The Drudge Report is one of the founding flag bearers of “new media”: a U.S.–based news aggregator founded in the late 1990s that has developed a reputation for breaking tomorrow’s news today. The site has become a powerful force in the U.S. media sphere and its founder was named one of Time Magazine’s most influential people in 2006. In existence for more than a decade, the Drudge Report makes an ideal case study for examining the “new media versus old media” argument. How dependent is such a “new media” aggregator on the “old media” it draws from, and how does it find its breaking stories? A cross–section of analytical techniques is used to demonstrate how to profile a news Web site, and finds that the Drudge Report relies heavily on wire services and obscure news outlets to find small stories that will break large tomorrow, making it highly dependent on mainstream “old media” sites.
In 2006, Time Magazine named Matt Drudge one of its Time 100, the “100 men and women whose power, talent, or moral example is transforming our world” (Cox, 2006). Later that year, the Washington Post noted that its “largest driver of traffic is Matt Drudge,” (Hirshman, 2006) while a 2005 CNET article chronicled the rise of Breitbart.com after the Drudge Report chose it as its preferred wire service provider (Sandoval, 2005). ABC News’ political director called the site “a force in the political news cycle for both the press and the campaigns,” while the 2004 Republican National Committee’s communications director stated that “no single person is more relevant to shaping the media environment in a political campaign” (Rutenberg, 2007). The site is often called an agenda–setter for the mainstream media, with one CBS executive claiming “Drudge is like a megaphone in the cyber–world. Other news organizations and Web sites take their cue from him” (Sappell, 2007). Its presence is felt even across the ocean, with the British newspaper the Telegraph reporting over one million hits to its site after Drudge linked to one of its stories (Thurman, 2007).
The Drudge Report is one of the founding fathers of the “new media” movement: a news aggregation site born in the late 1990s that has built a reputation on ferreting out the small stories of today that will go big tomorrow. Rather than reporting its own news stories like a citizen journalist or commenting on stories like a blogger, the Drudge Report collects stories from the mainstream press and packages them into a concise broadsheet of links. The site has become renowned for being the first to catch major breaking stories and piecing together coverage of them from all available outlets, updating constantly throughout the day. As the argument of “new media” versus “old media” has intensified in recent years, the Drudge Report’s decade–long tenure offers an ideal case study in how “new media” sites evolve and adjust to changes in the media sphere.
Despite the Drudge Report’s prominence in the 2008 U.S. presidential campaign (Rutenberg, 2007), there has been little formal academic study of the site. A 2005 study (Groseclose and Milyo, 2005) examined the ideological bias of the site, while Thurman (2007) touched on the site’s impact on newspaper Web sites in the United Kingdom. No substantive academic study has examined the Drudge Report itself as a media outlet and attempted to explore the patterns in its coverage structure. In particular, operating as an aggregator rather than a news reporting site, how is the Drudge Report affected by its reliance on mainstream media?
The Drudge Report
The visual layout of the Drudge Report has remained virtually unchanged over the past decade: a simple three–column page of links. All links are contained on the single front page, rather than being distributed across multiple pages, and the only graphical elements are advertising banners. At the top of the page is a large boldface link that is the “headline” of the moment, often followed by several smaller links set to the left of the headline that offer additional stories on the same major topic. These are considered together as the “headline” of the Report in this study. Unlike some sites, whose source code changes with every page refresh due to advertisements on the page, the Drudge Report is configured such that its source code does not change unless there is a modification to the link content of the page. This is important, as otherwise measures that count the number of updates per hour would be skewed by constant changes of banner ads.
The Drudge Report does not include the brief titles and short descriptive summary blurbs used by other news sites, relying instead entirely on links, with no descriptive text about each article other than the link text itself. The text of each link is written by the site’s editors and emphasizes the specific aspect of the story being focused on, rather than using the headline provided by the outlet. While the Drudge Report occasionally includes its own reporting, this analysis is focused on its external link activity as an aggregator.
In order to compile a historical profile of the Drudge Report, a suitable archive of the site stretching back several years was needed. One of the best known historical Web archives is the Internet Archive (http://www.archive.org/), a non–profit organization that has archived snapshots of the Web since 1996. The Internet Archive’s snapshot history of the Drudge Report is extremely small, with only 30 snapshots from 2003 and less than a third of the days in 2007 covered, including just two snapshots in July (Archive.org, 2009). Given that the Drudge Report is known for updating many times per day, this is far too coarse of a sample to explore meaningful patterns. Thankfully, the high profile of the Report led to the creation of the DrudgeReportArchives.com in 2001, a Web site that has taken a snapshot of the Drudge Report every two minutes since 18 November 2001. For the purposes of this study, the analytical period runs from 1 January 2002 through 31 December 2008, totaling 171,717 snapshots. The Drudge Report Archives only records a snapshot when the page changes, making it very easy to determine the site’s rate of change. Roughly half of the physical screen space of the Drudge Report page is devoted to links at the bottom of each column to sites that Drudge wishes to promote. This list has remained relatively unchanged during the seven years examined for this study. In addition, the Drudge Report also contains multiple advertisement links that generate the ongoing revenue of the site. Since these are static elements that are not part of the site’s news aggregation process, they were dropped from the analysis. Drudge also occasionally produces his own news reports and special features, but the rarity of these reports and the difficulty of integrating them with the rest of the analysis led to their exclusion.
A large part of the Drudge Report’s reputation lies in its fast update cycle, with new links posted and headlines updated continually on a 24/7 basis. Visitors tend to be extremely loyal, with Nielsen ratings showing the average visitor returning 20 times and spending more than an hour on the site each month (Sappell, 2007), the most of any online news Web site (Siklos, 2008). One of President Bush’s top political consultants was quoted as visiting the site “30 to 40 times a day” (ABC News, 2006). In a breaking news situation, individual news Web sites often lag several hours behind actual events, so the Drudge Report has become known for its ability to gather all the available reports together with wire updates, and provide a running narrative of what is known at the moment. Yet, on an average day, how many times a day does the Drudge Report actually update? Are updates concentrated during certain hours or days of the week, and does the update interval increase during periods of sustained major global events?
The Drudge Report archives report a total of 171,717 page updates to the Drudge Report from 1 January 2002 through 31 December 2008: an average of 67 updates per day. Examining the number of updates per year, 2007 is found to have the highest update rate, likely due to the early stages of the U.S. Presidential campaigning launched that year. Yet, 2008, where the majority of the campaigning took place, exhibited a nearly 10 percent reduction in updates, suggesting that Drudge focused his coverage on the early stages of the campaigning, when there was a full field of candidates battling for their party nominations . Perhaps most surprising is that 2003, with the launch of the U.S. invasion of Iraq, had the lowest number of updates of the past seven years. Drudge does not focus on “current events” the way other outlets do, so ordinary stories about the invasion would likely not be linked from his site. In an ongoing study, the author (Leetaru and Althaus, unpublished) found that the total volume of New York Times and Wall Street Journal coverage devoted to the invasion peaked at 20–30 percent of all news articles published in those outlets at the opening of the war, maintaining a steady state of 5–10 percent of their total coverage volume through the end of 2003. Front page coverage of the war in these papers peaked at 90 percent at the outset of the war and held at 20–40 percent through the remainder of the year. If other papers devoted similar percentages of their coverage to the war, this would dramatically reduce the space they dedicated to coverage of other events that Drudge focuses on, limiting the material he had for his site. As an aggregator, rather than a reporter, the Drudge Report “is vulnerable because he’s not producing anything. He’s just got muscle through his links to the work of others” (Sappell, 2007). In other words, as major global events displace the news coverage that an aggregator relies on, that aggregator is forced to reduce its update cycle to accommodate the reduction in stories to link to.
Figure 1: Drudge Report updates by year.
Figure 2: Drudge Report updates, percentage difference from the past year 2003–2008.
Updates by month
The year 2002 had the fourth highest number of updates, yet May 2002 had the highest number of updates of any single month. Revelations this month regarding early intelligence warnings of the 9/11 terrorist attacks spurred a blitz of media coverage and congressional inquiries (Burke and Vulliamy, 2002). While the topic was war–related, it also represented the kind of political scandal that Drudge has built his name upon, and so his site contained continual updates of the latest on the unfolding drama. Update volume decreased dramatically afterwards through a low point in the opening months of 2003, while March 2003, the start of the Iraq Invasion, represented a brief peak during this period. The first Democratic presidential candidate debate occurred in April 2007, while the first Republican debate occurred the following month. This period coincides with a leveling off of updates on the Drudge Report, suggesting a strong ramp–up to the first debates as the back stories of each candidate are vetted in the press, to a steady state level of coverage as the debates and formal campaigning began. The number of updates per month was already in sharp decline leading up to Obama’s nomination as the Democratic candidate in June 2008 and plummeted even further leading up to the November elections, suggesting that Drudge focused his attention on the early stories emerging about each candidate and by the time of the nomination and election, there were fewer “breaking stories.” Notably, the selection of Sarah Palin as John McCain’s running mate in August 2008 did not result in a substantial increase in updates.
Figure 3: Drudge Report updates, (body + headlines) by month 2002–2008.
Separating updates to the main body of the Drudge Report from updates to the headline section, two distinct curves emerge. Figure 4 shows update volume by month for body links: a curve nearly identical to Figure 3. Headline updates, however, decrease slowly from January 2002 to April 2005, before rebounding slightly and again descending through the end of 2008. This suggests that Drudge maintains a relatively consistent update cycle for his headlines, while his update cycle on body links varies based on the volume of material he finds each day.
Figure 4: Drudge Report updates, body updates by month 2002–2008.
Figure 5: Drudge Report updates, headline updates by month 2002–2008.
Day of week
Like many U.S. news outlets, the Drudge Report focuses the majority of its updates during the week, with significantly fewer updates during the weekend. Saturdays are its slowest day for updates, with Sunday representing a slight rebound. Tuesdays have the highest number of updates, with a ramp down through the rest of the week and a sharp decline in updates on Fridays. This largely mirrors the news cycle used by many U.S. outlets, again reinforcing the dependence of the Drudge Report on its sources.
Figure 6: Drudge Report updates, total updates by day of the week 2002–2008.
Updates by hour
The Drudge Report is updated 24 hours a day, seven days a week, with Matt Drudge updating the site in the morning and his assistant Andrew Breitbart taking the “afternoon shift” (Nolan, 2008). Figure 7 shows the total number of site updates by hour from 2002–2008 (using the Eastern Standard Timezone since Drudge is based in Florida). Update volume opens light each day, descending to a low point from 5 to 6 AM, and increasing sharply starting at 8 AM. The period with the most updates is 10–11 AM, with additional peaks from 3–4 PM and 8–9 PM. Breitbart is based in California, so these afternoon peaks would correspond to noon–1 PM and 5–6 PM Pacific Time.
Figure 7: Drudge Report updates by hour total 2002–2008 (EST time zone).
Figure 8 shows the hourly breakdown of each of the years individually. Starting with 2003, the years track each other fairly closely, with the early morning and late evening peaks increasing over time. The first year of the analysis period, 2002, stands out with a much flatter curve with dramatically higher numbers of early morning updates and a lower rampup in the late morning to afternoon without the third peak in the late evening. In particular, this was the only year that the midnight–6 AM period had significant numbers of updates: this has become a “dead period” in more recent years.
Figure 8: Drudge Report updates by hour yearly 2002–2008 (EST time zone).
Dating back at least to 2002, the Drudge Report has included an automatic refresh command in the source code of the site that causes visitors’ browsers to automatically reload the page at a set interval to display the latest updates. Currently set to three minutes, it has varied from three minutes up to four minutes over the last seven years, suggesting that the site expects fast–paced updates and wants readers to see the latest material. How often does an average link exist on the Drudge Report before being replaced, and how many links does the report include at a given time?
Figure 9 shows the total number of unique URLs linked to by the Drudge Report each month, while Figure 10 shows the average time in hours that a given link stays on the site during that period. The total number of non–advertising links on the Drudge Report has increased steadily during the past seven years, to a peak in late 2007, before decreasing at nearly the same rate through the end of 2008. The average lifespan of a given link decreased from 15 hours to a low of 10 hours in late 2004/early 2005, before beginning an ascent up to nearly 20 hours towards the end of 2008. Thus, more links are being posted on the Drudge Report and those links are staying up longer. On average, there were 1,232 unique links featured on the Drudge Report each month, and those links remained on the site for 13.2 hours each. Of note, the peak update volume of May 2002 seen earlier appears only as a small increase on these graphs. The Drudge Report on occasion includes brief commentary of a sentence or two with additional details on a news story or alerting the reader of imminent news. These are text–only and do not include links, so would be captured by the earlier graph that counted all updates to the site, while they would be ignored by the link–centric graphs below.
Figure 9: Drudge Report, number of unique links per month 2002–2008.
Figure 10: Drudge Report, average lifespan of a link in hours 2002–2008.
The Drudge Report has developed a reputation for “supply[ing] [the] links that guide [readers] through the ‘thicket’ of the day’s news” (Sappell, 2007). The Director of the Project for Excellence in Journalism noted that during the 2006 elections, “Drudge quickly sent his audience to the best destinations. ‘He had figured out in real time what we figured out more conclusively in hindsight.’” (Sappell, 2007) A 2005 CNET article asserted that “Drudge has shown little preference for any of the sites he links to” and that with the launch of Breitbart.com on 17 August 2005, Drudge is now “routinely post[ing] more story links from [Breitbart.com] than any other news source.” (Sandoval, 2005; Nolan, 2008) Besides Breitbart.com, does Drudge have any other favored outlets, and how many unique domains does he draw content from? Is there a specific geographic emphasis to the outlets he selects?
A domain like www.cnn.com is essentially a computer address that a Web browser looks up in a central database to determine which physical computer to request the page from. This address is much like a phone number: it directs a request to the proper place, but gives no information about who “lives” there. The Web equivalent to the phone directory is the DNS registration database. When a company registers a site like cnn.com, the domain is recorded in one of several central databases that contain contact and ownership information of the domain, connecting cnn.com to the Cable News Network in Atlanta, Georgia. Amazon.com offers a service called Alexa (http://www.alexa.com/), which aggregates this information from all domain registration databases and makes a free site available where visitors can type in any domain to see its registration information. It is important to understand that this information reflects the entity that owns and controls the domain, which may not always match a given physical location. For example, the Los Angeles Times is published in California, but is owned by the Tribune Company in Chicago, so the registration information for its site, www.latimes.com, displays Chicago. In a world where more and more newspapers are being bought by holding companies, this will gradually have a greater impact on the interpretation of such geographic information.
In addition to compiling domain registration information, Alexa offers a powerful database of traffic rank information, offering a way to rank sites by their “popularity.” For example, it may be useful to know that one news outlet receives more than a thousand times the visitors of another outlet. Alexa generates this data through its network of “several million” users who have installed its browser toolbar. This toolbar tracks every site that a user visits and transmits a log to Alexa that enables it to statistically compute the aggregate popularity of each site (Alexa, 2009). Like any sampled data, biases in the community of users allowing their actions to be tracked through the Alexa toolbar may impact the rankings, but the Alexa data offers a valuable measure of site popularity. Unlike many other commercial services, the Alexa data is also freely available to researchers, simply by visiting http://www.alexa.com/ and typing in a domain. Traffic ranks range from 1 for the most popular (currently Google) to at least 2.1 million. According to their documentation, any domain with a traffic rank larger than 100,000 is not reliable due to insufficient traffic (About Alexa, online). The site with the largest rank number linked from the Drudge Report during this period is the Canadian–based marketing firm Porter Novelli (porternovelli.ca), with a rank of 29,775,942. An additional 108 domains had no rank in the Alexa data due to extremely low traffic.
In the past seven years, the Drudge Report has linked to 103,472 stories on 2,744 different domains. The domain with the highest number of links is indeed breitbart.com, with 14.4 percent of all links during this period. Drudge notes that one of the reasons he favors Breitbart.com is that “for the wire stories, I’ve always looked for places with low graphics, without a lot of spinning Java tops on them … When I send my readers someplace, I want it to be convenient for them to get there.” (Sandoval, 2005) This may explain why the second highest site he links to (12.9 percent) is myway.com, a site which proclaims, “No banners. No pop–ups. No kidding.” The most popular newspaper cited by Drudge has been the Washington Post at #4, followed by the New York Times at #6, and the BBC at #7.
Table 1: Top 10 domains linked to by the Drudge Report 2002–2008. Rank Domain Total links Percentage of all links 1 breitbart.com 14,923 14.45 2 myway.com 13,423 12.99 3 yahoo.com 10,306 9.97 4 washingtonpost.com 6,471 6.26 5 reuters.com 5,133 4.96 6 nytimes.com 3,498 3.38 7 bbc.co.uk 1,508 1.46 8 ft.com 1,459 1.41 9 cnn.com 1,408 1.36 10 netscape.com 1,165 1.12
At least 49 countries are represented in the domains, through the United States dominates the list, with 2,086 domains (89.25 percent). The second highest is Canada with 64, followed by the United Kingdom with 57, and India with 13. Those four countries are the only ones with more than 10 domains, and only 28 countries have more than one domain represented. Locations for 407 domains were unavailable through Alexa. The strong representation of British outlets falls in line with a study that showed that a quarter of the traffic to a set of British newspaper Web sites came from the Drudge Report (Thurman, 2007).
Table 2: Top 10 countries linked to by the Drudge Report 2002–2008. Country Domain Percentage of all outlets United States 2,086 89.25 Canada 64 2.73 United Kingdom 57 2.43 India 13 0.55 Germany 9 0.38 United Arab Emirates 9 0.38 Italy 8 0.34 France 7 0.29 China 6 0.25 Israel 6 0.25
Figure 11: Countries by number of outlets linked by the Drudge Report 2002–2008.
Figure 12: Cities with outlets linked by the Drudge Report 2002–2008.
The Drudge Report linked to stories on an average of 828 different sites each year. In 2002, Drudge linked to domains in 17 countries, while in 2005 he linked to domains in 25 countries, and in 2008 he linked to 20 countries. India and Israel appear in each year’s lists, with many countries having a single domain each. The number of unique domains fluctuates each year, but does not appear to be steadily increasing or decreasing. While the vast majority of Drudge’s coverage comes from the U.S., U.K., and Canadian sources, the fact that many of the countries have only a single outlet linked and those countries often change each year, suggests that Drudge sources his outlets globally as needed, but focuses most of his coverage at home.
Figure 13: Number of unique domains linked by year 2002–2008.
Outlets by month
In January 2002, Drudge linked to articles on 123 domains, with yahoo.com accounting for 28.2 percent of that traffic, followed by the Washington Post (19.5 percent) and the New York Post (3.1 percent). The next five outlets were all U.K.–based: the Telegraph, BBC, Times of London, Reuters, and the Independent. Collectively, the top five outlets accounted for 55.6 percent of all links on the Drudge Report that month. In January 2008, Drudge linked to 224 domains, with Breitbart.com taking the top spot with 24 percent, followed by three other wire services, MyWay.com with 14.6 percent, Yahoo at 6.6 percent, and Reuters at 5.9 percent. In six years, the top four outlets went from one wire service and three papers to all wire services, reflecting a dramatic shift at the Drudge Report from aggregating traditional newspapers to focusing on breaking news in the wire reports. Despite this shift towards emphasizing a small stable of wire report sites, the diversity of the Drudge Report actually increased, with the average number of unique domains linked each month growing by nearly 50 percent.
Figure 14: Number of unique domains linked to by month 2002–2008.
The Washington Post held the title of most popular newspaper linked by the Drudge Report for nearly two years, and together with Yahoo.com held the top two spots on the Drudge Report every month from January 2002 through November 2003. In October 2003, a new wire service debuted on the Drudge Report, MyWay.com, in third place, and by the following month it was the #1 outlet on the Report, accounting for 17.6 percent of all links. In July 2005, Breitbart.com debuted on the Drudge Report at #6 with 2.9 percent of Drudge’s links, with MyWay still dominating with 19.3 percent, but the following month with its formal release, Breitbart.com became the top Drudge outlet, with 25 percent of all links. MyWay became the second highest linked outlet (14.2 percent), and Yahoo (9 percent) and Washington Post (7.3 percent) still retained top–five status. Another rising star on the Drudge Report has been Politico.com, a site focusing on U.S. national political news. Debuting in January 2007 (Politico, 2007), the first five links to it appeared on the Drudge Report that month. By the following January there were 49 links that month, and just two years later, Politico is now the 16th most linked outlet on the Report.
Of the 2,744 outlets Drudge linked to during this period, 1,415 of them (51.5 percent) are among the top 100,000 most popular sites in Alexa’s database. Only 108 (3.9 percent) are among the top 1,000 (such as cnn.com at 52 and foxnews.com at 266). Drudge thus overwhelmingly favors more obscure outlets over major news sites like the New York Times or CNN. In other words, Drudge finds stories when they are still small, reported by a local or niche outlet, and features them on his front page before they grow larger and are picked up by major newspapers.
The Drudge Report is known for its highly specialized focus on breaking political news, yet does not cover “policy … welfare … [or] healthcare …”, but rather focuses “on the most salacious aspects of American politics.” (Drudge Report Sets Tone, 2006) The InterMarkets advertiser package for the site suggests that 60 percent of his readers use the Web for political information (Intermarkets www.DrudgeReport.com, 2009). What are the core topics that the Drudge Report focuses on? How high did the 2008 U.S. presidential campaign rank in his coverage?
Analyzing topical focus
Examining the topical focus of a news outlet is a complex process, with many possible analytical techniques that can be employed. As noted earlier, the Drudge Report favors wire stories, which, due to licensing agreements, are usually expired from the Web after a period of 30 days. Many newspapers also expire articles off their sites after a preset time to force users to subscribe to access historical material. Preliminary estimates suggest that less than half of the articles linked by the Drudge Report over the last seven years remain online and that many of these tend to be from specific outlets, biasing the results. Thus, it was not possible to analyze the full text content of the articles, and the topical analysis instead focuses on the link text written by Matt Drudge and his assistants. Unlike other news aggregators, the Drudge Report does not reproduce article headlines in its links, but rather comes up with its own descriptive summary to use as the link text, “bend[ing] and shap[ing] the meaning of the story to meet [its] needs.” (Sappell, 2007) For example, a small article on efforts to remove the word “alien” as a term for undocumented immigrants was headlined by Drudge as “Bill Would Mandate Nicer Term for Illegals,” suddenly making a small article a big issue (Sappell, 2007). In this way, an analysis of the text Drudge creates, rather than the original article text, is more meaningful as a representation of the lens Drudge views the world through.
There are many computational methods for identifying topical focus of text, with state–of–the–art systems calculating the grammatical part of speech of each word and compiling strings of nouns into phrases approximating major topics in the text. However, the Drudge Report often phrases the same topic many ways (“President Bush”, “Bush”, “Pres Bush”, “W.”, etc), making such phrase–based analysis less useful here. For the purposes of this study, the most simplistic measure was used: breaking link text apart into single words. This eliminates the impact of phrasing differences and allows words to be considered individually.
Overall and yearly trends
In total, from 1 January 2002 through 31 December 2008, the Drudge Report used 31,801 unique words in its links, of which a total of 11,529 words (36.25 percent) were used only once. It averaged 12,856 unique words per year, with an average of 5,436 words used only once each year. From 2005 to 2007, the site experienced a nearly 20 percent increase in its vocabulary size, decreasing again in 2008.
Table 3: Unique words by year, Drudge Report 2002–2008. Year Unique words 2002 11,451 2003 11,470 2004 11,685 2005 12,281 2006 13,885 2007 14,855 2008 14,183
Table 4: Unique words used once by year, Drudge Report 2002–2008. Year Unique words 2002 5,240 2003 5,107 2004 5,078 2005 5,162 2006 5,680 2007 5,973 2008 5,812
The top 10 words of all time on the Drudge Report consist, as expected, primarily of grammatical structure words such as prepositions and articles. The only two nouns in the list are “bush” for President Bush and “us” for U.S. (United States), showing the strong focus on presidential issues in the Drudge Report during this period. (Drudge changes the capitalization of his words for emphasis, so all words are converted to lowercase for analysis.) After “US”, Iraq is the next most popular name, at #14 (2.8 percent), followed by Obama at #22 (2.07 percent), and Clinton at #32 (1.44 percent). Iran (#36; 1.32 percent) and China (#39; 1.15 percent) follow up the list.
Table 5: Top 10 words in Drudge Report links. Word Number links Percentage of all links to 19,726 19.06 in 18,044 17.43 of 11,453 11.06 for 9,997 9.66 on 8,437 8.15 the 6,190 5.98 at 4,940 4.77 bush 4,562 4.40 us 4,029 3.89 a 3,672 3.54
As a measure of how suddenly Barrack Obama emerged onto the U.S. political scene, “obama” was not mentioned in a single link in 2005, while in 2006 his name ranked 170th out of 192 (when words are grouped by the number of links mentioning each). In 2007, however, his name became the 26th most popular word, referenced in 1.93 percent of all links that year, and in 2008, his name ranked #5, with 9.56 percent of all links, becoming the first name to surpass Bush for the top spot during those seven years. In 2008, McCain’s name ranked #9 with 4.23 percent of links and Clinton was #10 with 3.28 percent.
As far back as 2005, media reports accused the Drudge Report of taking up the issue of global warming, focusing consistent and negative coverage on it, (David, 2005) and a 2008 article focused on what it called Drudge’s “hijack[ing] [of] headlines to sell global warming denial.” (Johnson, 2008). It is beyond the scope of this paper to analyze the tone of Drudge’s coverage of global warming, however the topic has been a consistent focus, with “warming” ranking 161st, “climate” 262nd, and “weather” 383rd, collectively totaling nearly half a percent of all links on the Drudge Report during this period.
Iraq has been another focal point of Drudge’s coverage, with 2.82 percent of all coverage devoted to the war: the 14th most popular term in those seven years. The invasion of Afghanistan, launched in 2001, was an ongoing conflict throughout the entire analysis period. Yet, “Afghanistan” appears in just 0.11 percent of all links, while Taliban appears in a mere 0.07 percent. Running for nearly a year and a half longer than Iraq, the Afghanistan invasion garnered only 4 percent of the coverage volume that Iraq did. One likely reason for this is that the Afghanistan war was less controversial at the time compared with the widespread global protest leading to the Iraqi invasion. Drudge emphasizes political scandal, not routine news reporting, so articles about torture or WMD intelligence would be far more likely to match his selection criteria than routine reports of troop movements.
The Drudge Report has built a reputation as a powerful agenda setter in the U.S. national political sector, with his site becoming a battleground of sorts in the 2008 Presidential campaign. Yet, rather than an all–powerful news outlet that controls the topic of the day, a picture emerges of a news aggregator whose ticket to success seems to be a particular knack for finding the small stories on the news wires and in obscure outlets that will grow big the next day, and having an extremely fast update rate that continually brings in the latest news on major events. His site is extremely dependent on the mainstream media he draws from, with his daily and hourly update cycles closely matching the update cycle of the U.S. outlets he draws almost 90 percent of his coverage from. He is also at the mercy of those outlets: when coverage of the Iraq invasion in 2003 dominated the news volume of major outlets, the update rate of the Drudge Report plummeted more than 20 percent. As Tom Rosenstiel, Director of the Project for Excellence in Journalism, noted in 2007, “The dirty little secret about Drudge is that he’s a gateway for conventional journalism.” (Sappell, 2007) In a time when mainstream media is coming under increasing pressure from “new” media, this portrait of one of the Web’s longest–lived and most commercially successful news aggregators paints an image not of a new media paradigm to replace the old, but rather of a symbiotic and highly dependent relationship between old and new.
About the author
Kalev Leetaru is Coordinator of Information Technology and Research at the University of Illinois Cline Center for Democracy, where he established and oversees its mass digitization center, Chief Technology Advisor to the Illinois Center for Computing in the Humanities, Arts, and Social Science, and Center Affiliate of the National Center for Supercomputing Applications. Among his research areas is the intersection of digital technologies and information management and he has recently completed a book manuscript on content analysis.
E–mail: leetaru [at] uiuc [dot] edu
1. Some outlets reported that Breitbart left the Drudge Report in early to late 2008, so this may also partially account for this decline (Cook, 2009).
ABC News, 2006. “Drudge Report sets tone for national political coverage,” (1 October), at http://abcnews.go.com/WNT/story?id=2514276&page=1, accessed 19 March 2009.
Alexa, 2009. “About the Alexa traffic rankings,” at hhttp://www.alexa.com/site/help/traffic_learn_more, accessed 14 March 2009.
Archive.org, 2009. “Search results for Jan 01, 1996 — Dec 27, 2008 (http://www.drudgereport.com,” at http://web.archive.org/web/*/http://www.drudgereport.com, accessed 25 March 2009.
Jason Burke and Ed Vulliamy, 2002. “Bush knew of terrorist plot to hijack U.S. planes,” Guardian (19 May), at http://www.guardian.co.uk/world/2002/may/19/terrorism.september11, accessed 22 March 2009.
John Cook, 2009. “Does Matt Drudge have a new understudy?” (24 March), at http://gawker.com/5182585/does-matt-drudge-have-a-new-understudy, accessed 27 March 2009.
Ana Marie Cox, 2006. “Matt Drudge” (30 April), http://www.time.com/time/magazine/article/0,9171,1186874,00.html, accessed 14 March 2009.
Laurie David, 2005. “Global warming: Drudge rebels against the dark side” (19 August), at http://www.huffingtonpost.com/laurie-david/global-warming-drudge-r_b_5912.html, accessed 14 March 2009.
Tim Groseclose and Jeff Milyo, 2005. “A measure of media bias,” Quarterly Journal of Economics, volume 120, number 4, pp. 1191–1237.http://dx.doi.org/10.1162/003355305775097542
David S. Hirshman, 2006. “‘Wash Post’ editor Downie: Everyone in our newsroom wants to be a blogger,” Editor & Publisher (6 October).
Intermarkets: www.DrudgeReport.com, 2009. At http://www.intermarkets.net/assets/PDF/drudgeReport.pdf, accessed 23 March 2009.
Brad Johnson, 2008. “Drudge hijacks headlines to sell global warming denial” (4 April), at http://wonkroom.thinkprogress.org/2008/04/04/drudge–sells–false–headlines/, accessed 19 March 2009.
Kalev Leetaru and Scott Althaus, Unpublished manuscript.
Hamilton Nolan, 2008. “Andrew Breitbart: Drudge’s human face” (3 September), at http://gawker.com/5044885/andrew–breitbart–drudges–human–face, accessed 14 March 2009.
Politico, 2007. “Mission statement,” at http://www.politico.com/aboutus/missionstatement.html, accessed 25 March 2009.
Jim Rutenberg, 2007. “Clinton finds way to play along With Drudge,” New York Times (22 October), at http://www.nytimes.com/2007/10/22/us/politics/22drudge.html, accessed 19 March 2009.
Greg Sandoval, 2005. “Breitbart.com has Drudge to thank for its success,” CNET (30 November), at http://news.cnet.com/Breitbart.com–has–Drudge–to–thank–for–its–success/2100–1025_3-5976096.html, accessed 14 March 2009.
Joel Sappell, 2007. “Hot links served up daily,” Los Angeles Times (4 August), at http://www.latimes.com/news/politics/la-fi-drudge4aug04,0,6016754.story, accessed 19 March 2009.
Richard Siklos, 2008. “The Drudge Report: Kickin’ it old school,” Toronto Globe and Mail (6 June).
Neil Thurman, 2007. “The globalization of journalism online: A transatlantic study of news websites and their international readers,” Journalism: Theory, Practice & Criticism, volume 8, number 3, pp. 285–307.http://dx.doi.org/10.1177/1464884907076463
Paper received 22 April 2009; accepted 14 June 2009.
Copyright © 2009, First Monday.
Copyright © 2009, Kalev Leetaru.
New media vs. old media: A portrait of the Drudge Report 2002–2008
by Kalev Leetaru.
First Monday, Volume 14, Number 7 - 6 July 2009