Framings of the Audience by Social Media Users (3-3)

This article examines the understandings and meanings of personal information sharing online using a predominantly symbolic interactionist analytic perspective and focusing on writers’ conceptions of their relationships with their audiences.

Study Design and Method

To operationalize this research, I used Google to find recently updated blogger-hosted blogs or LiveJournals from around London, UK, that offered personal information accessible to any Internet user. An online questionnaire was e-mailed to the authors of 237 of these blogs, both to solicit their permission to be interviewed and to request demographic information to aid in the selection of interviewees. From the 150 who answered, I chose a purposive sample (Chadwick, Bahr, & Albrecht, 1984) of 23 bloggers in an attempt to maximize relevant variety in interviewees. The number of bloggers interviewed was determined by the sense during the fieldwork that a point of saturation had been reached and by Bauer and Gaskell’s assertion (Bauer & Gaskell, 2000) that the upper limit for sensitive analysis by a single researcher is somewhere between 15 and 25 interviews.
Whereas the snowball samples other researchers have typically used to interview bloggers generally favor highly educated, middle-class users, surveys of blogging tend to suggest a more diverse blogging population (Lenhart & Fox, 2006). The sample constructed for this study therefore used the demographic information gathered to ensure a range of ages (between 16 and 64), educational backgrounds, self-reported social classes, occupations, and perceived audience sizes among interviewees.3 Semi-structured interviews lasting from 1 to 1.5 hours were conducted face-to-face. All were transcribed, and the resulting texts were analyzed thematically (Flick, 2006) through an interactive process of open
3 Brake (2007) provides more details on the methods used to construct a sample for this research, and an appendix of demographic profiles of interviewees can be downloaded from
http://davidbrake.org/intervieweesummarybloggers.xls International Journal of Communication 6 (2012) Who Do They Think They’re Talking To? 1061

and selective coding. Consistent with symbolic interactionist approaches to analysis, the focus was on understanding the different ways in which blog authors understood what types of practices they were engaged in—in particular with respect to their relationship with readers—and attempting to create a mutually exclusive and exhaustive typology (Lofland & Lofland, 1995).
The interviews followed three linked lines of questioning. Interviewees were asked what they knew about their blogs’ audiences, what evidence they drew on when making such an evaluation—either technical measurements such as Weblog analysis4 software, or online or off-line interactions—and what role their audience played in their Weblog practice.
4 Weblog analysis tools like Google Analytics (http://analytics.google.com) (as distinct from “weblogging” or “blogging” services and tools) allow Web publishers to track and measure (to some extent at least) how many people have visited their pages and where they come from.
5 A U.S. survey found that nearly half of bloggers surveyed had no idea how many people read their sites (Lenhart & Fox, 2006). When I surveyed personal bloggers to construct an interview sample (N = 150), only a third of respondents reported using traffic analysis tools, and a third of these said they checked their traffic monthly or less often.
6 Like all names given here, these are pseudonyms to protect the identities of the interviewees.
Findings
Knowledge of the Audience
As noted earlier, when writers share texts on the Web, uncertainty about who might read them is inherent. A strategic communication perspective like Goffman’s (1959) suggests that when the shared writing is personal—that is, imparting information normally shared only in a backstage setting—bloggers will attempt to reduce this uncertainty by monitoring their audiences as best they can, for example by using the available third-party Weblog analysis tools. My research, however, suggests their relationships with their audiences are more complex.
Though Weblog analysis tools are available free of charge through a number of websites, only limited evidence indicates that bloggers use such tools to track the number of visitors who read their pages. It moreover appears that few of those who do use these tools check their results regularly.5 It is possible that some of those without tracking tools would want them but are unaware that such tools exist or do not know how to make them work. In the case of my interviewees, however, most were asked if there were things they would like to be able to do with their blogs that they had not done, but only two (Betty, an artist in her mid 50s; and Renia, a 17-year-old student and writer6) evinced a desire to be able to better identify or track their readers.
Of the six interviewees who said they had used tracking tools, three were unenthusiastic about using them. Charles (early 20s, charity administrator) said he had stopped looking at his: “I did actually sort of have a hit counter because I was intrigued to see whether anyone was reading and I was you know horrified to find that pretty much nobody was.” Similarly, Nancy (early 20s, student) said: “I think I used to have one but . . . I don’t know . . . it’s not that interesting. It’s just numbers.” By contrast, Harriet (late 1062 David Russell Brake International Journal of Communication 6(2012)

20s, ad manager) said she was “always checking my stats” but was doing it less often than before because “I don’t quite have the same time to doss at work,” and Annie (early 30s, artist) was also a regular user of such tools.
Orientations Toward the Audience
As noted above, the interviewed bloggers often appeared to envision their readerships as they would like them to be, rather than attempting to discern exactly who they might be or what attitudes they might have. Meanwhile, no single kind of envisioned or desired audience relationship was common to the interviewees. I characterize their orientations toward the audience using two axes. The “anticipated direction of interaction flow” draws on analyses of mediated communication by Thompson (1995) and Ball-Rokeach and Reardon (1988), while the division of the intended audience into specific and indefinite readers is rooted in the CMC literature. Five different orientations of bloggers toward their audiences emerged from the analysis of interviews—narrowcast, broadcast, dialogic, telelogic, and self-directed (see Figure 1). While the first four imply interaction with the readers, as previous studies and symbolic interactionist approach would anticipate, the last, self-directed category emerged from the fieldwork and describes blogging practices that do not appear primarily to have an audience in mind.
Intended audience Anticipated direction of interaction flow One-way
Two-way Intrapersonal Friends (specific) Narrowcast Dialogic Self-directed (quasi-therapy, quasi-sociality, and blogging as an end in itself) Strangers (indefinite) Broadcast Telelogic