Computers in Human Behavior 62 (2016) 570e577

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Full length article

Celebrity's self-disclosure on Twitter and parasocial relationships: A

mediating role of social presence
Jihyun Kim a, *, 1, Hayeon Song b, **, 2
a
School of Communication Studies, Kent State University, Kent, OH 44242, USA
b
College of Business, Global Business Track, Gachon University, 1342 Seongnamdaueo, Sujeong-gu, Seongnam-si, Gyeonggi-do, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: This study investigated how celebrities' self-disclosure on personal social media accounts, particularly
Received 1 August 2015 Twitter, affects fans' perceptions. An online survey was utilized among a sample of 429 celebrity fol-
Received in revised form lowers on Twitter. Results demonstrated that celebrities' professional self-disclosure (e.g., sharing their
8 March 2016
work-related life), personal self-disclosure (e.g., sharing their personal life such as friends and family),
Accepted 31 March 2016
and fans' retweeting behavior, enhanced fans’ feeling of social presence, thereby positively affecting
parasocial interaction with celebrities. Further, the study found that the effects of self-disclosure and
retweeting on parasocial interaction were mediated by social presence. Implications and future research
Keywords:
Celebrities
directions are provided.
Parasocial interaction © 2016 Elsevier Ltd. All rights reserved.
Self-disclosure
Social media
Social presence
Twitter

1. Introduction As Ms. Silverman's story indicates, celebrities share a variety of

In September of 2013, Ms. Sarah Silverman's (best known as a
comedian) tweet gained much attention from fans when she
shared sad news about passing of her dog, Duck. Ms. Silverman
later posted a long and emotional tribute to Duck, an obituary type
story on her blog. By referring Duck as “my best friend”, she shared
her personal life story with millions of fans via her personal social
media. When Ms. Silverman posted this tweet, she has received
countless responses such as “… I don't know you personally, of
course, but I send you a virtual {{hug}}.” Ms. Silverman might not
have a real, personal relationship with most of those people who
responded to her tweet. However, her disclosure of aspects of her
personal life on her social media influenced her fans to relate their
emotions to Ms. Silverman.
* Corresponding author. School of Communication Studies, Kent State University,
P.O. Box 5190, Kent, OH 44242, USA.
** Corresponding author. College of Business, Gachon University, 1342 Seong-
namdaueo, Sujeong-gu, Seongnam-si, Gyeonggi-do, Republic of Korea.
E-mail addresses: jihyunkim218@gmail.com (J. Kim), hayeon.song@gmail.com
(H. Song).
1
Jihyun Kim (Ph.D., University of Wisconsin-Milwaukee, 2012) is an Assistant
Professor in the School of Communication Studies at Kent State University.
2
Hayeon Song (Ph.D., University of Southern California, 2008) is an Associate
Professor in the College of Business at Gachon University.
information about themselves on personal social media. For
example, celebrities share personal life stories about friends and
family as well as professional life to advertise upcoming events
(e.g., see Hambrick, Simmons, Greenhalgh, & Greenwell, 2010;
Kassing & Sanderson, 2010; Stever & Lawson, 2013). With easy
access to direct and interactive communication methods with ce-
lebrities, fans following celebrities through social media may feel
like they “know” the celebrity and experience increased intimacy
and a strong parasocial relationship.
There has been some research exploring the use of social media
between celebrities and fans. Thus far, the focus has been mostly on
either the content of celebrities' social media (e.g., Frederick, Lim,
Clavio, & Walsh, 2012; Kassing & Sanderson, 2010; Stever &
Lawson, 2013) or fans' motivations and reasons for following ce-
lebrities' social media from uses and gratification perspectives (e.g.,
Hargittai & Litt, 2011; Sanderson, 2011). However, almost little is
known about how celebrities' various self-disclosure styles on so-
cial media affect fans’ perceptions about those celebrities. Thus, this
study begins to fill this research gap.
Specifically, the current study examines how celebrities' self-
disclosure, particularly professional self-disclosure (revealing in-
formation about their professional life) and personal self-disclosure
(revealing information about their personal life), on personal social
media, particularly Twitter, influence fans' parasocial relationship

http://dx.doi.org/10.1016/j.chb.2016.03.083
0747-5632/© 2016 Elsevier Ltd. All rights reserved.
 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
with celebrities. Further, given that one of the unique features of
Twitter is the function of ‘retweeting’, the study also explores how
fans' act of retweeting celebrities' tweets would affect fans' per-
ceptions of parasocial relationship with celebrities. Lastly, a
considerable amount of research has revealed that social presence
is one of the key underlying mechanisms that facilitate people's
experiences in a mediated-communication environment (e.g., Jin,
2009, 2011; Jin & Park, 2009; Lee & Nass, 2005; Lee, Park, &
Song, 2005; Lee, Peng, Yan, & Jin, 2006; Song, Kim, & Lee, 2014).
Therefore, the study examines the role of social presence, particu-
larly as a mediator that links celebrities' self-disclosure and fans' act
of retweeting celebrities' tweets to fans' perceptions of parasocial
relationship.
2. Literature review
2.1. Celebrities’ use of personal social media and self-disclosure
2.1.1. Celebrities’ social media use
“Social media are inherently designed to facilitate human
connection” (Sanderson, 2011, p. 494). Because communicating
with fans on social media does not necessarily require a special
permission to access, social media have become an increasingly
popular option for celebrities who want to communicate directly
with their fans (Click, Lee, & Holladay, 2013; Stever & Lawson,
2013).
Traditionally, celebrities delivered their message to fans through
traditional media outlets (e.g., television, news paper) indirectly via
reporters that might have filtered the original messages. However,
social media platforms provide a personalized and unfiltered
method of communication. Consequently, these platforms have
significantly influenced communication patterns between celeb-
rities and fans. Many celebrities are establishing their own news
channel through their personal social media such as Twitter (Brau,
2013). A good number of celebrities use such platforms for
expression of dissent (Sanderson, 2009) and self-presentation
(Sanderson, 2008).
Similarly, Twitter has been used to provide athletes and their
fans with an opportunity to communicate more interpersonally and
to enhance the overall sport experiences (Frederick et al., 2012;
Kassing & Sanderson, 2010). Kassing and Sanderson found that
athletes used Twitter to share opinions and comments about the
sport event, to foster interactive communication with fans, and to
cultivate perspectives about the event among fans.
2.1.2. Celebrities’ self-disclosure
Social Penetration Theory (SPT) explains that human relation-
ships develop through revealing the self, and this process primarily
occurs through self-disclosure; the voluntary sharing of personally
relevant information such as feelings, thoughts, values, and beliefs
with another (Altman & Taylor, 1973). SPT holds a perspective that
the self consists of several layers from peripheral layers (i.e., de-
mographic information) to central layers (i.e., personal values, self-
concept). And, revealing each layer of the self to others, the act of
self-disclosure, is an important step in building a relationship
(Altman & Taylor, 1973). Given that self-disclosure is a complex
multidimensional concept explained by a variety of features and
characteristics of individuals, researchers have suggested and
investigated various types of self-disclosure. A majority of the SPT
scholars highlight two dimensions of self-disclosure, breadth and
depth (e.g., Altman & Taylor, 1973; Derlega, Metts, Petronio, &
Margulis, 1993; Jourard, 1971). Breadth refers to the variety of
topics of self-disclosure, thus a wide range of topics or a narrow
range may be disclosed. For example, topics related to profession or
work might be seen as professional self-disclosure, and topics
571
concerning family-, friends-related issues, or personal beliefs might
be viewed as personal self-disclosure. Depth refers to the degree of
disclosure in a specific area of an individual's life. That is, depending
on how deep an individual discloses him/herself about a certain
topic, depth of self-disclosure could vary from a superficial to
intimate level (Altman & Taylor, 1973; Derlega et al., 1993; Jourard,
1971).
Through social media, many contemporary celebrities share
their life with their fans (Marshall, 2010), and topics of stories that
celebrities share with fans might appear in various forms. Given
that celebrities need fans' support for their career success, some
celebrities might disclose their professional life more than other
parts of their life to promote their work (e.g., new album, new
movie). Stever and Lawson (2013) investigated a number of famous
celebrities' on Twitter. One of the main themes found in the study
was professional self-disclosure. Specifically, the study found that
Katy Perry, an American singer and songwriter, was mostly focused
on sharing her work-related information on her Twitter. In a similar
vein, another study (Hambrick et al., 2010) also reported that one of
the main themes found on sport celebrities’ tweets was concerned
with their work (e.g., information sharing or providing insights
about their team or game).
Some celebrities might mainly reveal personal life stories on
social media (e.g., Click et al., 2013; Hambrick et al., 2010; Marshall,
2010; Stever & Lawson, 2013). Stever and Lawson noted that Kristen
Chenoweth, an American actress and singer, was found to share a
significant portion of her personal life on Twitter. Similarly, some
sport celebrities also disclose non-sport-related life stories such as
dinner menu, favorite restaurants, and movies they want to see
(Hambrick et al., 2010). Politicians have also been found to disclose
personal life events on social media (Jackson & Lilleker, 2011; Park,
2010; Small, 2010).
2.2. Social presence
With its capacity of direct and interactive communication, social
media may provide fans with a strong feeling of ‘being together’ or
‘being connected’ with the celebrity (Stever & Lawson, 2013). That
is, through mediated communication, fans ‘meet’ celebrities and
learn about their daily life, whether it is one-way or two-way
communication. Through this interaction, fans might feel as if
those celebrities are socially present in their life.
Described as ‘being aware of the other person in online environ-
ments’ (Short, Williams, & Christie, 1976), social presence is one of
the important concepts to the understanding of mediated-
communication experiences (e.g., Biocca, Harms, & Burgoon,
2003; Short et al., 1976). Numerous scholars have explored this
notion since Short et al.’s research (e.g., Biocca et al., 2003; Biocca &
Nowak, 2001; Caspi & Blau, 2008; Edwards, Edwards, Spence, &
Westerman, 2015; Lee, 2013; Lee et al., 2005; Song et al., 2014;
Spence, Westerman, Edwards, & Edwards, 2014; Tu, 2000).
Although a definition of social presence has not been agreed upon
yet, social presence is generally understood as perceptual experi-
ence of being psychologically involved in the interaction with
another in a mediated environment (Biocca et al., 2003; Durlach &
Slater, 2000; Schroeder, 2006).
2.2.1. Self-disclosure and social presence
Several studies have examined what makes people to feel social
presence or related concepts such as presence (a broader category
of social presence; see Lee, 2004) (e.g., IJsselsteijn, Ridder, Freeman,
& Avons, 2000; Lee & Nass, 2005). Research identifies a few cate-
gories of causal factors for social presence: media/technology-
related factor, user factors (Lombard & Ditton, 1997), and social
factors (Lee & Nass, 2005). Media/Technology-related factors are
572 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
concerned with media/technology features such as image size,
quality, and vividness. User factors include media user character-
istics such as demographics, familiarity with media, and media use
patterns. Social factors, a particular interest of this investigation,
refer to users’ social response toward media.
Social factors are an important factor for an extended under-
standing of social presence research (Lee & Nass, 2005). Lee and
Nass’ theory driven research found that individuals hearing a
computer voice manifesting a personality similar to theirs experi-
enced stronger social presence compared to others hearing a voice
manifesting a dissimilar personality. Their study also found that
people felt stronger social presence when they heard an extrovert
voice than an introvert voice (Lee & Nass, 2005).
Moreover, the importance of social factors in fostering social
presence has been documented in the online education research. In
particular, Tu (2000) argues that social factors such as social and
informal relationships and trust influence a feeling of social pres-
ence among online students. Additionally, humor (Eggins & Slade,
1997), warmth (Rourke, Anderson, Garrison, & Archer, 2001), and
perceived self-disclosure (Rourke et al., 2001; Swan & Shih, 2005)
have been also addressed as important social factors that induce
strong social presence in online learning environments.
Compared to other causal factors for social presence, social
factors have received relatively less attention. Particularly, at the
time of writing this report, none has investigated how self-
disclosure, one of the social factors, influences social presence in
a context of celebrity-fan relationships. As discussed earlier, a sig-
nificant number of celebrities use personal social media and
communicate with fans on a variety of topics (e.g., Frederick et al.,
2012; Kassing & Sanderson, 2010; Stever & Lawson, 2013).
Acknowledging popular use of disclosing themselves on social
media and importance of social factors in fostering a feeling of
social presence, the current investigation examines the effect of
self-disclosure on fans’ feeling of social presence. Given that each
social media platform (e.g., Twitter, Facebook, Instagram) has its
own unique features, which might influence communication pat-
terns, the current study limits its scope to Twitter only.
H1a-b: Celebrities' (a) professional self-disclosure and (b) per-
sonal self-disclosure on Twitter increase fans' feeling of social
presence.
Twitter has capacity to provide various communication modes
between celebrities and fans. In most cases, a majority of fans are
likely to be observers of what celebrities have posted or/and what
others have responded to celebrities' tweets. However, some fans
engage in ‘retweeting,’ an act of posting or forwarding someone's
tweets to others. That is, some fans actively distribute celebrities'
self-disclosed messages (tweets) to others in their personal
network (e.g., their own followers) and share and discuss their
opinions about the disclosed messages.
Retweeting is an active communication behavior in a sense that
the act of retweeting represents an active reaction to someone's
communication messages by voluntarily reposting the message to
share with users' own followers. In addition, the act of retweeting
reflects users' active way of showing psychological and emotional
involvement with the content. This active form of communication
is well represented with celebrity Ellen DeGeneres's (a talk show
host) tweet. As of 2015, Ms. DeGeneres's tweet hit the record of the
most famous tweet to date with 3,344,002 retweets and 2,021,170
favorites (Bennett, 2015).
Previous studies indicate that involvement and active partici-
pation of media use induce strong social presence (McGreevy, 1993;
Witmer & Singer, 1998). That is, when media users are deeply
involved in an activity in the media, they tend to experience strong
social presence. In this regard, the current study proposes the
following hypothesis.
H2: The act of retweeting celebritie's tweets increases fans'
feeling of social presence.
2.2.2. Social presence and parasocial interaction experiences
Confirming the argument that social presence affects the nature
of the interaction (Short et al., 1976), social presence has been
identified to facilitate individuals’ psychological and physiological
responses in a mediated environment (Lombard & Ditton, 1997). In
particular, a great deal of empirical research reveals a positive
relationship between social presence and various relationship-
related outcomes such as relationship satisfaction (e.g., Moore &
Kearsley, 2005; Richardson & Swan, 2003) and perceived trust-
worthiness (Bente, Rüggenberg, Kra €mer, & Eschenburg, 2008).
In a particular context of the current study, given that celebrities
use social media to make their relationships with fans more real
(Stever & Lawson, 2013), it seems necessary to explore how social
presence is related to parasocial interaction (PSI) experiences.
Generally, PSI is defined as an imaginary relationship to a person
whom one knows well but who barely knows the other (Horton &
Wohl, 1956). That is, it is the relationship that individuals might
develop with media figures including celebrities, news hosts, and
even fictional characters (e.g., Horton & Wohl, 1956; Reeves & Nass,
1996; Rubin & McHugh, 1987). Because social media such as Twitter
allow people to respond to celebrities' messages, some people
might feel and act like they were in a typical interpersonal rela-
tionship. However, “users can follow one another's content without
reciprocal obligation” (Hargittai & Litt, 2011, p. 825), which is
especially common in the celebrity's social media use context. That
is, it would still be perceived as one-sided communication, as the
one party (celebrity) mostly controls their communication mes-
sages (Cohen & Perse, 2003). Thus, it still remains as a parasocial
relationship rather than a typical interpersonal relationship.
Previous research has indicated the association between social
presence and PSI (e.g., Lee, 2013; Lombard & Ditton, 1997; Rubin,
Perse, & Powell, 1985). In particular, research found that social
presence fosters a feeling of imagined intimacy, which is part of PSI
experience (Rubin et al., 1985). Directly germane to the current
study, Lee (2013) examined how people respond to a politician's
messages on media and found that heightened social presence
enhanced positive PSI experiences and overall evaluation of the
politician.
Although the celebrity-fan culture is pervasive, the under-
standing of the PSI in the social media domain and social presence
is still limited. Considering the importance of social presence in a
mediated environment (Short et al., 1976) and a strong need for
more understanding of PSI on the social media platform (Stever,
2009), it is important to investigate the association of these two.
Therefore, the study proposes the following hypothesis.
H3: Feeling of social presence enhances fans' parasocial interac-
tion experiences with celebrities.
A body of research has identified a variety of factors related to
PSI experiences (e.g., Rubin et al., 1985; Rubin & McHugh, 1987;
Rubin & Perse, 1987; Savage & Spence, 2014). A meta-analysis
(Schiappa, Allen, & Gregg, 2007) reported that TV viewing con-
sumption, desire for watching TV, perceived similarity with TV
characters, and communication apprehension are positively related
to PSI experiences. A study (Savage & Spence, 2014) found that
radio listeners tend to have favorable PSI experiences when a radio
host reveals personal stories during a show. However, what is not
so clear is how and why such factors are leading to favorable PSI
experiences.
 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
As discussed earlier, the current study predicts that celebrities'
self-disclosure and fans' involvement in retweeting lead to fans'
feeling of social presence of celebrities (H1a-b & H2 of this inves-
tigation). Then, the study also predicts that the feeling of social
presence would facilitate positive PSI experiences (H3). In this re-
gard, the current investigation argues that social presence mediates
the relationship between a celebrity-fan Twitter use (in particular,
celebrities' self-disclosure and fans’ retweeting) and PSI
experiences.
Although specific contexts are different, a mediating role of
social presence (or presence, a broader category of social presence;
see Lee, 2004) has been well documented in the current body of
research (e.g., Jin, 2009, 2011; Jin & Park, 2009; Lee & Nass, 2005;
Lee et al., 2005; Lee et al., 2006; Song et al., 2014). Germane to
the context of the current investigation, a study (Lee, 2013) found a
conditional mediating role of social presence in a PSI context. Lee
found that social presence mediated the relationship between
seeing a politician's TV interview (vs. Twitter messages) and
favorable PSI experiences among those with a low need for
cognition. In this regard, the current research proposes the
following hypothesis.
H4a-c: Social presence mediates the relationship between a ce-
lebrity-fan Twitter use e in particular, (a) professional self-disclo-
sure, (b) personal self-disclosure, (c) fans' retweeting behavior e
and parasocial interaction experiences.
3. Methods
3.1. Participants
A total of 572 undergraduate students from a large introductory
communication course at a public university in the United States
were contacted for this study. For the purpose of the current study,
celebrity followers on Twitter were only included to the sample of
the study. This created a sample of 429 participants. There were
more females (n ¼ 242: 56.4%) than males (n ¼ 172: 40.1%), and 15
individuals did not identify their biological sex (3.5%). The average
age was 19.24 years (SD ¼ 2.42). A majority of the participants were
Caucasian (n ¼ 355: 82.8%) followed by African American (n ¼ 27:
6.3%), Asian (n ¼ 22: 5.1%), and other ethnic groups (n ¼ 25: 5.8%).
3.2. Procedure
An online survey was conducted with university-licensed sur-
vey tools (www.qualtrics.com). The primary researcher contacted a
course director of the large introductory communication course
and asked whether it would be possible to provide students with
this research participation opportunity. Upon the director's
approval, a recruitment message that included a link to the survey
was sent to potential participants through university email. Once
participants accessed the survey, they were asked to read and
acknowledge the informed consent prior to completing the survey.
At the start, participants were given with a description of ce-
lebrities as following: “In general, ‘celebrity’ refers to a person or a
group of people who receive popular fame and public attention. For
example, celebrities would include actors, actresses, singers, co-
medians, models, and athletes.” Then, participants were asked to
think of one particular celebrity and to provide the celebrity's name
and primary occupation. These questions were asked to trigger
participants' experiences with the particular celebrity's tweets.
Then, participants were asked to answer the rest of the survey
questions based on the particular celebrity. Although political fig-
ures can be viewed as celebrities, in order to avoid any potential
effects of favoritism (or anti-favoritism) for a particular political
573
party, participants were asked not to think of political figures for
the purpose of this study.
When participants finished completing the survey, they were
redirected to a separate website, independent from the original
survey, where they could provide their name and class information.
This information was obtained only to provide participants with
research participation credit for their course. The use of separate
websites allowed the researcher to guarantee anonymity.
3.3. Measures
The survey questionnaire included a set of measures. Profes-
sional self-disclosure (Cronbach's a ¼ 0.73) was measured with
three items, developed for the study (e.g., “This celebrity talks
about her/his work schedule (e.g., concert, shows),” “…. talks about
joy, passion or enthusiasm for work,” “…. talks about work in
general”). Personal self-disclosure (Cronbach's a ¼ 0.82) was
measured with nine items (e.g., “This celebrity often discloses
personal things about herself/himself on Twitter,” “This celebrity
talks about her/his romantic partners,” “… talks about personal
habits,” “… talks about personal life history”). Two items were
adopted from Wheeless (1978) and seven items were developed for
the study. Responses for both professional and personal self-
disclosure were obtained on a 7-point Likert-type scale
(1 ¼ Never, 7 ¼ Always).
Retweeting was measured by asking the following question,
“How often do you retweet this celebrity's tweet in an average
week?” Response was obtained on a 5-point scale (1 ¼ None,
5 ¼ Very often). Social presence (Cronbach's a ¼ 0.87) was measured
with five items using a semantic differential scale adopted from
Lombard and Ditton (2000) and Short et al. (1976) (e.g., “I feel that
this celebrity is remote—immediate,” “… unemotional—
emotional,” “… impersonal—personal”). Responses for social pres-
ence were obtained on a 7-point semantic differential scale.
Parasocial interaction (Cronbach's a ¼ 0.82) was assessed with
six items selected from Rubin et al. (1985) and slightly modified for
the particular context of this research (e.g., “I feel like this celebrity
is my friend,” “This celebrity seems to understand the kinds of
things I want to know about her/himself,” “I like to share some of
my life (e.g., thoughts, opinions, hobbies) with this celebrity”).
Responses were obtained on a 7-point Likert-type scale
(1 ¼ Strongly Disagree; 7 ¼ Strongly Agree).
4. Results
4.1. Descriptive analyses
A set of simple descriptive analyses was conducted to assess
overall patterns of Twitter use in the sample (n ¼ 429). A majority
of participants (n ¼ 275; 64.1%) have used Twitter between 2 and 5
years. The rest of the participants reported Twitter use as following:
17 participants (4.0%) under 6 month; 34 participants (7.9%) be-
tween 6 months and 1 year; 72 participants (16.8%) between 1 and
2 years; and 31 participants (7.2%) over 5 years. On average, par-
ticipants reported spending 46.05 min (SD ¼ 35.08) for reading
tweets on Twitter and 11.74 min (SD ¼ 15.24) for writing tweets on
a daily basis. Regarding the particular celebrity that participants
identified for the current study, the following provides a variety of
primary job categories of the celebrity: actor/actress (n ¼ 77;
17.9%); singer (n ¼ 164; 38.2%); comedian (n ¼ 24; 5.6%); model
(n ¼ 17; 4.0%); athlete (n ¼ 84; 19.6%); and other (n ¼ 60; 14%).
4.2. Primary analyses
Before conducting hypothesis testing, correlation matrix was
574 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
first produced (Table 1). The highest correlation among the five
variables (professional self-disclosure, personal self-disclosure,
retweeting, social presence, and parasocial interaction) was 0.42
(correlation between social presence and parasocial interaction).
Thus, no multicollinearity problem was observed because the score
is lower than the recommended threshold of 0.7 (Tabachnick &
Fidell, 2001).
To test H1eH3, a path analysis was conducted using Lisrel 9.1
program. H1a-b predicted that celebrities' (a) professional and (b)
personal self-disclosure increase fans’ feeling of social presence.
Two types of self-disclosure were included in the model. Both
professional (b ¼ 0.22, t ¼ 4.15, p < 0.05) and personal self-
disclosure (b ¼ 0.11, t ¼ 2.17, p < 0.05) were found to have signifi-
cantly positive associations with social presence. Thus, H1a and
H1b were supported.
H2 was concerned with retweeting and social presence. Results
found that the act of retweeting a celebrity's tweets significantly
induced fans' feeling of social presence (b ¼ 0.09, t ¼ 1.80, p ¼ 0.05).
Thus, H2 was supported.
Next, H3 predicted a relationship between social presence and
parasocial interaction (PSI) experiences. Data confirmed that social
presence is positively related to PSI experiences with the celebrity
(b ¼ 0.42, t ¼ 9.02, p < 0.05). H3 was supported. See Fig. 1 for
H1eH3.
To test H4a-c concerning a mediating role of social presence, a
SPSS Macro, which produces sobel tests, bootstrap tests, and
regression tests, was used (see Preacher & Hayes, 2004, 2008).
While a sobel test assumes normal distribution, a bootstrap pro-
cedure imposes no distributional assumption. For a safe approach
to results, both tests were assessed. Unstandardized indirect effects
were computed with 3000 bootstrapped samples. Results for both a
sobel test and bootstrapping were interpreted with the 95% confi-
dence interval (CI). Further, regression coefficients were evaluated
to determine whether it is full or partial mediation. Full mediation
occurs when the effect of IV on DV becomes statistically non-
significant when controlling for a mediator. And, partial media-
tion occurs when the effect of IV on DV decreases although it still
remains significant when controlling for a mediator.
Regarding H4a, a sobel test indicated that social presence me-
diates the relationship between professional self-disclosure and PSI
experiences (b ¼ 0.10; z ¼ 4.67; p < 0.001; CI ¼ [0.06, 0.14]). A
bootstrap test confirmed social presence as a mediator with a mean
score for the indirect effect of 0.10 (SE ¼ 0.03; CI ¼ [0.06, 0.16]). The
results of subsequent evaluation of regression analysis indicated a
significant relationship between professional self-disclosure and
PSI experiences (b ¼ 0.28, t(415) ¼ 5.93, p < 0.001). When con-
trolling for social presence, although the relationship remained
significant, the coefficient decreased (b ¼ 0.18, t(415) ¼ 3.93,
p < 0.001). These findings indicate that partial mediation of social
presence occurs between professional self-disclosure and PSI ex-
periences. Overall, H4a was supported.
Table 1
Zero-order correlations, means, and standard deviations.
1 2 3 4
1 Professional SD 1
2 Personal SD 0.37** 1
3 Retweeting 0.18** 0.20** 1 the way celebrities disclose themselves might be a direct cause of
4 Social Presence 0.28** 0.21** 0.15** 1 PSI experiences. However, this study revealed that it is because of
5 Parasocial Interaction 0.28** 0.17** 0.32** 0.42**
M 4.46 3.64 1.64 4.9
SD 1.22 1.10 0.73 1.16
Note 1. *p < 0.05, **p < 0.01, ***p < 0.001.
Note 2. SD refers to self-disclosure.
As to H4b, a sobel test indicated that social presence mediates
the relationship between professional self-disclosure and PSI ex-
periences (b ¼ 0.97; z ¼ 3.95; p < 0.001; CI ¼ [0.05, 0.14]). A
bootstrap test also confirmed the finding with a mean score for the
indirect effect of 0.97 (SE ¼ 0.03; CI ¼ [0.04, 0.16]). Subsequent
evaluation of regression analysis indicated a significant relationship
between personal self-disclosure and PSI experiences (b ¼ 0.18,
t(404) ¼ 3.25, p < 0.01). However, when controlling for social
presence, the relationship was no longer significant (b ¼ 0.08,
t(404) ¼ 1.58, p > 0.05). Thus, full mediation of social presence
occurs between personal self-disclosure and PSI experiences. Data
supported H4b.
With regard to H4c, a sobel test indicated social presence as a
mediator between retweeting and PSI experiences (b ¼ 0.96;
z ¼ 2.94; p < 0.01; CI ¼ [0.03, 0.16]). A bootstrap test confirmed the
finding with a mean score for the indirect effect of 0.97 (SE ¼ 0.03;
CI ¼ [0.03, 0.17]). Evaluation of regression analysis indicated that
there is a significant relationship between retweeting and PSI ex-
periences (b ¼ 0.53, t(418) ¼ 6.85, p < 0.001). When controlling for
social presence, although the relationship remained significant, it
decreased (b ¼ 0.43, t(418) ¼ 6.03, p < 0.001). Overall, findings
indicated that partial mediation of social presence occurs between
retweeting and PSI experiences. H4c was supported. See Table 2 for
all steps of regression coefficients for H4a-c.
5. Discussion
5.1. Primary findings
The current investigation discovered new findings in the area of
research on celebrities' self-disclosure on Twitter, retweeting, social
presence, and parasocial interaction (PSI) experiences. First, find-
ings demonstrated that celebrities' self-disclosure on Twitter en-
hances fans' feeling of social presence. That is, when celebrities
share their life and directly communicate about theses experiences,
fans tend to feel as if those celebrities were socially present in their
life. Although the context is different, this finding is consistent to
previous research that examined self-disclosure and social pres-
ence in a virtual learning environment between a teacher and
students (e.g., Rourke et al., 2001; Swan & Shih, 2005). Additionally,
a retweeting act is found to positively influence social presence.
That is, actively involved fans in sharing celebrities’ self-disclosed
messages tend to experience strong social presence of celebrities.
Further, social presence is found to facilitate positive PSI expe-
riences. In the specific context of this investigation, results found
that when fans feel celebrities to be socially present, they tend to
develop imaginary interpersonal relationships with celebrities as if
they were friends in their real life. This finding on social presence
and PSI is consistent to previous research (e.g., Lee, 2013; Lombard
& Ditton, 1997; Rubin et al., 1985).
The most meaningful highlight of this investigation is a signif-
icant mediating role of social presence. In addition to the above-
mentioned associations with social presence, the study revealed
that social presence mediates the relationship between celebrities'
Twitter use and fans' PSI experiences. Importantly, personal self-
5
disclosure is found to be a full mediator. If merely the association
between celebrities’ self-disclosure and PSI experiences were
examined, the argument might simply observe and conclude that
1 social presence. That is, when taking social presence into the
4.40 equation, the relationship between personal self-disclosure and PSI
1.23 experience no longer exists. Further, although it is partial media-
tion, social presence also mediates the relationship between pro-
fessional self-disclosure and PSI, and the relationship between
 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577 575

Professional
SD 0.22*
0.11* 0.42*
Social Parasocial
Personal SD
Presence Interaction

Retweeting 0.09*

Note1. *p < .05

Note2. SD refers to self-Disclosure

Fig. 1. Path model (H1a-b, H2, and H3).

Table 2 explored how celebrities' self-disclosure on personal social media

Regression analyses for mediation tests (H4a-c). affects fans’ perceptions toward the celebrities, particularly in
b t
relation to social presence. In this regard, this research finding
provides a new understanding of communication phenomena by
Professional SD (X) YX 0.28 5.93*** Partial mediation
unveiling the understudied area of research.
MX 0.26 5.77***
YM.X 0.39 8.08*** Furthermore, the study makes meaningful contributions to so-
YX.M 0.18 3.93*** cial presence research. While media/technology factors have been
Personal SD (X) YX 0.18 3.25** Full mediation widely and extensively studied since Short et al.’s (1976) research,
MX 0.23 4.48***
social factors as a theoretical causal factor have not received much
YM.X 0.42 8.59***
YX.M 0.08 1.58
attention until Lee and Nass’ (2005) theoretical argument. By
Retweeting (X) YX 0.53 6.85*** Partial mediation investigating self-disclosure in relation to social presence, the
MX 0.24 3.14** current study strengthens the importance of social factors for
YM.X 0.40 8.76*** fostering social presence. The study makes a further contribution to
YX.M 0.43 6.03***
the importance of user factors. As found in this investigation, highly
Note 1. *p < 0.05, **p < 0.01, ***p < 0.001. involved social media users, in particular, those who engage in
Note 2. SD refers to self-disclosure.
retweeting, tend to feel stronger social presence. This finding pro-
Note 3. X (Independent variable); Y (Dependent variable) ¼ PSI (Parasocial inter-
action experience); M (Mediating variable) ¼ Social presence.
vides useful implications for user characteristics and social pres-
ence research.
The current study also makes contributions to self-disclosure
retweeting and PSI experiences. That is, the way professional self- research. The theoretical notion of self-disclosure research was
disclosure and retweeting influence PSI experience becomes first developed in an interaction setting where reciprocity is ex-
weak when considering the effect of social presence. Overall, these pected (Altman & Taylor, 1973). Thus, most self-disclosure research
findings indicate the importance of social presence in a context of has been done in contexts of interpersonal relationships in which
PSI experiences. reciprocity is expected such as romantic relationships and teacher-
student relationships. This investigation implies that self-
5.2. Implications and contributions disclosure needs to be further studied in a celebrity-fan relation-
ship where reciprocity might not be always expected or
This investigation reveals several contributions and implications guaranteed.
for research. First, the current research contributes to further The current study also provides practical implications and
expanding social media research to the PSI context. Since the rise of contributions for celebrities. First, celebrities may consider
social media, numerous scholars have examined the role of social communicating with their fans through social media platforms.
media in diverse communication contexts such as interpersonal Given that celebrities' success is largely based on how fans respond,
communication (e.g., Fox, Osborn, & Warber, 2014), communication it is important to consider what facilitates fans' favorable percep-
education (e.g., Evans, 2014), health communication (e.g., tions toward celebrities. The current study's findings suggest that
Antheunis, Tates, & Nieboer, 2013), and crisis communication (e.g., celebrities can benefit from appearing to self-disclose to fans,
Lachlan, Spence, Lin, Najarian, & Del Greco, 2016; Spence, Lachlan, whether it is about professional or personal life.
Lin, & del Greco, 2015; Spence, Lachlan, Edwards, & Edwards, 2016). Second, for celebrities who are reluctant to sharing their life on
However, social media use in PSI has received much less attention personal social media, the current study suggests that they directly
from scholars although celebrity culture is widely embedded in focus on fostering social presence. By helping fans feel like “they are
everyday life. In this regard, the current study contributes to the in my life; I feel so connected,” the study's finding indicates that
understanding of social media research in PSI. fans can enjoy positive PSI experiences, which would be important
Next, the study provides a new perspective for social media and to creating strong fan base.
celebrity research. In the limited extant literature of social media Lastly, the positive effect of professional self-disclosure found in
use among celebrities, much of the research has been focused in this investigation implies that celebrity's personal social media can
areas such as what celebrities share on their social media (e.g., be used as a cost-effective marketing tool. Typically, celebrities and
Frederick et al., 2012; Hambrick et al., 2010; Kassing & Sanderson, their agents employ a variety of media outlets when they promote
2010) and why fans follow celebrities on social media (e.g., events (e.g., movies, concerts, TV shows). In addition to using
Hargittai & Litt, 2011; Sanderson, 2011). However, little has traditional media outlets, which can be expensive, the study's
576 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
finding indicates that simple disclosure of celebrities' work or up-
coming events can attract fans' attention and create increased
support.
5.3. Limitations and future research directions
The directions for future research should be addressed with
limitations identified in this investigation. First, future researchers
may explore other aspects of self-disclosure. The current study
tested breadth of self-disclosure only, particularly topics related to
professional and personal life. Given that depth is also another
important core concept of self-disclosure (Altman & Taylor, 1973),
future researchers may test effects of depth in each professional
and personal self-disclosure on fans’ perceptions about celebrities.
Next, future researchers might explore how celebrities'
emotional (evaluative) versus factual (descriptive) self-disclosure
influence fans' perceptions and attitudes. Emotional self-
disclosure is focused on revealing personal feelings and emotions
(e.g., “My last break up hurts me so much”) rather than a simple
factual description of life events (e.g., “He broke up with me”)
(Morton, 1978; Reis & Patrick, 1996). Research has found that
emotional self-disclosure is more strongly related to relationship
closeness and satisfaction compared to merely sharing self-relevant
facts (e.g., Morton, 1978). That is, although the act of sharing self-
relevant information itself could contribute to fostering intimacy,
the particular aspect of emotional self-disclosure has stronger ef-
fects on intimacy development (e.g., Greenberg & Safran, 1987;
Laurenceau, Barrett, & Pietromonaco, 1998; Lin, 1992; Reis &
Patrick, 1996). Future research on emotional vs. factual self-
disclosure would help better understand how celebrities'
emotional or factual self-disclosure styles would affect fans’ atti-
tudes and perceptions.
Lastly, future researchers should also address effects of fans'
emotional responses to celebrities' self-disclosure. Previous
research suggests that the way receivers respond to the other
person's self-disclosure plays an important role of fostering in-
timacy (Reis, 2007). Taking this approach to the celebrity research,
future research should address how fans' liking or disliking a
celebrity's self-disclosed message influences fans' overall attitudes.
5.4. Conclusion
This study investigated how celebrities' self-disclosure on
Twitter affects fans' perceptions. Overall, the investigation reveals
that celebrities' professional and personal self-disclosure and fans'
act of retweeting celebrities' tweets enhance feelings of social
presence. Further, social presence is found to facilitate favorable PSI
experiences with celebrities. Social presence plays a role as a
mediator that links celebrities' self-disclosure and fans' retweeting
behavior with PSI experiences. Given that little is known in the
context of celebrities’ social media use, findings from the current
research are particularly meaningful. This exploratory investigation
calls for more research in this area.
Disclosure statement
No competing financial interests exist.
References
Altman, I., & Taylor, D. A. (1973). Social penetration: The development of interpersonal
relationships. New York: Holt, Rinehart & Winston.
Antheunis, M. L., Tates, K., & Nieboer, T. E. (2013). Patients' and health professionals'
use of social media in health care: motives, barriers and expectations. Patient
Education and Counseling, 92(3), 426e431. http://dx.doi.org/10.1016/
j.pec.2013.06.020.
Bennett, S. (January, 2015). Ellen, 1D and the president e the 3 most retweeted
Tweets in history. Social Times. Retrieved on May 2015 from http://www.
adweek.com/socialtimes/most-retweets-twitter/612320.
Bente, G., Rüggenberg, S., Kr€ amer, N. C., & Eschenburg, F. (2008). Avatar-mediated
networking: increasing social presence and interpersonal trust in net-based
collaborations. Human Communication Research, 34, 287e318. http://
dx.doi.org/10.1111/j.1468-2958.2008.00322.x.
Biocca, F., Harms, C., & Burgoon, J. K. (2003). Towards a more robust theory and
measure of social presence: review and suggested criteria. Presence: Tele-
operators and Virtual Environments, 12, 456e480. http://dx.doi.org/10.1162/
105474603322761270.
Biocca, F., & Nowak, K. (2001). Plugging your body into the telecommunication
system: mediated embodiment, media interfaces, and social virtual environ-
ments. Communication Technology and Society, 407e447.
Brau, M. (2013). Twitter kills the publicity star?: How social media is influencing the
business of celebrity PR. Germany: Grin Verlag.
Caspi, A., & Blau, I. (2008). Social presence in online discussion groups: testing three
conceptions and their relations to perceived learning. Social Psychology Edu-
cation, 11, 323e334. http://dx.doi.org/10.1007/s11218-008-9054-2.
Click, M. A., Lee, H., & Holladay, H. W. (2013). Making monsters: lady Gaga, fan
identification, and social media. Popular Music and Society, 36(3), 360e379.
http://dx.doi.org/10.1080/03007766.2013.798546.
Cohen, J., & Perse, E. M. (2003, May). Different strokes for different folks: an
empirical search for different modes of viewer-character relationships. In Paper
presented at the annual meeting of the international communication association,
san Diego, CA.
Derlega, V. J., Metts, S., Petronio, S., & Margulis, S. T. (1993). Self-disclosure. Newbury
Park, CA: Sage.
Durlach, N., & Slater, M. (2000). Presence in shared virtual environments and virtual
togetherness. Presence: Teleoperators and Virtual Environments, 9(2), 214e217.
Edwards, C., Edwards, A., Spence, P. R., & Westerman, D. (2015). Initial interaction
expectations with robots: testing the human-to-human interaction script.
Communication Studies, 1e12. http://dx.doi.org/10.1080/10510974.2015.1121899.
Eggins, S., & Slade, D. (1997). Analyzing casual conversation. Washington, DC: Cassell.
Evans, C. (2014). Twitter for teaching: can social media be used to enhance the
process of learning? British Journal of Educational Technology, 45(5), 902e915.
http://dx.doi.org/10.1111/bjet.12099.
Fox, J., Osborn, J. L., & Warber, K. M. (2014). Relational dialectics and social
networking sites: the role of Facebook in romantic relationship escalation,
maintenance, conflict, and dissolution. Computers in Human Behavior, 35,
527e534. http://dx.doi.org/10.1016/j.chb.2014.02.031.
Frederick, E. L., Lim, C. H., Clavio, G., & Walsh, P. (2012). Why we follow: an ex-
amination of parasocial interaction and fan motivations for following athlete
archetypes on Twitter. International Journal of Sport Communication, 5(4),
481e502.
Greenberg, L. S., & Safran, J. D. (1987). Emotion in psychotherapy. New York: Guilford
Press.
Hambrick, M. E., Simmons, J. M., Greenhalgh, G. P., & Greenwell, T. C. (2010). Un-
derstanding professional athletes' use of Twitter: a content analysis of athlete
tweets. International Journal of Sport Communication, 3(4), 454e471.
Hargittai, E., & Litt, E. (2011). The tweet smell of celebrity success: explaining
variation in Twitter adoption among a diverse group of young adults. New
Media and Society, 13, 824e842. http://dx.doi.org/10.1177/1461444811405805.
Horton, D., & Wohl, R. R. (1956). Mass communication and para-social interaction:
observations on intimacy at a distance. Psychiatry, 19, 185e206.
IJsselsteijn, W. A., Ridder, H., Freeman, J., & Avons, S. E. (2000). Presence: concept,
determinants and measurement. In , 3959. Proceedings of the SPIE, human vision
and electronic imaging V (pp. 520e529).
Jackson, N., & Lilleker, D. (2011). Microblogging, constituency service and impres-
sion management: UK MPs and the use of Twitter. The Journal of Legislative
Studies, 17, 86e105. http://dx.doi.org/10.1080/13572334.2011.545181.
Jin, S. (2009). Modality effects in second life: the mediating role of social presence
and the moderating role of product involvement. Cyber Psychology & Behavior,
12, 717e721. http://dx.doi.org/10.1089/cpb.2008.0273.
Jin, S. (2011). I feel present. therefore I experience flow: a structural equation
modeling approach to flow and presence in video games. Journal of Broadcasting
& Electronic Media, 55, 114e136. http://dx.doi.org/10.1080/
08838151.2011.546248.
Jin, S., & Park, N. (2009). Parasocial interaction with my avatar: effects of interde-
pendent self-construal and the mediating role of self-presence in an avatar-
based console game, Wii. Cyber Psychology & Behavior, 12, 723e727. http://
dx.doi.org/10.1089/cpb.2008.0289.
Jourard, S. M. (1971). Self-disclosure: An experimental analysis of the transparent self.
New York: Wiley.
Kassing, J. W., & Sanderson, J. (2010). Faneathlete interaction and Twitter tweeting
through the Giro: a case study. International Journal of Sport Communication,
3(1), 113e128.
Lachlan, K. A., Spence, P. R., Lin, X., Najarian, K., & Del Greco, M. (2016). Social media
and crisis management: CERC, search strategies, and Twitter content. Computers
in Human Behavior, 54, 647e652. http://dx.doi.org/10.1016/j.chb.2015.05.027.
Laurenceau, J. P., Barrett, L. F., & Pietromonaco, P. R. (1998). Intimacy as an inter-
personal process: the importance of self-disclosure, partner disclosure, and
perceived partner responsiveness in interpersonal exchanges. Journal of Per-
sonality and Social Psychology, 74(5), 1238e1251. http://dx.doi.org/10.1037/
0022-3514.74.5.1238.
 J. Kim, H. Song / Computers in Human Behavior 62 (2016) 570e577
Lee, K. (2004). Presence, explicated. Communication Theory, 14, 27e50. http://
dx.doi.org/10.1111/j.1468-2885.2004.tb00302.x.
Lee, E. J. (2013). Effectiveness of politicians' soft campaign on Twitter versus TV:
cognitive and experiential routes. Journal of Communication, 63(5), 953e974.
http://dx.doi.org/10.1111/jcom.12049.
Lee, K.-M., & Nass, C. (2005). Social-psychological origins of feelings of presence:
creating social presence with machine-generated voices. Media Psychology, 7(1),
31e45. http://dx.doi.org/10.1207/S1532785XMEP0701_2.
Lee, K. M., Park, N., & Song, H. (2005). Can a robot be perceived as a developing
creature?: effects of a robot's long-term cognitive developments on its social
presence and people's social responses toward it. Human Communication
Research, 31, 538e563. http://dx.doi.org/10.1111/j.1468-2958.2005.tb00882.x.
Lee, K. M., Peng, W., Yan, C., & Jin, S. (2006). Can robots manifest personality?: an
empirical test of personality recognition, social responses, and social presence
in human-robot interaction. Journal of Communication, 56, 754e772. http://
dx.doi.org/10.1111/j.1460-2466.2006.00318.x.
Lin, Y. C. (1992). The construction of the sense of intimacy from every- day social in-
teractions. Rochester, New York: University of Rochester. Unpublished doctoral
dissertation.
Lombard, M., & Ditton, T. B. (1997). At the heart of it all: the concept of presence.
Journal of Computer Mediated Communication, 3(2). http://dx.doi.org/10.1111/
j.1083-6101.1997.tb00072.x.
Lombard, M., & Ditton, T. B. (2000). Measuring presence: a literature-based
approach to the development of a standardized paper-and-pencil instrument.
In A Proceeding at Presence: The third international workshop on presence.
Marshall, P. D. (2010). The promotion and presentation of the self: celebrity as
marker of presentational media. Celebrity Studies, 1(1), 35e48. http://dx.doi.org/
10.1080/19392390903519057.
McGreevy, M. (1993). Virtual reality and planetary exploration. In A. Wexelblat
(Ed.), Virtual reality: Applications and explorations (pp. 163e197). Academic Press
Inc.
Moore, M. G., & Kearsley, G. (2005). Distance education: A systems view (2nd ed.).
New York: Wadsworth.
Morton, T. L. (1978). Intimacy and reciprocity of exchange: a comparison of spouses
and strangers. Journal of Personality and Social Psychology, 36, 72e81. http://
dx.doi.org/10.1037/0022-3514.36.1.72.
Park, S.-H. (2010). Twitter as personal political communities and networked public
sphere. Korean Political Communication Research, 18, 81e115. Retrieved from
http://www.riss.kr/link?id¼A82403373.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, &
Computers, 36, 717e731.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for
assessing and comparing indirect effects in multiple mediator models. Behavior
Research Methods, 40, 879e891. http://dx.doi.org/10.3758/BRM.40.3.879.
Reeves, B., & Nass, C. (1996). The media equation: How people treat computers,
television, and new media like real people and places. New York: Cambridge
University Press.
Reis, H. T. (2007). Steps toward the ripening of relationship science. Personal Re-
lationships, 14(1), 1e23. http://dx.doi.org/10.1111/j.1475-6811.2006.00139.x.
Reis, H. T., & Patrick, B. C. (1996). Attachment and intimacy: component processes.
In E. T. Higgins, & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic
principles (pp. 523e563). New York: Guilford Press.
Richardson, J. C., & Swan, K. (2003). Examining social presence in online courses in
relation to students' perceived learning and satisfaction. Journal of Asynchro-
nous Learning Networks, 7, 68e88.
Rourke, L., Anderson, T., Garrison, D. R., & Archer, W. (2001). Assessing social
presence in asynchronous text-based computer conferencing. Journal of Dis-
tance Education, 14, 1e18.
Rubin, R. B., & McHugh, M. P. (1987). Development of parasocial interaction re-
lationships. Journal of Broadcasting & Electronic Media, 31, 279e292. http://
dx.doi.org/10.1080/08838158709386664.
577
Rubin, A. M., & Perse, E. M. (1987). Audience activity and soap opera involvement a
uses and effects investigation. Human Communication Research, 14(2), 246e268.
Rubin, A. M., Perse, E. M., & Powell, R. A. (1985). Loneliness, parasocial interaction,
and local television viewing. Human Communication Research, 12, 155e180.
http://dx.doi.org/10.1111/j.1468-2958.1985.tb00071.x.
Sanderson, J. (2008). You are the type of person that children should look up to as a
hero: parasocial interaction on 38pitches.com. International Journal of Sport
Communication, 1(3), 337e360.
Sanderson, J. (2009). Thanks for fighting the good fight: cultivating dissent on
blogmaverick.com. The Southern Communication Journal, 74, 390e405. http://
dx.doi.org/10.1080/10417940802650153.
Sanderson, J. (2011). To tweet or not to tweet: exploring division I athletic de-
partments' social-media policies. International Journal of Sport Communication,
4, 492e513.
Savage, M. E., & Spence, P. R. (2014). Will you listen? an examination of parasocial
interaction and credibility in radio. Journal of Radio & Audio Media, 21(1), 3e19.
http://dx.doi.org/10.1080/19376529.2014.891214.
Schiappa, E., Allen, M., & Gregg, P. B. (2007). Parasocial relationships and television:
a meta-analysis of the effects. Mass Media Effects Research: Advances Through
Meta-Analysis, 301e314.
Schroeder, R. (2006). Being there together and the future of connected presence.
Presence: Teleoperators and Virtual Environments, 15(4), 438e454. http://
dx.doi.org/10.1162/pres.15.4.438.
Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommuni-
cation. London: Wiley.
Small, T. (2010). Canadian politics in 140 characters: Party politics in the Twitter-
verse. Canadian Parliamentary Review, 33, 39e45. Retrieved from http://www.
revparl.ca/english/issue.asp?param¼199&art¼1397.
Song, H., Kim, J., & Lee, K. M. (2014). Virtual body vs. real body in exergames:
reducing social physique anxiety in exercise experiences. Computers in Human
Behavior, 36, 282e285. http://dx.doi.org/10.1016/j.chb.2014.03.059.
Spence, P. R., Lachlan, K. A., Edwards, A., & Edwards, C. (2016). Tweeting fast mat-
ters, but only if I think about it: information updates on social media.
Communication Quarterly, 1e17. http://dx.doi.org/10.1080/
01463373.2015.1100644.
Spence, P. R., Lachlan, K. A., Lin, X., & del Greco, M. (2015). Variability in Twitter
content across the stages of a natural disaster: implications for crisis commu-
nication. Communication Quarterly, 63(2), 171e186. http://dx.doi.org/10.1080/
01463373.2015.1012219.
Spence, P. R., Westerman, D., Edwards, C., & Edwards, A. (2014). Welcoming our
robot overlords: initial expectations about interaction with a robot. Communi-
cation Research Reports, 31(3), 272e280. http://dx.doi.org/10.1080/
08824096.2014.924337.
Stever, G. (2009). Parasocial and social interaction with celebrities: classification of
media fans. Journal of Media Psychology, 14(3), 1e39.
Stever, G. S., & Lawson, K. (2013). Twitter as a way for celebrities to communicate
with fans: implications for the study of parasocial interaction. North American
Journal of Psychology, 15(2), 339e354.
Swan, K., & Shih, L. F. (2005). On the nature and development of social presence in
online course discussions. Journal of Asynchronous Learning Networks, 9(3),
115e136.
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Boston:
Allyn & Bacon.
Tu, C. (2000). On-line learning migration: from social learning theory to social
presence theory in a CMC environment. Journal of Network and Computer Ap-
plications, 23, 27e37. http://dx.doi.org/10.1006/jnca.1999.0099.
Wheeless, L. R. (1978). A follow-up study of the relationships among trust, disclo-
sure, and interpersonal solidarity. Human Communication Research, 4(2),
143e157. http://dx.doi.org/10.1111/j.1468-2958.1978.tb00604.x.
Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: a
presence questionnaire. Presence: Teleoperators and Virtual Environments, 7(3),
225e240. http://dx.doi.org/10.1162/105474698565686.