Applying conversation analysis methods to online talk: A literature review

Discourse, Context and Media 12 (2016) 1–10

Contents lists available at ScienceDirect

Discourse, Context and Media journal homepage: www.elsevier.com/locate/dcm

Applying conversation analysis methods to online talk: A literature review Trena Paulus a,n, Amber Warren b, Jessica Nina Lester c a

University of Georgia, Qualitative Research Program, 310 River's Crossing, 850 College Station Road, Athens 30606, GA, USA Indiana University School of Education, Literacy, Culture & Language Education, 201 N. Rose Ave #3044, Bloomington 47405-1006, IN, USA c Indiana University School of Education, Counseling and Educational Psychology, Indiana University, 201 North Rose Avenue, 4060, Bloomington, IN 47405-1006, USA b

art ic l e i nf o

a b s t r a c t

Article history: Received 18 November 2015 Received in revised form 23 March 2016 Accepted 2 April 2016 Available online 8 April 2016

While researchers have used conversation analysis (CA) methods to understand online talk since the 1990s, to date there has been no systematic review of these studies to better understand this methodological development. This article presents a comprehensive literature review of 89 peer-reviewed journal articles reporting findings of empirical studies using CA to understand social interaction online. In this review, we describe who is conducting this type of research, the contexts in which CA has been used to make sense of text-based online talk, and where such studies are being published. We also identify the “fundamental” conversational structures researchers are drawing upon in making sense of online talk as social interaction. Findings show that studies are using CA to understand “mundane” conversational contexts, as well as institutional talk from educational, counseling and workplace settings. The number of such studies are increasing and are being conducted by an international network of researchers across a variety of disciplines. The data is most often described as synchronous or asynchronous, with a slow increase in attention to social media data. Publication outlets are mostly languagebased and/methodological journals. Analysis revealed four main aims: (1) comparing online and face-toface talk, (2) understanding how coherence is maintained, (3) understanding how participants deal with trouble, and (4) understanding how social actions are accomplished asynchronously. This review contributes to the overall understanding of the methodological development of CA, offering useful insights for those interested in using it to understand social interaction as it occurs online. & 2016 Elsevier Ltd. All rights reserved.

Keywords: Conversation analysis Computer mediated communication Ethnomethodology Language-based methods Literature review Online talk

1. Introduction In a recent literature review on notions of embodiment, Nevile (2015) noted that, “as analysts of social interaction, we are interested in how people, together in real time, make sense to do whatever it is they are doing, with whatever resources are available, including talk, body, objects, and the surrounding environment” (p. 141). For the past 25 years, these resources have included the Internet, computing devices and the various kinds of social interaction they support. In this paper, we report on a review of journal articles exploring how conversation analysis (CA) has been used to understand social interactions that occur in the form of text-based online talk. Known by a variety of terms, including computer-mediated communication (Herring, 1996), computer-mediated discourse n

Corresponding author. E-mail addresses: [email protected] (T. Paulus), [email protected] (A. Warren), [email protected] (J.N. Lester). http://dx.doi.org/10.1016/j.dcm.2016.04.001 2211-6958/& 2016 Elsevier Ltd. All rights reserved.

(Herring, 2004), and electronic discourse (Meredith and Potter, 2013), online talk takes place through a range of modalities. These modalities have evolved over time from Usenet groups and Internet Relay Chat to discussion forums and instant messenger to Facebook, Twitter, and YouTube. As new ways of interacting online have proliferated, so have the research approaches being used to understand them. These include a variety of language-based methodologies, such as linguistics (Crystal, 2006; Georgakopoulou, 2011; Zappavigna, 2012), sociolinguistics (Androutsopoulos, 2006; Thurlow and Mroczek, 2011), pragmatics (Herring et al., 2013; Yus, 2011), and discourse analysis (Herring, 2004; Myers, 2010). Studies of online talk from a language perspective have included a focus on play and performance (Georgakopoulou, 2011), communities (Seargeant and Tagg, 2014; Stommel, 2008), selfpresentation and identity work (Androutsopoulous, 2006; Seargeant and Tagg, 2014), stories (Page, 2012), and gender (Herring and Stoerger, 2014), to name a few. Meredith and Potter (2013) have argued that “electronic discourse should be seen as electronic interaction” (p. 374) and, as such, requires a method such as CA to understand it. While CA has

2

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

roots in sociology, generally, and in ethnomethodology more specifically, fields such as education (Gibson, 2009a, 2009b), educational technology (Mazur, 2004), journalism (Steensen, 2014), and counseling (Stommel and Lamerichs, 2014) have not only used CA to understand online talk, but have also published methodological pieces explicating its use. Scholars such as Hutchby (2001), ten Have (1999, 2000), Reed and Ashmore (2000), and Reed (2001) were early proponents of using CA to investigate online interaction; however, others such as Schegloff (2006), argued that “computer chats” should not be considered “talk” at all (p. 90). In this paper, we use the phrase ‘online talk’ to be consistent with how the researchers themselves characterize this sort of communication (e.g. Meredith and Potter, 2013). Regardless of Schegloff's claim, researchers have been drawing upon CA to study online talk since the late 1990s. For example, Baym published the first systematic study of agreement in “computer-mediated discussions” in 1996, drawing upon preference structure (Pomerantz, 1984) and making comparisons with oral and written conversations. Disciplines such as psychology are particularly interested in how to treat online talk as research data (Gernsbacher, 2014; Holtz et al., 2012; Jowett, 2015). Yet, Greiffenhagen and Watson (2005) criticized some CA/ethnomethodological approaches to online talk for failing to consider its use from the participants’ perspectives; that is, as local and situated; instead, much of the research treats online talk as if it were monolithic. Furthermore, they noted that most studies have exclusively relied on “logfiles” – transcripts of the interactions – rather than videotaping the users as they engage in the talk. This they call the missing body (p. 7), that is, ignoring the activities that may be going on at the same time as talk is being exchanged online. They also critiqued the tendency of researchers to treat CA concepts, such as Sacks et al. (1974) description of turn-taking, as a “model” that online talk should be judged against, highlighting that the turn-taking system was never meant to be an “ideal type construction from which real cases are seen as departing” (Greiffenhagen and Watson, 2005, p. 16). Beyond the academic literature, the interest in exploring such methodological issues can be seen within professional conversations focused on research methodologies for understanding online talk. For example, in 2007, the International Pragmatics Conference included a section on “data and methods in computermediated discourse analysis” which was then developed into a special issue for Language@Internet in 2008 (Androutsopoulos and Beißwenger, 2008). Further, in 2010 and 2011, the same journal featured a two-part special issue on “computer-mediated conversation” (Herring, 2010; Herring, 2011), including articles on e-mail, instant messaging, blogs, and 3D virtual worlds. In 2014, the International Conference on Conversation Analysis panel and a forthcoming special journal issue focused on orders of interaction in mediated settings, including video-conferencing, blogging, and instant messaging interactions. Since 2013, the Microanalysis of Online Data (MOOD) international network has held an annual symposium, bringing together scholars from around the world to engage in such discussions (Giles et al., 2015). Even though many online talk researchers have oriented to turn-taking as originally described by Sacks et al. (1974), some scholars have begun to explore how other CA concerns, such as transcription methods (Meredith and Potter, 2013) and repair (Meredith and Stokoe, 2013), might function in these new contexts. Nonetheless, Giles et al. (2015) pointed to the ongoing need for extensive methodological discussions around refining CA approaches for use with online talk. They noted that: the move from an uncritical ‘digitized’ application of CA to a customized version of CA for specific use with online interaction requires the reworking of a number of tenets of CA in the

light of the challenges posed by electronic communication technology. (Giles et al., 2015, p. 47) Subsequently, because so many disciplines have an interest in online talk, it is difficult to identify all of the relevant publications in which such research appears. A more thorough understanding is needed of which disciplines have been using CA and in what ways these applications can offer useful insights to those desiring to take up this methodology. To date, however, no comprehensive literature review of such studies has been conducted. Therefore, we engaged in a systematic review of empirical studies published in peer-reviewed journals to explore how and to what extent CA has been applied to the everyday and institutionalized talk that occurs online. Our research questions were: 1. In which disciplinary journals are CA studies of text-based online talk being published? 2. In which countries are researchers using CA to understand this talk? 3. In which years were they published? 4. Which types of online talk are being analyzed using CA? 5. Which foundational structures of conversation (as described in Sidnell and Stivers 2013) are being used to understand the talk?

2. Methods We initially reviewed over 200 articles, identifying 89 peerreviewed journal articles in which authors self-identified using CA to study text-based online talk, with some of these articles also drawing upon discursive psychology (DP). We chose to focus on text-based online talk because this is still the most common type of online communication and most relevant to our own areas of research in education, health, and psychology (Lester and Paulus, 2011; Paulus and Lester, 2013). We excluded conference proceedings and book chapters from our corpus, because as a field it is important to know how issues and topics are represented in peerreviewed journals, often considered the “gold standard” of publications (Woods et al., in press). Books, book chapters and conference proceedings are important to consider for their historical perspective on the emergence of the field, and thus they served to contextualize our work and are cited in both the introduction to and discussion of our analysis. We also limited our review to studies published in English, as we did not have a budget for translation services. To locate relevant articles, we searched eight scholarly databases and conducted direct searches of 24 journal websites. Keyword searches were conducted using “conversation analysis” in conjunction with: blogs, online, discussion forum, computermediated communication, Twitter, Facebook, social media, chat, YouTube, e-mail, synchronous, Internet, and computer. Article references pointed us to additional relevant studies for review. We imported these articles into ATLAS.ti version 7 for analysis. Our analysis proceeded through several steps. First, we identified the discipline of the journal in which the study was published, the country location of the first author, and publication date. Second, we identified the type of data being analyzed and the context from which it was drawn. Next, we identified the CA features that were drawn upon in the studies. These features were then categorized according to “fundamental structures” as outlined in the Handbook of Conversation Analysis (Sidnell and Stivers, 2013). Finally, we did an intensive reading across articles to identify the interpretations and claims that were being made about online talk as a result of CA. We structured our discussion of the findings around these analytical focal points.

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

3. Findings

3

Table 2 Number of publications by journal.

We present our findings in two sections, beginning with a focus on the broad characteristics of the articles, as we believe this provides important context about where, when, and how CA is being applied to the study of online talk. Second, we discuss the claims and interpretations that were made about online talk, focusing explicitly on the authors' aims for employing CA. 3.1. Broad characteristics of the articles When reviewing the articles, we sought to make sense of the general characteristics of the articles, as previously noted. Specifically, we considered the journal discipline, the country location of the first author, publication date, data type, research context, and CA's fundamental structures. We present the findings from our analyses of each of these characteristics in turn. 3.1.1. Journal discipline We found that the majority of studies appeared in journals focused on language and communication (n ¼54). Table 1 outlines the number of publications organized by journal discipline. While many articles are being published for an audience already interested in the study of language, other disciplines, such as health studies, education, and the social sciences more broadly, are also interested in the use of CA for understanding online talk. Table 2 outlines the titles of journals that published two or more studies. The full list of journals is available upon request. These findings provide further evidence that most studies are being published in venues that are inherently interested in language-based approaches to research. Many studies of online talk from a CA perspective are being published in methodological journals, with some being published in health and psychologyfocused journals. 3.1.2. Country location of first author First authors in the USA and UK accounted for over half of the published papers. A total of 19 countries were represented. See Table 3. While over half of the articles were published in the UK or US, it is promising that scholars publishing their studies using CA to understand text-based online talk in peer-reviewed journals came from 17 additional countries on six continents. 3.1.3. Publication date Table 4 shows the number of peer-reviewed journal publications by date range, with the earliest being published in 1994 (McKinlay et al., 1994). While only a handful of studies appeared in the first decade after 1994, the number has increased steadily since. While elsewhere Table 1 Number of publications by discipline. Discipline

Number

Discourse and language studies Communication and media studies Health Linguistics and language teaching Gender studies Social sciences broadly Education Medical education Qualitative methodology Psychology Organizational studies TOTAL

33 15 12 6 5 4 3 3 3 3 2 89

Journal

Number of publications

Discourse Studies Language@Internet Journal of Computer-Mediated Communication Journal of Pragmatics Qualitative Health Research Research on Language and Social Interaction Communication & Medicine Forum: Qualitative Social Research Text & Talk Appetite Australian Journal of Communication British Journal of Social Psychology CALICO Journal Discourse & Communication Feminism & Psychology Journal of the American Society for Information Science and Technology

7 7 6 6 4 4 3 3 3 2 2 2 2 2 2 2

Table 3 Number of countries with first author. Discipline

Number

USA UK Netherlands Australia Spain Sweden Norway Singapore Canada Columbia Finland France Germany India Italy Scotland South Africa Taiwan Thailand TOTAL

24 23 12 5 5 5 2 2 1 1 1 1 1 1 1 1 1 1 1 89

Table 4 Date published. Publication date

Number

1994–1999 2000–2005 2006–2010 2011–2015 Total

6 9 25 49 89

scholars have noted that conversation analysts’ interests in online talk appeared to have diminished in recent years (Giles et al., 2015), our review found that the number of such studies have been increasing. Among the studies we located, the earliest focused on illustrating how CA could be a potentially fruitful means of approaching the study of online talk or even justifying its use (Baym, 1996; Negretti, 1999; ten Have, 2000). For example, ten Have (2000) looked at how people use categories to find suitable chat partners in MUDs and MOOs. While not relying on extended extracts, he offered a collection of chatroom names, profile descriptions, and nicknames, describing how these worked to categorize participants. Later studies have applied CA in a number

4

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

of ways, which we describe more fully throughout the remainder of this paper. 3.1.4. Data type Table 5 illustrates that nearly half of the studies (n ¼ 41) analyzed synchronous data such as instant messaging. Some studies analyzed types of data, such as observations, surveys, or interviews. Finally, with a few exceptions (e.g. Meredith and Stokoe, 2013), we noted that most researchers still rely on transcripts, rather than screen recordings of the interaction as it unfolds in real time. Although the distinction between synchronous and asynchronous communication has become less useful over time as modalities have converged (Herring, 2013), deciding whether online talk is more similar to speaking or to writing was an early focus of language-based investigations into its use (Herring, 2007).We retained the distinction between synchronous and asynchronous data because many of the authors did so. We created separate categories for e-mail because of its particular interest in workplace studies and for social media as the most rapidly growing type of online talk. Many of the earliest studies explored ‘quasi-synchronous’ communicative settings (Garcia and Jacobs, 1999), such as in instant messaging (IM) (Garcia and Jacobs, 1998, 1999; Rintel and Pittam, 1997), or chatrooms (ten Have, 2000). However, a few early studies also focused on asynchronous textual communication, such as in message boards (Baym, 1996) and email (Harrison, 2003). As forums became more prevalent, they became a significant site of focus (e.g., Flinkfeldt, 2014; Stommel and Meijman, 2011) and with the development of social media, interest in understanding communication in those settings developed as well. Our findings show that CA studies of social media posts began around 2009 with a study of Flickr (Adkins and Nasarczyk, 2009), and have included studies of MySpace (Goodings, 2011), YouTube (Hall et al., 2012a; Pihlaja, 2014), and Facebook (Frobenius and Harper, 2015; Meredith and Stokoe, 2013). 3.1.5. Data Context As illustrated in Table 6, most of the studies analyzed data drawn from what we categorized as leisure or mundane settings (n ¼26) and academic contexts (n ¼25). Data from leisure/mundane contexts included comments on social media posts on YouTube and Facebook (e.g., Pihlaja, 2014), Table 5 Data type.a Data type

Number

“Quasi”-synchronous chat/messaging Asynchronous blog/forum posts and comments Email Social media posts/comments

41 33 10 7

a Some studies analyzed more than one type of data so the total is not included.

Table 6 Data context. Data context

Number

Leisure/mundane University/school/academia Online support groups Counseling Language learning Workplace Total

26 25 20 8 6 4 89

text-based comments during online video games (Moore et al., 2007), comments on news and magazine sites (Hall et al., 2013; Lazaraton, 2014), and conversations in a variety of chat rooms. Data in academic contexts, including those with a language-learning focus, most often included conversations that were required as part of a formal course and included both synchronous and asynchronous interactions on blogs, discussion forums, and chat environments (Bou-Franch et al., 2012; Gibson, 2009a, 2009b; Negretti, 1999). Support group contexts were most often asynchronous discussion forums for those with mental health concerns (Giles and Newbold, 2011), or on topics such as bullying (Osvaldsson, 2011), veganism (Sneijder and te Molder, 2004, 2005, 2009), or conditions such as celiac disease (Veen et al., 2010). Several studies looked at how counseling is done through chat and email, focusing on how the design of the turns impacts the counseling activity (Danby et al., 2009; Stommel, 2012; Stommel and Koole, 2010). The workplace studies included library chat reference studies (Koshik and Okazawa, 2012) and a study of emoticons in workplace email (Skovholt et al., 2014). 3.1.6. CA's “fundamental structures” We initially analyzed the studies in order to identify which CA “features” were used by researchers to understand the social actions accomplished in the online talk. We began by using the Handbook of Conversation Analysis (Sidnell and Stivers, 2013) as a flexible framework, categorizing the features according to the “fundamental structures” outlined in the Handbook, as well as by attending to how the authors labeled the structures they were investigating. In so doing, we decided to include a breakdown of studies which focused specifically on adjacency pairs and/or openings/closings as subsets of sequence organization, since numerous studies were focused specifically on one or both of these aspects of sequence organization. Unsurprisingly, we found that researchers often attended to more than one structure as part of their analysis. The most frequent structures of conversation taken up by analysts are outlined in Table 7. While MCA was not included in the Handbook, we chose to include it because it was quite prevalent in studies of asynchronous data from a discursive psychology (DP) perspective and frequently in conjunction with storytelling. As Levinson (2013) noted, Sacks described both MCA and sequence analysis as major frameworks, but only sequence analysis has really been developed by CA researchers. However, Stokoe (2012) observed that MCA has been “enjoying something of a renaissance within'the discipline’ of CA itself” and may allow scholars “with a primary interest in categorial or ‘topical,’ rather than sequential, issues an empirically tractable method for studying those issues” (p. 279). Furthermore, although CA and MCA may have “developed largely in isolation of one another, in more recent years there has been a move to rethink CA and MCA together” (Plunkett, 2009, p. 24). Indeed, our search found Table 7 Fundamental structures of conversation.a Structure

Number

Sequence organization (including 12 articles focused on openings/closings and 9 focused on adjacency pairs) Turn design Turn allocation/turn-taking/turn-construction units (TCUs) Membership category analysis (MCA) Repair Preference structure Storytelling/second stories

32 30 26 16 10 5 4

a Some studies included more than one structure, so the total is not included.

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

that 13 of the 16 articles using MCA to understand online talk simultaneously claimed to be using CA, or DP informed by CA. In our readings of the articles, we found that some scholars creatively utilized CA alongside other approaches to explore online talk. For example, several studies applied CA as a “first stage” of analysis, using CA features as a kind of coding scheme to inform descriptive statistics (Berglund, 2009; Bou-Franch et al., 2012; Lipinski-Harten and Tafarodi, 2012), content analysis (Steensen, 2014) or quasi-experimental designs (Condon and Čech, 2010; McKinlay, et al., 1994), rather than using CA to study naturallyoccurring talk. For example, Steensen (2014) counted the frequency of adjacency pairs in journalist-hosted synchronous chat and identified their characteristics to conduct a “CA inspired content analysis” (p. 1198), with an explicit focus on understanding how journalists establish and maintain their legitimacy in online settings. As another example, Berglund (2009) studied coherence in instant messaging, specifically coding for disrupted turn adjacency (among other conversational features). As part of her findings, she reported the frequencies with which the selected conversational features were deployed in the data. A number of studies also combined CA with other forms of analysis, such as various approaches to discourse analysis (Castañeda, 2012; Giles and Newbold, 2011; Smithson et al., 2011b). 3.2. Research aims for using CA to understand online talk In our intensive review of the articles, we noted that the included studies focused on four primary aims: (1) comparing faceto-face (F2F) with online talk; (2) understanding how online talk is coherent to participants; (3) understanding how participants deal with trouble in online talk; and (4) understanding how participants accomplish social actions in asynchronous environments. Although we have chosen to organize our findings this way, the aims of the studies sometimes overlapped as conversational structures serve multiple functions and most studies utilized more than one in their analysis. In the following discussion, we address each aim in turn, describing which CA structures were frequently drawn upon within each focus through the identification of key studies and provision of illustrative examples from our review. In highlighting these four aims, we illustrate the predominant uses of CA to understand text-based online talk and argue for continued conversation around this methodological focus. 3.2.1. Comparing face-to-face with online talk Across the literature, we noted that researchers often compared online talk with phone or F2F conversations. For example, some of the earliest studies using CA to understand online talk were focused on openings and/or closings, and these studies focused mainly on synchronous chat, such as instant messaging (e.g., Negretti, 1999; Rintel et al., 2001; Rintel and Pittam, 1997). Studies concerned with openings and/or closings had a more or less explicit focus on the comparison of online and F2F data (Danby et al., 2009; Markman, 2009; Negretti, 1999; Raclaw, 2008; Rintel et al., 2001; Rintel and Pittam, 1997; Stommel and te Molder, 2015). Overall, however, these researchers concluded that textbased online chat looked much like voice-based F2F or telephone conversations, with some effect due to the conversational medium (Raclaw, 2008; Rintel et al., 2001; Rintel and Pittam, 1997). This concern with comparison is still evident in some recently published studies (e.g., Frobenius and Harper, 2015), although it is somewhat less prominent as a focus overall. Further, some recent publications have called for moving beyond the need for explicit comparison (e.g., Meredith and Potter, 2013), recalling Lamerichs and te Molder's (2003) call to analyze online interactions as “social practices in their own right” (p. 461). Ong (2011), notably, showed that we must extend and/or modify

5

the ‘rules’ for preference structure and pursuing a non-response so that they apply to chat contexts, too, drawing upon findings around the use of ellipses-only and blank turns in chat. Relatedly, Gonzalez-Lloret (2011) argued that “rather than imposing existing structures on the new medium, we should examine the ways in which participants achieve different sequence types in the new medium …” (p. 313). Collister's (2011) study of the use of an asterisk (*) as a repair morpheme, for example, focused on this online-only phenomenon as there is no spoken English counterpart. Similarly, Meredith and Stokoe (2013) argued that “online interaction should be treated as an adaptation of an oral speechexchange system” (p. 1), with systematic differences designed to accomplish specific actions. These findings echo the earlier findings of Garcia and Jacobs (1999), Schönfeldt and Golato (2003), and others who established that CA was an appropriate methodology for examining online talk. We noted that within the studies of asynchronous data, such as online support groups, there was a greater focus on justifying the use of CA for this type of data, rather than making comparisons with F2F or telephone interactions. This may be because of how vastly different asynchronous data can be from spoken talk. 3.2.2. Understanding how online talk is coherent to participants Another aim of CA for online talk was to closely examine how participants performed social actions despite what was perceived as a lack of coherence due to the disruption of face-to-face turntaking systems. The most prominent of these studies were tied to the concept of “disrupted turn adjacency” (Herring, 1999) in which the spatio-temporal relationship of first- and next-turn pair parts is interrupted. In these studies, the concern was how participants organized their talk, usually in synchronous modalities, to be coherent as interpreted through various aspects of sequence organization – in particular turn-taking and adjacency pairs. Sacks, et al. (1974) established rules for turn-taking offline, which were as follows: (a) one party speaks at a time; (b) at the end of a TCU the next selected speaker should take a turn; (c) if a new speaker has not been selected someone can self-select, and; (d) if no one is selected or self-selected the original speaker can continue. However, as Herring (1999) and others have observed, these rules do not apply in text-based online talk. Thus, a number of studies were concerned with the degree to which turn-taking rules still hold true in the online medium as a function of maintaining coherence. In one frequently cited study, Simpson (2005) observed that disrupted turn adjacency was likely due to the fact that turns cannot be seen until they are sent and that paralinguistic clues are missing. He further suggested that “these observations on disrupted turn adjacency tend to support the view that applying models of turn-taking in spoken conversation directly to [online talk] is not profitable” (p 344). Thus, he proposed an “alternative perception of cohesion” (p. 344) by introducing the notion of the conversational floor – citing Edelsky's (1981) seminal work on holding the floor – noting that online chat is more akin to multi-party dinner parties rather than the twoparty conversations that are often studied in CA. Overwhelmingly, too, the articles we identified as having an explicit focus on adjacency pairs were concerned with coherence. These studies could perhaps be understood as asking: when firstand second-pair parts are no longer physically adjacent, what does this do to coherence? While early studies (e.g., Garcia and Jacobs, 1998, 1999) found that adjacency pairs could not be relied upon for establishing coherence in online talk, later studies (e.g., Berglund, 2009; Bou-Franch et al., 2012; Gibson, 2009a) tended to find that while there might potentially be difficulties in establishing coherence, participants generally did not orient to it as problematic. This, they noted, was partially accomplished through establishing additional means of coherence (e.g., lexical

6

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

relationships). The later articles also seem to work with what might be called an expanded notion of adjacency, in which the adjacency of talk is established not solely by physical co-presence in which the second-pair part immediately follows the first-pair part. Instead, they tended to recognize that second-pair part responses were generally readable as such wherever they physically occurred in the sequence. For example Schönfeldt and Golato (2003) suggested that adjacency in quasi-synchronous chat is “an achievement of the participants’ reading of, and selection from, a quickly changing stream of messages addressed to them,” a phenomenon which they termed “virtual adjacency” (p. 251). Garcia and Jacobs (1998, 1999), too, alluded to the potentially distinct nature of online talk when they identified “phantom adjacency pairs,” defined as utterances that look like adjacency pairs, but in which the second-pair part does not actually belong to the first. By comparing transcripts of interactions with videotapes of the same interactions, they were able to identify that some apparent second-pair parts were begun before the apparent first-pair parts. They used this to argue that an important part of coherence is lost in quasi-synchronous talk. Other scholars concerned with coherence, though, found less communicative collapse. Berglund (2009), for example, used the notion of affordances (Gibson, 1977) to identify whether and how disrupted turn adjacency led to interactional breakdown by focusing on sections of chat transcripts where messages were intertwined or ambiguous. She found that the sequential structure of interaction was generally clear, and that participants used strategies, such as lexical repetition, substitution, and timing to establish coherence. A number of studies emphasized the strategies participants used to maintain coherence, such as the use of explicit clarifications and/ or repairs (Nilsen and Mäkitalo, 2010), repetition, addressivity, and posting short messages to hold the floor (Gonzales-Lloret, 2011), using ellipsis-only or blank turns to accomplish turn elicitation (Ong, 2011), floor holding (Simpson, 2005), and waiting to take a turn when it is clear that the other person is typing a response (Gibson 2014; Nilsen and Mäkitalo, 2010). Overwhelmingly, the studies found that while initially participants may struggle with the lack of coherence, they quickly adapt and are able to communicate without any notable difficulty (Berglund, 2009; Bou-Franch et al., 2012; Negretti 1999; Nilsen and Mäkitalo, 2010). 3.2.3. Understanding how participants deal with trouble in online talk While coherence was not found to be particularly problematic to participants in online talk, troubles do, at times, arise. Researchers have thus also explored how the invisibility of the interlocutors can function as a source of interactional trouble managed through repair. As discussed previously, these studies frequently drew upon what is known about repair from F2F and telephone conversations, concluding that while participants do draw upon repair strategies familiar in oral conversations, they do so in ways that are shaped by the medium. Repair is defined as “the set of practices whereby a cointeractant interrupts the ongoing course of action to attend to possible trouble in speaking, hearing, or understanding the talk” (Kitzinger, 2013, p. 229). Within studies of repair, we identified two main strands: studies focused on how repairs are accomplished, and those focused on the consequences of repairs within a specific context. In the first strand, concern with how repairs are accomplished included questions of who initiates repairs and who performs them. Among these studies (e.g., Collister, 2011), the work of Schönfeldt and Golato (2003) was frequently cited. Their study compared multiparty chat to the interactional organization of offline conversation, attending to turn-taking, adjacency pairs, and sequence organization in addition to repair. They found that “participants have developed a system for dealing with the

limitations of the medium” (p. 252), noting among other things a preference for self-completed repairs and the different functions of silence online, attributable to server problems or the overlooking of one message in a steady stream of turns. Similarly, Meredith and Stokoe's (2013) study focused on screen captures of two party chat in Facebook, wherein they noted how both visible repairs (designed to correct production errors) and invisible repairs (made before posting the turn) are designed to accomplish particular actions. They argued, similar to Garcia and Jacobs (1999) and Schönfeldt and Golato (2003), that online interaction should be understood a “particular speech-exchange system” (p. 2), with the differences designed to accomplish specific actions. The second strand of research focused on the consequences of repairs in specific contexts. Koshik and Okazawa (2012), for instance, examined how repairs function to show sources of trouble in library chat reference interactions, while Tanskanen and Karhukorpi (2008), Tudini (2013), and Castañeda (2012) used CA to understand repair in the context of language learning. The three studies focused on language learning concentrated on the ways that repairs function in negotiating affiliation (Castañeda, 2012; Tanskanen and Karhukorpi, 2008) and in acquisition of language (Tudini, 2013). In contrast, Koshik and Okazawa (2012) used Schönfeldt and Golato's (2003) findings as a starting point to study repair in library chat reference, which they argued was a useful starting place for training librarians and thereby improving practice. 3.2.4. Understanding how participants accomplish social actions in asynchronous environments A number of the studies of asynchronous data, such as discussion forums and blogs, situated their analysis in DP, drawing on CA to understand issues of accountability (Antaki et al., 2005; Flinkfeldt, 2011, 2014; Paulus and Lester, 2013; Sneijder and te Molder, 2004, 2005) and stake management (Durrheim et al., 2015; Vayreda and Antaki, 2011). Lamerichs and te Molder (2003) wrote one of the first articles making the case for taking a DP approach to understanding asynchronous online talk, arguing that it “require(s) an approach in which text and talk are analysed as part of social practice” (p. 452) and that “in order to understand how identity categories work in actual online discourse, we have to reformulate the dominant cognitivist notion of identity into a participant-centered and action-oriented account” (p. 452). Given that asynchronous data is much different from the spoken data studied by CA researchers, these articles spent time justifying their use of CA to understand long passages of written texts exchanged online. These articles typically used CA to understand broader social practices, exploring topics such as how identity construction (Flinkfeldt, 2014; Giles and Newbold, 2011; Goodings, 2011; Sneijder and te Molder, 2006, 2009; Vayreda and Antaki, 2009, 2011), relationship development (Adkins and Nasarczyk, 2009; Harris et al., 2012; Stommel, 2012), advice giving (Frith, 2013; Smithson et al., 2011; Vayreda and Antaki, 2009, 2011), learning (Lester and Paulus, 2011; Paulus and Lester, 2013; Smithson et al., 2012), racism (Durrheim et al., 2015), and legitimate community participation (Stommel and Koole, 2010; Stommel and Meijman, 2011; Vayreda and Antaki, 2011) are interactionally managed through talk. In this group of studies, analysis often focused on what various design features of the turn accomplished. Design features that were analyzed included but were not limited to what modality choices, oh-prefaced news receipts and other change of state tokens, script formulations, descriptions, humor/emoticons, assessments, extreme case formulations, mitigation devices and conditional structures accomplished in the talk. Sneijder and te Molder (2004), for example, illustrated how script-formulations and expressions of modality supported blame-allocation in an Internet forum on veganism, and through a line-by-line analysis illustrated:

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

how a vegan diet and ways to solve individual problems are implicitly constructed as an integral part of ordinary life, i.e. as mundane achievements. Participants are thus able to undermine potential negative inferences about the nature of veganism as a complex lifestyle in terms of protecting one's health. (p. 614.) In addition to turn design, researchers drew on MCA and storytelling in their investigation of how broader social actions are accomplished in asynchronous contexts. The original idea of MCA came out of work by Sacks (1972a, 1972b, 1992) who was concerned with “how categories of member generically figured in the common sense worlds at issue” (Schegloff, 2007, p. 466). The bulk of studies identified using MCA focused on how membership categories (e.g., metrosexuality) are constructed online. For instance, in a series of four articles, Hall and colleagues explored the construction of ‘metrosexuality’ in various online contexts (e.g., online men's magazines, YouTube tutorials) (Hall and Gough, 2011; Hall et al., 2012a, 2013, 2012b). Their research looked at how participants identifying as “metro” use categories, category-bound activities, category predicates, and MCDs to construct, define, and defend the category of metrosexual. Other studies have focused on constructing the categories of “vegan” (Sneijder and te Molder, 2009), “housewife” (Flinkfeldt, 2014), and “Christians” (Pihlaja, 2014). A second focus of studies using MCA was how categories are used in online communities, often as part of understanding membership or belonging. Most frequently, these studies occurred in the context of online support groups (Giles, 2006; Giles and Newbold, 2013; Smithson et al., 2011a; Stommel and Meijman, 2011). For example, Stommel and Koole (2010) explored how categories related to illness are made relevant and how these categories and category predicates either normalize or ‘make deviant’ the participants in a web forum about health, ultimately policing group membership through legitimizing some, but not other, presentations of identity. Interest in storytelling comes from Sacks' (1992) work with sequence organization and storytelling as it occurs in everyday talk. In his Lectures, this work was mainly concentrated on how a speaker projects a coming story in order to maintain control for several turns, how the co-participants accept the longer turn-attalk and participate in the telling (e.g., nodding), and how the speaker organizes the close of a story to invite a next turn-at-talk. Studies of storytelling online, though, have concentrated little – if at all – on these aspects, perhaps because all the articles we located took their data from asynchronous sites where maintaining the conversational “floor” was not of concern. The four articles that we found that drew upon storytelling shared a concern with social relationships. All included a focus on second stories (Adkins and Nasarczyk, 2009; Giles and Newbold, 2013; Osvaldsson, 2011; Veen et al., 2010) and two also used MCA in their analysis (Giles and Newbold, 2013; Osvaldsson, 2011). For example, Veen et al. (2010) looked at how participants in a celiac disease support group, through second stories, shared information and support, and established “dietary compliance as a matter of course” (p. 27). Specifically, through scripting diet frustrations as something not out of the ordinary and reformulating decisions to quit the diet as an action within the bounded-activities of maintaining the diet, participants were able to “exclude quitting as an option” (p. 35).

4. Discussion The purpose of this study was to better understand researcher use of CA to examine online talk. Our findings show that such studies are increasing and are being conducted by an international

7

network of researchers across a variety of disciplines who are publishing their findings mainly in language-based and/methodological journals. Researchers have retained the synchronous/ asynchronous distinction in their description of data and are only beginning to study social media data. Data sources included the mundane conversations typical of traditional CA studies, as well as institutional talk from school, counseling, and workplace settings. We agree with Giles et al. (2015) that scholars taking up CA to understand online talk should “collect as much context data as necessary to understand the (technical) conditions in which the data came into existence” (p. 50). Meredith and Potter (2013) and Meredith and Stokoe (2013) have made a strong case as to why data in CA studies of online talk should include screen recordings of synchronous chat participants’ real-time interactions (and new forms of transcribing), and we support their arguments. There is indeed a need for continued study of the range of human activities involved in producing online talk. Perhaps unsurprisingly, a great deal of attention has been paid to how online talk differs from F2F talk as it has been described over the years by conversation analysts. While at times this attention has been to justify the use of CA for understanding new forms of social interaction, it has also been an attempt to document the ways in which social actions are accomplished in these new contexts. A greater understanding of this can help contribute to the methodological development of the use of CA for understanding online talk (Greiffenhagen and Watson, 2005). More specifically, researchers have studied how turn-taking systems and other structures of conversation vary when people's talk is mediated by various sorts of online systems, when people are not physically together (and thus invisible) and/or or not interacting in real-time, and how troubles are attended to by the participants. The resulting studies not only provide valuable insights into what strategies people use in online talk to accomplish social actions, but they are also useful examples of how CA has been both “digitized” and reconstructed as a “natively digital” method (Giles et al., 2015). In the future, however, it would certainly be valuable to pursue research that combined CA with multimodal forms of analysis to further extend the conversation beyond CA's primary focus on sequentiality. Additionally, while a variety of “foundational structures” of CA were used to understand online talk, few studies drew upon deviant cases or specifically mentioned unmotivated looking or next-turn proof as ways of establishing the trustworthiness of their work. Similarly, only a few studies dealt at all with non-verbal features (e.g. Skovholt et al., 2014), even though such features are not necessarily absent in online talk. Despite its genesis alongside other CA concepts, MCA has received more attention in ethnomethodological circles (e.g., Hester and Eglin, 1997; Housley and Fitzgerald, 2007). However, as Stokoe (2012) keenly observed, for researchers primarily interested in categorical or topical issues, MCA can be a useful means for “studying those issues as members’, rather than analysts’, categories” (p. 279). It is mainly this capacity in which its uptake in studies of online talk has been prominent. In fact, the use of MCA by researchers of online talk is perhaps contributing to its development. While MCA has been enthusiastically embraced by those studying asynchronous online talk, epistemics (Heritage, 2012) has yet to be explored and may be a fruitful area of analysis, especially in settings such as health and education. Finally, there were limitations to our review, which influenced the findings we could report. First, we faced a challenge when determining the final set of articles to analyze in large part because of the extensive overlap between CA and other languagebased methods and approaches. Second, some studies selfidentified as doing “conversation analysis” but were not actually drawing upon conversation analysis as an established methodological approach (i.e. they were using the phrase as a generic

8

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

description for analyzing an online conversation). For example, these studies did not cite key CA scholars nor did they engage in a line-by-line analysis of the data. We excluded such studies from our review. It is also possible that other studies which did not self-identify as CA were in fact using this method but were not located for our review.

5. Conclusion Future research should investigate in more depth how specific fields, most notably counseling, education, the workplace, and language learning, are taking up CA to understand what is happening in their discipline as counseling, business meetings, and teaching continue to migrate online. Expanding our review to include newer forms of multimodal data is a natural extension of this work. Finally, the prevalence of studies that draw upon naturally-occurring online talk in conjunction with other sorts of data (e.g. interviews, surveys) and/or use CA in conjunction with other forms of analysis may be worth exploring in more depth to better understand the questions that can be answered through innovative research designs. Focusing specifically on how CA, in conjunction with DP, is being used to examine online support groups is another fruitful area of research.

References Adkins, B.A., Nasarczyk, J., 2009. Asynchronicity and the ‘time envelope’ of online annotation: the case of the photosharing website, Flickr. Aust. J. Commun. 35 (3), 115–140. Androutsopoulos, J., 2006. Introduction: sociolinguistics and computer-mediated communication. J. Socioling. 10 (4), 419–438. http://dx.doi.org/10.1111/ j.1467-9841.2006.00286.x. Androutsopoulos, J., Beißwenger, M., 2008. Introduction: data and methods in computer-mediated discourse analysis. Language@Internet 5 (2) http://www. languageatinternet.org/articles/2008/1609. Antaki, C., Ardévol, E., Núñez, F., Vayreda, A., 2005. “For she who knows who she is:” managing accountability in online forum messages. J. Comput.-Mediat. Commun. 11 (1), 114–132. http://dx.doi.org/10.1111/j.1083-6101.2006.tb00306.x. Baym, N.K., 1996. Agreements and disagreements in a computer-mediated discussion. Res. Lang. Soc. Interact. 29 (4), 315–345. http://dx.doi.org/10.1207/ s15327973rlsi2904_2. Berglund, T.Ö., 2009. Disrupted turn adjacency and coherence maintenance in instant messaging conversations. Language@Internet 6 (2), 1–25 http://www. languageatinternet.org/articles/2009/2106. Bou-Franch, P., Lorenzo-Dus, N., Blitvich, P.G.-C., 2012. Social interaction in YouTube text-based polylogues: a study of coherence. J. Comput.-Mediat. Commun. 17 (4), 501–521. http://dx.doi.org/10.1111/j.1083-6101.2012.01579.x. Castañeda, A., 2012. EFL women-learners construction of the discourse of egalitarianism and knowledge in online-talk-in-interaction. Colomb. Appl. Linguist. J. 14 (1), 163–179. Collister, L.B., 2011. *-repair in online discourse. J. Pragmat. 43 (3), 918–921. http: //dx.doi.org/10.1016/j.pragma.2010.09.025. Condon, L.S., Čech, G.C., 2010. Discourse management in three modalities. Language@Internet 7 (6) http://www.languageatinternet.org/articles/2010/2770. Crystal, D., 2006. Language and the Internet. Cambridge University Press, Cambridge, England. Danby, S.J., Butler, C.W., Emmison, M., 2009. When 'listeners can't talk': comparing active listening in opening sequences of telephone and online counselling. Aust. J. Commun. 36 (3). Durrheim, K., Greener, R., Whitehead, K.A., 2015. Race trouble: attending to race and racism in online interaction. Br. J. Soc. Psychol. 54 (1), 84–99. http://dx.doi. org/10.1111/bjso.12070. Edelsky, C., 1981. Who's got the floor? Lang. Soc. 10 (3), 383–421. http://dx.doi.org/ 10.1017/S004740450000885X. Flinkfeldt, M., 2011. ‘Filling one's days’: managing sick leave legitimacy in an online forum. Sociol. Health Illn. 33 (5), 761–776. http://dx.doi.org/10.1111/ j.1467-9566.2011.01330.x. Flinkfeldt, M., 2014. Making equality relevant: gender, housework, and sick leave legitimacy in online interaction. Fem. Psychol. 24 (3), 295–313. http://dx.doi. org/10.1177/0959353513515295. Frith, H., 2013. ‘CONGRATS!! You had an orgasm’: constructing orgasm on an internet discussion board. Fem. Psychol. 23 (2), 252–260. http://dx.doi.org/ 10.1177/0959353512463558. Frobenius, M., Harper, R., 2015. Tying in comment sections: the production of meaning and sense on Facebook. Semiotica 2015 (204), 121–143.

Garcia, A.C., Jacobs, J.B., 1998. The interactional organization of computer mediated communication in the college classroom. Qual. Sociol. 21 (3), 299–317. Garcia, A.C., Jacobs, J.B., 1999. The eyes of the beholder: understanding the turn-taking system in quasi-synchronous computer-mediated communication. Res. Lang. Soc. Interact. 32 (4), 337–367. http://dx.doi.org/10.1207/S15327973rls3204_2. Georgakopoulou, A., 2011. Computer-mediated communication. In: Ostman, J., Verschueren, J. (Eds.), Pragmatics in Practice. John Benjamins Publishing Company, Philadelphia, PA, pp. 93–110. Gernsbacher, M.A., 2014. Internet-based communication. Discourse Process. 51 (5– 6), 359–373. http://dx.doi.org/10.1080/0163853X.2014.916174. Gibson, J.J., 1977. The theory of affordances. In: Shaw, R.E., Bransford, J. (Eds.), Perceiving, Acting, and Knowing. Lawrence Erlbaum Associates, Hillsdale, NJ, pp. 67–82. Gibson, W., 2009a. Intercultural communication online: conversation analysis and the investigation of asynchronous written discourse. Forum: Qual. Soc. Res. 10 (1) http://www.qualitative-research.net/index.php/fqs/article/view/1253/2730. Gibson, W., 2009b. Negotiating textual talk: conversation analysis, pedagogy and the organisation of online asynchronous discourse. Br. Educ. Res. J. 35 (5), 705–721. Gibson, W., 2014. Sequential order in multimodal discourse: talk and text in online educational interaction. Discourse Commun. 8 (1), 63–83. http://dx.doi.org/ 10.1177/1750481313503222. Giles, D.C., 2006. Constructing identities in cyberspace: the case of eating disorders. Br. J. Soc. Psychol. 45 (3), 463–477. http://dx.doi.org/10.1348/014466605  53596. Giles, D.C., Newbold, J., 2011. Self- and other-diagnosis in user-led mental health online communities. Qual. Health Res. 21 (3), 419–428. http://dx.doi.org/ 10.1177/1049732310381388. Giles, D.C., Newbold, J., 2013. ‘Is this normal?’ The role of category predicates in constructing mental illness online. J. Comput.-Mediat. Commun. 18 (4), 476–490. http://dx.doi.org/10.1111/jcc4.12022. Hester, S., Eglin, P. (Eds.), 1997. Culture in Action: Studies in Membership Categorization Analysis. International Institute for Ethnomethodology and University Press of America, Boston, MA. Thurlow, C., Mroczek, K. (Eds.), 2011. Digital Discourse: Language in the New Media. Oxford University Press, Oxford, UK. Giles, D.C., Stommel, W., Paulus, T.M., Lester, J.N., Reed, D., 2015. Microanalysis of online data: the methodological development of “digital CA”. Discourse Context Media 7, 45–51. Gonzalez-Lloret, M., 2011. Conversation analysis of computer-mediated communication. CALICO J. 28 (2), 308–325. Goodings, L., 2011. The dilemma of closeness and distance: a discursive analysis of wall posting in MySpace. Forum: Qual. Soc. Res. 12 (3) http://www.qualitativeresearch.net/index.php/fqs/ article/view/1576. Greiffenhagen, C., Watson, R., 2005. ‘Teoria’e ‘Método’na CMC: Identidade, género e tomada-de-turno-uma abordagem etnometodológica e analíticoconversacional1. In: Braga, A. (Ed.), CMC, identidades e género: Teoria e método. Universidade da Beira Interior: Colecção Estudos em Comunicação, Covilhã, PT, pp. 89–114. Hall, M., Gough, B., 2011. Magazine and reader constructions of ‘metrosexuality’ and masculinity: a membership categorisation analysis. J. Gend. Stud. 20 (01), 67–86. http://dx.doi.org/10.1080/09589236.2011.542023. Hall, M., Gough, B., Seymour-Smith, S., 2012. “I'm METRO, NOT gay!”: a discursive analysis of men's accounts of makeup use on YouTube. J. Men's Stud. 20 (3), 209–226. http://dx.doi.org/10.3149/jms.2003.209. Hall, M., Gough, B., Seymour-Smith, S., 2013. Stake management in men's online cosmetics testimonials. Psychol. Market. 30 (3), 227–235. http://dx.doi.org/ 10.1002/mar.20600. Hall, M., Gough, B., Seymour-Smith, S., Hansen, S., 2012. On-line constructions of metrosexuality and masculinities: a membership categorization analysis. Gend. Lang. 6 (2), 379–403. Harris, J., Danby, S., Butler, C.W., Emmison, M., 2012. Extending client-centered support: counselors' proposals to shift from e-mail to telephone counseling. Text Talk 32 (1), 21–37. Harrison, S., 2003. Repair in email discussions. Discourse Analysis Online. Retrieved from: 〈http://extra.shu.ac.uk/daol/articles/closed/2003/004/harrison2003004. html〉. Heritage, J., 2012. Epistemics in action: action formation and territories of knowledge. Res. Lang. Soc. Interact. 45, 1–29. Herring, S., Stein, D., Virtanen, T., 2013. Pragmatics of Computer-Mediated Communication. De Gruyter Mouton, Boston, MA. Herring, S.C., 1996. Computer-Mediated Communication. John Benjamins Publishing Company, Philadelphia, PA. Herring, S.C., 1999. Interactional coherence in CMC. J. Comput.-Mediat. Commun. 4 (4). http://dx.doi.org/10.1111/j.1083-6101.1999.tb00106.x. Herring, S.C., 2004. Computer-mediated discourse. In: Tannen, D., Schiffrin, D., Hamilton, H. (Eds.), Handbook of Discourse Analysis. Blackwell, Oxford, pp. 612–634. Herring, S.C., 2007. A faceted classification scheme for computer-mediated discourse. Language@Internet 4 (1) http://www.languageatinternet.org/articles/ 2007/761. Herring, S.C., 2010. Computer-mediated conversation: introduction and overview. Language@Internet 7 http://www.languageatinternet.org/articles/2010/2801. Herring, S.C., 2011. Computer-mediated conversation part II: introduction and overview. Language@Internet 8 (2) http://www.languageatinternet.org/arti cles/2011/Herring.

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

Herring, S.C., 2013. Discourse in web 2.0: familiar, reconfigured and emergent. In: Tannen, D., Tester, A.M. (Eds.), Georgetown University Roundtable on Languages and Linguistics 2011: Discourse 2.0: Language and new media. Georgetown University Press, Washington, DC. Herring, S.C., Stoerger, S., 2014. Gender and (a)nonymity in computer-mediated communication. In: Ehrlich, S., Meyerhoff, M., Holmes, J. (Eds.), The Handbook of Language, Gender, and Sexuality, 2nd ed. John Wiley & Sons, Ltd., Chichester, UK. Holtz, P., Kronberger, N., Wagner, W., 2012. Analyzing internet forums: a practical guide. J. Media Psychol. 24 (2), 55–66. Housley, W., Fitzgerald, R., 2007. Categorization, interaction, policy, and debate. Crit. Discourse Stud. 4 (2), 187–206. http://dx.doi.org/10.1080/ 17405900701464840. Hutchby, I., 2001. Technologies, texts and affordances. Sociology 35 (2), 441–456. http://dx.doi.org/10.1177/s0038038501000219. Jowett, A., 2015. A case for using online discussion forums in critical psychological research. Qual. Res. Psychol. 12 (3), 287–297. http://dx.doi.org/10.1080/ 14780887.2015.1008906. Kitzinger, C., 2013. Repair. In: Sidnell, J., Stivers, T. (Eds.), The Handbook of Conversation Analysis. Wiley-Blackwell, West Sussex, UK, pp. 229–256. Koshik, I., Okazawa, H., 2012. A conversation analytic study of actual and potential problems in communication in library chat reference interactions. J. Am. Soc. Inf. Sci. Technol. 63 (10), 2006–2019. http://dx.doi.org/10.1002/asi.22677. Lamerichs, J., te Molder, H.F.M., 2003. Computer-mediated communication: from a cognitive to a discursive model. New Media Soc. 5 (4), 451–473. http://dx.doi. org/10.1177/146144480354001. Lazaraton, A., 2014. Aaaaack! The active voice was used! Language play, technology, and repair in the Daily Kos weblog. J. Pragmat. 64, 102–116. http://dx.doi.org/ 10.1016/j.pragma.2014.02.002. Lester, Jessica N., Paulus, Trena M., 2011. Accountability and public displays of knowing in an undergraduate computer-mediated communication context. Discourse Stud. 13 (6), 671–686. http://dx.doi.org/10.1177/1461445611421361. Levinson, S.C., 2013. Action formation and ascription. In: Sidnell, J., Stivers, T. (Eds.), The Handbook of Conversation Analysis. Wiley-Blackwell, West Sussex, UK, pp. 103–130. Lipinski-Harten, M., Tafarodi, R.W., 2012. A comparison of conversational quality in online and face-to-face first encounters. J. Lang. Soc. Psychol. 31 (3), 331–341. http://dx.doi.org/10.1177/0261927  12446601. Markman, K.M., 2009. “So what shall we talk about” openings and closings in chatbased virtual meetings. J. Bus. Commun. 46 (1), 150–170. Mazur, J.M., 2004. Conversation analysis for educational technologists: theoretical and methodological issues for researching the structures, processes and meaning of on-line talk. In: Jonassen, D.H. (Ed.), Handbook of Research on Educational Communications and Technology, 2nd ed. Lawrence Erlbaum Associates, Mahwah, NJ, pp. 1073–1098. McKinlay, A., Procter, R., Masting, O., Woodburn, R., Arnott, J., 1994. Studies of turntaking in computer-mediated communications. Interact. Comput. 6 (2), 151–171. http://dx.doi.org/10.1016/0953-5438(94)90022-1. Meredith, J., Potter, J., 2013. Conversation analysis and electronic interactions: methodological, analytic and technical considerations. In: Lim, H.L., Sudweeks, F. (Eds.), Innovative Methods and Technologies for Electronic Discourse Analysis. IGI Global, Hershey, PA, pp. 370–393. Meredith, J., Stokoe, E.H., 2013. Repair: comparing facebook ‘chat’ with spoken interaction. Discourse Commun. 8 (2), 181–207. Moore, R.J., Ducheneaut, N., Nickell, E., 2007. Doing virtually nothing: awareness and accountability in massively multiplayer online worlds. Comput. Support. Co-op. Work 16 (3), 265–305. http://dx.doi.org/10.1007/s10606-006-9021-4. Myers, G., 2010. Stance-taking and public discussion in blogs. Crit. Discourse Stud. 7 (4), 263–275. Negretti, R., 1999. Web based activities and SLA: a conversation analysis approach. Lang. Learn. Technol. J. 3 (1), 75–87. Nevile, M., 2015. The embodied turn in research on language and social interaction. Res. Lang. Soc. Interact. 48 (2), 121–151. http://dx.doi.org/10.1080/ 08351813.2015.1025499. Nilsen, M., Mäkitalo, Å., 2010. Towards a conversational culture? How participants establish strategies for coordinating chat postings in the context of in-service training. Discourse Stud. 12 (1), 90–105. http://dx.doi.org/10.1177/ 1461445609346774. Ong, K.K.W., 2011. Disagreement, confusion, disapproval, turn elicitation and floor holding: Actions as accomplished by ellipsis marks-only turns and blank turns in quasisynchronous chats. Discourse Stud. 13 (2), 211–234. http://dx.doi.org/ 10.1177/1461445610392138. Osvaldsson, K., 2011. Bullying in context: Stories of bullying on an internet discussion board. Child. Soc. 25 (4), 317–327. http://dx.doi.org/10.1111/ j.1099-0860.2011.00383.x. Paulus, Trena M., Lester, Jessica N., 2013. Making learning ordinary: ways undergraduates display learning in a CMC task. Text Talk 33 (1), 53–70. http://dx.doi. org/10.1515/text-2013-0003. Page, R., 2012. Stories and Social Media: Identities and Interaction. Routledge, New York. Pihlaja, S., 2014. “Christians” and “bad Christians”: categorization in atheist user talk on YouTube. Text Talk 34 (5), 623–639. http://dx.doi.org/10.1515/ text-2014-0020. Plunkett, R., 2009. Fashioning the feasible: categorisation and social change. Aust. J. Commun. 36 (3), 23.

9

Pomerantz, A., 1984. Giving a source or basis: the practice in conversation of telling ‘how I know’. J. Pragmat. 8 (5), 607–625. Raclaw, J., 2008. Two patterns for conversational closings in instant message discourse. Colo. Res. Linguist. 21 (1), 1–21. Reed, D., Ashmore, M., 2000. The naturally-occurring chat machine. M/C: J. Media Cult. 3 (4). Reed, D., 2001. ‘Making conversation’: sequential integrity and the local management of interaction on Internet newsgroups. In: Proceedings of the 34th Hawaii Conference on System Sciences. Retrieved from: 〈http://citeseerx.ist.psu.edu/ viewdoc/download?doi ¼ 10.1.1.96.9001&rep¼rep1&type ¼ pdf〉. Rintel, E.S., Pittam, J., 1997. Strangers in a strange land: interaction management on internet relay chat. Hum. Commun. Res. 23 (4), 507–534. http://dx.doi.org/ 10.1111/j.1468-2958.1997.tb00408.x. Rintel, E.S., Mulholland, J., Pittam, J., 2001. First things first: internet relay chat openings. J. Comput.-Mediat. Commun. 6 (3). http://dx.doi.org/10.1111/ j.1083-6101.2001.tb00125.x. Sacks, H., 1972a. An initial investigation of the usability of conversational data for doing sociology. In: Sudnow, D.N. (Ed.), Studies in Social Interaction. Free Press, New York, pp. 31–74. Sacks, H., 1972b. On the analyzability of stories by children. In: Gumperz, J.J., Hymes, D. (Eds.), Directions in Sociolinguistics: The Ethnography of Communication. Basil Blackwell, New York, pp. 325–345. Sacks, H., 1992. Lectures on Conversation. Blackwell, Oxford. Sacks, H., Schegloff, E.A., Jefferson, G., 1974. A simplest systematics for the organization of turn-taking for conversation. Language 50 (4), 696–735. Schönfeldt, J., Golato, A., 2003. Repair in chats: a conversation analytic approach. Res. Lang. Soc. Interact. 36 (3), 241–284. http://dx.doi.org/10.1207/ S15327973RLSI3603_02. Schegloff, E.A., 2006. Interaction: the infrastructure for social institutions, the natural ecological niche for language, and the arena in which culture is enacted. In: Tannen, D., Hamilton, H.E., Schiffrin, D. (Eds.), The Handbook of Discourse Analysis. Wiley Blackwell, West Sussex, UK, pp. 70–96. Schegloff, E.A., 2007. Sequence Organization in Interaction. Vol. 1. Cambridge University Press, Cambridge. Seargeant, P., Tagg, C., 2014. The Language of Social Media: Identity and Community on the Internet. Palgrave Macmillan, Hampshire, UK. Sidnell, J., Stivers, T., 2013. The Handbook of Conversation Analysis. WileyBlackwell, Chichester, West Sussex, UK. Simpson, J., 2005. Conversational floors in synchronous text-based CMC discourse. Discourse Stud. 7 (3), 337–361. http://dx.doi.org/10.1177/1461445605052190. Skovholt, K., Grønning, A., Kankaanranta, A., 2014. The communicative functions of emoticons in workplace e-mails. J. Comput.-Mediat. Commun. 19 (4), 780–797. http://dx.doi.org/10.1111/jcc4.12063. Smithson, J., Jones, R.B., Ashurst, E., 2012. Developing an online learning community for mental health professionals and service users: a discursive analysis. BMC Med. Educ. 12 (1), 12–22. Smithson, J., Sharkey, S., Hewis, E., Jones, R., Emmens, T., Ford, T., Owens, C., 2011. Problem presentation and responses on an online forum for young people who self-harm. Discourse Stud. 13 (4), 487–501. http://dx.doi.org/10.1177/ 1461445611403356. Smithson, J., Sharkey, S., Hewis, E., Jones, R.B., Emmens, T., Ford, T., Owens, C., 2011b. Membership and boundary maintenance on an online self-harm forum. Qual. Health Res. 21 (11), 1567–1575. http://dx.doi.org/10.1177/ 1049732311413784. Sneijder, P., te Molder, H.F.M., 2004. ‘Health should not have to be a problem’: talking health and accountability in an internet forum on veganism. J. Health Psychol. 9 (4), 599–616. http://dx.doi.org/10.1177/1359105304044046. Sneijder, P., te Molder, H.F.M., 2005. Moral logic and logical morality: attributions of responsibility and blame in online discourse on veganism. Discourse Soc. 16 (5), 675–696. http://dx.doi.org/10.1177/0957926505054941. Sneijder, P., te Molder, H.F.M., 2006. Disputing taste: Food pleasure as an achievement in interaction. Appetite 46 (1), 107–116. http://dx.doi.org/10.1016/ j.appet.2005.10.002. Sneijder, P., te Molder, H.F.M., 2009. Normalizing ideological food choice and eating practices. Identity work in online discussions on veganism. Appetite 52 (3), 621–630. http://dx.doi.org/10.1016/j.appet.2009.02.012. Steensen, S., 2014. Conversing the audience: a methodological exploration of how conversation analysis can contribute to the analysis of interactive journalism. New Media Soc. 16 (8), 1197–1213. Stokoe, E.H., 2012. Moving forward with membership categorization analysis: Methods for systematic analysis. Discourse Stud. 14 (3), 277–303. Stommel, W., 2008. Conversation analysis and community of practice as approaches to studying online community. Language@Internet 5 (5), 1–22 http:// www.languageatinternet.org/articles/2008/1537. Stommel, W., 2012. Salutations, closings and pronouns: some aspects of recipient design in online counselling. Commun. Med. 9 (2), 145–158. http://dx.doi.org/ 10.1558/cam.v9i2.145. Stommel, W., Koole, T., 2010. The online support group as a community: a microanalysis of the interaction with a new member. Discourse Stud. 12 (3), 357–378. http://dx.doi.org/10.1177/1461445609358518. Stommel, W., Meijman, F.J., 2011. The use of conversation analysis to study social accessibility of an online support group on eating disorders. Glob. Health Promot. 18 (2), 18–26. http://dx.doi.org/10.1177/1757975911404764. Stommel, W., Lamerichs, J., 2014. Interaction in online support groups: advice and beyond. In: Hamilton, H., Chou, W.S. (Eds.), Handbook of Language and Health Communication. Routledge, New York.

10

T. Paulus et al. / Discourse, Context and Media 12 (2016) 1–10

Stommel, W., te Molder, H., 2015. Counseling online and over the phone: when preclosing questions fail as a closing device. Res. Lang. Soc. Interact. 48 (3), 281–300. http://dx.doi.org/10.1080/08351813.2015.1058605. Tanskanen, S.K., Karhukorpi, J., 2008. Concessive repair and negotiation of affiliation in e-mail discourse. J. Pragmat. 40 (9), 1587–1600. http://dx.doi.org/10.1016/j. pragma.2008.04.018. ten Have, P., 1999. Structuring writing for reading: hypertext and the reading body. Hum. Stud. 22 (2-4), 273–298. ten Have, P., 2000. Computer-mediated chat: ways of finding chat partners. M/C: J. Media Cult. 3 (4) http://journal.media-culture.org.au/0008/partners.php. Tudini, V., 2013. Form-focused social repertoires in an online language learning partnership. J. Pragmat. 50 (1), 187–202. http://dx.doi.org/10.1016/j.pragma. 2012.12.005. Vayreda, A., Antaki, C., 2009. Social support and unsolicited advice in a bipolar disorder online forum. Qual. Health Res. 19 (7), 931–942.

Vayreda, A., Antaki, C., 2011. To vaccinate or not? The disqualification of commercial sources of health advice in an online forum. Commun. Med. 8 (3), 273–282. Veen, M., te Molder, H.F.M., Gremmen, B., van Woerkum, C., 2010. Quitting is not an option: an analysis of online diet talk between celiac disease patients. Health 14 (1), 23–40. http://dx.doi.org/10.1177/1363459309347478. Woods, M., Paulus, T., Atkins, D.P., Macklin, R., 2016. Advancing qualitative research using qualitative data analysis software (QDAS)? Reviewing potential versus practice in published studies using ATLAS.ti and NVivo, 1994–2013. Soc. Sci. Comput. Rev. . http://dx.doi.org/10.1177/0894439315596311 (in press) Yus, F., 2011. Cyberpragmatics: Internet-mediated Communication in Context. John Benjamins Publishing Company, Philadelphia, PA. Zappavigna, M., 2012. Discourse of Twitter and Social Media: How we use Language to Create Affiliation on the Web. Continuum, London, UK.