via letters to the editor, online forum, or online poll).

"Readers will be curious" received the most frequent first-place ranking at 44.4 percent. "Readers will likely respond" received first-place ranking from 33.3 percent of respondents. The other two choices, "readers will be educated" and "readers will be entertained," received 11.1 percent each as first-choice selections. One respondent remarked that newspapers needed to be more responsive to their readers' preferences: "I think the paper chooses what readers should want, instead of what they've shown they want. We can't ignore what readers are actually reading, if we are going to survive as print newspapers. Frequently an inside Metro story goes on my home page to high readership." This comment well represents the changing gatekeeping process due to the audience data now available. This editor had statistical support for the "success" of a local (Metro) story that drew in the Web audience. Similar follow-through data of a story's success is not available for the print edition; one could only speculate that print readership in this case was lower because of the story's inside placement.

The survey also asked about the number of staff involved in deciding which stories would receive major play in the online edition. The range of responses was one to three people with one editor noting that one individual made the decisions but that this was a different individual every day; in total, seven people made decisions during each week. Another respondent noted that only the new media editor made decisions because of Newspaper Guild contract requirements.

Other comments about the process came from an open-ended question on how decision-making for the online edition differed from that of the print edition. One online staffer described a "streamlined decision process." Remarkably, at some newspaper Web sites, although a codified system of editorial conferences exists for deciding the primary content of the print edition, the Web edition decisions may be left to a few or, in some cases the one Web editor on duty. One survey respondent wrote, "Well, there's only two of us as opposed to 200 of them [print staff members]. A lot simpler!"

Selected Content Analysis of Daily Newspaper Web Sites

Clearly, online editors have a fairly good sense of what copy attracts their readers with a natural leaning of the medium toward local interest and breaking news. How, then, does this detailed knowledge of what the audience wants affect the traditional gatekeeping role of giving readers what they should have - an unpleasant dose, at times, of all the news that editors judge fit to print? To gather data on variable content selection, media researchers may examine how different newspapers present the same news day across the country. In fall 1999, I developed and applied a content analysis rubric for assessing the degree of local news prominence on newspaper Web sites and the mix of news versus feature stories that might attract readers. The analysis covered a range of medium and large-sized, daily newspaper organizations throughout the United States on two occasions, one in September and one in October.

Within this limited study, "local interest or community issue" was defined as: 1) the who, what, or where of the news being local; 2) regional news that a local audience would follow because of proximity (such as N.C. residents in the center of the state tracking hurricane damage on their coastline); or 3) a "localized" story providing the local angle on a regional or national issue. Analysis on one of the selected news days, October 1, 1999, covered ten daily newspaper Web sites: Arizona Republic, Baltimore Sun, Chicago Tribune, Cincinnati Enquirer, The Dallas Morning News, Detroit Free Press, Philly.com (comprised of the Philadelphia Inquirer and Daily News), St. Petersburg Times, San Diego Union-Tribune, and Seattle Post-Intelligencer.

All these newspaper Web sites displayed at least four main, on-screen stories while most contained five and one as many as eight. For evaluating content selection, the study limited the "top story" designation to the top four stories in a top-to-bottom screen scroll. That day's analysis revealed that local news comprised 75 percent or more of the top four stories in 70 percent of the Web sites. And of that group, two sites used 100 percent local stories. Of the remainder, two of the sites used 50 percent local news, and one site had no local news among the top four stories. "No local stories" among the top four did not mean a complete absence of local coverage on that site; though, most daily newspaper pages have categorized menus of headline links. For example, the Baltimore Sun carried two national and two international stories as its top four but offered a rail (display of options along one side of the screen) of other news links, including a list of Maryland stories. Local coverage clearly dominated the news day's offerings on October 1, 1999.

An investigation of content style offered a wider range of results. For this analysis, "news stories" were defined as those on current events or issues and breaking news; "feature stories" included articles that were not time sensitive, reviews and entertainment-related pieces, and stories that provided feature approaches to issues within "sections" (such as health). The results for October 1 were an even split: 50 percent of the Web sites reviewed placed only "news stories" in their top four; the other half offered from 25 to 75 percent "feature stories" in the top displayed content. Half of the sites also offered a headline link list of breaking news or "top news" articles. Even though these "top news" stories received only a headline, instead of the four or five lines of text and accompanying art that the main stories received, the top news menus typically appeared in a prominent location on the screen.

The October 1 analysis revealed higher news than feature content. Thus, the content style examination of the Web pages did not immediately suggest that editors were selecting a high percentage of "softer" feature copy to entice their online audiences.

The other news day reviewed was more difficult to analyze because of the item content itself and its impact on space use and selection decisions. September 17, 1999, was a "heavy" news day with continuing coverage of Hurricane Floyd up and down the East Coast and a two-day-old church shooting in Fort Worth, Texas. Though, the sample of newspaper Web sites was selected that day, in part, to reflect the influence of these news stories. The papers reviewed were: The Charlotte Observer, Chicago Tribune, The Dallas Morning News, Detroit Free Press, Fort Worth Star-Telegram, Kansas City Star, Philly.com, and two more national newspapers, The New York Times and The Washington Post.

Local versus non-local content analysis of this somewhat aberrant news day could easily be skewed because of the dominance of major regional stories. The Post and the Times only gave prominent display to three stories, rather than the common four "top stories" of other sites. The Star-Telegram and the Morning News, newspapers in the Dallas-Fort Worth area, dedicated the whole of their article space to the recent local shooting.

Three of the four eastern newspaper sites placed Hurricane Floyd or resultant flooding stories as their lead articles (only the Times gave it second position). The Fort Worth paper showcased fourteen links to news stories and news/feature sidebars related to the shooting; the Dallas paper's sixteen links included recorded 911 tapes and connections to television video coverage on the church shooting. Interestingly, both of the Texas newspaper Web sites observed tapped into the distinctively interactive qualities of the electronic medium. The Star-Telegram offered an online forum for comments on the shooting, and both sites provided a deposit link for users to type in condolences for church members and victims' families.

Although these intermittent analyses of Web site content do not offer statistically significant patterns, they do help focus the discussion around content selection factors and the changing nature of editorial gatekeeping as newspaper content moves from page to screen. During this technological sea change, journalism has become a much more audience-centered universe nurturing newspaper Web sites that are innately more interactive than their print siblings.

Effects on the Gatekeeping Process

Online newspaper Web site editors not only crunch through much more data about their audience's reading habits but also hear more from their public and hear it more often. Many reporters and news staff are receiving more feedback these days, especially with the popularity of e-mail and a push to print e-mail addresses next to writers' bylines.

Newspaper sites often include immediate-response reader polls and forums in which users simply click a "yes" or "no" to the day's question and choose a separate link to view vote totals. (Although there's no way to enforce one person/ one ballot, users are restricted from clicking twice in a row.) In online forums, readers may view a cumulative list of comments on an issue or story the newspaper Web site has run. Anyone may select the forum link and scroll through his or her neighbors' comments. (Some site editors have had to delete offensive or obscene comments from this open depository.)

Beyond providing outlets for response, online editors and staff may be affected by the feedback they receive and the telling numbers, or "hit reports," that arrive on their desks. Esther Dyson, of the Electronic Frontier Foundation, likens the online editor to a "virtual bartender," whose value is in listening as well as talking. She adds, "The more successful online services put a high premium on interacting with readers."[13] The other voices that online editors may be listening to are wary advertisers and their own anxious publishers who have been dumping dollars down a gaping online pit. So far, the cyber venture has not been profitable. In 1998, Editor & Publisher reported that newspaper Web sites would likely lose more than $80 million that year.[14]

While publishers wait out more significant advertising numbers, they increase revenue through sales of archived stories, Internet Service Provider packages, and subscriptions (some sites offer a modicum of information gratis and charge for a full or premium subscription).[15] Ian Murdock is director of new media at the San Antonio Express-News, which requires readers to subscribe for access to most of its site. Murdock told the American Journalism Review that he seeks a balance between how much to offer free and how much to withhold for paying users: "'We need to have enough value for the subscribers,' says Murdock, 'but enough eyeballs for the advertisers.'"[16]

To give the advertisers the number of page views they're paying for and keep the readers clicking through, have editors become less professional sifters, selecting and honing what copy to publish, and more shovelers,[17] feeding the public belly only what it craves? If readers, directly through feedback or indirectly through their clicking habits, are dictating what will satisfy them, then who's driving the proverbial media bus?

Those within the media circle vary in their interpretations but identify a recasting of the editor's role. On a more restrained note, Scott Whiteside, director of strategy and technology/new media for Cox Enterprises, suggests editors may be "so accountable to the interactive reader that a degree of editorial control is ceded to the readers themselves."[18]

J. D. Lasica, an editor at the Sacramento Bee, finds the old model no longer applies: "…gatekeeper suggests that journalists be guardians of what comes into the public sphere - an antiquated notion in an age of information overload and reader-directed news."[19] He suggests changing the metaphor: "Simply put: the gate is gone."[20]

Away from the pundits and back in the 24-hour newsroom, the beleaguered online editor ponders traditional news value as well as cybermarketing issues: What do I publish, where do I place it, and, now, how many page views will it get? As the sender (editor) / receiver (reader) paradigm evolves into a less passive, more audience-dictated framework, journalists gradually may be abandoning their guardian status.

Notes

[1] Kurt Lewin, "Frontiers in Group Dynamics," Human Relations 1 (1947): 145-153, quoted in Steven E. Clayman and Ann Reisner, "Gatekeeping in Action: Editorial Conferences and Assessments of Newsworthiness," American Sociological Review 63, no. 2 (1998): 178-199, (ProQuest -- [cited 4 August 1999]), p. 2 of 21.

[2] Paul F. Lazarsfeld, Bernard Berelson, and Hazel Gaudet, "The People's Choice" (New York: Free Press, 1944), quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 67.

[3] Elihu Katz and Paul F. Lazarsfeld, "Personal Influence" (New York: Free Press, 1955), quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 65.

[4] Bernard Cohen, "The Press and Foreign Policy" (Princeton, NJ: Princeton University Press, 1963), quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 105.

[5] Maxwell McCombs and Donald Shaw, "The Agenda-setting Function of Mass Media," Public Opinion Quarterly 36 (1972): 176-187, quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 105.

[6] Marilyn S. Roberts and Philemnon Bantimaroudis, "Gatekeepers in International News," Harvard International Journal of Press/Politics 2, no. 2 (1997): 62-77, (EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999]), accession no. 9705016477, p. 4 of 15.

[7] Alicia C. Shepard, "Gatekeepers Without Gates," American Journalism Review 21, no. 2 (1999): 22-30, (EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999]), accession no. 1715781, p. 3 of 14.

[8] Steven E. Clayman and Ann Reisner, "Gatekeeping in Action: Editorial Conferences and Assessments of Newsworthiness," American Sociological Review 63, no. 2 (1998): 178-199, (ProQuest -- [cited 4 August 1999]), p. 16 of 21.

[9] Jonathan Alter, "The New Powers That Be," Newsweek (18 January 1999),
24-25, (NewsBank InfoWeb. -- [cited 4 August 1999]), record no. 005510DAC753C0B4A97E0, p. 2 of 3.

[10] Ibid., p. 2-3 of 3.

[11] J. D. Lasica, "Net Gain," American Journalism Review 18, no. 9 (1996): 20-34, (EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999]), accession no. 9611170710, p. 24 of 27.

[12] Felicity Barringer, "Narrowing the Electronic News Gulf: Papers Beefing Up
Reporting on Line as TV Sites Aim for Depth," New York Times, 7 June 1999, section C, p. 14.

[13] Lasica, p. 14 of 27.

[14] Robert Neuwirth, "Race into Cyberspace Gushes $80M Red Ink," Editor & Publisher, 19 December 1998, 12-13, (ProQuest -- [cited 4 August 1999]), p. 1 of 3.

[15] Scott Kirsner, "Profits in Site?" American Journalism Review 19, no. 10 (1997): 40-44, (ProQuest -- [cited 4 August 1999]).

[16] Ibid., p. 7 of 10.

[17] Concept of "shovelers" from Lasica.

[18] Scott Whiteside, "Web Redefines Who an Editor Is," American Editor, no. 777, (July/August 1996), 5.

[19] Lasica, p. 13 of 27.

[20] Ibid., p. 14 of 27.

Works Cited

Alter, Jonathan. "The New Powers That Be." Newsweek, 18 January 1999, 24-25. NewsBank InfoWeb. -- [cited 4 August 1999], record no.005510DAC753C0B4A97E0.

Barringer, Felicity. "Narrowing the Electronic News Gulf: Papers Beefing Up Reporting on Line as TV Sites Aim for Depth." New York Times, 7 June 1999, sec. C, p. 14.

Berger, Charles R., and Steven H. Chaffee, eds. Handbook of Communication Science. Newbury Park, Calif.: Sage, 1987.

Clayman, Steven. E. and Ann Reisner. "Gatekeeping in Action: Editorial Conferences and Assessments of Newsworthiness." American Sociological Review 63, no. 2 (1998): 178-199. ProQuest -- [cited 4 August 1999].

Cohen, Bernard. The Press and Foreign Policy. Princeton, NJ: Princeton University Press, 1963. Quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 105.

Katz, Elihu and Paul F. Lazarsfeld. Personal Influence. New York: Free Press, 1955. Quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 65.

Kirsner, Scott. "Profits in Site?" American Journalism Review 19, no. 10 (1997): 40-44. ProQuest -- [cited 4 August 1999].

Lasica, J. D. "Net Gain." American Journalism Review 18, no. 9 (1996): 20-34. EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999], accession no. 9611170710.

Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet. The People's Choice. New York: Free Press, 1944. Quoted in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987), 67.

Lewin, Kurt. "Frontiers in Group Dynamics." Human Relations 1 (1947): 145-153. Quoted in Steven. E. Clayman and Ann Reisner, "Gatekeeping in Action: Editorial Conferences and Assessments of Newsworthiness," American Sociological Review 63, no. 2 (1998): 178-199. ProQuest -- [cited 4 August 1999].

McCombs, Maxwell and Donald Shaw. "The Agenda-setting Function of Mass Media." Public Opinion Quarterly 36 (1972): 176-187. Discussed in Charles R. Berger and Steven H. Chaffee, eds., Handbook of Communication Science (Newbury Park, Calif.: Sage, 1987).

Neuwirth, Robert. "Race into Cyberspace Gushes $80M Red Ink." Editor & Publisher, 19 December 1998, 12-13. ProQuest -- [cited 4 August 1999].

Roberts, Marilyn S., and Philemnon Bantimaroudis. "Gatekeepers in International News." Harvard International Journal of Press/Politics 2, no. 2 (1997): 62-77. EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999], accession no. 9705016477.

Shepard, Alicia C. "Gatekeepers Without Gates." American Journalism Review 21, no. 2 (1999): 22-30. EBSCOhost, MasterFILE Premier database -- [cited 4 August 1999], accession no. 1715781.

Whiteside, Scott. (1996,). "Web Redefines Who an Editor Is." American Editor, no. 777, July/August 1996, 4-5.