With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television models
First Monday

With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television models by Matthew Pittman and Alec C. Tefertiller



Abstract
Social networking services like Twitter have changed the way people engage with traditional broadcast media. But how social is “second screen” activity? The purpose of this study is to determine if patterns of connected viewing (augmenting television consumption with a second screen) and co-viewing (watching television together) are different for traditionally broadcast, “appointment” television shows versus streaming, asynchronous television releases. This study explores this phenomena of “co-connected viewing” — a combination of connected and co-viewing — on Twitter for four programs that were all released within seven days of each other: Parks and Recreation, Downton Abbey, House of Cards, and Unbreakable Kimmy Schmidt. Complete datasets (over 200,000 tweets) from 72 hours’ worth of Twitter activity for four television programs, two traditional and two streaming, were collected and analyzed. In terms of co-connected viewing, the study found that despite radically different broadcast models and corresponding shapes in Twitter activity, the ratios of social to non-social tweets were nearly identical. Additionally, the study found that the asynchronous, streaming Netflix shows saw more engagement from active Twitter users. Finally, implications are discussed for viewers, fans, advertisers, and the television industry, as well as directions for future research.

Contents

Introduction
Literature review
Method
Results
Discussion
Conclusion

 


 

Introduction

In 2013, more than 36 million Americans tweeted about television programs and events (Nielsen Social, 2014). The Pew Internet & American Life Project also found that younger (18–24 years old) adults, in particular, use their mobile devices “to connect directly with the programming content — and to those who are interested in the same content” [1]. Twitter makes co-viewing (traditionally defined as individuals in the same household watching television together) possible across geographic distance for fans of the same content. Through the use of applications and web-enabled devices, connected co-viewing (“co-connected” viewing) has the potential to unite viewers around the globe not only to the shows they enjoy, but also to each other. Because the relationship remains unclear, this study aims to determine the ratio of co-connected viewing to overall Twitter activity for television shows of different genres and broadcast methods.

For broadcasters and advertisers, the real value in connected viewing and platforms such as Twitter is in their ability to keep audiences engaged and active for the duration of a broadcast. This increased attentiveness to a program — as opposed to merely “having the television on,” perhaps even in the other room — may lead to increased attentiveness to its commercials as well. Twitter is of particular interest as a platform, as it provides “some of the most enriched data sets available, especially on a minute-by-minute basis, which is the traditional framework for linear advertising stories” [2]. Twitter provides a platform to both facilitate and measure active engagement with live television, crucial for advertisers and broadcasters alike. However, alternatives to the traditional broadcast model introduced by streaming services have turned the concept of the live broadcast on end.

Over the past few decades, the traditional appointment model of broadcasting to a live television audience has been challenged by time-shifting technologies like the VCR (Video Cassette Recorder), Pay-Per-View or VOD (Video on Demand), and DVR (Digital Video Recorder). More recently, however, popular streaming services such as Netflix have given consumers a way to experience television programs that is truly asynchronous: the consumer, not the broadcaster, may now choose the time and amount of viewing. In 2012, Netflix released all episodes of their first original program Lilyhammer at one time, rather than rolling out episodes at the traditional pace of once per week. Explaining the reasoning behind this new broadcasting model, Netflix chief content officer Ted Sarandos explained, “We are trying to give our members what they want: choice and control” [3]. Netflix’s strategy has not been ineffective, as they are poised to release more original content in 2015 than ever before — over 320 hours, compared to only 200 hours released by cable giant HBO (O’Keefe, 2015). At the same time, live television is struggling, as both television ratings and advertising dollars are slipping, prompting some in the industry to argue that the television advertising sales business may have peaked (Lieberman, 2015). Simply put, “While audiences are watching more television than ever, they’re watching less of it live, causing ad revenue to shrink” [4].

While Twitter becomes more salient as a means to facilitate connected viewing, co-viewing, and television as a social experience, trends suggest that the live-viewing audience continues to erode, giving way to models introduced by asynchronous streaming services such as Netflix. The purpose of this study is to examine what effect this shift from live appointment broadcasting to time-shifted broadcasting is having on online social interaction. Using data derived from Twitter regarding four popular television programs (Downton Abbey and Parks and Recreation broadcast via the appointment model, and House of Cards and the Unbreakable Kimmy Schmidt broadcast via asynchronous, Netflix model) this study seeks to better understand how different broadcast models affect the nature and quantity of social interaction.

 

++++++++++

Literature review

Connected viewing

The use of a smartphone or “second screen” to augment television consumption is now a common practice. In a phenomenon now known as “connected viewing,” over half of all adults use a mobile device while engaging with televised content (Smith, 2012) for a variety of reasons, including alleviating boredom during commercials, fact checking statements in the program, or exchanging opinions about the program online. Connected viewing refers to “a multiplatform entertainment experience, and relates to a larger trend across the media industries to integrate digital technology and socially networked communication with traditional screen media practices” [5]. The advent of connected viewing means vast research potential for television scholars and producers.

According to Lee and Andrejevic (2014), the promise of the second-screen is the promise of an aggregate audience, since “the interactive affordances of second-screen apps typically rely on live viewing” [6]. Lee and Andrejevic argue that rather than creating a democratic surround, where users have greater freedom in shaping their particular information and entertainment experience, instead what is being created is a digital enclosure, or marketing surround, where audience preferences are captured, scraped, saved, and used for consumer purposes. Live viewing coupled with second-screen interaction provides marketers with both a captive audience and a source of real-time data. However, while the second-screen may create an aggregate audience, Lee and Andrejevic (2014) suggest that second screen viewing might be reconfiguring viewing as a social ritual, where viewing becomes a shared event.

Extant research has examined connected viewing centered on news (Deller, 2011; Diakopoulos and Shamma, 2010; Himelboim, 2014; Wohn and Na, 2011) and entertainment programs (Lochrie and Coulton, 2011; Doughty, et al., 2014; Giglietto and Selva, 2014; Schirra, et al., 2014), as well as the possibility of using television-centered Twitter activity in studying viewership ratings and patterns (Wakamiya, et al., 2011). Harrington, et al. (2013) discuss how Twitter serves as a backchannel for fans of television to discuss their favorite programs as they are broadcast in real time. This audience-created running commentary may, provided it contains the appropriate hashtags for referencing, serve as an additional media stream for other fans of the same program to consume on their own second screens at home.

People tweet about programs or events that matter to them, both at (literally through use of an “@”) and with (through use of a “#”) others who share a similar disposition. Ji and Raney (2015) found that “the content of naturally occurring, freely offered, live-tweeted reactions to entertainment” [7] offers researchers potentially more valid data than could traditionally be collected in more artificial settings. In other words, certain narrative occurrences might precipitate a surge of Twitter activity. Schirra, et al. (2014) found that moments of sadness or loss, character growth, and humor were particular triggers for fans’ tweets. The same study also confirmed the social aspect of connected viewing, particularly for audience members watching television alone. Even when others were physically present in the home, if “friends and spouses have different television interests,” [8] then live-tweeting offers feelings of sociability and affirmation.

Xu and Yan (2011) assert that the feeling of connection to others via Twitter may be stronger than parasocial interaction because “it is easier for a viewer to feel connected to real people than to fictional characters” [9]. Furthermore, the same study also explored a scale of “feeling connected television viewing” (FCTV), which found three factors of sociability: viewers live-tweeted a program to bond and subsequently communicate with an immediate social circle, share in a sense of global community, and communicate with distant unknown others on Internet forums.

Additionally, the genre of a television show might influence what kind of connection Twitter users experience. Doughty, et al. (2012) used social network analysis in determining that different programs reflect different Twitter activity — a reality show had more audience members connecting to the celebrities it featured, while a current affairs programs had more audience members engaging in dialogue with friends. However, it is unclear whether or not this distinction holds up for time-shifted, streaming television. Regardless, the use of a second screen to tweet at and with other viewers of a television program is bringing greater enjoyment to audiences, and this connected viewing phenomena is changing the way viewers engage with their favorite shows.

Terms like “fan” and “fandom” carry with them connotations of emotional commitment that do not necessarily exist for all viewers. While definitions of fandom vary, Baym [10] notes that most definitions involve a “collective of people organized socially and their shared appreciation of a pop culture object or objects”. This study proceeds from this definition, assuming fans can simultaneously exercise their social collectivity and shared appreciation through use of hashtags on Twitter (connected viewing).

Of course, every viewer who tweets about a television show is not necessarily a fan. Jonathan Gray (2003) notes the presence of “non-fans” and “anti-fans” in construction of and participation in the ongoing dialogue of any cultural text. He describes non-fans as “those viewers or readers who do view or read a text, but not with any intense involvement” [11]. Non-fans make up a casual audience whose marginal affinity for a show is enough to watch it when it happens to be on, but not enough to seek it out on a regular basis. Non-fans are less likely with asynchronous content, a broadcast method that requires active agency, which implies a stronger connection to, or at least curiosity about, a show.

Anti-fans, on the other hand, possess stronger feelings about a show or its fan base (Alters, 2007) and are thus more likely to express these feelings through action. Gray [12] describes anti-fandom as “the realm not necessarily of those who are against fandom per se, but of those who strongly dislike a given text or genre, considering it inane, stupid, morally bankrupt and/or aesthetic drivel.” More colloquially, this might constitute what has come to be known as “hate-watching”: “Everybody does it. We obsessively watch TV shows that we kind of despise, and then we mock them, and complain about them on the Internet afterwards.” This anti-fandom ranges from possible secret enjoyment as a form of guilty pleasure (Harman and Jones, 2013) to outright “fan wars” (Recuero, et al., 2012) through use of negative or pejorative hashtags, spoiling information, or general trolling. While full exploration of anti-fan Twitter activity is beyond the scope of this paper, it remains relevant as a possible component of any connected viewing.

Co-viewing

The social utilities of television have been recognized since researchers first turned their attention to audience uses and gratifications of media. The uses and gratifications approach posits that consumers are active in their consumption of media, and they make choices about how media is used to gratify needs (Blumler, 1979; Katz, et al., 1974; Rubin, 1993). Much attention has been given to television using a uses and gratifications approach (Palmgreen, et al., 1988). Early examinations of television that explored the gratifications of the medium in comparison to other media, such as newspapers, radio, and the cinema, discovered that television scored highest in its ability to build connections with family, in addition to being a means of providing social integration (Katz, et al., 1973), a finding supported by Peled and Katz (1974).

Morrison and Krugman (2001) determined that the television could facilitate much of a household’s social activity, as television viewing is often experienced with multiple family members. In fact, more than the television alone, Morrison and Krugman discovered that the VCR was the most socializing form of television technology, as it enabled special co-viewing events, such as family movie nights or co-viewing of a rented program. It may be the case that the asynchronous viewing experience created by streaming services makes co-viewing possible in a manner similar to the VCR. For example, Haridakis and Hanson (2009) determined that co-viewing was one of the motivations for viewing and sharing videos on YouTube. It is likely that services such as Netflix, with its emphasis on user-controlled asynchronous viewing, are also useful in facilitating geographically independent co-viewing, in much the same way that the introduction of VCRs as a household television technology created new opportunities for shared television experiences.

Social media based, second-screen co-viewing is a key indicator of involvement with television. As Laursen and Sandvik (2014) have demonstrated, consumers interact online with live television, conversing not only with the marketing channels connected to content creators, but also in one-way and two-way conversations involving both small and large circles of online community. Social media co-viewing of live television can provide valuable opportunities for individuals who typically watch television alone to interact with others (Cohen and Lancaster, 2014). Social media sites such as Twitter provide an outlet for such interactions, with a spike in activity corresponding with a program’s broadcast (Lochrie and Coulton, 2011).

In addition, the nature of the live program often dictates the depth of responses (Doughty, et al., 2011). Tussey (2014) suggested that users favor social applications such as Twitter over applications provided by content providers because the content provider applications are often restrictive in their efforts to keep audiences tuned in, and users can take advantage of active social circles available through their existing social media networks. Holt and Sanson (2014) acknowledge that while ideological principles such as privacy have been fundamentally altered by corporate efforts to monitor and manage online interactions, ubiquitous digital connectedness has given people the ability to create strong social communities capable of interacting in ways never before possible. In spite of efforts to harvest online activity as a digital commodity, users find ways to build vibrant communities of social interaction.

In addition to providing a platform for co-viewing, television has been acknowledged as being a vital source of social integration. Rubin (1983) determined that the need to be able to talk about things that are going on, a form of information-seeking, continued to be an important gratification of television, more so than co-viewing and direct social interaction. In fact, television’s main gratifications were originally identified as entertainment, killing time, and information seeking, with the promise of future conversations and social integration being a part of the information seeking process. Taking the social aspect of television beyond co-viewing, it is important to note that the social integration gratification does not orient television as a social ritual, but rather emphasizes the acquisition of knowledge from television that can be used at a later date, in particular to facilitate conversations about cultural events (Matrix, 2014).

While notions of co-viewing — a term which implies two or more individuals simultaneously viewing television — have been associated with connected viewing as a live appointment-viewing phenomenon (Lee and Andrejevic, 2014), it is important to acknowledge the role asynchronous streaming television plays, not just in co-viewing as a household phenomenon (Morrison and Krugman, 2001), but as an important source of social integration (Matrix, 2014). Thus we propose the term of “co-connected” viewing that includes connected viewers who may co-view the same program at the same time (as with traditional appointment broadcast) or at different times (as with asynchronous streaming broadcast). It is possible that social interaction and social integration may be alive and well on Twitter for programs broadcast through non-traditional, streaming channels, and not just the exclusive domain of legacy networks that continue to promote and operate a live broadcast model.

Understanding Twitter communication

It has been suggested that Twitter is superior to previous social media networks in its ability to facilitate the live reporting of real-time events in addition to providing a rich platform for ongoing discussion (Bruns and Burgess, 2012). The text-centric nature of Twitter means minimal bandwidth, reception, and processing power are required to engage with others, thereby diminishing barriers to its use. Conversations on Twitter include information sharing and news reporting as well as direct conversations with other users (Java, et al., 2007). One particular type of activity specific to the Twitter network is the retweet function, which allows users to quickly and easily share another user’s tweet that is of interest. Retweets may serve different purposes for different individuals, but most notably they allow users to amplify and endorse particular content of interest to them (Bruns and Burgess, 2012), in keeping with the information sharing function of the site.

Of particular interest to researchers is the use of hashtags. Users utilize a hashtag (#) followed by an unbroken string of text to connect their tweet to a larger network of tweets using the same hashtag (e.g., ##MarriageEquality, #Oscars). Researchers have used hashtags to identify tweets related to particular events or topics of interest (Deller, 2011; Diakopoulus and Shamma, 2010; Doughty, et al., 2011; Giglietto and Selva, 2014; Lochrie and Coulton, 2011). The use of hashtags allows users the “flexibility and ability to rapidly form discursive communities around breaking news” [13] without having to designate specific follower/following connections. Users are able to immediately interact with other users regarding a particular topic of interest. Finally, it is important to note that not all Twitter users are equally active with the platform. Krishnamurthy, et al. (2008) found that Twitter users with more than 250 followers were much more active on the network, posting much more often (and about more topics) than those with less than 250 followers.

Research questions

As more and more individuals are augmenting their television experience with a second screen, television consumption — be it traditional appointment viewing or asynchronous streaming viewing — is rapidly becoming what constitutes connected viewing. However, while patterns of Twitter use for live television have been examined (Lochrie and Coulton, 2011), patterns of Twitter use for streaming, asynchronous television (and how they differ from live use patterns) have yet to be explored. Therefore, we seek to determine if different broadcast models of television have different levels of connected viewing:

RQ1: Do different broadcast models generate different patterns of corresponding Twitter activity?

Furthermore, Chen (2011) determined that tweeting to someone (using @replies) mediates the relationship between active Twitter use and the need for connection. Of the audience members who tweet using a program’s hashtag, it is not clear how many are co-viewing (tweeting to another user with the @), and how many are engaging in connected viewing (tweeting with the hashtag alone). In addition, it is possible that the different broadcast models may engage Twitter users who are active with the network at different levels:

RQ2: Do different broadcast models result in different levels of co-viewing?

Finally, are active or casual Twitter users more likely to practice connected viewing? Using a threshold of 250 followers (Krishnamurthy, et al., 2008), active users — who are probably tweeting a lot anyway — seem more likely to be the main participants in connected viewing, as opposed to casual Twitter users, who are more apt to be information-seekers:

RQ3: Do different broadcast models attract active or casual Twitter users?

 

++++++++++

Method

To address the research questions, we performed an analysis of tweets regarding both traditional, appointment broadcast and asynchronous, Web-streaming broadcast television shows released during a two-week period (23 February–9 March 2015). Four television programs were chosen, including two half-hour comedies and two hour-long dramas, with one show of each genre representing both broadcast models. The two streaming shows were House of Cards and Unbreakable Kimmy Schmidt, both distributed by Netflix; the two broadcast shows were the series finale of Parks and Recreation (NBC) and the season finale of Downton Abbey (PBS). Since all episodes of the streaming shows are released at once (including what is technically a season premiere and season finale), it was deemed acceptable to include the season finales of the traditionally broadcast shows, as they represent significant events in the course of the television season. Seventy-two hours’ worth of tweets were harvested for each show, beginning at 3PM EST the afternoon before each show’s release. This allowed for the capture of tweets that were written in anticipation of each release as well as ongoing chatter, as the streaming model allows users to continue to watch subsequent episodes as well as choose their own viewing times. Tweets were harvested using twarc (https://github.com/edsu/twarc), a command line tool for collecting and archiving Twitter JSON (JavaScript Object Notation) data using Twitter’s API (Application Programming Interface).

Hashtags were chosen to identify tweets for each show: #ParksandRec, #DowntonAbbey, #HouseofCards, and #UnbreakableKimmySchmidt. Hashtags closely related to the title of the show were chosen for consistency, and the popularity of each hashtag was confirmed using the Twitter analytics tool Topsy (http://topsy.com/). Capturing tweets that use a particular hashtag has its limitations, as others have noted (Bruns and Burgess, 2012). It is ostensibly possible to collect all tweets pertinent to a certain subject or event, provided those tweets include the corresponding hashtag. However, there may be many other tweets on the subject that do not use the hashtag, and those would be missed. For example, replies to tweets containing a hashtag that do not themselves include the hashtag will not be captured in the sample. Most Twitter research is limited by this self-selection bias: people motivated enough to tweet about a given topic are not necessarily representative of all other Twitter users. For the purposes of this study, however, we are only concerned with tweets that are specifically directed at and about a given television program’s audience through the use of that show’s hashtag.

 

++++++++++

Results

To suggest answers to the research questions, the complete datasets of Twitter activity from all four television programs were analyzed and visually represented

Data

A total of 204,327 tweets were archived from 125,538 unique user accounts. The hashtag #ParksandRec produced 40,394 tweets from 27,902 unique users, #HouseofCards produced 122,093 tweets from 81,438 users, #DowntonAbbey produced 21,807 tweets from 10,344 users, and #UnbreakableKimmySchmidt produced 20,033 tweets from 10,842 users. Each tweet was then coded as being directly social (original tweet containing an @reply), a retweet (contains an @reply, but is not original tweet), or neither (an original tweet with no @reply). In addition, each tweet was coded in accordance with the activity level of the user, using the threshold established by Krishnamurthy, et al. (2008). All tweets from users with fewer than 250 followers were coded as coming from low-activity Twitter users, and all tweets from users with 250 followers or more were coded as coming from high-activity Twitter users.

Research question one

To answer research question one, line graphs detailing all four programs’ Twitter activity based on their respective hashtags were created for the length of the 72-hour sample (See Figures 14). For both appointment broadcast shows, Parks and Recreation and Downton Abbey, non-social tweets, retweets, and social tweets spiked at times corresponding with the programs’ airtime, followed by steep drop-offs in activity soon after. For both programs, the airtime spike in Twitter activity was dominated by original, non-social chatter, followed by retweets, and then social tweets. Beyond the air-time spike, Parks and Recreation had one notable yet comparatively small spike in retweet activity on Thursday, 26 February, at 5PM EST when the official Parks and Recreation Twitter account (@parksandrecnbc) tweeted an image of one of the show’s popular characters, Ron Swanson, played by actor Nick Offerman, which accounted for 454 of the 602 retweets generated during that hour. With the exception of that small spike, neither Parks and Recreation nor Downton Abbey saw spikes exceeding 600 tweets from any category after their initial airtime spikes in activity.

 

Tweets per hour by tweet type for #ParksandRec
 
Figure 1: Tweets per hour by tweet type for #ParksandRec; 72 hours are represented, beginning at 3PM EST/2PM PST on 24 February. The episode aired at 10PM EST, which corresponds with Hour 8 on the graph.

 

 

Tweets per hour by tweet type for #DowntonAbbey
 
Figure 2: Tweets per hour by tweet type for #DowntonAbbey; 72 hours are represented, beginning at 3PM EST/2PM PST on 1 March. The episode aired at 9PM EST, which corresponds with Hour 7 on the graph.

 

 

Tweets per hour by tweet type for #HouseofCards
 
Figure 3: Tweets per hour by tweet type for #HouseofCards; 72 hours are represented, beginning at 3PM EST/2PM PST on 26 February. The complete season was released at 3AM EST, which corresponds with Hour 13 on the graph.

 

Tweets per hour by tweet type for #UnbreakableKimmySchmidt
 
Figure 4: Tweets per hour by tweet type for #UnbreakableKimmySchmidt; 72 hours are represented, beginning at 3PM EST/2PM PST on 5 March. The complete season was released at 3AM EST, which corresponds with Hour 13 on the graph.

 

The Twitter pattern for the asynchronous streaming broadcast shows was markedly different than that of the traditional broadcast shows. Rather than demonstrating one steep spike, tweets ebbed and flowed with daylight and evening hours for both programs, with decreases in activity during the late night/early morning hours, and then gradual builds to high activity in the evenings across the 72-hour time period. While House of Cards saw a spike in activity around 3AM EST on Friday, its release time, its sharpest peak came Saturday evening. Though The Unbreakable Kimmy Schmidt did not see a major peak at its release time (possibly because of its status as a new program, whereas House of Cards was in its third season), it did reach a peak on Friday evening. As with the traditionally broadcasted shows, both programs saw primarily original, non-social chatter, followed by retweets, and then social tweets across the time-span, with few deviations.

Research question two

We propose that a tweet constitutes “co-connected” viewing if it contains a program’s hashtag (thus representing connected viewing) as well as an “@” symbol (indicating communication directed at another person). We did not count retweets as co-connected viewing; while they automatically contain an @username of the person being retweeted, they are not necessarily original, social content. Research question two was addressed by analyzing the social nature of all of the tweets for each program within the 72-hour period. The percentage of tweets that were non-social, retweets, and co-connected (social) were determined for each program (see Table 1). Of the 40,394 #ParksandRec tweets and 11,359 #DowntonAbbey tweets, only 10.6 percent (4,279) and 11.8 percent (2,483), respectively, were social @reply tweets. Of the 69,392 #HouseofCards tweets and 12,641 #UnbreakableKimmySchmidt tweets, 11.3 percent (13,185) and 12.6 percent (2,529), respectively, were social @reply tweets. For all four shows, there was a difference of only two percentage points in co-connected Twitter activity.

 

Table 1: Summary of tweet type by television show hashtag.
Note: Non-social tweets were neither a retweet nor contained an @reply.
Tweet type#ParksandRec#DowntonAbbey#HouseofCards#Unbreakable KimmySchmidt
Non-social tweets17,63611,35969,39212,641
 (43.7%)(53.9%)(56.8%)(63.1%)
 
Reweets18,4797,24538,8864,863
 (45.7%)(34.4%)(31.8%)(24.3%)
 
@Reply tweets4,2792,48313,8152,529
 (10.6%)(11.8%)(11.3%)(12.6%)
 
Totals40,39421,087122,09320,033

 

Research question three

To answer the third research question, the activity level of each tweet’s user across the 72-hour period for each show was measured (see Table 2). The two streaming shows saw more engagement from active Twitter users, who were responsible for an average of 58 percent of the Twitter activity, while for appointment shows, they were responsible for an average of 52 percent. All but one of the programs saw more activity from active Twitter users than low-activity users, though the difference in activity did not exceed twenty percentage points. The biggest difference came from House of Cards at 59.3 percent for highly-active users and 40.7 percent for low-activity users, a difference of 18.6 percentage points, compared to 12.4 percentage points for Unbreakable Kimmy Schmidt and 10 percentage points for Downton Abbey. The notable exception was Parks and Recreation, where 51.8 percent of its Twitter activity came from low-activity users. This might be explained by the fact that Parks and Recreation’s season finale episode was also the series finale, which may have attracted attention and activity from those who would normally be less likely to engage in connected viewing.

 

Table 2: Summary of tweets by user activity level by television show hashtag.
Note: High activity level users have 250 or more followers, low activity level users have less than 250 followers.
User activity level#ParksandRec#DowntonAbbey#HouseofCards#Unbreakable KimmySchmidt
High activity19,47411,59572,34611,252
 (48.2%)(55.0%)(59.3%)(56.2%)
 
Low activity20,9209,49249,7478,781
 (51.8%)(45.0%)(40.7%)(43.8%)
 
Totals40,39421,087122,09320,033

 

 

++++++++++

Discussion

The purpose of this study was to determine if patterns of connected viewing, co-viewing, and co-connected viewing — a term we propose to indicate combination of connected and co-viewing — are different for traditionally broadcast, appointment television shows versus streaming, asynchronous television releases. The first research question asked if there were differences in Twitter activity between traditional and streaming programs. The second research question asked if there were different patterns of co-viewing for each broadcast model. The third research question asked if there were different levels of activity from highly active Twitter users for different broadcast models. Complete datasets of Twitter activity surrounding hashtags for 72 hours of four television programs, two traditional and two streaming, were employed to find answers to the research questions. Findings suggest that, while patterns of activity were different for each type of program, as well as for appointment and all-at-once streaming, the amount of direct social activity (co-connected viewing) was similar, and differences in involvement by active versus casual Twitter users was negligible.

The Twitter activity for appointment and asynchronous broadcast models differs (RQ1) the way one might expect: appointment shows see a large spike of Tweets that quickly dissipates once the broadcast is over, whereas streaming shows see a smaller-but-steady flow of Tweets that last for the entire 72-hour period following initial release. Both Parks and Recreation and Downton Abbey saw spikes in activity during their broadcast windows that constituted the vast majority (over 95 percent) of their Twitter activity, with minute pulses of trace activity occurring the following day. These pulses may have come from viewers who were watching a network rerun of the program, a time-shifted recording of their own, or watching on-demand online. These subsequent minor spikes were the only times when retweets surpassed original tweets in frequency, perhaps because fans were retweeting the show’s actors during humorous or dramatic moments of the show (Schirra, et al., 2014). It is also possible that fans were retweeting other fans, or the “fictional” accounts of a show’s characters, a Twitter practice not unusual for fans of a program (Bore and Hickman, 2013).

With viewers watching shows at different times, the possibility exists — particularly for the asynchronous model — that second screen activity could contain spoilers, or “advance information of what will happen in the plot” [14]. While researchers are exploring the possibility of a content-oriented filter that acts as a mute button on Twitter (Golbeck, 2012), it is more likely that as one’s affinity for a text increases, so does the likelihood that he or she will simply avoid Twitter entirely until fully caught up with the show. For fans of a show, simply discussing a narrative element may spoil it for others. For anti-fans of a show, spoiling key plot twists for fans or non-fans would require very little effort. In this study, however, the impact of spoilers appeared to be negligible for both broadcast models: on average, 0.77 percent of all tweets for a show contained the word “spoiler” or “spoilers.” Of course, the number of those tweets that actually spoil is fewer than that, as some of those tweets do not contain actual spoilers but prohibitions of them (e.g., “Hey stop talking about House of Cards #spoilers!”). Future research should examine these spoiler tweets in greater detail and determine their relationship to the nature of and concern over spoiling in general. When Jason Mittell and Jonathan Gray surveyed Lost fans, a theme they noticed is that while some viewers avoided spoilers, many actually sought them out. As they note: “for spoiler fans, having already discovered what will happen freed them to concentrate on the formal pleasures of innovative narration and inventive presentation” [15]. Subsequent studies need to examine the role of spoilers in negotiating mediated spaces between fans, anti-fans, and non-fans for different genres and broadcast models.

Despite the different patterns of Twitter activity for the appointment and asynchronous broadcast models, all four shows had relatively similar co-connected viewing (connected viewers co-viewing with each other, RQ2). This difference of only two percentage points between the all four shows is remarkable given the differences in genre (comedy and drama), program length (30 and 60 minutes), and broadcast model (appointment and streaming). One might think that the appointment shows — with their live audiences — would precipitate more friends tweeting @ each other in real time as the show aired. Yet, contrary to research that found different levels of co-connected viewing for different genres (Doughty, et al., 2012), the data in this study revealed that people practice co-connected viewing (tweeting @ their friends and using a #) just as much when binging a show on their own time, even over the course of a few days, as when watching a single episode of a live show at the network’s appointed time, in one evening.

While there did not appear to be a large overall difference in engagement level from active and casual Twitter users (RQ3) for either broadcast model, the discrepancies bear mentioning, as they suggest that more active Twitter users are also more active with other Internet platforms such as Netflix. The fact that House of Cards had the highest proportion of its Twitter activity come from more active Twitter users (almost 60 percent) should perhaps not be surprising, given that the show received heavy promotion from Netflix in the days and weeks leading up to its release. The more time one spent online, the greater the likelihood that one would be exposed to advertisements for and discussions about House of Cards’ third season.

Parks and Recreation, on the other hand, saw the highest proportion of its Twitter engagement (51.8 percent) come from casual users. It was the only show where casual Twitter users contributed more than half of all the tweets. This may be attributed to the fact that it was the finale of not only a season, but the entire series itself, and thus more fans felt the need to participate in the final celebratory dialogue of a popular show. Indeed, the spike of #ParksandRec tweets during the show’s broadcast included the highest percentage of retweets (46 percent) by a wide margin (#DowntonAbbey was next with 34 percent). Many fans retweeted tweets from actors in the show — most of whom are younger adults and very active on social media — as well as other celebrities who are also fans of the show, whose tweets attracted even more attention.

As previously noted, research involving Twitter hashtags has its limitations (Bruns and Burgess, 2012). Each television program has additional hashtags employed by its fans on Twitter (e.g., “ParksandRecFinale”, “ParksFarewell”). In addition, not all users utilize hashtags when discussing events and programs. As such, it cannot be assumed that a sample of tweets based on a hashtag is a complete representation of all Twitter activity concerning that subject.

Furthermore, it is possible for anti-fans to co-opt a show’s hashtag in order to influence trending topics against the wishes of fans (Recuero, et al., 2012), so there is no guarantee that tweets using a show’s hashtag are positive. Yet even anti-fans’ ostensibly pejorative use of a show’s hashtag might inadvertently contribute to its trending on Twitter. Future research should employ more hashtags and search terms in an attempt to build a more complete sample.

In addition, while the use of @replies as an indicator of social engagement does provide valuable insight into how users are engaging in conversations on Twitter, it does not necessarily constitute a complete picture of social activity. Users may reply to any tweet, even one that is not an explicit initiation of a conversation; similarly, if a reply does not include the appropriate hashtag, it will not be captured in the overall sample. Future research should specifically examine the actual conversations that do or do not take place after an initial tweet, including tweets that are not directly social in nature.

While this study’s sample of four programs was unique in that all four shows were released within the same two-week time period, eliminating possible seasonal differences in viewing, future research could include more programs as well as a longer sampling period. It would also be helpful to sample entire seasons of television programs along with a similar timeline for all-at-once, streaming programs. For example, how long would it take for #HouseofCards activity to drop below one thousand tweets per hour for a full 24 hour period? While this study focused specifically on scripted dramas and comedies, future studies might explore different genres of programming, including live events, such as award shows and sporting events, and reality programming.

While more research is needed, this study suggests that traditional appointment viewing does not create an online social space any more than asynchronous, streaming viewing. Streaming viewing merely distributes the articulated social space through time (diachronic) over a longer period than a single temporal (synchronic) event. Although this study did not look at the course of an entire season, it is expected that the appointment model would produce spikes of Twitter activity each week, while the asynchronous, all-at-once model will produce a more expanded surge in activity that gradually declines as all interested fans eventually get around to watching the show.

The benefit for advertisers with weekly appointment viewing is the regular brand recognition over a period of months, but the “cost” consumers pay in ceding control of their viewing experience may offset this potential gain. However, an all-at-once release model may be viable for a traditional, commercial advertising approach. Perhaps a better comparison would be between all-at-once model releases and one-time events such as Oscars, Grammys, or the Super Bowl. Advertisers regularly pay high prices for ads during these events, which may generate a burst of Twitter activity and attention. A popular and well-promoted show released using an all-at-once model could generate a similar burst of attention, so a market is certainly possible where advertisers find novel ways to sponsor asynchronous streaming content. To some degree, this market already exists in the form of “limited commercial interruption” television. When Hulu, a rival of Netflix, gave users the option to choose between normal commercial breaks and one longer commercial up front, customers overwhelmingly chose the latter (Kelly, 2009). The potentialities for a mutually satisfactory equilibrium between producers, advertisers, and consumers continue to evolve with each new platform, technology, and broadcast model.

Additionally, because they generate such a rapid increase in Twitter activity, appointment shows may be more likely to “trend” on Twitter and thus draw extra attention. This attention may result in more fans initially and more advertising dollars later on. When topics are trending they receive special mention and increased attention within the Twitter platform. However, Twitter is not specific on what exactly is required for a #topic to be trending, as they discuss on their FAQ page:

Trends are determined by an algorithm and, by default, are tailored for you based on who you follow and your location. This algorithm identifies topics that are popular now, rather than topics that have been popular for a while or on a daily basis, to help you discover the hottest emerging topics of discussion on Twitter that matter most to you.

The algorithm used by Twitter had declared topics to be trending both before and after their peak of activity (O’Neill, 2013), so it is unclear what threshold (of total Tweets or Tweets per minute) must be reached to achieve this notoriety. On the other hand, if the algorithm recognizes sustained activity instead of a single spike, then Twitter activity from asynchronous streaming shows may be more likely to trend. The water-cooler conversation of the past is now being reconfigured (Matrix, 2014): the dialogue is no longer limited to the day after the program aired, only with the co-workers or friends who saw it; now the discussion continues as long as the content is available and there are people who want to watch and discuss it. As such, an asynchronous, all-at-once model may be very beneficial for producers hoping to create an ongoing conversation that drives audiences to their programs.

Finally, producers whose shows are able to turn casual viewers or non-fans into fans who then engage in connected viewing will reap the benefits of a “paratext” that is generated by the Twitter content those fans create. As Jonathan Gray explains, “If we imagine the triumvirate of Text, Audience, and Industry as the Big Three of media practice, then paratexts fill the space between them, conditioning passages and trajectories that criss-cross the mediascape, and variously negotiating or determining interactions among the three.” [16] A Twitter user unaware of a primary text (e.g., House of Cards) might still be directed to it by stumbling onto the paratext (#HouseOfCards dialogue). Because it consists of individuals talking to one another, co-connected viewing may give a paratext the additional quality of appearing sociable or popular, providing additional impetus for unfamiliar viewers to check out a show. Paratextual cultivation by fans, consumption by non-fans, and attempted undermining by anti-fans are all important phenomena that continue to evolve in a dynamic media landscape and warrant further study.

 

++++++++++

Conclusion

While patterns of viewing differ dramatically based on its release model, the levels of co-connected viewing (connected co-viewing) are strikingly similar. Our data suggest that, by tweeting at their friends using a television program’s designated hashtag, co-connected viewers will account for just over 10 percent of all Twitter activity for that show. Whether watching a single episode at a time appointed by the network or “binging” on an entire season over several days, the mediated social component of a viewer’s experience on Twitter remains fairly constant, for comedies and dramas. The fact that both models generate very similar levels of social activity, though their patterns vary, is consequential for both advertisers and consumers. It is important to note that Netflix programs do not include advertisements but use a subscription model to generate revenue. However, both models depend on viewers: more viewers typically equate to more subscriptions for Netflix and more advertising revenue for the networks.

The good news for content creators and marketers is that consumers proved more than willing to interact online with the content creators (Laursen and Sandvik, 2014) of the shows they like. As previously mentioned, NBC’s Twitter account for Parks and Recreation tweeted a photo of Ron Swanson that fans retweeted enough to temporarily outnumber original #ParksandRec tweets. The thought of this phenomenon — fans echoing a network’s public-facing statements — occurring before the advent of social media is absurd. Yet now, with the ubiquity of social networking sites and their easy currency of endorsements (a “like”, a retweet, or a follower), brands and producers who can find creative, entertaining ways to engage their viewers will be rewarded.

For fans, new technologies and platforms present novel ways to consume, discuss, and interact with the television shows they enjoy. When he was awarded a knighthood for his work in 2003, Tim Berners-Lee, inventor of the World Wide Web, said of its intended purpose, “The original idea of the Web was that it should be a collaborative space where you can communicate through sharing information.” In the case of co-connected viewing, this collaborative space in one in which not only information is shared, but also the laughter, joys, and fears between friends and the shows they watch. Furthermore, the sharing of these emotions and opinions by fans creates a paratext that might serve to engage other fans (potentially making new friends) or draw non-fans to the primary text (potentially making new fans).

In the past, the conversation between friends co-viewing television in a room together might have facilitated various social gratifications (Katz, et al., 1973; Rubin, 1983) yet the experience was bound by (inhabiting the same) space and time. With the advent of co-connected viewing, the conversation between friends now occurs from different points around the globe and through different moments in time. Furthermore, the conversation is not only between them, but all and any other viewers who might enjoy the show while using a second screen. Henry Jenkins notes the importance of this phenomenon: “there is a new kind of cultural power emerging as fans bond together within larger knowledge communities, pool their information, shape each other’s opinions, and develop a greater self-consciousness about their shared agendas and common interests” [17]. What an exciting age in which we live, where a fan can potentially affect how others view his or her favorite show! Whether an individual is connecting with all other viewers of a show in real time or just one friend in particular during commercial breaks, co-connected viewing is an evolving and dynamic force that will continue to influence television production and consumption of all genres and broadcast formats. End of article

 

About the authors

Matthew Pittman is a Ph.D. student in the School of Journalism and Communication at the University of Oregon.
E-mail: mpittman [at] uoregon [dot] edu

Alec C. Tefertiller is a Ph.D. student in the School of Journalism and Communication at the University of Oregon.
E-mail: alect [at] uoregon [dot] edu

 

Notes

1. Smith, 2012, p. 7.

2. Flomenbaum, 2015, para. 4.

3. Roettgers, 2012, para. 3.

4. Wolk, 2015, para. 3.

5. Holt and Sanson, 2014, p. 1.

6. Lee and Andrejevic, 2014, p. 47.

7. Ji and Raney, 2015, p. 6.

8. Schirra, et al., 2014, p. 2,447.

9. Xu and Yan, 2011, p. 201.

10. Baym, 2007, p. 1.

11. Gray, 2003, p. 74.

12. Gray, 2003, p. 70.

13. Bruns and Burgess, 2012, p. 804.

14. Gray, 2010, p. 20.

15. Gray, 2010, p. 150.

16. Gray, 2010, p. 23.

17. Gray, et al., 2007, p. 363.

 

References

D.F. Alters, 2007. “The other side of fandom: Anti-fans, non-fans, and the hurts of history,” In: J. Gray, C. Sandvoss, and C. Lee Harrington (editors). Fandom: Identities and communities in a mediated world. New York: New York University Press, pp. 344–356.

N. Baym, 2007. “The new shape of online community: The example of Swedish independent music fandom,” First Monday, volume 12, number 8, at http://firstmonday.org/article/view/1978/1853, accessed 18 June 2015.

J.G. Blumler, 1979. “The role of theory in uses and gratifications studies,” Communication Research, volume 6, number 1, pp. 9–36.
doi: http://dx.doi.org/10.1177/009365027900600102, accessed 18 June 2015.

I.-L.K. Bore and J. Hickman, 2013. “Studying fan activities on Twitter: Reflections on methodological issues emerging from a case study on The West Wing fandom,” First Monday, volume 18, number 9, at http://firstmonday.org/article/view/4268/3740, accessed 18 June 2015.
doi: http://dx.doi.org/10.5210/fm.v18i9.4268, accessed 18 June 2015.

A. Bruns and J. Burgess, 2012. “Researching news discussion on Twitter: New methodologies,” Journalism Studies, volume 13, numbers 5–6, pp. 801–814.
doi: http://dx.doi.org/10.1080/1461670X.2012.664428, accessed 18 June 2015.

G.M. Chen, 2011. “Tweet this: A uses and gratifications perspective on how active Twitter use gratifies a need to connect with others,” Computers in Human Behavior, volume 27, number 2, pp. 755–762.
doi: http://dx.doi.org/10.1016/j.chb.2010.10.023, accessed 18 June 2015.

E.L. Cohen and A.L. Lancaster, 2014. “Individual differences in in-person and social media television coviewing: The role of emotional contagion, need to belong, and coviewing orientation,” Cyberpsychology, Behavior, and Social Networking, volume 17, number 8, pp. 512–518.
doi: http://dx.doi.org/10.1089/cyber.2013.0484, accessed 18 June 2015.

N.A. Diakopoulos and D.A. Shamma, 2010. “Characterizing debate performance via aggregated twitter sentiment,” CHI ’10: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 1,195–1,198.
doi: http://dx.doi.org/10.1145/1753326.1753504, accessed 18 June 2015.

R. Deller, 2011. “Twittering on: Audience research and participation using Twitter,” Participations, volume 8, number 1, pp. 216–245, at http://www.participations.org/Volume%208/Issue%201/PDF/deller.pdf, accessed 18 June 2015.

M. Doughty, S. Lawson, C. Linehan, D. Rowland, and L. Bennett, 2014. “Disinhibited abuse of othered communities by second-screening audiences,” TVX '14: Proceedings of the 2014 ACM International Conference on Interactive Experiences for TV and Online Video, pp. 55–62.
doi: http://dx.doi.org/10.1145/2602299.2602311, accessed 18 June 2015.

M. Doughty, D. Rowland, and S. Lawson, 2012. “Who is on your sofa? TV audience communities and second screening social networks,” EuroiTV ’12: Proceedings of the 10th European Conference on Interactive TV and Video, pp. 79–86.
doi: http://dx.doi.org/10.1145/2325616.2325635, accessed 18 June 2015.

M. Doughty, D. Rowland, and S. Lawson, 2011. “Co-viewing live TV with digital backchannel streams,” EuroITV ’11: Proceedings of the 9th International Interactive Conference on Interactive Television, pp. 141–144.
doi: http://dx.doi.org/10.1145/2000119.2000147, accessed 18 June 2015.

A. Flomenbaum, 2015. “Thanks, Nielsen, for that Twitter study — But it’s time we ‘total it up’ with you,” LostRemote (12 March), at http://www.adweek.com/lostremote/thanks-nielsen-for-that-twitter-study-but-its-time-we-total-it-up-with-you/50907, accessed 1 May 2015.

F. Giglietto and D. Selva, 2014. “Second screen and participation: A content analysis on a full season dataset of tweets,” Journal of Communication, volume 64, number 2, pp. 260–277.
doi: http://dx.doi.org/10.1111/jcom.12085, accessed 18 June 2015.

J. Gray, 2010. Show sold separately: Promos, spoilers, and other media paratexts. New York: New York University Press.

J. Gray, 2003. “New audiences, new textualities: Anti-fans and non-fans,” International Journal of Cultural Studies, volume 6, number 1, pp. 64–81.
doi: http://dx.doi.org/10.1177/1367877903006001004, accessed 18 June 2015.

J. Gray, C. Sandvoss, and C. Lee Harrington (editors), 2007. Fandom: Identities and communities in a mediated world. New York: New York University Press.

P. Haridakis and G. Hanson, 2009. “Social interaction and co-viewing with YouTube: Blending mass communication reception and social connection,” Journal of Broadcasting & Electronic Media, volume 53, number 2, pp. 317–335.
doi: http://dx.doi.org/10.1080/08838150902908270, accessed 18 June 2015.

S. Harman and B. Jones, 2013. “Fifty shades of ghey: Snark fandom and the figure of the antifan,” Sexualities, volume 16, number 8, pp. 951–968.
doi: http://dx.doi.org/10.1177/1363460713508887, accessed 18 June 2015.

S. Harrington, T. Highfield, and A. Bruns, 2013. “More than a backchannel: Twitter and television,” Participations, volume 10, number 1, pp. 405–409, at http://www.participations.org/Volume%2010/Issue%201/30%20Harrington%20et%20al%2010.1.pdf, accessed 18 June 2015.

I. Himelboim, 2014. “Political television hosts on Twitter: Examining patterns of interconnectivity and self-exposure in Twitter political talk networks,” Journal of Broadcasting & Electronic Media, volume 58, number 1, pp. 76–96.
doi: http://dx.doi.org/10.1080/08838151.2013.875017, accessed 18 June 2015.

J. Holt and K. Sanson (editors) 2014. Connected viewing: Selling, sharing, & streaming media in a digital age. New York: Routledge.

A. Java, X. Song, T. Finin, and B. Tseng, 2007. “Why we twitter: Understanding microblogging usage and communities,” WebKDD/SNA-KDD ’07: Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 Workshop on Web Mining and Social Network Analysis, pp. 56–65.
doi: http://dx.doi.org/10.1145/1348549.1348556, accessed 18 June 2015.

Q. Ji and A.A. Raney, 2015. “Morally judging entertainment: A case study of live tweeting during Downton Abbey,” Media Psychology, volume 18, number 2, pp. 221–242.
doi: http://dx.doi.org/10.1080/15213269.2014.956939, accessed 18 June 2015.

E. Katz, J.G. Blumler, and M. Gurevitch, 1974. “Utilization of mass communication by the individual,” In: J.G. Blumler and E. Katz (editors). The uses of mass communication: Current perspectives on gratifications research. Beverly Hills, Calif.: Sage, pp. 19–32.

E. Katz, M. Gurevitch, and H. Haas, 1973. “On the use of the mass media for important things,” American Sociological Review, volume 38, number 2, pp. 164–181.
doi: http://dx.doi.org/10.2307/2094393, accessed 18 June 2015.

J.-P. Kelly, 2009. “Brought to you with limited commercial interruption: The temporalities of online TV,” In Media Res (18 February), at http://mediacommons.futureofthebook.org/imr/2009/02/17/brought-you-limited-commercial-interruption-temporalities-online-tv, accessed 18 June 2015.

B. Krishnamurthy, P. Gill, and M. Arlitt, 2008. “A few chirps about Twitter,” WOSN ’08: Proceedings of the First Workshop on Online Social Networks, pp. 19–24.
doi: http://dx.doi.org/10.1145/1397735.1397741, accessed 18 June 2015.

D. Laursen and K. Sandvik, 2014. “Talking with TV shows: Simultaneous conversations between users and producers in the second-screen television production Voice,” Northern Lights, volume 12, number 1, pp. 141–160.
doi: http://dx.doi.org/10.1386/nl.12.1.141_1, accessed 18 June 2015.

H.J. Lee and M. Andrejevic, 2014. “Second-screen theory: From the democratic surround to the digital enclosure,” In: J. Holt and K. Sanson (editors). Connected viewing: Selling, sharing, & streaming media in a digital age. New York: Routledge, pp. 40–61.

D. Lieberman, 2015. “Networks still nagged by weak ad growth as up front market approaches,” Deadline (5 March), at http://deadline.com/2015/03/tv-advertising-sales-growth-weak-upfront-market-1201386655/, accessed 18 June 2015.

M. Lochrie and P. Coulton, 2011. “Mobile phones as second screen for TV, enabling inter-audience interaction,” ACE ’11: Proceedings of the 8th International Conference on Advances in Computer Entertainment Technology, article number 73.
doi: http://dx.doi.org/10.1145/2071423.2071513, accessed 18 June 2015.

S. Matrix, 2014. “The Netflix effect: Teens, binge watching, and on-demand digital media trends,” Jeunesse, volume 6, number 1, pp. 119–138.
doi: http://dx.doi.org/10.1353/jeu.2014.0002, accessed 18 June 2015.

M. Morrison and D.M. Krugman, 2001. “A look at mass and computer mediated technologies: Understanding the roles of television and computers in the home,” Journal of Broadcasting & Electronic Media, volume 45, number 1, pp. 135–161.
doi: http://dx.doi.org/10.1207/s15506878jobem4501_9, accessed 18 June 2015.

Nielsen Social, 2014. “Maximum impact through social TV,” at http://www.nielsensocial.com/data-and-tools/, accessed 18 June 2015.

K. O’Keefe, 2015. “Why Netflix is about to bury the competition — and blow up bigger than ever,” Arts.Mic (6 March), at http://mic.com/articles/111994/why-netflix-s-quality-is-beating-network-quantity, accessed 18 June 2015.

J. O’Neill, 2013. “How many tweets make a trend?” Cision (28 August), at http://www.cision.com/uk/blog/how-many-tweets-make-a-trend/, accessed 18 June 2015.

P. Palmgreen, P.L. Cook, J.G. Harvill, and D.M. Helm, 1988. “The motivational framework of moviegoing: Uses and avoidances of theatrical films,” In: B.A. Austin (editor). Current research in film: Audiences, economics, and law, volume 4. Norwood, N.J.: Ablex, pp. 1–23.

T. Peled and E. Katz, 1974. “Media functions in wartime: The Israel home front in October 1973,” In: J.G. Blumler and E. Katz (editors). The uses of mass communication: Current perspectives on gratifications research. Beverly Hills, Calif.: Sage, pp. 49–69.

R. Recuero, A. Amaral, and C. Monteiro, 2012. “Fandoms, trending topics and social capital in Twitter,” Selected Papers of Internet Research, ir 13, at http://spir.aoir.org/index.php/spir/article/view/7, accessed 18 June 2015.

J. Roettgers, 2012. “Binge viewers rejoice: Netflix releases Lilyhammer,” Gigaom (6 February), at https://gigaom.com/2012/02/06/netflix-original-content-binge-viewing/, accessed 18 June 2015.

A.M. Rubin, 1993. “Audience activity and media use,” Communication Monographs, volume 60, number 1, pp. 98–105.
doi: http://dx.doi.org/10.1080/03637759309376300, accessed 18 June 2015.

A.M. Rubin, 1983. “Television uses and gratifications: The interactions of viewing patterns and motivations,” Journal of Broadcasting, volume 27, number 1, pp. 37–51.
doi: http://dx.doi.org/10.1080/08838158309386471, accessed 18 June 2015.

S. Schirra, H. Sun, and F. Bentley, 2014. “Together alone: Motivations for live-tweeting a television series,” CHI ’14: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 2,441–2,450.
doi: http://dx.doi.org/10.1145/2556288.2557070, accessed 18 June 2015.

A. Smith, 2012. “The rise of the ‘connected viewer’,” Pew Research Center (17 July), at http://www.pewinternet.org/2012/07/17/the-rise-of-the-connected-viewer/, accessed 18 June 2015.

E. Tussey, 2014. “Connected viewing on the second screen: The limitations of the living room,” In: J. Holt and K. Sanson (editors) 2014. Connected viewing: Selling, sharing, & streaming media in a digital age. New York: Routledge, pp. 202–216.

S. Wakamiya, R. Lee, and K. Sumiya, 2011. “Towards better TV viewing rates: Exploiting crowd’s media life logs over Twitter for TV rating,” ICUIMC ’11: Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, article number 39.
doi: http://dx.doi.org/10.1145/1968613.1968661, accessed 18 June 2015.

D.Y. Wohn and E.-K. Na, 2011. “Tweeting about TV: Sharing television viewing experiences via social media message streams,” First Monday, volume 16, number 3, at http://firstmonday.org/article/view/3368/2779, accessed 18 June 2015.

A. Wolk, 2015. “When Netflix and other on-demand services killed the TV ad golden goose,” Guardian (5 February), at http://www.theguardian.com/media-network/2015/feb/05/netflix-subscription-services-television-ad-revenues, accessed 18 June 2015.

H. Xu and R.-N. Yan, 2011. “Feeling connected via television viewing: Exploring the scale and its correlates,” Communication Studies, volume 62, number 2, pp 186–206.
doi: http://dx.doi.org/10.1080/10510974.2010.550380, accessed 18 June 2015.

 


Editorial history

Received 8 April 2015; revised 28 May 2015; accepted 25 June 2015.


Creative Commons License
“With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television models” by Matthew Pittman and Alec Tefertiller is licensed under a Creative Commons Attribution 4.0 International License.

With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television models
by Matthew Pittman and Alec C. Tefertiller.
First Monday, Volume 20, Number 7 - 6 July 2015
http://firstmonday.org/ojs/index.php/fm/article/view/5935/4663
doi: http://dx.doi.org/10.5210/fm.v20i7.5935





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.