The online audience as gatekeeper: The influence of reader metrics on news editorial selection

Abstract

This study provides a snapshot of the hierarchy-of-influences model in the new media environment through examining the effects of audience web metrics on editors. Surveying 318 gatekeepers, the study found that audience metrics influence editors in gatekeeping. Editors’ likelihood to monitor web metrics is affected by their journalism training. Gatekeepers who attach the importance of high readership to economic benefits are more likely to have different news decisions based on web metrics. The study suggests a revision of the hierarchy-of-influences model with more emphasis being placed on the role of the audience.

Keywords

Web metrics audience gatekeepers hierarchy of influences online newsroom

In the highly technological world of today, at any given time journalists can easily pull out figures on such audience behaviors as how many web cruisers are reading a story, what stories they prefer, or whether they comment, email, “Facebook,” or Tweet a story. At The Washington Post, for example, detailed web metrics are displayed on a television screen for the entire newsroom. Editorial staff at The Post also receive emails everyday with specific data on web traffic. In other news organizations, web metrics have become important indicators of how well newsrooms perform (Peters, 2010). Indeed, the Internet has afforded journalists and news audiences an unprecedented interactivity, which did not exist in the traditional print media context.

A number of studies have found evidence of audience influence on the gatekeeping process. For example, scholars have discovered that it is becoming common for online journalists or gatekeepers to monitor web metrics and track readers’ behaviors, hoping to learn more about their audiences (MacGregor, 2007; McKenzie et al., 2011). However, little is known about how much the availability of such data affects editorial decisions in newsrooms or whether journalists are willing to accommodate readers’ content preferences, or what kinds of editorial adjustment to news selection gatekeepers are willing to adopt to get more subsequent online traffic. Unveiling these issues is the aim of this research.

Surveying news editors across the United States, this study examines the extent to which web metrics affect online news editors’ decision-making, and ultimately, news coverage. It investigates factors that lead online editors to monitor online traffic and other web analytics. Most importantly, using quantitative data, this study revisits the hierarchy-of-influences theory (Reese, 2011; Shoemaker and Reese, 1996), a model created in the traditional media environment, when gatekeepers had limited access to information about readers’ news preferences, through comparing the importance of audience against other factors in influencing editorial decision-making in newsroom today. Because the gatekeeping role is occupationally journalistic, the study provides an analysis of how technology has fueled the repositioning or sharing of this role by professional gatekeepers who increasingly allow some audience influence on their editorial responsibility. In doing so, the study is expected to also provide some real-world implications by shedding light on how editorial practices have evolved in the new media environment to adapt to audience tastes.

Gatekeeping and hierarchy of influences

Traditional gatekeeping research has focused largely on the role of gatekeepers in controlling what should be presented to audiences and what should not. In the original study, gatekeeping was described as “highly subjective,” with the gatekeeper exercising his control with biases (White, 1950: 386). Although this whole process of selection or rejection of news stories was based on the gatekeeper’s judgment and experience, news decisions are considered intuitive. Responding to the question: “What is news?” David Brinkley, a well-known TV journalist said news is “What I say it is,” (Rowe, 2005). For Arthur McEwen, an editor of the San Francisco Chronicle “News is anything that makes a reader say ‘Gee Whiz!’” (Boorstin, 1961). These two examples illustrate that there is no absolute rule in determining what make a news story, and thus confirming White’s conjecture about the gatekeeping process.

Communication scholars contended that White’s original assumption of the media gatekeeping was too simplistic. Gatekeeping has been shown to also include “writing, editing, positioning, scheduling, repeating, and otherwise massaging information to become news” (Shoemaker et al., 2008: 73). Along with this process many other factors can come into play. For example, Sylvie and Huang (2008) found that an editor’s decision on passing or rejecting a piece of news can be influenced by his or her own background and the organization he or she represents. Shoemaker and Reese (1991) argued that apart from the common newsworthiness’ constituents, social significance, complexity, and personal relevance, other factors, either being intrinsic or coming from outside to the communication workers, could also affect the process of transforming information from news sources to media messages.

Shoemaker and Reese (1996) delineated a five-level model of micro and macro influences on gatekeepers, arguing that gatekeeping no longer happens at only one gate of an individual communication worker as White did, but at multiple gates. The first level – the lowest – focuses on the individual factors of the communicator (e.g., personal background, experiences, attitudes, beliefs, etc.). The second level is media routines (e.g., audience orientation, newsroom routines). The third level of the model concentrates on the organizational influences (e.g., internal structure, ownership, goal, and policy). Extramedia forces or factors extrinsic to media organizations constitute the fourth level (e.g., sources, advertisers, audience, government control, market competition, technology). The last level — is media ideology.

Shoemaker and Reese (1996) posited that although each level in the concentric model has its own range of influences, the lower is subsumed by the higher one(s), which must take the lower ones into account. In the original model the audience factor, however, did not receive adequate attention, and thus impressionably deemphasizing the weight of such influence on the gatekeeping process.

Contextually, the hierarchy-of-influences model was created in the traditional media environment with journalists being notably distant from their audience. Yet a question still exists about the extent to which the influence of factors in this model exerted on gatekeeping work in the new media era, especially when declining readership, tough competition, fragmented audiences, plummeting advertising revenues, and downsizing newsrooms are darkening the picture of the journalism industry (Lowrey and Woo, 2010). Scholars have begun to explore possible changes of professional journalistic gatekeeping (McKenzie et al., 2011; Shoemaker et al., 2011; Singer, 2011). What is missing from the literature are the ways online audiences may have affected those newsroom practices.

Audience role

Journalists have often been accused of being aloof from their audiences in the context of traditional media (MacGregor, 2007). Gieber asserts (1960: 204) that “news selection has no direct relationship to the wants of readers.” Gans (1979) supports this view, arguing that journalists pay little to no attention to audience feedback, but mainly put together content based on what they think would interest their audiences. One of journalists’ excuses was technological constraints.

In order to learn about their audiences, news organizations used to rely on marketing companies (e.g., Nielsen) for audience research, which provided them with information on readers’ general interests. However, this kind of research did not “come often enough to help” news producers and editors adjust their daily editorial decision-making (Shoemaker and Reese, 1996: 105). Consequently, news became a product of a top-down, centralized model, with stories produced independently of news audiences (Schultz, 1999). With the Internet and two of its key elements –interactivity and immediacy (Chung, 2008) –learning sooner about audiences and their news consumption experiences is no longer a problem for journalists and news organizations. Interactivity, according to Chung (2008: 660), is a unique feature of digital technology that facilitates medium-to-human (e.g., commenting system, customizability of web pages) and human-to-human interactions (e.g., emailing links, reposting on social media sites). Immediacy allows for “publishing stories as quickly as possible” (Domingo, 2008: 692) and also encourages immediate communication with and among audiences (Chung and Yoo, 2008).

Technologies have made it possible for news organizations to record, in quantifiable details, audience news consumption – through keeping track of numbers of clicks, amount of viewing time, number of shares, or degree of engagement (e.g., commenting and “like”) – to gather a better view of audience preferences (Napoli, 2010). In a qualitative study, MacGregor (2007) interviewed online journalists and found that gatekeepers monitor audience data to re-weigh their editorial priorities. According to this study, in response to a story with more clicks, some journalists would expand the coverage of it, provide additional analysis for it, or publish stories of the same type.

Other research on the issue has found mixed results. Lowrey and Woo (2010), for example, discovered that uncertainties in newsrooms caused by recent financial woes within the industry have driven journalists’ attention to audience information. Boczkowski and Peer (2011), comparing placement of stories on news sites with most-viewed articles on these web pages, found that journalists’ and audience members’ choices of news do not intersect. Through a time-lag analysis of stories on the most-viewed lists and news sites’ agendas, Lee and Lewis (2012) argued that audience preferences are influencing editorial judgments of news stories more than the other way around. Other studies also found abundant evidence of journalists monitoring traffic to their news sites (Domingo, 2008; Lowrey and Woo, 2010; McKenzie et al., 2011). However, what is not well understood is why journalists are adopting this new routine of monitoring web metrics. Thus, it would be necessary to ask:

RQ1: What are the primary reasons an editor monitors online traffic to his/her news site?

For decades, newspapers have depended greatly on audience and advertisers for revenue generation. Also, the latter source often requires “sufficiently sizable readership” (Gabszewicz et al., 2002: 319). The competition for advertisements to appease that readership becomes tougher as news websites face a large number of competitors such as Craigslist offering no-cost advertisements. It is undeniable that newspapers or news sites, especially major metro dailies, play an important role in covering news in local communities and setting the agenda for other media from local TV news sites to blogs. However, statistics on page views and numbers of visitors are also important for these sites to sustain themselves financially. These statistics can be monetized by attracting more advertisements to news sites (Benbunan-Fich and Fich, 2004; Napoli, 2010). Now, more than ever, being able to keep and engage audiences is seen as an important way to salvage the in-peril news industry.

From reader standpoints, technologies have provided audiences with ample choices for selecting what kinds of news they want in a borderless world of abundant news providers. Audiences today are not mere recipients of news. Some have become active participants in news production and distribution (Shoemaker et al., 2011; Singer, 2006, 2011). Besides consuming news, audiences now contribute to content production of news sites (Kperogi, 2011; Lewis et al., 2010), or help spread the news (Baek et al., 2011; Baresch et al., 2011).

Under pressures of driving more traffic to their sites, online news organizations have found different ways to attract and engage their audiences. Professional gatekeepers now are more willing to give up their autonomy by passing off some of their tasks to audience members (Singer, 2011). Besides allowing audiences to personalize settings such as creating profiles, tracking stories of their interests, or changing layouts of the websites (Bucy, 2004; Singer, 2011), news sites also offer web applications that enable readers to leave feedback and engage in news production and delivery through emailing journalists. Site applications also allow participation and feedback with the reader capabilities of generating news on their own, commenting, sharing, and customizing content to their needs (Singer, 2011). In the increasingly interactive media environment, news has become more of a socially shared experience according to the Pew Project For Excellence in Journalism (2010a). Singer (2011: 4) argued that as news media become more interactive, for any news story to emerge from the vast pool of online articles, “the gatekeeping role must necessarily be shared far more broadly than in a traditional media environment.” According to Boczkowski (2004: 183), digitizing news is somewhat equal to moving it from “being mostly journalist-centered, communicated as a monologue … to also being increasingly audience-centered, part of multiple conversations.”

Theoretically, this new phenomenon of sharing the gatekeeping role is clear evidence of a transformation in journalism practices, blurring the line between gatekeepers and audience members, challenging the notion of journalistic professionalism (Singer, 2003). Any influence these statistics might have on news production is, therefore, economically and journalistically important. That influence demonstrates a possible reconfiguration of the relationship between audiences and journalists (Bucy, 2004), altering “the sociology of news production,” which suggests changes in the gatekeeping process (Hoffman, 2006: 59). It shows new tensions that arise from the negotiation between journalism professional control and open participation, challenging the boundaries of the profession (Lewis, 2012). This change has drawn significant attention from media scholars. From an audience perspective, studies have focused on the increasing choices offered by news sites (Chung, 2008; Shoemaker et al., 2011; Singer, 2011). Envisioning the change from news workers’ standpoints, scholars have found that this evolution does have an impact on journalists (Cassidy, 2008; Lee and Lewis, 2012; McKenzie et al., 2011). What is unclear here is the importance of the audience among other factors proposed in Shoemaker and Reese’s model, in exerting influences on the gatekeeping process, especially by using quantitative data. Therefore, RQ2 asks:

RQ2: Compared to other factors involved in editorial decision-making, how strong is audience appeal?

Empirical gatekeeping studies have found evidence of online audience metrics influencing journalists (MacGregor, 2007; McKenzie et al., 2011). For example, in an ethnographic study, Anderson (2011: 561) reported that at Philly.com audience metrics are “primary ingredients” for news judgment. However, like MacGregor’s (2007) and Domingo’s (2008) studies, the qualitative nature of the study made its results ungeneralizable to a larger population. More importantly, what gatekeeping tasks journalists are willing to compromise to audience choice remain a myth. Thus, it is relevant to ask:

RQ3: Based on audience content preferences, what kinds of editorial adjustments regarding article selection are editors more likely to make?

Recent economic woes have hit journalism hard, with newsrooms across the country reporting declining advertising revenues and layoffs (Kirchhoff, 2010; Napoli, 2010; Perez-Pena, 2008; Seelye, 2006). Of different types of news media, newspapers experienced the biggest economic problem and did not see any promising trend since the beginning of the last decade. In 2010, for example, newspaper circulation in the United States declined by 10.6% compared to a year earlier. Other media sectors being in the same boat were magazines, local TV, and network TV news (Pew Project for Excellence in Journalism, 2010b). However, local, cable, and network TV news saw recovery signs with viewership in 2011 increasing by 1.4%, 1%, and 4.5% for each of the sectors, respectively. The picture was not any brighter for newspapers and magazines (Pew Project for Excellence in Journalism, 2012). The growing online readership may be the salvation. Pew estimated that, in 2012, about 100 news websites have moved to digital subscription as a matter of survival. Uncertain about which directions to go in order to increase readership, journalists tend to check whether their editorial decisions resonate with audiences’ preferences (Lowrey and Woo, 2010). In addition, according to Shoemaker and Reese’s model, economic, organizational, and market competition pressures can influence journalists. Therefore, assessing economic influences on gatekeepers’ routines, including monitoring traffic and changing editorial decisions on news selection, is the focus of the following research questions:

RQ4: Does editors’ perception of economic benefits of getting high readership predict their adoption of new routines including (4a) monitoring web traffic and (4b) making editorial adjustments to article selection?

Method

Sample and instruments

This study employed a survey not only to collect data but also to assess views of editors of daily American newspapers and news sites on the influence of reader metrics on news selection. A list of newspaper editors across the United States, including their contact details, was provided by the media relations division of a public university. Editors from daily newspapers with a circulation of less than 10,000 were excluded from the population, because those editors are more likely to have tasks other than editorial as a result of recent financial cutbacks and technological changes (Anderson, 2011). Compared to the 2009 edition of Editors & Publisher, which lists 770 daily newspapers in the United States with a circulation of 10,000 and above, the database used in this study consists of 720 daily newspapers. The database was chosen because it provides direct contact to editors, who are directly involved in content production and placement across the country rather than just managing editors.

One of the goals of this research is to look at how declining readership and dwindling revenues have stricken the news organizations, forcing them to find better way to draw in audiences. This study chose to survey only newspapers’ editors, not online broadcast news sites, for the reason that for most broadcast institutions, television channels still play the major role in generating revenues. According to the Pew Project for Excellence in Journalism (2012), revenues generated from circulation and advertising for newspapers had fallen sharply, reaching a loss of 43% in 2011 compared to 2000. Advertising revenues for broadcast had posted a gradual increase for cable news, a fluctuation for local TV news, and a decline for network TV. The economic pressure for TV news sites’ gatekeepers may presumably be different from that for their online and print fellows.

Because the purpose of the study is to look at editorial decision-making on news articles, journalists with titles like photo editor, editor of photography, columnist, or editorial assistants were also excluded. The sampling frame consisted of 7012 editors of online news sites or newspapers. A systematic random sampling, which takes every other subject on the list, was conducted to select 3506 editors to participate in a web-based survey.

The survey questionnaire, containing 31 questions, was pretested several times online and offline on current and former editors before going through a standard institutional review board (IRB) process. The survey was administered online through Qualtrics, an online site. An invitation to participate in the survey was sent to 3506 editors in the list.

Thirty-three emails were no longer valid as the recipients had either changed their jobs, or retired. The final sample included 3473 editors. After three weeks, a total of 396 editors completed the survey, yielding a response rate of 11.4%. Although this response rate was low, it is not uncommon for web-based surveys. For example, when conducting a meta-analysis of organizational research in 17 refereed management and behavioral sciences journals in 2000 and 2005, Baruch and Holtom (2008) recorded web surveys with a response rate of as low as 10.6%. When using online methods to survey medical practitioners in Australia, Aitken et al.’s study (2008) on pharmacotherapy prescription posted an overall response rate of 8.7%. Mass communication research has also experienced the problem of declining response rate. Gibbs et al. (2006) received a response rate of 14.3% for their study on self-presentation in Internet dating. The rate seems to be even lower when surveying news workers. Aiming for a national sample of sport journalists, Wigley and Meirik (2008) had 13% out of 3021 subjects complete their survey. The present study adopted measures that are found to boost survey response rates, such as limiting the time to complete the survey to less than 15 minutes, keeping the questionnaire format simple, sending email reminders, emphasizing the academic benefits of the study to journalism research, guaranteeing confidentiality of respondents, and offering an incentive (Andrews et al., 2003). Out of 396 editors who responded to the survey, 78 of them skipped too many questions and were excluded from the analysis. The final sample size for analysis was 318. It was, however, large enough for analysis (Dillman, 2007; Keeter et al., 2000).

Measures

The survey asked respondents how much they were involved in editorial decision-making with regard to assignment of articles, selection of articles for sections, placement of articles on the homepage or the front page of the print newspaper, and/or selection of such additional elements as videos, photos, or audio. This filter question allowed for exclusion of those who were not involved in editorial decision-making. The survey also asked editors whether they monitor online traffic and how they did so.

Control variables

Demographics: editors were also asked to provide information on their age and journalism training. The age of respondents was assessed using five categories (18–25; 26–35; 36–45; 46–55; and 56 or older). Journalism education was measured using four categories (no training; short courses; undergraduate college training; and graduate training). The database used for sampling contained editors’ newspapers’ circulations. Previous research has found, in many cases, influence on journalists in their day-to-day work comes from various demographic factors such as their age, education, or their publication’s circulation (Herscovitz, 2004; Lowrey and Woo, 2010; McKenzie et al., 2011; Meyer, 2011; Schultz, 2002, Sylvie and Huang, 2008).

Major variables

Reasons to monitor traffic: editors’ reasons for monitoring traffic were measured through an open-ended question on why they monitor online traffic. Answers were then coded into five different categories based on what respondents had explicitly stated, including (1) audience scrutiny (e.g., “To see what people are looking at and what is keeping them on our site,” “to gauge the stories that interest and involve readers,” “to see what kinds of stories, videos, databases draw traffic and readers’ preferences”); (2) content adjustment (e.g., “to help in constantly adjusting mix and display on site,” “to help determine coverage; how to allocate our resources,” “to get an idea of how to play future stories; to pursue follows that have high readership; to decide what kinds of stories to cover”); (3) audience resonation (e.g., “just try to get all the feedback I can to make sure we’re doing what we should,” “to make sure that what we assign resonates with readers”); (4) corporate pressure (e.g., “Corporate wants us to,” “to please corporate”); and (5) advertising and/or marketing purpose (e.g., “to compile data for advertising”).

Two coders, who were graduate students, coded 40 (22.1%) of the answers for the intercoder reliability test. The Cohen Kappa result was .96, showing a strong agreement.

Factors of influences were measured by asking editors nine questions corresponding to the five levels of influences in Shoemaker and Reese’s (1996) model. The questions were on a four-point scale range: “not likely,” “somewhat likely,” “likely,” and “very likely.” For these questions, editors were asked about their likelihood of running an article if: (1) “Many readers might read it”; (2) “You believe readers need to know about it”; (3) “The sources are newsworthy”; (4) “Your newspaper/site usually covers it,” (5) “You think your competitor is going to run something similar”; (6) “It would get your organization more advertisement”; (7) “It promotes democracy or the greater good”; (8) “Your boss likes it”; and (9) “It’s in line with your personal values.”¹

Editorial adjustments are identified in this study as changes in editorial work according to audience appeals. Examples of adjustments are changing the placement of stories (e.g., increase or decrease the prominence of stories on their sites based on audience preferences), or updating most-viewed/most-read stories more frequently. Editorial adjustments were measured through seven four-point scale questions: “not likely,” “somewhat likely,” “likely,” and “very likely.” The question asked editors about what they would do based on audience metrics. Scenarios included (1) “Running articles of the same kinds as the most-viewed/ most-read;” (2) “Look for possible follow-up articles for the most-viewed/most-read ones;” (3) “Try to update most-viewed/most-read articles more often to attract audiences;” (4) “Try to look for possible editorials for most-viewed/most-read articles;” and (5) “Look for possible additional elements (video, pictures, sounds, etc.) for most-viewed/most-read articles.” Two items were about placement of articles: (1) “Make articles that drive more traffic more prominent on the homepage or on the front page;” and (2) “Make articles with low hits less prominent.”

These scenarios were partially adapted from Cox (2010). Choices ranged from “not likely,” to “somewhat likely,” “likely,” and “very likely.” The responses for these seven items were then summed to form an interval index to make it a dependent variable on editorial adjustments. The index showed a Cronbach’s Alpha coefficient of .84.

Traffic monitoring routines were measured by the question: “How often do you monitor traffic to your new site?” Answer choices comprised: (1) daily, (2) two to three times a week; (3) once a week; (4) two to three times a month; (5) less than once a month; and (6) never. Those who picked the first choice were asked how many times a day they monitored the traffic. These responses were then turned into numerals for analysis.

Editors’ perceived economic benefits were measured by asking them to rate their agreement on a seven-point scale statement: “Getting more readers is necessary because more readers mean high revenues,” with one meaning strongly disagree and seven mean strongly agree.

Results

The sample size was 318 editors across the United States. Most editors (60%) were responsible for both online and print content. About one-fifth (22%) worked only for the print version, while only a small number (2%) were online editors. In addition, 17% of respondents said their job titles were more specific (e.g., editorial page editor, features editor, travel editor, etc.). The inclusion of these respondents was based on whether they were involved in content decision-making (e.g., assignment of articles, selection of articles for sections, homepage article placement, etc.).

In terms of age, the largest were those in the 46–55 age group (38%). More than one-third of the respondents (35%) were 56 or older. One-fifth of the editors (20%) were between 36 and 45 years of age, whereas 7% were between 26 and 35. Only 1% was younger than 25.

Of the 312 respondents who answered the question about journalism training, 68% held an undergraduate in journalism. Nearly one-fifth (17%) had graduate training in journalism. Some editors (8%) did not receive any journalism training. A small number of editors (6%) took short courses in journalism.

In comparison to the characteristics of American journalists in a national survey by Weaver et al. (2007), the sample of this study was similar with regards to gender with 66% online and 66.4% print journalists being male. In this study 64% of the respondents were male and 36% were female. The median age for online and print journalists in Weaver et al.’s study was 39 and 41, respectively. The majority of American journalists (63%) did not major in journalism. In this study, the largest age group was between 46 and 55 (38%), and most respondents (85.3%) majored in journalism. Our sample skewed high in terms of age and education. A possible explanation is while Weaver et al. looked at all journalists without differentiating their positions, this study aimed at editors, who are more senior and experienced in news organizations.

Regarding circulation, the largest circulation was over 2 million, while the smallest was 10,274. The mean was 130,236 (SD = 212,728). The median was 52,510.

Of the 248 who were responsible for online or both online and print content, 84% said they monitored web traffic on a regular basis. Among those who monitored web traffic, more than half (52%) said they did so daily. Some said they checked online metrics hourly, whereas others did so either once or several times a day.

RQ1 asked about the reasons editors monitor web traffic. Responses to an open-ended question reveal: Almost three-fifth of editors (58%) monitor web traffic to scrutinize readers’ behavior; nearly one-third (31%) of editors said they use online metrics to plan content production. The remaining editors use online metrics (1) to check their news judgment against audience preferences (2%), (2) to follow corporate requests (6%), or (3) to prepare reports for advertising and marketing (3%).

To answer RQ2, which focused on comparing influences of different factors on the gatekeeping process, the study reported descriptive data on editors’ likelihood of running an article. We also used repeated measures analysis of variance (ANOVA) with Greenhouse–Geisser and Bonferroni corrections. Because this study tested nine factors and compared them pairwise, the use of these two statistical correction procedures would help minimize problems with sphericity violation and family-wise errors. Results of the ANOVA test with Greenhouse–Geisser correction showed that the mean of editors’ likelihood differed significantly between the nine scenarios (F(6.24, 1529.54) = 197.1, p<.001, ŋ² = .45). Post hoc tests using Bonferroni correction showed that editors’ likelihood to run an article because many readers might read it, M = 3.7, is significantly higher than how likely it is that they would publish a story based on the assumptions that (a) “The sources are newsworthy,” M = 3.4, p < .001; (b) “Your newspaper/site usually covers it,” M = 3.3, p < .001; (c) “You think a competitor is about to run something similar,” M = 2.8, p < .001; (d) “It would get your organization more advertisement;” M = 1.7, p < .001; (e) “It promotes democracy of the greater good,” M = 3.0, p < .001; (f) “Your boss likes it,” M = 2.7, p < .001; and (g) “It’s in line with your values,” M = 2.4, p < .001. The only comparison that was not statistically significant was between the two scenarios “Many readers might read it” and “Readers need to know about it,” M = 3.7, p = 1. This means when deciding on whether or not to run an article, editors are most influenced by the two assumptions about readers: “Many readers might read it” and “Readers need to know about it.”

RQ3 was concerned with the kinds of editorial adjustments that gatekeepers are more likely to make based on audience appeals. Of the seven practices included in the survey, editors are most likely to “make articles that drive more traffic more prominent (on the homepage or on the front page of the newspaper)” (M = 2.9, SD = .92) (see Table 1). Looking for “possible editorials for most-viewed/most-read articles” is the least likely change that editors would make (M = 2.0, SD = .98).

Table 1.

Editors’ likelihood of making editorial changes.

Changes	Mean	SD
“Make articles that drive more traffic more prominent on the homepage or on the front page.”	2.9	.92
“Look for possible follow-up articles for the most-viewed/most-read ones.”	2.7	.88
“Look for possible additional elements (video, pictures, sounds etc., for most-viewed/most-read articles.”	2.7	.92
“Try to update most-viewed/most-read articles more often to attract audiences.”	2.6	.9
“Run articles of the same kind as the most-viewed/most-read ones.”	2.3	.86
“Make articles with low hit less prominent.”	2.1	.85
“Try to look for possible editorials for most-viewed/most-read articles.”	2.0	.98

RQ4 is about the relationships between editors’ perceived economic benefits and two of their routines adopted as part of online newsroom practices, including monitoring traffic (RQ4a) and making editorial adjustments to news selection and placement (RQ4b; see Table 2). Control variables were age, journalism training, and newspaper circulation. In terms of predicting editors’ monitoring traffic, results of a multiple regression indicated that only journalism training was a negatively significant predictor of respondents’ attention to web traffic (β = –.14, p < .05). This means that those with journalism training will be less likely to monitor traffic. Perceived economic benefits, however, was not a statistically significant predictor for monitoring traffic.

Table 2.

Editors’ perceived economic benefits as a predictor for monitoring traffic and making editorial adjustments.

Variable	Traffic monitoring		Editorial changes
	Model 1a1	Model 2a1	Model 1a2	Model 2a2

	β	β	β	β
Block 1
Age	.08	.07	−.04	−.05
Journalism training	−.14*	−.15*	−.03	−.05
Newspaper circulation	.00	.01	−.06	−.04
Block 2
Economic benefits		.03		.23***
R²	.03	.04	.01	.06***

Notes: *p < .05, ***p < .001.

In assessing the relationship between editors’ perceived economic benefits and their likelihood of making editorial adjustments based on audience appeal, regression results showed that none of the demographics saw significant association with the dependent variable. However, editors’ perceived economic benefit of getting high readership was a significant predictor of their willingness to change their decisions on content production and/or presentation based on audience metrics (β = .23, p < .001). This demonstrates that the more editors feel that getting high readership brings economic benefits to the organizations, the more likely they are to make editorial changes based on online audience web analytics. In the last model (Model 2a2 in Table 2), R² is significant at .06, p < .001, indicating that the model accounts for 6% of the variance.

Discussion and conclusion

This study investigated the influence of online audience metrics on editorial decision-making. It found that to some extent, editors are willing to adjust their editorial decision-making based on web metrics. This willingness is influenced by their perceived economic benefits of getting readership.

In further examining this tendency of audience influence on newsroom practices, this research offered preliminary information on different reasons for editors to monitor web traffic. Still, most editors said that they monitor web metrics to only scrutinize audience behavior. However, nearly one-third also explicitly said that online metrics helps them plan future content production and/or placement. This, perhaps, is because the journalistic occupational pride of sustaining autonomy against any kind of non-professional influences made it harder for editors to admit that their editorial decision-making is affected by audience metrics. It is important to note that those who said they monitor web metrics only to learn about audience did not elaborate on the reason why they check online traffic statistics to scrutinize audience behavior. But just like an editor wrote about his motivation to track audience content preferences – “To judge what readers want, which is then balanced against what readers need” – this new routine in newsroom does raise a question on whether following web metrics would eventually affect editorial decision-making.

Findings of this study have theoretical implications. It provides an update of the hierarchy-of-influences model, comparing the importance of each factor in the model on the gatekeeping process in the new media environment. Specifically, the study found that gatekeeping today is very much audience-centric, with editors saying their decision to run an article is most affected by audience factors, either they think “readers need to know,” or “many readers might read it.” In the original hierarchy-of-influences model, audience is just one of many other extramedia factors including advertisers, sources, competing media, government controls, and economics. However, in the context of greater audience fragmentation coupled with the instantly available readers’ metrics, the audience factor has become more influential to online gatekeepers.

Findings of this study suggest a revision of the model with more emphasis being placed on the audience role when journalists practice gatekeeping. Perhaps, because of its importance to newsroom practice, audience factor should be given a separate level for how influential readers are to gatekeepers.

Another important contribution of this study is that it went a step further in explicating what sort of gatekeeping tasks in terms of news selection and placement journalists are willing to compromise to the audience’s wants manifested through the number of clicks on stories. Previous research mostly concentrated on investigating what affects gatekeepers in doing their job (McKenzie et al., 2011), or whether online audience behaviors have an impact on editorial decision-making (Lowrey and Woo, 2010). Knowledge on the degree to which audiences are involved in shaping news content today is generally short, especially in terms of quantitatively generalizable results. An insightful finding of this research shows most editors reported a relatively strong likelihood of making editorial changes based on web metrics ranging from “somewhat likely” to “likely.” This finding is in line with a stream of qualitative research on changes in newsroom practices. It demonstrates that, to a certain degree, journalists in digital newsrooms are increasingly exercising their gatekeeping selection-based audience interests. New tasks range from changing placement, adding extra editing or analysis, seeking additional web elements to running similar stories to what readers want to read.

Statistical analyses show associations between editors’ perceived economic benefits and their willingness to make editorial adjustments based on audience web metrics. A rather confounded finding is that there was a difference in the predictability of the economic pressure variable. It was found that editors’ perceived economic benefits of getting high readership do not drive them to monitor web metrics. However, this perception was a predictor of editors’ likelihood to make editorial changes. Perhaps this finding shows such an uncertainty in newsrooms today: journalists are trying to cope with rampant financial problems that have hit the industry, but are still unsure whether allowing deeper audience interference on their professional turf is the solution.

A similar complication was also seen with editors’ demographics. Results from this study show that journalism training does have influence on what new routines editors have adopted. However, there was an inconsistency in its predictive power. Editors with less journalism training tend to attend to web metrics more than those with higher journalism degrees. However, this tendency was not detected in its relation with editors’ willingness to change their decisions on content production and/or presentation. These relationships (i.e., between journalism education and economic benefits with the dependent variable) are therefore open for further exploration. As Lowrey and Woo (2010) point out, when having economic uncertainty, journalists tend to monitor traffic more. It is logical to believe that gatekeepers with journalism training would be more confident in making editorial judgment than those who did not gain the expertise through formal training. In terms of the relationship between economic pressure and editorial changes, a possible explanation is that occupational pride might, perhaps, have made it harder for journalists to admit that news content is increasingly becoming more like other products: catered to the tastes of consumers. Thus, they tend to downplay the importance of economics when self-reporting the influences on their decision-making.

This research supports the findings of previous studies such as Lee and Lewis (2012), Anderson (2011) and Domingo (2008), which stipulate that in the digital newsroom today, audiences have a significant role influencing editorial practices. Methodologically, this study found strong evidence on this influence by providing quantitative results through surveying current editors in newsrooms across the United States. The study proposes an update to the theory of influences in the new media environment with an increasingly important audience role. Also, findings of this study suggest that it is necessary to assess whether what online gatekeepers are willing to concede would match with what audience wants.

This study is not without limitations. Most notably, its response rate was low, weakening its representation of editors in the United States. Secondly, it uses only quantitative data, which might not help capture all the nuances of such a complex issue. However, it suggests several directions for future research. For example, studies should look deeper at factors that motivate editors to monitor traffic. As said, because of their occupational pride, some journalists might not admit economic influences when they do self-reporting about influences on their work. Another research study in the form of an experiment might help to discern such issues when these factors are manipulated.

This study offers insights into ongoing changes to the journalism industry, which eventually affect the quality of news. It explores an emerging issue within newsrooms across the United States, or perhaps the globe. That is, the tension between news users as content consumers and journalists as content producers in repositioning the gatekeeping control in the new media environment. Continuing efforts in examining such tension is important to journalism to help redefine journalism identity in the age of increasingly blurring boundaries between news users and producers.

Footnotes

Acknowledgements

The author would like to thank Renita Coleman, Thomas Johnson, George Sylvie, Iris Chyi, Paula Poindexter, and the two anonymous reviewers for their helpful comments on earlier versions of this article.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biography

Hong Tien Vu is a doctoral student at the School of Journalism, University of Texas at Austin. His research focuses on international/intercultural communication, agenda setting theory, and changes in the newsroom in the new media environment. He has published articles in peer-reviewed journals, including the Journalism & Mass Communication Quarterly and the Asian Journal of Communication.

References

Aitken

Power

Dwyer

(2008) A very low response rate in an online survey of medical practitioners. Australian and New Zealand Journal of Public Health 32(3): 288–289.

Anderson

(2011) Between creative and quantified audiences: Web metrics and changing patterns of newswork in local US newsrooms. Journalism 12(5): 550–566.

Andrews

Nonnecke

Preece

(2003) Electronic survey methodology: A case study in reaching hard-to-involve Internet users. International Journal of Human-Computer Interaction 16(2): 185–210.

Baek

Holton

Harp

. (2011) The links that bind: Uncovering novel motivations for linking on Facebook. Computers in Human Behavior 27(6): 2243–2248.

Baresch

Knight

Harp

. (2011) Friends who choose your news: An analysis of content links on Facebook. In: the ISOJ: the official research journal of international symposium on online journalism, Austin, TX.

Baruch

Holtom

(2008) Survey response rate levels and trends in organizational research. Human Relations 61(8): 1139–1160.

Benbunan-Fich

Fich

(2004) Effects of web traffic announcements on firm value. International Journal of Electronic Commerce 8(4): 161–181.

Boczkowski

(2004) Digitizing the News: Innovation in Online Newspapers. Cambridge, MA: The MIT Press.

Boczkowski

Peer

(2011) The choice gap: The divergent online news preferences of journalists and consumers. Journal of Communication 61(5): 857–876.

10.

Boorstin

(1961) The Image: A Guide to Pseudo-events in America. New York: Harper & Row.

11.

Bucy

(2004) Second generation net news: Interactivity and information accessibility in the online environment. International Journal on Media Management 6(1–2): 102–113.

12.

Cassidy

(2008) Outside influences: Extramedia forces and the newsworthiness conceptions of online newspaper journalists. First Monday 13(1) (accessed 8 July 2012).

13.

Chung

(2008) Interactive features of online newspapers: Identifying patterns and predicting use of engaged readers. Journal of Computer-Mediated Communication 13(3): 658–679.

14.

Chung

Yoo

(2008) Audience motivations for using interactive features: Distinguishing use of different types of interactivity on an online newspaper. Mass Communication and Society 11(4): 375–397.

15.

Cox

(2010) The changing role of the gatekeeper with online news. Unpublished master’s thesis. University of Kansas, Lawrence.

16.

Dillman

(2007) Mail and Internet Surveys: The Tailored Design Method. New York: Wiley.

17.

Domingo

(2008) Interactivity in the daily routines of online newsrooms: Dealing with an uncomfortable myth. Journal of Computer-Mediated Communication 13(3): 680–704.

18.

Gabszewicz

Laussel

Sonnac

(2002) Press advertising and the political differentiation of newspapers. Journal of Public Economic Theory 4(3): 317–334.

19.

Gans

(1979) Deciding What’s News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time. Chicago, IL: Northwestern University Press.

20.

Gibbs

Ellision

Heino

(2006) Self-presentation in online personals: The role of anticipated future interaction, self-disclosure, and perceived success in internet dating. Communication Research 33(2): 152–177.

21.

Gieber

(1960) How the “gatekeepers” view local civil liberties news. Journalism & Mass Communication Quarterly 37(2): 199–205.

22.

Herscovitz

(2004) Brazilian journalists’ perceptions of media roles, ethics and foreign influences on Brazilian journalism. Journalism Studies 5(1): 71–86.

23.

Hoffman

(2006) Is Internet content different after all? A content analysis of mobilizing information in online and print newspapers. Journalism & Mass Communication Quarterly 83(1): 58–76.

24.

Keeter

Miller

Kohut

. (2000) Consequences of reducing nonresponse in a national telephone survey. Public Opinion Quarterly 64(2): 125–148.

25.

Kirchhoff

(2010) The U.S. Newspaper Industry in Transition. Washington, DC: Congressional Research Service.

26.

Kperogi

(2011) Cooperation with the corporation? CNN and the hegemonic cooptation of citizen journalism through iReport. com. New Media & Society 13(2): 314–329.

27.

Lee

Lewis

(2012) Audience preference and editorial judgment: A study of time-lagged influence in online news. In: the international symposium of online journalism, Austin, TX, 20–21 April.

28.

Lewis

(2012) The tension between professional control and open participation: Journalism and its boundaries. Information, Communication & Society 15(6): 836–866.

29.

Lewis

Kaufhold

Lasorsa

(2010) Thinking about citizen journalism. Journalism Practice 4(2): 163–179.

30.

Lowrey

Woo

(2010) The news organization in uncertain times: Business or institution? Journalism & Mass Communication Quarterly 87(1): 41–61.

31.

MacGregor

(2007) Tracking the online audience. Journalism Studies 8(2): 280–298.

32.

McKenzie

Lowrey

Hays

. (2011) Listening to news audiences: The impact of community structure and economic factors. Mass Communication & Society 14(3): 375–395.

33.

Meyer

(2011) Highlights from the 2011 journalists engagement survey. Available at: http://www.rjionline.org/news/highlights-2011-journalists-engagement-survey (accessed 10 July 2012).

34.

Napoli

(2010) Audience Evolution: New Technologies and the Transformation of Media Audiences. New York: Columbia University Press.

35.

Perez-Pena

(2008) New York Times plans to cut 100 newsroom jobs, The New York Times, 14 February. Available at: http://www.nytimes.com/2008/02/14/business/media/14cnd-times.html?_r=0 (accessed 12 August 2012).

36.

Peters

(2010) Some newspapers, tracking readers online, shift coverage, The New York Times, 5 September. Available at: http://www.nytimes.com/2010/09/06/business/media/06track.html?_r=1 (accessed 18 November 2012).

37.

Pew Project For Excellence in Journalism (2010a). Understanding the participatory news consumer: News gets personal, social, and participatory. Avaliable at: http://www.journalism.org/analysis_report/news_gets_personal_social_and_participatory (accessed 22 July 2012).

38.

Pew Project For Excellence in Journalism (2010b). State of news media 2011. Available at: http://stateofthemedia.org/2010/overview-3/key-findings/ (accessed 5 August 2012).

39.

Pew Project For Excellence in Journalism (2012). State of news media 2012. Available at: http://stateofthemedia.org/2012/overview-4/major-trends/ (accessed 28 November 2012).

40.

Reese

(2011). Journalism research and the hierarchy of influences model: A global perspective. Brazilian Journalism Research 3(2): 29–42.

41.

Rowe

(2005) Broadcast News Writing for Professionals. Oak Park, IL: Marion Street Press.

42.

Schultz

(1999) Interactive options in online journalism: A content analysis of 100 US newspapers. Journal of Computer-Mediated Communication 5(1). Available at: http://jcmc.indiana.edu/vol5/issue1/schultz.html.

43.

Schultz

(2002) Does education matter? Characteristics of journalists who went to graduate school. Journalism 3(2): 223–238.

44.

Seelye

(2006) Washington Post to cut 80 newsroom jobs. The New York Times, 11 March. Available at: http://www.nytimes.com/2006/03/11/business/media/11post.html?pagewanted=print&_r=0 (accessed 16 July 2012).

45.

Shoemaker

Johnson

Seo

. (2011) Readers as gatekeepers of online news: Brazil, China, and the United States. Brazilian Journalism Research 6(1): 55–77.

46.

Shoemaker

Reese

(1991) Mediating the Message: Theories of Influences on Mass Media Content. White Plains, NY: Longman.

47.

Shoemaker

Reese

(1996) Mediating the Message. New York: Longman.

48.

Shoemaker

Vos

Reese

(2008) Journalists as gatekeepers. In: Wahl-Jorgensen

Hanitzsch

(eds) Handbook of Journalism Studies. New York: Routledge, pp.73–87.

49.

Singer

(2003) Who are these guys? The online challenge to the notion of journalistic professionalism. Journalism 4(2): 139–163.

50.

Singer

(2006) Stepping back from the gate: Online newspaper editors and the co-production of content in campaign 2004. Journalism & Mass Communication Quarterly 83(2): 265–280.

51.

Singer

(2011) Crowd control: Collaborative gatekeeping in a shared media space. In: the Association for Education in Journalism and mass communications annual conference, St. Louis. 10–13 August.

52.

Sylvie

Huang

(2008) Value systems and decision-making styles of newspaper front-line editors. Journalism & Mass Communication Quarterly 85(1): 61–82.

53.

Weaver

Randall

Brownlee

. (2007) The American Journalist in the 21st Century: U.S. News People at the Dawn of a New Millennium. Mahwah, NJ: Lawrence Erlbaum.

54.

White

(1950) The gatekeeper: A case study in the selection of news. Journalism Quarterly 27(4): 383–390.

55.

Wigley

Meirick

(2008) Interactive media and sports journalists: The impact of interactive media on sports journalists. Journal of Sports Media 3(1): 1–25.