The Imagined Audience for and Perceived Quality of News Comments: Exploring the Perceptions of Commenters on News Sites and on Facebook

Abstract

A number of news organizations have begun shifting commenting from their websites to Facebook, based on the implicit assumption that commenting on Facebook is an equivalent (or preferred) substitute. Using survey data from 317 online news commenters, and drawing on the concept of imagined audience, this article examines this assumption by comparing news commenters’ perceptions of imagined audiences for comments on news organizations’ websites and on Facebook. While news commenters had mostly different imagined audiences between the two platforms, they had similar evaluations of the personal dimensions of their audiences and the quality of news comments. News commenters on Facebook, for example, did not perceive their audiences to be any more reasonable, intelligent, or responsive—or any less aggressive—than did commenters on news organizations’ websites. Facebook commenters also did not perceive comments to be of any greater quality than did commenters on news organizations’ websites. Thus, it appears that at least in the context of aiming to elevate the quality and civility of civic discourse, news commenters do not perceive Facebook to be demonstrably better than news organizations’ websites. Implications for journalism, social media, and future research are discussed.

Keywords

audience digital journalism Facebook imagined audience news commenting

Introduction

In August 2016, NPR announced that it would no longer allow reader comments at the bottom of news stories. Instead, it was shifting the conversation to NPR’s Facebook pages. “After much experimentation and discussion,” NPR’s managing editor for digital news said, “we’ve concluded that the comment sections on NPR.org stories are not providing a useful experience for the vast majority of our users” (Montgomery, 2016). The majority of news organizations’ websites still maintain comment features at the bottom of news articles. However, given the difficulty of moderating those comments and amid the rapid growth of social media sites, some news organizations—including NPR, USA Today, Reuters, Bloomberg, Vice, The Verge, and Recode—have dropped comment sections from their websites and have encouraged users to move the conversation to social media platforms (Bode, 2016; Finley, 2015).

This shift toward encouraging news comments on Facebook is based on the implicit assumption that commenting on social media platforms, especially Facebook, is an equivalent—or even better—substitute for commenting on news organizations’ websites (e.g., Sonderman, 2011). Many news organizations have suffered from irrelevant, uncivil, abusive, and aggressive comments on their websites, in part, because of the anonymous/pseudonymous character of their comment sections (Coe, Kenski, & Rains, 2014). Facebook, in contrast, is less anonymous, a space with ostensibly “real names” and a system for notifying users of others’ likes and comments, altogether giving commenters greater accountability for their posts (Rowe, 2015a).

Shifting news commenting to Facebook requires readers to have Facebook accounts to comment and to reveal their account identity when commenting. The shift also makes it more likely that commenters’ Facebook friends may see those comments. In other words, the potential audience shifts from less personal, anonymous readers of news websites to a more intimate group of Facebook users, including one’s family and friends linked through the social network. What might this change mean for the imagined audience for news comments?

Despite a growing body of literature examining the content of news comments (e.g., to gauge the relative civility of comments), only a few studies have sought to understand the perceptions of news commenters themselves, particularly across different platforms. Considering the importance of news comment sections as spaces shown to influence how audiences evaluate the news (e.g., Prochazka, Weber, & Schweiger, 2016), it is essential to examine whether distinct platforms would perform comparably in terms of news commenters’ perceptions and behaviors.

To fill this gap in previous literature, this study addresses the following questions: (1) How do news commenters feel about the spaces in which they comment and the audience they are trying to reach with their comments? (2) Do news commenters perceive Facebook to be an equivalent or better substitute for commenting on news organizations’ websites? Specifically, through a survey of 317 news commenters, this study investigates differences in perceptions between respondents who primarily comment on news organizations’ websites and those who comment on news organizations’ Facebook pages. Drawing on the concept of “imagined audience” (Litt, 2012; Marwick & boyd, 2010), we explore how news commenters envision those reading their comments. We compare these perceptions across platforms to consider the implications of the trend to shift news commenting to Facebook.

Literature Review

Online News Commenting

For many years, virtually the only way for readers to publicly comment on the news was to do so via letters to the editor in newspapers (Wahl-Jorgensen, 2002). With the development of the Web and news websites in the late 1990s, and particularly amid the participatory (Web 2.0) turn during the 2000s (Domingo et al., 2008; Lewis, 2012), news organizations have provided comment sections on their sites, generally at the end of articles, to allow readers to express their opinions and perspectives, and to interact with other users (Rowe, 2015b). Like other forms of Web use over that time period, the growth in news comments has been extraordinary. As of 2016, more than half of American adults had reported leaving news comments online and nearly 80% said they read comments, according to a report by the Engaging News Project (Stroud, Van Duyn, & Peacock, 2016). Because of the growing popularity of leaving and reading comments online, these spaces for public discourse have become an important topic for communication research. Scholars have examined news comment sections to understand the impact of interactivity and other content features (Weber, 2014; Ziegele, Breiner, & Quiring, 2014), the quality of news comments (Coe et al., 2014; Rowe, 2015a, 2015b; Santana, 2014), motivations for commenting (Springer, Engelmann, & Pfaffinger, 2015; Stroud et al., 2016), personal characteristics of news commenters (Wu & Atkin, 2017), and the influence of news comments on users’ evaluation of news articles and social issues (Anderson, Brossard, Scheufele, Xenos, & Ladwig, 2014; Prochazka et al., 2016) and their future commenting behaviors (Rösner & Krämer, 2016). In addition, from the perspective of journalists, there have been a number of studies of how reporters and editors react to and negotiate the value of news comments (e.g., Loke, 2012; Nielsen, 2014; Robinson, 2010).

Many journalists have been and remain quite critical of comment sections (Bergström & Wadbring, 2015). Journalists perceive comment sections as dens of incivility and prefer to ignore or wish them away—much as journalists for decades have preferred to keep audiences at arm’s length (Lowrey, 2009), writing for each other rather than with the audience in mind (Boczkowski, 2010). Nevertheless, news comment sections remain widely used and potentially generative forms of public discourse and democratic debate (Ruiz et al., 2011). If designed and managed well, such commenting spaces may be valuable both for news organizations and the communities they seek to serve, not merely as brute feedback mechanisms or chaotic public spaces but as possible extensions of the public sphere (Zamith & Lewis, 2014). Beyond simply signaling their preferences toward a given news article, individuals can post additional information regarding news events, propose an alternative perspective, or attempt to correct information described in a news article or in other readers’ comments (Robinson, 2015; Rowe, 2015b). While reporters rarely comment on their own stories in response to readers, scholars nevertheless have emphasized how journalists and news organizations have a positive influence on comment sections by moderating and interacting with news commenters (Springer et al., 2015; Stroud, Scacco, Muddiman, & Curry, 2014; Ziegele & Jost, 2016).

Although news commenters function as important contributors for news organizations, there have not been many studies on the characteristics or perceptions of news commenters as they compare across platforms—that is, between a news organization’s own website and its presence on Facebook, for example—except a few studies on motivations for news commenting (Springer et al., 2015; Wu & Atkin, 2017). Wu and Atkin (2017) found that motivations of informing, exhibitionism, and obtaining feedback are significant predictors of the frequency of news commenting on news sites. Of note, only the motivation for social connection was found to be a significant predictor of the frequency of news commenting on social media, which suggests that motives for news commenting may vary across platforms.

Given the conversational features of news commenting and the importance of social motives for news commenting, research is needed to examine (1) how news commenters perceive their audiences (i.e., those who share the same platform and might read their comments) and (2) how (if at all) such perceptions influence their news commenting behaviors.

The Imagined Audience for News Comments in Online and Social Media Environments

Every form of communication has intended or imagined audiences in mind (Marwick & boyd, 2010). When individuals engage in a dialogical conversation, there is a specific other who reacts or responds to a given message. Authors construct an image of their audiences when crafting novels (Ong, 1975), and journalists and television producers have their audiences in mind when developing narratives (Matthew, 2008). Theories about self-performance indicate that when message producers have direct interaction with their audiences, they may use audiences’ actions and responses as feedback to adjust their behaviors (Goffman, 1959). However, when message producers do not have any direct interaction with audiences, an image of the audience contributes to shaping the character of messages generated (Gans, 1957; Matthew, 2008; Ross, 2014). Such construction of mass audiences, however, is not the exclusive purview of professional communicators. Indeed, as a type of message producers, news commenters also have an intended or imagined audience in mind when leaving comments.

In contrast to face-to-face conversations, news commenting behaviors are mediated in asynchronous digital environments. Although news commenters sometimes engage in direct conversation with a specific other (e.g., in a threaded conversation with a fellow news commenter), the majority of news comments are not oriented toward a particular individual. Thompson (1995) conceptualized this type of interaction as quasi-mediated interaction, which refers to interaction mediated by mass communication technologies (e.g., radio, newspapers), thus having an unlimited and undefined range of potential recipients. Since news comments on news organizations’ websites and social media could be revealed to an indeterminate set of people, including anyone who visits news websites or scans news content on social media, news commenters presumably leave comments on news based on their imagined audiences. Previous research has shown that perceptions about the nature and reaction of audiences could affect a message producer’s behaviors (Gans, 1957; Goffman, 1959; Ross, 2014). Thus, we examine news commenters’ perceptions of audiences so as to understand their role in news commenting behaviors.

In recent years, scholars have deployed this concept of the imagined audience to study a range of social behaviors online, such as users’ privacy management or personal branding on social media (e.g., Baym & boyd, 2012; Brake, 2012; Duffy, Pruchniewska, & Scolere, 2017; Litt, 2012; Litt & Hargittai, 2016; Marwick & boyd, 2010). The imagined audience is defined as “the mental conceptualization of the people with whom we are communicating” (Litt, 2012, p. 331), which describes both the people assumed to be on the receiving end of communication and the perceived personal qualities and characteristics of individuals in the audience. In general, online and social media users have diverse perceptions of their audiences, ranging from a particular person to a large group of strangers (Brake, 2012; Litt & Hargittai, 2016; Marwick & boyd, 2010). Specifically, Litt and Hargittai (2016) found that roughly half of social media users had broad, abstract imagined audiences for their social media posts, presuming that anyone might see them. When people have a specific group as their imagined audience for their social media posts, most of those perceptions are based on personal ties (e.g., to family or friends; Litt & Hargittai, 2016; Vitak, Blasiola, Patil, & Litt, 2015). Duffy et al. (2017) showed that creative workers (e.g., bloggers/writers, designers, or marketers) have different perceptions regarding demographics, the level of professionalism, personality, and other qualities of their imagined audiences across platforms.

While a number of factors influence online and social media users’ perceptions of audiences—ranging from their personality and motivations, to a platform’s policies and social norms (Brake, 2012; Duffy et al., 2017; Litt, 2012)—this study focuses on how individuals’ perceptions of imagined audience vary across different platforms, in the specific context of news commenting.

In general, scholars have argued that the characteristics or affordances of a platform can influence individuals’ perceptions of their audiences (Baym & boyd, 2012; Litt, 2012). In communication, affordances describe “what material artifacts such as media technologies allow people to do” (Bucher & Helmond, 2018, p. 3). As such, the algorithms that a social media platform uses to display users’ posts, or the manner in which the platform offers audience feedback (e.g., likes or comments), could affect individual users’ perceptions of their actual or potential audiences (Litt, 2012). Likewise, and particularly relevant for news commenting (Rowe, 2015a; Santana, 2014; Wallsten & Tarsi, 2014), a site or platform’s relative acceptance of user anonymity may be a key factor in users’ imagined audience (Litt, 2012).

Bringing these concepts together, Nagy and Neff’s (2015) notion of “imagined affordance” sheds light on how platform-based characteristics influence users’ perceptions and behaviors. Imagined affordance refers to the properties of technologies (or platforms) that are imagined by users, which include users’ (mis)perceptions, attitudes, and expectations about the qualities of technologies. Nagy and Neff (2015) emphasize that imagined affordances emerge out of a complex interaction between technologies and users’ perceptions of their characteristics. To imagine an audience in connection with media technologies, therefore, is not simply to conceive of possible recipients to communication, but to imagine both who those individuals are and what they might do according to the enabling and constraining features of various platforms (cf. Duffy et al., 2017).

Thus far, with regard to news commenters and news comment sections, few studies have attempted to understand the form and influence of imagined-audience perceptions across different platforms. Kim (2016) conducted a series of interviews with news commenters to ask whom they perceived when they left comments on news articles, both those posted on news organizations’ websites and on social media. News commenters who primarily commented on news organizations’ websites perceived their audience as public and broader in scope, while news commenters on Facebook imagined their audience as friends or family, rather than strangers, and thus felt they would be more connected with them. Interestingly, it seems that news commenters on Facebook believed that because of a lack of anonymity, Facebook would be a more civil place than news organizations’ websites. These findings suggest that the characteristics of the platforms can make a difference in how news commenters conceptualize their audiences.

News Commenting on Different Platforms

There has long been discussion about the role of social media platforms in hosting public debate, particularly in the case of news content (Braun & Gillespie, 2011). While many news organizations still maintain commenting sections on their websites, the gradual shift toward social platforms, particularly Facebook, as the mode for hosting news comments developed out of news organizations’ frustration with managing thousands of comments—most of them anonymous/pseudonymous, some of them vulgar and vitriolic. So far, there have been few studies on the differences between news organizations’ websites and their Facebook pages in terms of the quality of news comments. Unlike studies that compare two conditions—for example, commenting spaces with versus without anonymity—on the same news sites (e.g., Santana, 2014), there has been conflicting evidence regarding the quality of news comments across the two platforms (Hille & Bakker, 2014; Rowe, 2015a, 2015b). Rowe (2015a) compared news comments on political news between the Washington Post site and the Washington Post Facebook page and found that comments on Facebook showed more civility and politeness than the news site. However, Rowe (2015b) found that comments on the news site were more likely to be relevant to the topic, be ideologically balanced, offer alternative perspectives, include sources for their arguments, and pose questions for getting more clarity, compared to news comments on its Facebook page. Hille and Bakker (2014) also found that commenters on news sites provided more elaborate comments, but Facebook news commenters provided more personalized comments on news. These mixed results suggest that the platform is not the only factor affecting news commenters’ behaviors.

In all, scholars have examined news comments’ quality by conducting content analyses of news comments rather than exploring news commenters’ perceptions of quality. Indeed, despite a widespread belief that anonymity has a bad influence on the quality of news comments, some news commenters still prefer anonymous comment sections (Hille & Bakker, 2014). They argue that anonymity helps facilitate free speech among news commenters by reducing the feeling of any threat or the fear of social disapproval or isolation (Cho & Kwon, 2015). Hille and Bakker (2014) showed that there was a low number of comments on news organizations’ Facebook pages compared to their news sites, which means that removing anonymity might make news commenters hesitant to comment on news. Similarly, Stroud et al. (2016) explained that news commenters have different perceptions about the impact of anonymity on their comments. Sometimes, news commenters believe that anonymity guarantees high-quality comments, especially for controversial topics. Ultimately, it is unclear how news commenters perceive their imagined audience and the relative quality offered by their fellow commenters on a given platform for news commenting.

Research Questions

With the shift in comment sections from news organizations’ websites to Facebook, there is an implicit assumption that Facebook would be an equivalent (and perhaps better) place for news commenting. Because of the difficulty in reaching news commenters (it is far easier to content-analyze their postings than to track them down and interview/survey them), scholars have examined the quality of news comments according to dimensions such as civility or politeness across platforms. From existing research, however, it is not entirely clear how news commenters perceive each space for commenting and the imagined audiences they are seeking to reach as a factor that influences their actual comments.

Previous studies on imagined audiences found that online and social media users tend to develop a variety of perceptions about their audiences, ranging from a particular person to a large group of strangers (Brake, 2012; Litt & Hargittai, 2016; Marwick & boyd, 2010). Although studies that examine differences in imagined audience perceptions across platforms have been missing, studies of imagined affordances suggest that features of a given platform could change users’ perceptions and attitudes toward the platform (Nagy & Neff, 2015). Given different features (e.g., anonymity) of news organizations’ websites relative to Facebook as a place for news commenting, news commenters’ perceptions may change as news organizations move their comment sections from their websites to Facebook.

Thus, this study compares news commenters’ perceptions of two key elements: (1) imagined audiences (who they are and what type of people they are) and (2) news comments (their relative quality), depending on commenters’ preferences for news organizations’ websites or Facebook as their platform of choice for news comments.

Three research questions are posed as follows:

RQ1: Do commenters on news organizations’ websites imagine different audiences for their comments than do commenters on Facebook?

RQ2: Do news commenters’ evaluations of the personal dimensions of their imagined audiences differ between news organizations’ websites and Facebook?

RQ3: Are news commenters’ evaluations of the quality of others’ comments different between news organizations’ websites and Facebook?

Methods

Survey Procedures and Participants

To examine news commenters’ perceptions across different platforms, this study draws from a partial sample of survey data that examined news commenters, lurkers, and non-engaged respondents. In all, 648 adult respondents from the United States were recruited through an online Qualtrics panel between September 15 and 24, 2016, and between October 12 and 20, 2016. For this particular study, the population includes individuals who reported reading news online and commenting on news on news organization’s websites and/or Facebook. Given the relatively low number of self-identified commenters on news organizations’ websites and news organizations’ Facebook pages, this study relied on a purposive sample as means to capture a very finite population while balancing research costs. Once participants completed the consent form, they were presented with a deception question about use of streaming media services. Following the deception question, individuals were asked to answer regarding whether they had read online news in the past week. Those who selected “yes” were then moved through subsequent screening (quota) questions regarding their commenting or lurking behaviors on news organizations’ websites and news organizations’ Facebook pages during the past 2 months. More than 5,100 individuals attempted the survey, of which 2,639 individuals qualified based on the initial qualifying question.

The two quota questions were used to gather a roughly equal number of individuals in each group: news commenters on news organizations’ websites, news commenters on Facebook and other social media, those who only read comments (lurkers), and those who neither comment on news nor read comments (non-engaged respondents). When a group quota was filled, participants were terminated from the survey. Given the proposed analytical test, cost, and time to complete the given survey, we collected answers from 648 respondents.

News commenters were asked to answer which platform they most frequently used to post comments on news stories. For lurkers and non-engaged respondents, they simply were asked whether they had read comments posted on news organizations’ websites or news stories posted to Facebook in the past 2 months. In total, there were 158 news commenters on news organizations’ websites or their applications (24.38%), 159 news commenters on Facebook and 16 news commenters on other social media such as Reddit or Tumblr (27.01%), 158 lurkers (24.38%), and 157 non-engaged respondents (24.23%). Respondents who successfully completed the survey in their quota group were granted monetary compensation for their participation.

Because this study sought to examine perceptions of news commenters across different platforms, only data from news commenters on news organizations’ websites (or their applications; N = 158) and Facebook (N = 159) were used for analysis. Because there were only 16 news commenters on other social media sites besides Facebook, those news commenters were not included. The demographics of the resulting sample are described in the Appendix.

Measures

Perception of Each Group of Imagined Audience for News Comments

News commenters were asked to indicate the extent to which they perceive a particular group of people as their audience when leaving comments in their primary place for commenting, on a 7-point scale: other commenters on the website or platform (M = 4.40, standard deviation [SD] = 1.55), other people on the website or platform who read comments but do not post comments themselves (i.e., lurkers; M = 4.43, SD = 1.56), journalists and/or news editors (M = 3.28, SD = 1.81), family and/or friends (M = 4.25, SD = 1.92), those who are interested in the issues that I commented on (M = 4.85, SD = 1.42), politicians and/or governmental officials (M = 2.89, SD = 1.81), and trolls (i.e., people who want to ruin the commenting experience for others; M = 3.51, SD = 1.89). The categories of groups reflect all the groups mentioned by news commenters in a previous interview study concerning news commenters’ imagined audience perception (Kim, 2016).

Evaluation of Personal Dimensions of the Imagined Audience

News commenters would not only perceive their imagined audiences as a particular group but also evaluate them positively or negatively (Kim, 2016). This study operationalizes such evaluations according to how news commenters assess people who read their comments according to several personal dimensions. News commenters were asked to evaluate, on a 7-point scale, the extent to which these 11 dimensions apply to the people who read their comments: thoughtful, reasonable, intelligent, uninteresting [reverse-coded], similar to me, open-minded, opinionated [reverse-coded], aggressive [reverse-coded], honest, sarcastic [reverse-coded], and responsive (Cronbach’s α = .80, M = 4.61, SD = 0.72). These personal dimensions were based on interview data with news commenters in Kim (2016).

Evaluation of the Quality of News Comments

News commenters were asked to indicate their level of agreement with five statements regarding the overall quality of news comments. The scale ranged from strongly disagree (1) to strongly agree (7). These qualities include civil, unfriendly [reverse-coded], valuable, informative, and helpful (Cronbach’s α = .82, M = 5.19, SD = 0.82).

Data Analysis

To control for demographic influence on news commenters’ perceptions (see Table 1 for the demographic differences between the two groups of news commenters), this study conducted analyses of covariance (ANCOVAs) to assess news commenters’ comparative perceptions of their imagined audiences (and such audiences’ perceived personal dimensions) and their evaluation of the quality of news comments. Gender, age, race, income, and education were included as covariates for all the analyses.

Table 1.

Preliminary Analysis of News Commenters in Online and Social Media.

	News commenters on news organizations’ websites (N = 158)	News commenters on Facebook (N = 159)
	M (SD)	M (SD)	t(315)
Age (years)	53.80 (16.79)	52.96 (15.42)	0.46
Gender (male = 1)	0.63 (0.48)	0.39 (0.49)	4.45^***
Race (White = 1)	0.91 (0.29)	0.88 (0.33)	0.71
Education	4.67 (1.45)	3.92 (1.56)	4.46^***
Income	3.67 (1.47)	3.11 (1.35)	3.56^***

Note. The degree of freedom for education is 313.668, since the equal variances were not assumed. SD = standard deviation.

***

p < .001.

Results

News Commenters’ Perceptions of their Imagined Audience

RQ1 and RQ2 were asked to examine how news commenters perceive their imagined audience differently across platforms. To answer RQ1, a series of ANCOVAs were conducted for each category of imagined audience.

As seen in Table 2, news commenters on news organizations’ websites (M_website = 4.68, SD = 1.54) perceived “other news commenters” as their audience more than news commenters on Facebook (M_Facebook = 4.13, SD = 1.53), F(6, 310) = 6.69, p = .01. In contrast, news commenters on Facebook (M_Facebook = 4.95, SD = 1.59) showed that they considered “family/friends” as their audiences more than news commenters on news organizations’ websites (M_website = 3.54, SD = 1.97), F(6, 310) = 61.00, p < .001. It appears that family and friends were the main group perceived as an audience for comments among news commenters on Facebook, while other news commenters were the main audience group for news commenters on news organizations’ websites. Altogether, this showed clear differences in news commenters’ perceptions between the platforms.

Table 2.

Effects of Platforms on News Commenters’ Imagined Audience Perceptions.

	News commenters on news organizations’ websites (N = 158)	News commenters on Facebook (N = 159)
	M (SD)	M (SD)	F(6, 310)
Imagined audience
Other commenters	4.68 (1.54)	4.13 (1.53)	6.69^*
Lurkers	4.51 (1.60)	4.36 (1.52)	0.63
Journalists/news editors	3.64 (1.91)	2.92 (1.63)	8.59^**
Family/friends	3.54 (1.97)	4.95 (1.59)	61.00^***
People who are interested in same issue	4.77 (1.43)	4.93 (1.40)	1.88
Politicians	3.22 (1.97)	2.57 (1.58)	5.52^*
Trolls	3.87 (1.90)	3.16 (1.82)	9.52^**

Note. SD = standard deviation.

p < .05.

These differences reflect previous findings that social media users tend to perceive their audiences based on personal ties (Litt & Hargittai, 2016) and also that comments on Facebook tend to be more personal than those on news organizations’ websites (Rowe, 2015b). Although we did not examine the relationship between news commenters’ perceptions of imagined audiences and their actual comments, the result shows that the differences in news commenters’ imagined audience across the platforms might affect their actual news commenting behaviors.

In addition, news commenters on news organizations’ websites (M_website = 3.64, SD = 1.91) perceived that “journalists” would read their comments more than news commenters on Facebook (M_Facebook = 2.92, SD = 1.63), F(6, 310) = 8.59, p = .004. News commenters on news organizations’ websites (M_website = 3.22, SD = 1.97) also perceived “politicians/governmental officials” as their audiences for news comments more than news commenters on Facebook (M_Facebook = 2.57, SD = 1.58), F(6, 310) = 5.52, p = .02. These differences indicate that news commenters on news organizations’ websites, more so than those on Facebook, hope to interact directly with journalists or politicians and also aim to participate in politics and the journalistic process through commenting activities.

News commenters on news organizations’ websites (M_website = 3.87, SD = 1.90) perceived “trolls” as their audiences more than news commenters on Facebook (M_Facebook = 3.16, SD = 1.82), F(6, 310) = 9.52, p = .002. That is, news commenters on Facebook were less likely to be concerned about being trolled by other commenters when they made comments on news as compared to news commenters on news organizations’ websites. Given that one of the reasons for turning to Facebook is to reduce aggressive and uncivil comments by making news commenters use their “real names,” the lack of anonymity presumably would help reduce trolling—at least according to the perceptions of news commenters.

There were no significant differences in the categories of “lurkers” (M_website = 4.51, SD = 1.60 and M_Facebook = 4.35, SD = 1.52) and “people who are interested in the same issue they commented on” (M_website = 4.77, SD = 1.43 and M_Facebook = 4.87, SD = 1.40) between news commenters on news websites and on Facebook, respectively. Still, news commenters on both platforms tended to believe strongly that these two groups were, in fact, their audiences.

To answer RQ2, an ANCOVA was conducted to examine news commenters’ evaluation of personal dimensions of the imagined audience. Although news commenters on Facebook (M_website = 4.66, SD = 0.65) showed a slightly more positive evaluation of the quality of their imagined audience than news commenters on news organizations’ websites (M_Facebook = 4.56, SD = 0.78), there was no significant difference between them, F(6, 310) = 1.80, p = .18.

When we conducted a series of ANCOVAs for each aspect of imagined audiences, however, we found that news commenters on Facebook were more likely to perceive their audiences as interesting and similar to themselves than commenters on news organizations’ websites (see Table 3). These two aspects might be less related than others to commenters’ evaluation of their audiences as a discussion partner. It appears that news commenters on Facebook, at least, were more likely to evaluate their audiences as similar to themselves, and thus feel more closeness with their audiences than news commenters on news organizations’ websites.

Table 3.

Effects of Platforms on News Commenters’ Evaluation of Imagined Audiences and Perceived Quality of Others’ Comments.

	News commenters on news organizations’ websites (N = 158)	News commenters on Facebook (N = 159)
	M (SD)	M (SD)	F(6, 310)
Aspect of imagined audiences	4.66 (0.65)	4.56 (0.78)	1.80
Thoughtful	5.06 (1.24)	5.11 (1.04)	0.26
Reasonable	5.08 (1.22)	5.19 (1.52)	1.29
Intelligent	5.20 (1.20)	5.19 (1.02)	0.10
Interesting	4.58 (1.37)	4.95 (1.26)	4.65^*
Similar to me	4.84 (1.31)	5.07 (1.07)	3.95^*
Open-minded	4.94 (1.44)	5.04 (1.17)	1.18
Non-opinionated	2.58 (1.21)	2.66 (1.25)	0.21
Non-aggressive	3.90 (1.48)	4.11 (1.36)	1.86
Honest	5.08 (1.26)	4.99 (1.08)	0.21
Non-sarcastic	3.80 (1.45)	3.72 (1.41)	0.58
Responsive	5.09 (1.13)	5.18 (1.03)	0.24
Quality of news comments	4.78 (1.11)	4.88 (1.33)	0.27
Civil	4.78 (1.52)	4.48 (1.49)	1.52
Friendly	4.20 (1.69)	4.43 (1.61)	1.11
Valuable	4.87 (1.49)	4.77 (1.24)	0.78
Informative	4.97 (1.42)	4.98 (1.20)	0.19
Helpful	4.89 (1.46)	4.88 (1.22)	0.6

Note. SD = standard deviation.

p < .05.

News Commenters’ Evaluation of the Quality of News Comments

RQ3 was posed to assess differences in news commenters’ evaluations of the quality of news comments between platforms. An ANCOVA was conducted for news commenters’ evaluations of other comments. There was no significant difference between news commenters on news sites and news commenters on Facebook (M_website = 4.78, SD = 1.11 and M_Facebook = 4.88, SD = 1.33), F(6, 310) = 0.27, p = .61. As Table 3 shows, there was no difference in individual categories.

This result showed that Facebook commenters did not perceive comments to be of any greater quality than did commenters on news organizations’ websites, regardless of the lack of anonymity on Facebook. In part because we only analyzed data from active news commenters (as opposed to non-engaged respondents), comments and commenters generally were evaluated positively across platforms.

Discussion

As news organizations consider shifting commenting from their websites to social media platforms (namely Facebook), the emerging academic literature on news commenting generally, and differences between commenting across platforms specifically, means that assumptions underlying this decision remain largely untested. The desire to shift comments from news organizations’ websites to Facebook is based on the implicit assumption that commenting on Facebook is an equivalent (and perhaps better) substitute. This article examined this assumption from the perspective of news commenters and their relative perceptions.

Commenters on Facebook imagined that trolls were less likely to be among their audience than did commenters on news organizations’ websites. Interestingly, though, the perception of fewer trolls did not translate to a better overall assessment of the personal dimensions of imagined audiences or of the quality of the discourse. News commenters on Facebook, for example, did not perceive fellow commenters to be any more reasonable, intelligent, or responsive—or any less aggressive—than did commenters on news organizations’ websites.

These findings suggest that while Facebook commenters may be slightly less concerned about being trolled by fellow commenters, commenters generally do not share either news organizations’ or researchers’ conclusions that the lack of anonymity on Facebook produces a more civil forum for news commenting. Thus, it appears that at least in the context of aiming to elevate the quality and civility of the civic discourse, news commenters do not perceive any differences between news websites and Facebook as spaces for commenting on news.

However, in terms of perceptions of their imagined audiences, it seems that news commenters did not perceive the audiences across the two platforms to be equivalent, which could potentially mean a change in commenting behaviors. Family and friends tended to be perceived as the main audience for news commenters on Facebook, while news commenters on news sites were more likely to perceive other news commenters as their primary audience. In addition, news commenters on news organizations’ websites were more likely to perceive journalists and politicians as their imagined audience than were news commenters on Facebook.

Those differences likely would affect news commenters’ motivations, satisfaction with the commenting features, frequency of their commenting, and even their actual comments on news, according to previous studies. For example, Springer et al. (2015) discovered that news commenters’ motivations for interacting with journalists or other users could increase the frequency of commenting. And news commenters’ motives for interacting with journalists could affect their satisfaction with commenting affordances. Those findings, in combination with ours, suggest that news organizations’ websites may be better places for eliciting more comments and satisfying news commenters, given that in our study, news commenters on news organizations’ websites were more likely to perceive their audience as journalists.

Of course, Wu and Atkin (2017) showed that news commenting on distinct platforms may be driven by distinct motives—for example, the motive of informing others or getting feedback on news organizations’ websites and the motive of connecting with others (mostly friends or family) on Facebook. Because news commenters on news organizations’ websites perceive other commenters as their audiences, they might use the comment sections for the purpose of discussion. In contrast, Facebook commenters appear more inclined to use the platform for connecting or interacting with their friends or family, consistent with their imagined audience perceptions.

These findings are also consistent with Rowe’s (2015a) conclusions that comments on the Washington Post’s Facebook page are more personal, while comments on the newspaper’s website are less civil but also more topic-based, elaborate, and ideologically balanced (Rowe, 2015b). Those on Facebook are more likely to imagine their friends and family when commenting, whereas commenters on news organizations’ websites are more likely to perceive that they can discuss a news event, reach, and potentially influence decision-makers—and thus might be more likely to invest effort in elaborating on their comments.

In terms of audience engagement in public dialog and participatory news-making processes, there are several implications from these findings. First, considering the potential for commenting spaces on news websites to act as an extension of the public sphere (Zamith & Lewis, 2014), the shift to Facebook may provide less opportunity for citizens to interact with journalists or other users regarding a given news item. In contrast, on Facebook, because news commenters tend to perceive their friends and family as the main audience, motivations for commenting and the nature of comments likely would become more personal rather than public, altogether diminishing the potential for publicly oriented conversation. Even more, from the perspective of participatory journalism, news comment sections have been an important means of allowing users to offer additional perspectives, suggest corrections to news stories, and express their feelings toward journalists or news organizations (Robinson, 2015). If news commenters were less likely to believe that reporters and editors would see their comments, those participatory journalism behaviors presumably would decrease. Finally, differences in imagined audience perceptions between news organizations’ websites and Facebook pages not only change the relationship between audiences and journalists but also potentially transform relationships among audience members. In short, the sense of community that may exist in comment spaces could change according to revised notions of the imagined audience. Lin (2008) showed that participation in an online forum over time could make users view the forum as a community in which loyalties develop, leading to greater participation, and Meyer and Carey (2014) similarly found that a (virtual) sense of community was critical to encouraging engagement in the comment sections of news websites. Turning to Facebook changes the sense of audience and, in turn, the sense of community.

In all, this study contributes important insights into the differences that news commenters perceive—and do not perceive—between news organizations’ websites and Facebook as forums for commenting on the news. This study is limited using a non-random, purposive sample. Thus, we cannot be certain how well our findings reflect the general population of news commenters, nor can we examine the relative preference among news commenters for commenting on news organizations’ websites versus commenting on Facebook. Given that very few news readers comment on the news, drawing a sufficiently large, random sample of news commenters is cost-prohibitive for most studies, which makes multiple, triangulated studies of news commenting that much more important.

Beyond future replications of this study’s findings concerning the imagined audience, future studies should consider other important perceptions of commenters across commenting platforms: Is one forum perceived as being more inclusive than the other? How is the frequency of news commenting associated with imaginations about the audience? How do perceptions of and preferences for different types of imagined audiences influence different choices for news commenting platforms? (We did find that a majority of commenters on news organizations’ websites were male, while the majority of commenters on Facebook were women. Overall, the gender breakdown of commenters was roughly 50/50. Is Facebook perceived as safer for women who comment on the news?)

Despite its limitations, this study contributes a better understanding of whom news commenters imagine as their potential audiences, and how imagined audiences differ between platforms. Given that previous studies on the imagined audience concept have been conducted using qualitative interviews or case studies, this study suggests an alternative way to measure the imagined audience by providing several categories of imagined audiences, ranging from other commenters to trolls. Future studies might also consider how imagined audiences affect news commenting behaviors and the content of what commenters post. Generally, though, we find that at least in terms of the perceptions of commenters’ imagined audiences, Facebook is not an equivalent substitute for commenting on news organizations’ websites. More research is needed to verify existing conclusions about news commenting and to further challenge some of the assumptions implicit in decisions that news organizations are making about how to handle news comments.

Footnotes

Appendix

Descriptive Statistics of News Commenters.
	All news commenters (N = 317)
	M (SD), %
Age (years)	53.38 (16.10)
18-29	5.7
30-49	32.7
50-65	32.4
65+	29.1
Gender
Male	51.1
Female	47.9
Other	0.9
Race
White, non-Hispanic	89.3
Black or African American	4.1
Asian or Asian American	3.8
Native American	0.3
Other	2.5
Education
High school graduates	16.1
College graduates	62.2
Advanced degree (master’s, PhD, or JD)	21.8
Income
<$15K	3.5
$15K to <$50K	32.8
$50K to <$75K	22.1
$75K to <$100K	14.8
More than $100K	26.2
Prefer not to answer	0.6
Political ideology (1 = very liberal)	4.07 (1.77)
News use	4.34 (1.16)

Note. When comparing this table with demographics of news commenters on a news site, app, or social media page (n = 600) collected in Stroud et al. (2016), our sample includes more news commenters older than 65 years (10.5% in Stroud et al., 2016) and White (63% in Stroud et al., 2016), but fewer news commenters whose age is below 29 years (27.5% in Stroud et al., 2016). The differences might result from the fact that our population is composed of those who not only comment on news but also reported having read online news in the past week. Overall, however, demographics of news commenters between our study and theirs are roughly similar on measures such as gender, education, and income.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Jisu Kim (MA, Korea University), is a doctoral candidate in the Hubbard School of Journalism & Mass Communication at the University of Minnesota—Twin Cities (USA). Her research focuses on how digital technologies change audiences’ engagement in the journalistic process, with particular expertise in a computational research approach.

Seth C. Lewis (PhD, University of Texas at Austin), is Shirley Papé Chair in Emerging Media in the School of Journalism and Communication at the University of Oregon (USA). His research interests include journalism studies and boundary work, social media and journalist–community dynamics, and the emerging roles of algorithms, automation, and artificial intelligence in how news is constructed, circulated, and consumed.

Brendan R. Watson (PhD, University of North Carolina at Chapel Hill), is an assistant professor of journalism innovations at Michigan State University. His research interests include community information needs and civic engagement; how emerging digital and participatory communication technologies are or are not changing the availability of, and the public’s interaction with, public affairs news and information; and quantitative research methods, particularly content analysis.

References

Anderson

A. A.

Brossard

Scheufele

D. A.

Xenos

M. A.

Ladwig

(2014). The “Nasty effect”: Online incivility and risk perceptions of emerging technologies. Journal of Computer-Mediated Communication, 19, 373-387. doi:10.1111/jcc4.12009

Baym

N. K.

boyd

(2012). Socially mediated publicness: An introduction. Journal of Broadcasting & Electronic Media, 56, 320-329. doi:10.1080/08838151.2012.705200

Bergström

Wadbring

(2015). Beneficial yet crappy: Journalists and audiences on obstacles and opportunities in reader comments. European Journal of Communication, 30, 137-151. doi:10.1177/0267323114559378

Boczkowski

P. J.

(2010). News at work: Imitation in an age of information abundance. Chicago, IL: University of Chicago Press.

Bode

(2016, December 27). Vice joins trend of killing news comments because giving a damn about your site’s community is just too hard. Techdirt. Retrieved from https://www.techdirt.com/articles/20161227/09365436348/vice-joins-trend-killing-news-comments-because-giving-damn-about-your-sites-community-is-just-too-hard.shtml

Brake

D. R.

(2012). Who do they think they’re talking to? Framing of the audience by social media users. International Journal of Communication, 6, 1056-1076.

Braun

Gillespie

(2011). Hosting the public discourse, hosting the public: When online news and social media converge. Journalism Practice, 5, 383-398. doi:10.1080/17512786.2011.557560

Bucher

Helmond

(2018). The affordances of social media platforms. In Burgess

Poell

Marwick

(Eds.), The SAGE Handbook of Social Media (pp. 233-255), London: SAGE.

Cho

Kwon

K. H.

(2015). The impacts of identity verification and disclosure of social cues on flaming in online user comments. Computers in Human Behavior, 51, 363-372. doi:10.1016/j.chb.2015.04.046

10.

Coe

Kenski

Rains

S. A.

(2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication, 64, 658-679. doi:10.1111/jcom.12104

11.

Domingo

Quandt

Heinonen

Paulussen

Singer

J. B.

Vujnovic

(2008). Participatory journalism practices in the media and beyond: An international comparative study of initiatives in online newspapers. Journalism Practice, 2, 326-342. doi:10.1080/17512780802281065

12.

Duffy

B. E.

Pruchniewska

Scolere

(2017). Platform-specific self-branding: Imagined affordances of the social media ecology. Paper presented at the 8th International Conference on Social Media & Society, New York, NY. doi: 10.1145/3097286.3097291

13.

Finley

(2015, October 8). A brief history of the end of the comments. Wired. Retrieved from https://www.wired.com/2015/10/brief-history-of-the-demise-of-the-comments-timeline/

14.

Gans

H. J.

(1957). The creator-audience relationship in the mass media: An analysis of movie making. In Rosenberg

White

D. M.

(Eds.), Mass culture: The popular arts in America (pp. 314-324). Glencoe, IL: The Free Press.

15.

Goffman

(1959). The presentation of self in everyday life. Garden City, NJ: Doubleday.

16.

Hille

Bakker

(2014). Engaging the social news user: Comments on news sites and Facebook. Journalism Practice, 8, 563-572. doi:10.1080/17512786.2014.899758

17.

Kim

(2016). To whom are they speaking? The imagined audience of online news comments. Paper presented at the Association for Education in Journalism and Mass Communication, Minneapolis, MN.

18.

Lewis

S. C.

(2012). The tension between professional control and open participation: Journalism and its boundaries. Information, Communication & Society, 15, 836-866. doi:10.1080/1369118X.2012.674150

19.

Lin

H.-F.

(2008). Determinants of successful virtual communities: Contributions from system characteristics and social factors. Information & Management, 45, 522-527. doi:10.1016/j.im.2008.08.002

20.

Litt

(2012). Knock, knock. Who’s there? The imagined audience. Journal of Broadcasting & Electronic Media, 56, 330-345. doi:10.1080/08838151.2012.705195

21.

Litt

Hargittai

(2016). The imagined audience on social network sites. Social Media + Society, 2(1), 1-12. doi:. doi:10.1177/2056305116633482

22.

Loke

(2012). Old turf, new neighbors: Journalists’ perspectives on their new shared space. Journalism Practice, 6, 233-249. doi:10.1080/17512786.2011.616649

23.

Lowrey

(2009). Institutional roadblocks: Assessing journalism’s response to changing audiences. In Papacharissi

(Ed.), Journalism and citizenship: New agendas (pp. 44-67). Mahwah, NJ: Lawrence Erlbaum.

24.

Marwick

A. E.

boyd

(2010). I tweet honestly, I tweet passionately: Twitter users, context collapse and the imagined audience. New Media & Society, 13, 114-133. doi:10.1177/1461444810365313

25.

Matthew

(2008). A missing link? The imagined audience, news practices and the production of children’s news. Journalism Practice, 2, 264-279. doi:10.1080/17512780801999402

26.

Meyer

H. K.

Carey

M. C.

(2014). In moderation: Examining how journalists’ attitudes toward online comments affect participation and the creation of community. Journalism Practice, 8, 213-228. doi:10.1080/17512786.2013.859838

27.

Montgomery

(2016, August 17). Beyond comments: Finding better ways to connect with you. NPR. Retrieved from http://www.npr.org/sections/npr-extra/2016/08/17/490208179/beyond-comments-finding-better-ways-to-connect-with-you

28.

Nagy

Neff

(2015). Imagined affordance: Reconstructing a keyword for communication theory. Social Media + Society, 1(2), 1-9. doi:10.1177/2056305115603385

29.

Nielsen

C. E.

(2014). Coproduction or cohabitation: Are anonymous online comments on newspaper websites shaping news content? New Media & Society, 16, 470-487. doi:10.1177/1461444813487958

30.

Ong

W. J.

(1975). The writer’s audience is always a fiction. Publications of the Modern Language Association of America, 90, 9-21.

31.

Prochazka

Weber

Schweiger

(2016). Effects of civility and reasoning in user comments on perceived journalistic quality. Journalism Studies, 19, 62-78. doi:10.1080/1461670X.2016.1161497

32.

Robinson

(2010). Traditionalists vs. convergers: Textual privilege, boundary work, and the journalist-audience relationship in the commenting policies of online news sites. Convergence, 16, 125-143. doi:10.1177/1354856509347719

33.

Robinson

(2015). Redrawing borders from within: Commenting on news stories as boundary work. In Carlson

Lewis

S. C.

(Eds.), Boundaries of journalism: Professionalism, practices and participation (pp. 152-168). New York, NY: Routledge.

34.

Rösner

Krämer

N. C.

(2016). Verbal venting in the social web: Effects of anonymity and group norms on aggressive language use in online comments. Social Media + Society, 2(3), 1-13. doi:10.1177/2056305116664220

35.

Ross

(2014). Were producers and audiences ever separate? Conceptualizing media production as social situation. Televison & New Media, 15, 157-174. doi: 10.1177/1527476412454686.

36.

Rowe

(2015a). Civility 2.0: A comparative analysis of incivility in online political discussion. Information, Communication & Society, 18, 121-138. doi:10.1080/1369118X.2014.940365

37.

Rowe

(2015b). Deliberation 2.0: Comparing the deliberative quality of online news user comments across platforms. Journal of Broadcasting & Electronic Media, 59, 539-555.

38.

Ruiz

Domingo

Micó

J. L.

Díaz-Noci

Meso

Masip

(2011). Public sphere 2.0? The democratic qualities of citizen debates in online newspapers. The International Journal of Press/Politics, 16, 463-487. doi:10.1177/1940161211415849

39.

Santana

A. D.

(2014). Virtuous or vitriolic: The effect of anonymity on civility in online newspaper reader comment boards. Journalism Practice, 8, 18-33. doi:10.1080/17512786.2013.813194

40.

Sonderman

(2011, August 18). News sites using Facebook comments see higher quality discussion, more referrals. Poynter. Retrieved from http://www.poynter.org/2011/news-sites-using-facebook-comments-see-higher-quality-discussion-more-referrals/143192/

41.

Springer

Engelmann

Pfaffinger

(2015). User comments: Motives and inhibitors to write and read. Information, Communication & Society, 18, 798-815. doi:10.1080/1369118X.2014.997268

42.

Stroud

N. J.

Scacco

J. M.

Muddiman

Curry

A. L.

(2014). Changing deliberative norms on news organizations’ Facebook sites. Journal of Computer-Mediated Communication, 20, 188-203. doi:10.1111/jcc4.12104

43.

Stroud

N. J.

Van Duyn

Peacock

(2016). News commenters and news comment readers. Engaging News Project. Retrieved from https://mediaengagement.org/research/survey-of-commenters-and-comment-readers/

44.

Thompson

J. B.

(1995). The media and modernity: A social theory of the media. Cambridge, UK: Polity Press.

45.

Vitak

Blasiola

Patil

Litt

(2015). Balancing audience and privacy tensions on social network sites. International Journal of Communication, 9, 1485-1504.

46.

Wahl-Jorgensen

(2002). Understanding the conditions for public discourse: Four rules for selecting letters to the editor. Journalism Studies, 3, 69-81. doi:10.1080/14616700120107347

47.

Wallsten

Tarsi

(2014, August 19). It’s time to end anonymous comments sections. The Washington Post. Retrieved from https://www.washingtonpost.com/news/monkey-cage/wp/2014/08/19/its-time-to-end-anonymous-comments-sections/?utm_term=.eceb85fde395

48.

Weber

(2014). Discussions in the comments section: Factors influencing participation and interactivity in online newspapers’ reader comments. New Media & Society, 16, 941-957. doi:10.1177/1461444813495165

49.

T.-Y.

Atkin

(2017). Online news discussions: Exploring the role of user personality and motivations for posting comments on news. Journalism & Mass Communication Quarterly, 94, 61-80. doi:10.1177/1077699016655754

50.

Zamith

Lewis

S. C.

(2014). From public spaces to public sphere. Digital Journalism, 2, 558-574. doi:10.1080/21670811.2014.882066

51.

Ziegele

Breiner

Quiring

(2014). What creates interactivity in online news discussions? An exploratory analysis of discussion factors in user comments on news items. Journal of Communication, 64, 1111-1138. doi:10.1111/jcom.12123

52.

Ziegele

Jost

P. B.

(2016). Not funny? The effects of factual versus sarcastic journalistic responses to uncivil user comments. Communication Research. Advance online publication. doi:10.1177/0093650216671854