Individual Differences in In-Person and Social Media Television Coviewing: The Role of Emotional Contagion,Need to Belong,and Coviewing Orientation

Abstract

The popularity of social media television coviewing is growing, but little is known about why people engage in these connected viewing experiences or how they differ from in-person coviewing. This study investigated how engaging in in-person and social media coviewing is predicted by individual differences: emotional contagion, need to belong, and three dimensions of a coviewing orientation scale created for this research (need for company, need for solitude, and audience monitoring). On Amazon Mechanical Turk, 451 people were recruited for an online survey. The mean age was 34.64 years (SD=13.16 years), and 52% of the sample was female. Emotional contagion predicted in-person coviewing only. Need to belong predicted several mediated co-viewing activities. Need for solitude negatively predicted in-person coviewing, but need for company positively predicted in-person coviewing. Results indicate that viewers have different motivations for engaging in various coviewing activities. Findings also suggest that social media coviewing can provide valuable opportunities for social connection among viewers who watch television in physical solitude.

Introduction

Even before television sets became a living room fixture, television facilitated social connection by bringing people together physically to share their viewing experiences. Increasingly, viewers are also using social media to connect with others who are physically remote during viewing. These mediated “social TV” practices permit television viewers to connect with, monitor, and express themselves to entire communities during viewing.^1–3 A survey found that 11% of U.S. mobile phone users use smartphones to monitor other people's reactions to programs online, 11% post program-related comments online, and nearly a quarter of mobile phone users exchange text messages with other viewers.⁴ Users of social networking sites report connected viewing experiences in even greater numbers. A recent report shows that close to a quarter of adults between 18 and 34 years of age use some form of social network technology to post a comment about a TV show.⁵

The growing popularity of social media coviewing might have positive psychological implications for audience members who stand to gain more control over how and when they interact with coviewers. The psychological and relational benefits of television coviewing with physically present others are well documented. Television viewing in the presence of another provides viewers opportunities to converse,^6–10 and allows them to provide comfort to one another^11–12 and practice social identity roles.⁸ But notably, even when viewers do not interact, viewing with others helps people feel connected by virtue of being together.^8,9,13,14 Some preliminary evidence suggests that computer-mediated coviewing can mimic these in-person coviewing experiences. Viewers who watch televised events in separate locations but still maintain awareness of each other with technology experience social presence, and report feeling increased attraction toward their coviewers.^15,16 Interestingly, Wohn and Na³ found coviewing experiences via Twitter usually occur without communication between viewers, suggesting that digitally connected coviewers do not seek out these experiences to connect directly with anyone in particular and converse about the program. Instead, we suggest that people are drawn to coviewing on Twitter because these experiences offer a sense of belonging with no interaction strings attached. Social media coviewing may be desirable for some because it gives viewers the opportunity to observe, feel part of, and share experiences with a much larger community without necessitating any direct socializing.

Collectively, this research suggests that both in-person and mediated coviewing can be a rewarding social experience for audiences, but little is known about what draws people to share media experiences with others in either context. Uses and gratifications research has demonstrated that viewers self-report a variety of social uses of television, such as being among company or having something to talk about,¹⁷ but more fundamental differences in how people are oriented to process and respond to social situations could affect whether individuals are drawn to different types of connected viewing experiences. To shed light on dispositional motivations for connected viewing experiences, this research examines three individual differences that may account for similarities and differences of coviewing in nonmediated and mediated contexts: emotional contagion, people's susceptibility to “catching” other's emotions; need to belong, the importance people place on feeling socially accepted; and coviewing orientation, the extent to which viewers are predisposed to use television watching as a social activity.

Emotional contagion

Emotional contagion refers to the tendency to pick up on and experience or express another person's mood or emotions. People with this emotional susceptibility tend to be better at attending to and interpreting others' emotional expression, and they see themselves as being interconnected with others. Those who exhibit more emotional contagion also mimic others' nonverbal expressions and tend to have conscious emotional experiences that are susceptible to feedback.¹⁸

Coviewing is theorized to heighten affective reactions to content because the presence of others permits emotions to spread contagiously.^19,20 A handful of scholars suggest that this process increases enjoyment of the viewing experience.^21–23 Furthermore, some limited experimental support found that emotional contagion is correlated with enjoyment, and people who view programs together have more synchronized enjoyment experiences.^22,23

The link between emotional contagion and coviewing suggests that this variable might encourage people to watch television with others physically present, but it is less clear if emotional contagion will influence social media coviewing. Social information processing theory²⁴ contends that the reduction in nonverbal cues of emotion available in computer-mediated environments can slow processes that facilitate interpersonal relationship development. Hence, it is understood that digitally connected coviewers can express emotional reactions to television content,³ but it is unknown if these digital displays of emotion are capable of the same contagion effects found in nonmediated, nonverbal cue-rich environments. Given that past research suggests that emotional contagion is related to media enjoyment in face-to-face contexts, we suspect that people's susceptibility to catching others' emotions predicts nonmediated coviewing. However, it is less clear if emotional contagion will play a similar role in whether or not people engage in mediated coviewing. Accordingly, the following hypothesis and research question were developed:

H1: Emotional contagion will positively predict frequency of nonmediated coviewing.

RQ1: How will emotional contagion be related to frequency of mediated coviewing?

Need to belong

Gardner et al. argue that people are hardwired to monitor their social environment for external cues that help them assess how well they fit in with others.²⁵ This social monitoring system is activated by social events. For instance, these researchers found that after being socially rejected, people were able to recall information related to social interaction more accurately, suggesting that they were prompted to engage in social monitoring. But they also posit that differences in how active a social monitoring system is can be a matter of individual difference in the need to belong. Rooted in humans' fundamental desire to form attachments with others,²⁶ those with a greater need to belong have a stronger motivation to feel socially accepted and included. Consequently, individuals with a high need to belong tend to be more vigilant about monitoring their social surroundings,²⁷ a disposition that should manifest when they are given opportunities to monitor responses from other audience members in coviewing contexts. Possibly the best means of satisfying a need for belonging is through regular, positive interaction with others.²⁸ As such, individuals with a greater need to belong may appreciate television viewing with physically co-present others. Need to belong is also predictive of social networking site^29,30 and mobile phone use,³¹ suggesting that this individual difference might also predict mediated social TV experiences, in which people text, instant message, or post to or monitor social networking sites to express their views about television programming. Accordingly, two hypotheses were proposed:

H2: Need to belong will positively predict nonmediated coviewing.

H3: Need to belong will positively predict computer-mediated coviewing.

Coviewing orientation

We propose that there are individual differences in how people are predisposed to consume traditional mass communication media, such as newspapers, radio, and—the focus of the current project—television. These media have traditionally been conceived as being nonsocial,³² perhaps because they are not interactive like new media technologies, which provide a two-way channel for users to communicate with each other (e.g., social networking sites). However, these types of mass communication can provide a catalyst for social interaction when people consume the media together or socialize about media content. Still, despite the social opportunities that media such as television offer, people should vary in how prone they are to view and use television watching as a social activity. For instance, some viewers may appreciate the presence of others as a supplementation to the overall viewing experience,²³ while others may find the emotional expressions of others annoying, distracting, or even distressing. Similarly, although multitasking with media (e.g., texting and watching television) is emotionally gratifying to many,³³ individuals who prefer to focus their full attention on their television programs should find these social activities more problematic. To profile better the extent to which individuals seek out or benefit from connected viewing experiences, we propose considering individuals' coviewing orientation, a term we conceptualize as referring to how inclined they are to share media experiences with others and experience television watching as a social activity. Because no other scales that measure social dispositions to different types of media could be identified, a coviewing orientation scale was devised for this study.

Presumably, coviewing orientation positively predicts both types of connected viewing experiences— nonmediated and social media coviewing situations. Those who have a tendency to appreciate being in social contexts during television viewing should be unlikely to discriminate between whether the social contact is mediated or not. Likewise, those who have a preference for more solitary viewing situations should be unlikely to seek out company in any context. As such, the following hypotheses were proposed:

H4: Coviewing orientation will positively predict nonmediated coviewing.

H5: Coviewing orientation will positively predict computer-mediated coviewing.

Method

Data were collected with a web-based questionnaire posted to Amazon Mechanical Turk (MTurk), a crowdsourcing Web site that connects people needing workers to complete online tasks with individuals wanting to be hired to complete the tasks. Workers registered on MTurk browse for posted tasks, and receive monetary compensation after they complete the tasks and the work requester approves their work. In the past few years, social scientists have begun using MTurk as a convenient and inexpensive way to recruit diverse groups of participants for online survey and experimental research. Although MTurk provides a convenience sample of respondents who actively seek out opportunities to participate in online research for low wages, a growing number of studies on MTurk find it provides diverse samples that are more representative of the general population than Internet samples collected via other means, and that psychometric properties and effect sizes in findings with MTurk samples match those found with more traditional subject pools.^34–40 By virtue of their use of MTurk, we also presume that these workers are more familiar with technology than the general population, permitting us to capture a wider range of computer-mediated coviewing activities.

An invitation to participate in the study was posted to MTurk for 2 weeks in October, 2012. A total of 451 workers from 33 states completed the online survey. Each worker received 35 cents as compensation for responding to the questionnaire. Respondents' ages ranged from 18 to 71 years (M=34.64, SD=13.16). More females (52.6%) than males (47.4%) completed the questionnaire. Most respondents (74%) identified themselves as white/Caucasian. The highest level of education reported was 17.8% with a high school diploma or less, 33.7% with some college, 36.7% with a college degree, and 11.7% with a graduate or professional degree.

Measures

Dependent measures: frequency of connected viewing experiences

Single items were used to measure each of the five connected viewing frequency variables. Respondents indicated the frequency they performed four different mediated coviewing behaviors on a scale ranging from 0=“never” to 5=“most of the time.” As a measure of in-person coviewing, respondents were asked to report how often they watched TV with another person present (M=3.35, SD=1.02). Respondents also indicated the frequency in which they sent text messages (M=1.90, SD=1.01) or instant messages (M=1.59, SD=0.94), shared content, or posted status updates on social networking sites (M=1.99, SD=1.08) related to the show they were watching, or monitored social networking sites (M=2.08, SD=1.13) for what other people were posting about the show they were watching.

Independent measures: individual differences

For independent variable measures, respondents indicated their agreement to statements on 7-point scales ranging from 0=“disagree strongly” to 6=“agree strongly.” Doherty's 15-item Emotional Contagion Scale measured respondents' tendency to pick up others' emotions (e.g., “Being with a happy person picks me up when I'm feeling down”; α=0.89; M=3.01, SD=1.05).⁴¹ The 10-item Need to Belong Scale assessed participants' individual differences in motivation to form attachments (e.g., “My feelings are easily hurt when I feel that others do not accept me”; α=0.85; M=3.65, SD=1.00).⁴² Finally, 12 items were developed as indicators of viewers' orientation to coviewing situations. Individual items were designed as a one-dimensional gauge of how much respondents prefer to attend to others socially while viewing. A principal axis analysis with Varimax rotation was used to identify and remove items that were poor indicators. One item (“When I watch television, I often become so absorbed that I don't notice other people around”) was dropped because it did not load on any factor. Notably, although coviewing orientation was conceived as a one-dimensional measure of how predisposed respondents are to consume television socially, three interpretable factors emerged when the analysis was performed again. Examination of the subscales suggests that they tap into different dimensions of coviewing orientation by gauging people's tendency to monitor others socially while coviewing, their preference for being alone during viewing, and their proclivity to have the company of others during viewing. Each dimension has a high level of face validity. The dimensions were labeled as follows: audience monitoring (four items, α=0.79; M=3.35, SD=1.34), need for solitude (three items, α=0.76; M=3.63, SD=1.31), and need for company (four items, α=0.77; M=2.68, SD=1.18). Scale items are displayed in Table 1. All factor loadings were 0.53 or higher, and no cross loadings exceeded 0.23.

Table 1.

Coviewing Orientation

Exploratory factor structure	E	H	H
Audience monitoring
I enjoy watching other people's reactions when we watch television shows together.	0.76	0.03	0.19
I like watching television with people who express their emotions in response to the shows.	0.76	0.06	0.24
When I watch television shows, I appreciate the opportunity to talk to other people about what's going on.	0.56	0.16	0.27
I feel frustrated when people I am watching television with don't seem to enjoy shows as much as I do.	0.55	−0.38	0.23
Need for solitude
Having people around ruins the television viewing experience.	0.05	0.78	−0.07
It is better to watch television shows alone.	0.09	0.69	0.19
It's hard for me to focus on television shows when other people are around.	−0.01	0.66	0.05
Need for company
Television is better as a social event.	0.26	0.12	0.75
I feel that I get more out of television shows when I watch them with other people.	0.34	0.17	0.66
I like to share my television time with other people.	0.36	0.32	0.54
I often feel lonely when I watch television shows by myself.	0.12	−0.19	0.54
Cronbach's alpha	0.79	0.76	0.77

Results

Five hierarchal regression analyses were conducted to predict the frequency of each coviewing behavior: nonmediated coviewing and types of coviewing through social media (text, instant message, social networking sites, and social networking site monitoring). Table 2 displays the results for each of the analyses. Age was unrelated to nonmediated coviewing but strongly negatively correlated to the frequency of each type of technology-mediated coviewing behaviors. As age increased, respondents reported engaging in less text coviewing (b=−0.03, p<0.001), instant message coviewing (b=−0.02, p<0.001), social networking site coviewing (b=−0.03, p<0.001), and social networking site monitoring (b=−0.03, p<0.001). Sex positively predicted frequency of all types of coviewing except for instant message coviewing. Compared to males, females were engaged in more nonmediated coviewing (b=0.18, p=0.03), text coviewing (b=0.19, p=0.04), social networking site coviewing (b=0.20, p=0.05), and social networking site monitoring (b=0.20, p=0.05). We probed to see if sex interacted with need to belong, emotional contagion, or the three coviewing dimensions to predict any of the dependent variables. Because no significant interactions emerged, these interaction terms were not included in the regression models.

Table 2.

Regressions Predicting Connected Viewing

	Nonmediated coviewing		Text coviewing		Instant message coviewing		Social networking site coviewing		Social networking site monitoring
	b	R²	b	R²	b	R²	b	R²	b	R²
Demographics		0.02^*		0.10^***		0.07^***		0.12^***		0.11^***
Age	−0.01		−0.03^***		−0.02^***		−0.03^***		−0.03^***
Sex (0=female)	0.23^*		0.19^*		−0.02		0.20^*		0.20^*
Need to belong	0.01	0.00	0.10^*	0.01^*	0.07	0.01	0.12^*	0.01^*	0.16^***	0.02^***
Emotional contagion	0.18^*	0.02^**	0.09	0.01	0.04	0.00	0.07	0.01	0.07	0.01
Coviewing orientation		0.21^***		0.02^*		0.02^*		0.02^*		0.03^**
Need for solitude	−0.19^***		0.06		0.08^*		0.06		0.13^**
Audience monitoring	0.06		0.13^**		0.03		−0.09^*		0.12^*
Need for company	0.29^***		0.01		0.08		0.05		0.01
Adjusted R²	0.23^***		0.12^***		0.08^***		0.14^***		0.15

p≤0.05; ^**p≤0.01; ^***p≤0.001 (two-tailed tests).

Hypothesis 1 predicted that emotional contagion would positively predict nonmediated coviewing, and this prediction was confirmed (b=0.18, p=0.002). Those with more emotional contagion susceptibility reported participating in more nonmediated coviewing. RQ1 asked how emotional contagion would be related to frequency of computer-mediated coviewing behaviors. Emotional contagion was unrelated to all four social media coviewing behaviors.

Hypotheses 2 and 3 predicted that need to belong would predict all of the coviewing behaviors. Hypotheses 2 was not supported. Need to belong did not predict nonmediated coviewing (b=0.01, p=0.78). Hypothesis 3 was partially supported. Need to belong was a positive predictor of all forms of technology mediated coviewing, except for instant message coviewing (b=0.07, p=0.12). Need to belong was associated with more text message coviewing (b=0.10, p=0.04), more social networking site coviewing (b=0.12, p=0.12), and more social networking site monitoring (b=0.16, p=0.001).

As stated in hypothesis 4, coviewing orientation was expected to predict the frequency of all coviewing behaviors positively, but the analysis revealed a different pattern of findings for the different dimensions of coviewing. Need for company positively predicted nonmediated coviewing (b=0.29, p<0.001), and need for solitude was also a negative predictor (b=−0.19, p<0.001). That is, orientation toward coviewing as a social event was associated with increased nonmediated coviewing, while the tendency to prefer watching television alone was associated with reduced nonmediated coviewing. On the contrary, need for solitude had a positive association with two social media variables. A greater need for solitude was associated with more instant message (b=0.08, p=0.03) and social networking site monitoring coviewing (b=0.13, p=0.003). Audience monitoring positively predicted text coviewing (b=0.13, p=0.005), and social networking site monitoring (b=0.12, p=0.02), and negatively predicted social networking site coviewing (b=−0.09, p=0.05).

Discussion

Consistent with arguments made in past research^19,20 and the first hypothesis, those with greater susceptibility to catching others' emotions coview with greater frequency, suggesting that this tendency encourages in-person connected viewing, where emotional experiences can be transmitted easily. However, regarding the first research question, emotional contagion did not predict social media coviewing, suggesting that emotional contagion processes that enhance in-person viewing²⁴ do not operate during mediated coviewing. Features such as the live news feed on Facebook and Twitter, which can be organized by hashtags related to live television programs, permit people to monitor a constant trickle of information about other viewers' affective states,⁴³ but these digital emotional cues may be less transmittable than nonverbal cues that occur in face-to-face contexts, even among people who score higher in emotional contagion. Single emotional experiences could also be harder to “catch” in these social media environments because users are exposed to a multitude of emotional reactions. Emotional contagion also did not predict more participatory behavior (e.g., sharing program-related comments through text message), suggesting that online interaction could disrupt emotional contagion processes. In face-to-face contexts, emotional contagion is thought to occur automatically and unconsciously when people peripherally detect others' emotional cues.¹⁸ Emotional contagion processes in mediated contexts might require a more deliberate observation, taking up some of viewers' limited attentional resources. The transfer of these emotions could be limited if viewers focus attention on sharing their own reactions instead of absorbing the emotional climate created by others.

Consistent with our predictions in hypothesis 3, need to belong predicted all but one form of mediated coviewing, but contrary to hypothesis 2, it was not a predictor of nonmediated coviewing. This is consistent with the finding that people use social media during television viewing to feel like part of a group.^1,3 These results also speak to social media's ability to act as extensions of people's social monitoring system, permitting them to scan for information that helps them gauge how well they fit into larger communities. Compared to in-person coviewing, social media users generally have access to more audience members and thus a larger pool of social information to process during the televised event. This social information can help individuals with a strong need to belong by creating a sense that they are part of a larger community of viewers.

This research is the first of its kind to examine social orientation to television use as a disposition. Three dimensions of coviewing orientation were uncovered: need for solitude, need for company, and audience monitoring. Individual predictions for these specific dimensions of coviewing were not made a priori before the scales were factored, but the finding that need for solitude negatively predicts nonmediated coviewing in the presence of others seems rather intuitive. More revealing is that need for solitude positively predicts instant message coviewing and social network monitoring. This finding runs counter to the logic behind our prediction in hypothesis 5 that a preference to use television as a catalyst for social interaction should predict coviewing—even computer-mediated coviewing. Taken together, these findings suggest that need for solitude may be best thought of as a measure of how much people prefer to be physically alone when they watch television, but not necessarily a measure of how much they like to be without a computer-mediated presence of another. The frequency of some mediated coviewing for those with greater solitude needs suggests that mediated coviewing provides a viable alternative to in-person coviewing for those who dislike the distractions of physically present coviewers, or simply prefer to control the coviewing environment. New television technologies, such as time-shifting devices and on-demand programming, give audiences greater control over what they watch and when they watch it, as well as who they watch with and how they coview. Computer-mediated coviewing offers coviewers a range of options, including whether to watch synchronously or asynchronously, with one person or many, or with friends or strangers.

This study also found that while audience monitoring was unrelated to in-person coviewing, it predicted the same types of coviewing that are positively associated with need to belong: text coviewing, and social networking site coviewing and monitoring. This pattern of results lends some credence to the claim that social networking sites and mobile phones with text capabilities are ideal coviewing platforms for individuals with high belonging needs because they expand the opportunities people have for monitoring their social environment. Perhaps because social media technologies offer more opportunities to monitor and participate in broad social networks, those who appreciate audience feedback during viewing tend to engage in more mediated connected viewing.

Limitations and future research

Several limitations of this research should be noted. First, this study relied on single-item rating scales to measure frequencies of different types of coviewing behaviors, which may not gauge respondents' actual coviewing behaviors accurately. Future research should consider using ratio-level measures or behavioral indexes that can capture greater variance in the time spent on these activities.

Relative to other convenience sampling methods, MTurk provides a diverse sample.³⁷ The current sample, for instance, includes respondents from a wide range of ages, helping to shed some preliminary light on coviewing motivations that could vary by age. However, the sample is not representative of American television viewers. Almost half of respondents had a college degree, which is far more than the general television audience.⁴⁴ Respondents were also probably more technologically savvy and likely to engage in mediated connected viewing experiences than the general population evidenced by their use of MTurk. Thus, these findings cannot be overly generalized.

This research makes important progress to understanding the motivations behind connected viewing experiences, but it is limited in its ability to explain the social mechanisms that underlie coviewing. Additional survey and experimental research is needed to unpack coviewing as both a social process and a media effect. For instance, the findings from this study strongly suggest that a greater need to belong should motivate individuals to seek out social TV experiences with social networking sites because their natural social insecurities increase their need to monitor social cues in their network. Support for this claim will best be provided from experimental research that provides a stronger link between desire for social bonding as a state and mediated coviewing.

This study is the first of its kind to compare motivations for sharing different types of viewing experiences with others, providing insight into some psychological gratifications these experiences provide that viewers may not be entirely aware of. Importantly, these findings indicate that connected viewing experiences are not all alike, and viewers have different intrinsic emotional and social motivations for coviewing in-person versus through social media.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Highfield

, Harrington

, Bruns

. Twitter as a technology for audiencing and fandom. Information, Communication & Society, 2013; 16:315–339.

Norman

. Saturday night's alright for Tweeting: cultural citizenship, collective discussion, and the new media consumption/production of Hockey Day in Canada. Sociology of Sport Journal, 2012; 29:306–324.

Wohn

, Eun-Kyung

. Tweeting about TV: sharing television viewing experiences via social media message streams. First Monday, 2011; 16:1.

Smith

. (2012) The rise of the “connected viewer.”. http://pewinternet.org/∼/media//Files/Reports/2012/PIP_Connected_Viewers.pdf (accessed November 29, 2012).

Nielsen (2012) Social media report. www.nielsen.com/us/en/reports/2012/state-of-the-media-the-social-media-report-2012.html (accessed August 2, 2013).

Alberts

, Yoshimura

, Rabby

, et al. Mapping the topography of couples' daily conversation. Journal of Social & Personal Relationships, 2005; 22:299–322.

Fallis

, Fitzpatrick

, Friestad

. Spouses' discussion of television portrayals of close relationships. Communication Research, 1985; 12:59–81.

Finucane

, Horvath

. Lazy leisure: a qualitative investigation of the relational uses of television in marriage. Communication Quarterly, 2000; 48:311–321.

Gantz

. Exploring the role of television in married life. Journal of Broadcasting & Electronic Media, 1985; 29:65–78.

10.

Lull

. The social uses of television. Human Communication Research, 1980; 6:197–209.

11.

Buijzen

, Walma Van Der Molen

, Sondij

. Parental mediation of children's emotional responses to a violent news event. Communication Research, 2007; 34:212–230.

12.

Zillmann

, Weaver

, Mundorf

, et al. Effects of an opposite-gender companion's affect to horror on distress, delight, and attraction. Journal of Personality & Social Psychology, 1986; 51:586–594.

13.

Brody

, Stoneman

, Sanders

. Effects of television viewing on family interactions: an observational study. Family Relations, 1980; 29:216–220.

14.

Kubey

. (1990) Television and family harmony among children, adolescents, and adults: results from the experience sampling method. In Bryant

, ed. Television and the American family. Hillsdale, NJ: Lawrence Erlbaum, pp. 73–88.

15.

Huijnen C

, sselsteijn

, Markopoulos

, et al. Social presence and group attraction: exploring the effects of awareness systems in the home. Cognition, Technology & Work, 2004; 6:41–44.

16.

Markopoulos

, Ijsselsteijn

, Huijnen

, et al. Sharing experiences through awareness systems in the home. Interacting with Computers, 2005; 17:506–521.

17.

Rubin

. An examination of television viewing motivations. Communication Research, 1981; 8:141.

18.

Hatfield

, Cacioppo

, Rapson

. Emotional contagion. Current Directions in Psychological Science, 1993; 2:96–99.

19.

Hocking

. Sports and spectators: intra-audience effects. Journal of Communication, 1982; 32; 100–108.

20.

Wenner

, Gantz

. (1998) Watching sports on television: audiences, experiences, gender, fanship, and marriage. In: Wenner

, ed. MediaSport. New York: Routledge, pp. 233–251.

21.

Denham

. Toward an explication of media enjoyment: The synergy of social norms, viewing situations, and program content. Communication Theory, 2004; 14:370–387.

22.

Lin

, McDonald

. (2006) Emotional contagion as a function of individual responses and emotional contagion. Paper presented to the Information System Division of the International Communication Association, Dresden, Germany.

23.

Lin

, McDonald

. (2007) Peer coviewing and individual difference effects on comedy viewing. Paper presented to the Mass Communication Division of the International Communication Association.

24.

Walther

. Interpersonal effects in computer-mediated interaction: a relational perspective. Communication Research, 1992; 19:52–90.

25.

Gardner

, Pickett

, Brewer

. Social exclusion and selective memory: how the need to belong influences memory for social events. Personality 7 Social Psychology Bulletin, 2000; 26:486–496.

26.

Baumeister

, Leary

. The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 1995; 117:497–529.

27.

Pickett

, Gardner

, Knowles

. Getting a cue: the need to belong and enhanced sensitivity to social cues. Personality & Social Psychological Bulletin, 2004; 30:1095–1107.

28.

Baumeister

, Leary

. The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 1995; 117:497–529.

29.

Nadkarni

, Hofmann

. Why do people use Facebook?. Personality & Individual Differences, 2012; 52:243–249.

30.

Sun

, Wu

. Traits, predictors, and consequences of Facebook self-presentation. Social Science Computer Review, 2012; 30:419–433.

31.

Walsh

, White

, Cox

, et al. Keeping in constant touch: the predictors of young Australians' mobile phone involvement. Computers in Human Behavior, 2011; 27:333–342.

32.

O'Sullivan

. (2003). Mass personal communication: Communication technology and transcendent theoretical frameworks. Presented at the annual conference of the International Communication Association, San Diego, CA.

33.

Wang

, Tchernev

. The “myth” of media multitasking: reciprocal dynamics of media multitasking, personal needs, and gratifications. Journal of Communication, 2012; 62:493–513.

34.

Buhrmester

, Kwang

, Gosling

. Amazon's Mechanical Turk: a new source of inexpensive, yet high-quality, data?. Perspectives on Psychological Science, 2011; 6:3–5.

35.

Gardner

, Brown

, Boice

. Using Amazon's Mechanical Turk website to measure accuracy of body size estimation and body dissatisfaction. Body Image, 2012; 9:532–534.

36.

Holden

, Dennie

, Hicks

. Assessing the reliability of the M5-120 on Amazon's Mechanical Turk. Computers in Human Behavior, 2013; 29:1749–1754.

37.

Mason

, Suri

. Conducting behavioral research on Amazon's Mechanical Turk. Behavior Research Methods, 2012; 44:1–23.

38.

Rand

. The promise of Mechanical Turk: how online labor markets can help theorists run behavioral experiments. Journal of Theoretical Biology, 2012; 299:172–179.

39.

Sprouse

. A validation of Amazon Mechanical Turk for the collection of acceptability judgments in linguistic theory. Behavior Research Methods, 2011; 43:155–167.

40.

Simons

, Chabris

. Common (mis)beliefs about memory: a replication and comparison of telephone and Mechanical Turk survey methods. Plos ONE, 2012; 7:1–5.

41.

Doherty

. The emotional contagion scale: a measure of individual differences. Journal of Nonverbal Behavior, 1997; 21:131–154.

42.

Leary

, Kelly

, Cottrell

, et al. Individual differences in the need to belong: mapping the nomological network. Unpublished manuscript, Duke University, 2007.

43.

Mitra

. Creating a presence on social networks via narbs. Global Media Journal: American Edition, 2012; 9:1–18.

44.

Nielsen . 2013. Advertising and audiences: the state of the media. www.nielsen.com/content/dam/corporate/us/en/reports-downloads/2013%20Reports/Advertising-Audiences-Report-April-2013.pdf (accessed April 27, 2014).