Using the News: An Examination of the Value and Use of News Sources in CMC
Address: Steve Jones, Faculty of Communication, University of Tulsa, 600 S. College Ave, Tulsa OK 74104, USA.
This study examines one facet of the penetration of personal computers into everyday life. It seeks to discover how members of a Usenet newsgroup value and use news sources. Electronic news sources predominated. An important finding is that media use was not tied to the user's local geographic. The study raises several questions for future research: What are the rhetorical dimensions of media use in electronic communities? How might our understanding of readers and communities be affected by new patterns of media use in electronic communities?
The dichotomy between mass society and community permeates the discourse concerning mass media and mass culture. A significant thread in that discourse concerns the transformation of mass society into tightly-knit communities by way of electronic communication. Carey (1989) finds this to be “an increasingly prevalent and popular brand of the futurist ethos, one that identifies…computers and information with a new birth of community”. An important element in the futurist ethos is the penetration of personal computers and their concomitant communication technologies into everyday life. And it is equally as possible that computer-mediated communication will separate participants and content by way of compartmentalizing messages and users, as well as disinhibiting users, thus leading to heightened interpersonal (computer-mediated) conflict, thereby playing a role in the death of community.
This study examines one facet of the penetration of personal computers into everyday life. It seeks to discover how members of a Usenet newsgroup value and use news sources. At a time when the news media are making efforts to have an online presence, and communication service providers are working to bring online services to consumers, an important area of study is the nature of news in the context of computer-mediated communication (CMC).
This study of media use in soc.culture.yugoslavia brings together ideas from two separate research strands: the study of media use, and the study of new communication forms. The most developed tradition of media use research relates media use to community ties. The Minnesota community research program (Tichenor, Donohue and Olien, 1980 and 1970; Donohue, Tichenor and Olien, 1975; Olien Donohue and Tichenor, 1978) concluded that “strong community ties are major forces leading to reading the local newspaper” (Tichenor et al., 1980: 57), though subsequent studies showed fewer direct ties (Stamm, 1985). The Minnesota research tells us about how communities use the media. They offer one account of how media are linked to communities. It also investigated how mass media are used as means of social control by powerful elites in the community. In the research cited definitions of community play a lesser role than the news media, and in some instances (most notably Tichenor et al., 1980) community goes undefined. A new type of community, an electronic, or virtual community has been forecast, and one can see some form of it in the variety of computer networks (Internet, Usenet) and electronic on-line services (CompuServe, America Online) now available (Jones, 1995). In regard to this study, definition of community as “a variable of social relations…a complex of ideas and sentiments” (Calhoun, 1980: 107) is most appropriate.
Another thread in media use research is related to the choices people make when faced with an array of media to use for their own communication (Fulk, Steinfield, Schmitz and Power, 1987). As Baym (1995: 162) notes in the context of another newsgroup,
the ways in which people have appropriated the commercial and noncommercial networks demonstrate that CMC not only lends itself to social uses but is, in fact, a site for an unusual amount of social creativity.
(See also Rice and Love (1987) for discussion of creativity.)
New communication forms problematize the concept of “media use.” In previous research, such as Tichenor's, “media use” is largely considered “reading,” and that activity is tied mostly to newspapers. CMC users have appropriated news sources for their own purposes. Schweitzer (1991) includes in media use other media consumption activities, such as television viewing. Delener and Neelankavil (1990) consider media consumption behaviors solely on the basis of time spent attending to a particular medium.
Indeed, there is something profoundly alienating about traditional definitions of media use. Though they may provide opportunities for measurement of audience activity related to the media, they paint, overall, a generally passive picture of the consumer, leaving out the active processes of meaning-making in which people engage. It is necessary to consider media use a particularly active process in electronic contexts, in part because CMC provides hypertext-like discourse and also because traditional media (newspapers, television, etc.) are fodder for conversation and discourse via CMC (one glimpse at Usenet provides ample evidence). As Rogers (1986) claims, computer-mediated communication is distinctively interactive when compared to traditional media. Thus, to put it another way, media content becomes part of CMC content, and “readership” is not nearly as interesting an issue as is authorship based, in turn, on readership.
In essence, media users in an electronic community choose to redistribute (or “re-author’) news stories. A study of personal computer owners and media use showed that “adopters of new information technologies do not necessarily give up more traditional communication media in order to adopt the new communication forms” (Schweitzer, 1991: 689). Since a wide variety of electronic news sources are available to computer users, including the AP, UPI, Reuters news wire, as well as USA Today‘s Decisionline, will members of an electronic community predominantly use electronic news sources? It is important to discover what news sources matter to network users. What are the implications of changes in users' valuation of news and its redistribution?
Jeffres, Dobos and Lee (1988) made some effort at answering a similar question, as related to television and, again, only as regards the consumption of a media text. If media use is redefined to include the consumption and production of texts, and those activities occur on a worldwide scale thanks to the Internet and Usenet, how does this use of news affect our ideas about media use and community ties?
The group studied is soc.culture.yugoslavia on Usenet. The reasons for the selection of soc.culture.yugoslavia are several. First, it is a group mired in conflict, mirroring the wars in the former country of Yugoslavia. The electronic messages that were captured from the group for this study were collected daily between June 1991 and July 1992, during a time of escalation in the ethnic wars there. Group members would therefore be likely to desire more information about the status of the war, especially since until June, 1992 when the war commanded international attention, coverage was relatively sparse.
Since reporting of that conflict had been limited, information about the situation in the former Yugoslavia was often passed from person to person, or from one person to many, electronically. Its passage, by way of citation and comment, reflects the attitudes toward media, and related information sources among community members.
For instance, postings of news stories can precipitate antagonistic discussions of both individual bias and media bias. One typical exchange was initiated after a letter to the TorontoGlobe and Mail was published in early 1993. A portion of that letter was posted to the newsgroup:
The blanket condemnation of the Serbian nation and the misrepresentation of the Serbian position would not be so deeply trenched in the media if people like my friend were not so willing to endorse any position or stereotype, as long as it ostensibly leads to peace. An unreflective and sensational press–the electronic media in particular–has provided a well- meaning public with simplistic solutions to a complex situation. The person who posted this message claimed the letter “explains rather well Serbian position” [sic].
However, the ensuing exchange in message threads related to that posting focused on media representations of the sides in the war, typified by comments like this one:
I have seen plenty of new stories about Serbs civilians being victimized. But please, Bosnian Serbs are not the main victims, unless CNN, ABC, NBC have it all wrong.
What is particularly interesting about these postings is that they point to an active readership making clear its constructions. In that regard, soc.culture.yugoslavia is a site at which to view how audiences interpret media.
A total of 6,192 electronic messages were collected from the soc.culture.yugoslavia group on Usenet during the period June 1991 to July 1992. Although some messages were duplicated, first sent and perhaps re-sent to confirm delivery, only original, and not duplicate, messages were analyzed. Some messages were sent from other groups by way of “cross-posting,” a form of sending one message to multiple groups. Cross-posted messages were included in the analysis as they were considered part of the discourse on soc.culture.yugoslavia. While there is the possibility that some messages were misrouted by the numerous computers in the Usenet chain and are missing from this analysis, the chances of this are very slim. It is likely that few messages suffered such a fate, and since all messages were analyzed and no form of sampling was involved, any such loss should be acceptable. Duplicate messages created by group members unsure of how to use Usenet software were discarded.
A content analysis of soc.culture.yugoslavia messages was undertaken following methods established by Krippendorf (1980), Holsti (1969), and Rosengren (1981). The procedure followed was to capture messages in ASCII text form on an IBM-PC. Each message was read to determine whether or not it contained reference to a news source. In all, 6,087 messages were analyzed; 1,120 messages mentioned a news source. These 1,120 messages were then coded using the following categories: news source name; news source type (print, broadcast, electronic); news source location; message sender location (determined from the sender's electronic mail address). Each message that contained mention of a news source, information directly attributed to a news source (paraphrased or quoted), or text re-posted directly from a news source was coded. A news source was defined as any institutionalized, mass-mediated source of information. The news source was coded once for each message in which it was mentioned. If a news source was mentioned multiple times in a message it was only coded once. If a message posted in the newsgroup quoted a previous poster's message referencing a news source, the quoting in the second message was not considered a new instance of a news source mention. The content analysis was performed by the author.
The 1,120 messages analyzed were posted by 131 individuals. A check of all messages analyzed revealed that 211 individuals contributed messages to the newsgroup. Since it is possible to hold more than one electronic mail address, it is also possible that these numbers are inflated. It should also be noted at the outset that members of this particular group are part of a larger social group of business people, academicians, independent scholars, college and university students, and computer professionals who use computer network services. Nevertheless, since what is of interest is media use situated in electronic discourse, the number and type of individuals should not be a primary concern, though study of group members may be of interest for future research. Since this study uses content manifest in posted messages, it is important to remember that it reflects media use among those who post messages, and not among those who read the newsgroup without posting their own messages. It is also important to note that the posters set the discourse agenda for the (presumably) larger audience of “lurkers” (that is, those who read the newsgroup but do not themselves post messages).
An analysis of news sources mentioned in messages on soc.culture.yugoslavia showed that Vreme, an electronic digest of a Belgrade, Serbia-based periodical, led all sources with 141 messages in which it was mentioned, representing 12.59 percent of total news sources mentioned (Figure 1). Vreme is available to those with Internet access, and copies of some of the stories from it were regularly posted on the newsgroup.
An aggregated analysis of news sources illustrated in Figure 2 showed that electronic news sources predominated among those mentioned on soc.culture.yugoslavia (46%), followed by print news sources (36%) and, lastly, broadcast news sources (18%).
In light of the electronic nature of the newsgroup this spread is not surprising. It does point, however, to a heavy, or at least heavier, use of electronic news sources like news wires available on the Internet or other BBSs. This finding supplements Schweitzer's (1991: 689) assertion that personal computer use “did not lead to radical changes in use of traditional news media” because it shows that, while users did not abandon traditional media, they did adopt new media as news sources. Although Schweitzer based his claim on analysis of readership surveys and not on analysis of news sources mentioned in, say, conversation, rendering comparisons problematic, these findings do point to a change in use of traditional news media.
Further analysis broke down the location, or geographic origin, of news sources mentioned. A comparison with message sender location reveals that most news sources mentioned originated in US “media centers” (New York and Washington). Most (just over two-thirds) of the individuals posting messages mentioning news sources are based in the US as well. However, upon closer examination it is clear that there is little similar connection between message sender location and news source location, and a further breakdown of message senders by state reveals that news sources are not linked to geographic location of the message sender. The analysis of messages with news sources mentioned demonstrated no significant correlation (r = .021) between message sender location and news source location.
A questionnaire was posted to the newsgroup in November, 1993, initially in an effort to gain demographic information about its users once data had been collected for content analysis. 128 responses were received. It is difficult to discern a “response rate” in this instance, insofar as the questionnaire was publicly posted. Information that has since been compiled about many newsgroups is not available for the time during which the questionnaire was posted on soc.culture.yugoslavia. Moreover, though the questionnaire was posted well after the sample of messages was taken for the previous section of this study, it is believed that the high proportion of nonstudents among posters means that user turnover was infrequent. Given the general availability of the Internet, and hence Usenet, to primarily academic communities at the time of the survey, results should be approached cautiously.
From the questionnaire responses a random sample of 14 respondents was chosen for on-line interviews. It was at this point that attitudes toward news sources and their usefulness became most clear. Most of the respondents interviewed stated that the Internet had become a major news source in its own right. One respondent, for instance, said, “I really appreciate the electronic mail sources for news,” in reference to being able to reach USA Today, and that “the Usenet newsgroups are a second source of news.” Another respondent stated that he would not know very much about events in the former Yugoslavia were it not for Usenet. Still another stated that Usenet is useful because it “digestifies” news reports, often including ones unavailable locally (via newsstands or subscription) to many people. This can account, too, for the high number of messages with reference to Vreme and other electronic news sources.
Making clear their critical reading of the news and newsgroup, all 14 respondents believed that the media were biased in their coverage of the war. They also believed that those posting messages to the newsgroup that contained stories from traditional news sources were doing so less to inform the newsgroup but rather to use the traditional stories to convince others that their arguments were correct.
Perhaps indicative of the predominance of PhDs among the newsgroup's posters, some respondents were critical of objectivity itself, as expressed by one user who identified himself as a poster: “If a source appears objective, that only means that it best approximates the reader's opinions or else the average viewpoint of all the other sources the reader sees.” For this user the mass media were reinforcers of already-held beliefs.
Another respondent said that the newsgroup “helps to cross-check information from different sources.” One respondent's comment typified those of the others who believed that by having access to wire service reports they thus had access to a more objective news source than if they had to rely on those reports as edited and altered by local media.
It is possible that the predominance of electronic news sources stems simply from the relative ease with which their text can be captured by a computer and redistributed to other users and communities, as opposed to the (relatively) laborious process of transcribing a broadcast news story or a newspaper story. But the nature of data used in this study, made manifest in discourse, may nevertheless provide a more profound and penetrating measure of media use than the type of self-reports generated from questionnaires and surveys in traditional media use studies. Moreover, the mention of particular news sources in the newsgroup points to their value (whether positive or negative) and tells us about which news sources were worthy of mention.
In the case of the Usenet newsgroup soc.culture.yugoslavia there is both a local aspect to media use as it has been defined in earlier studies (Tichenor et al., 1980) and evidence of desire for nonlocal media, particularly insofar as it appeared that users believed that wire services were more objective than local media. There is thus a tension between desire for knowledge of local “angles” other than ones available to newsgroup users in their own local (geographic) region and a desire for more objective news stories.
An important finding is that media use was not tied to the user's local geographic community, a finding opposite to expectations one might have from previous studies like Tichenor et al. (1980), Stamm (1985), and Schweitzer (1991) that did not examine media use in CMC groups. And, in what may be the most important finding, we have evidence of cosmopolitan media “elites” (Rogers, 1962). Tichenor et al. (1980) also find that media “elites” are implicated in information flow in communities, though they defined those “elites” as community leaders and news professionals. In the case of soc.culture.yugoslavia it may be argued that posters are community leaders and lurkers community members, but the ability to join the information flow is unrestricted, and thus lurkers may at any time “cross over” to become posters, a highly unlikely scenario in traditional media use studies.
As a Times-Mirror survey showed, “computer-users tend to read more newspapers, books and magazines than others” (Davis, 1994: 24) and indeed there seemed to be quite an appetite among newsgroup users for news and information. It is absolutely imperative to note that these same users are not simply “consuming” news but are engaging in its critical analysis as well as passing it along as part of their own postings, often in service of their own opinions. Consequently, they fit the description of an “interpretive community” actively constructing meaning (Lindlof, 1989). Interestingly, during that process, few breaches of Usenet standards of conduct (McLaughlin et al., 1995) are found, even when particularly vehement criticism of a posting occurred. Indeed, “flame wars” (MacKinnon, 1995: 129-130) were largely absent. Tichenor et al. (1980) also investigated the use of mass media by powerful elites, and in particular noted that:
A large portion of the information available…depends on an information delivery system which reflects the pluralistic organization and vested interests of the society in which it exists. Information appears to be generated and disseminated as a result of joint activity of professionals within the mass media channels and professionals who have advocacy functions for interdependent special interest groups (p. 15).
One of the ways to interpret newsgroup user interest in varied forms of information is to “break through” this form of “joint activity” and engage in media activity of their own. Interestingly, this action leads to the type of “social control” Tichenor et al. identify, as each user utilizes news and news sources to express and augment their opinions. However, given the nature of CMC, and the “personae” (MacKinnon, 1995) created on Usenet, the exchanges in the newsgroup critical of news sources, messages, and so forth, provide a kind of dialogue for which traditional social control paradigms, which focus on the “one-way” quality of news media, do not account.
It is possible that the news sources represented in this newsgroup are simply ones of sufficient international stature to be of interest to the group. However, how would one then explain the absence of newspapers like the Washington Post and L.A. Times, or the absence of the three U.S. broadcast networks? One would indeed expect major international newspapers to have greater representation, and yet they do not. It is apparent that members of the newsgroup made greater use of electronic news sources like Vreme and directly accessible wire services over other news sources. Newsgroup users posted little if any news from local sources such as newspapers in their hometowns.
These findings provide a basis for the assertion that local media are unimportant to readers except for reporting of local news. The heavy use of electronic news sources like the wire services may mean that members of this community have an understanding of how news “works,” that is, they use their computer access to bypass local sources to get straight to “the” source, say, the Associated Press.
Second, and more importantly, community members use news sources to provide a foundation for their own arguments and opinions expressed within the discourse in the news group. Consequently, the most useful sources are the ones that seem most unbiased, such as the wire services themselves (since those sources most often feature stories filed directly by reporters in the field). Jensen (1991) argues that American social thought is “concerned about ‘modern’ public opinion being formed by the mass media, [deemed to be] an untrustworthy, self-serving source” (p. 77). In this instance, it is individuals who may be using the media in a self-serving fashion to support and elevate their own opinions.
One might as well have expected to find news sources from areas where there is a substantial population of Yugoslav immigrants, as there is in Germany, or in U.S. cities like Cleveland, Pittsburgh, and Chicago. Nothing was found in this study to support such an expectation. An immigrant press, or at least one sensitive to immigrants in its area, seems unimportant to this electronic community, and yet the immigrant press is claimed to have “established more personal, helpful ties with (its) communities than did the majority press” (Reed, 1990: 77). What happened to such ties in this newsgroup?
This study raises several questions for future research: What are the rhetorical dimensions of media use in electronic communities? How might our understanding of readers and communities be affected by new patterns of media use in electronic communities? Future research should also employ readership surveys in an attempt to understand readership patterns and to better collect demographic information.
What is particularly interesting about these findings is that they run counter to claims that the media of communication eviscerate history. Clifford Christians (1976) notes that Jacques Ellul wrote:
…the unrelenting flow of news inebriates human memory, a loss Ellul laments….”There is no politics where there is no grasp of the past, where there is no continuity, where there is no analysis of errors or capacity to understand the present through that analysis and in that continuity.” Man aids in that evaporation and consequent weakening of his political order by driving events into oblivion, that is, actively forgetting for the sake of maintaining sanity (p. 13).
Those posting to soc.culture.yugoslavia are tinged with a deep sense of continuity with past Usenet exchanges, and postings often included quotes and replies several messages back in a thread. Though one can filter messages on particular topics or from particular senders using “kill files” (MacKinnon, 1995), message threads can continue for weeks, months, even years.
One of the consequences of the externalization of memory (Ong, 1982) in such a public (and print-based) fashion as this is that one's words do not evaporate. As Aycock and Buchignani (1995: 223-224) note:
Interpersonal discourses confer upon the individuals who engage in them an immediate sense of authorship and authority…. Print-based discourses delimit texts that are forever fixed in place and identified with a specified author or authors…. Computer-based discourses partakes of some of the characteristics of interpersonal and print-based discourses…
Like Aycock and Buchignani, who found some evidence of authority being questioned, critical readership is evident among those who posted to soc.culture.yugoslavia. However, such critical readership is tempered by authorship that is often uncritical, and it is that tension between readership and authorship in soc.culture.yugoslavia that fuels both the need for information and, ultimately, discourse within the newsgroup.
The author is greatly indebted to Professor Joli Jensen and Professor Joseph Schmitz in the Faculty of Communication at the University of Tulsa for their valuable suggestions and insights.