SHARING THE POSITIVE OR THE NEGATIVE?
UNDERSTANDING THE CONTEXT, MOTIVATION AND
CONSEQUENCE OF EMOTIONAL DISCLOSURE ON
FACEBOOK
LIN HAN
LIN HAN
School of Humanities and Social Sciences
A thesis submitted to the Nanyang Technological University
in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
2014
Abstract
The recent emergence of Social Networking Sites (SNSs), such as Facebook,
provides substantial opportunities for emotional disclosure. A relevant and timely
question arises concerning how such a context alters as well as supports emotional
disclosure. Three studies were conducted to address this issue. Study 1 compares
individuals’ emotional disclosure in the context of Facebook versus diary writing and
real life. Results show that emotional disclosure on Facebook is more frequent than in
diaries, but less frequent than in real life, supporting the notion that emotional
disclosure is determined by the sociality of the context. Using a social network
analysis, Study 2 further tests how emotional disclosure is associated with social
network properties—network size and density. It further demonstrates that network
size is associated with disclosure of positive emotions because of need for impression
management, while network density is associated with both disclosure of positive
emotions and disclosure of negative emotions through the need for emotional
expression. Study 3 extends the concerns to the influence of emotional disclosure on
well-being. Results reveal that emotional disclosure is related to fulfillment of both
needs. Fulfillment of need for impression management accounts for the positive
association between positive emotional disclosure and well-being, whereas fulfillment
of neither need mediates the negative association between negative emotional
disclosure and well-being. The set of studies demonstrates how the motivational needs
play roles in the associations between emotional disclosure, context, and well-being.
Altogether, this research offers a compelling account of emotional disclosure in a
complicated social context and sheds light on leveraging SNSs for well-being.
i
ACKNOWLEDGEMENTS
I would like to express my deepest gratitude to a number of individuals who
play a role during the completion of this dissertation. First, I would like to thank my
advisor, Dr. Qiu Lin, for guiding me to this exciting field of research on social media.
The four and half years’ training has given me the courage and endurance for
academic pursuit.
Meanwhile, I am highly appreciative of the reviewers for their constructive
suggestions that have improved this dissertation substantially. I am also very grateful
for the oral defense committee, Dr. Xu Hong and Dr. Ito Kenichi. I would sincerely
thank the Chair, Dr. Wan Ching, who gave me insights from the beginning of writing
research proposal to the end of the completing this dissertation. I am also thankful to
the other two members in my confirmation examination panel, Dr. Rebecca Ang and
Dr. Venus Sau-lai Lee; their valuable opinions made this dissertation possible.
In addition, the completion of this dissertation could not be possible without
the assistance of Pearlyn Kwang, Cassie Yung, Meiryl Stevianna Rusil, and Cao
Zhiguang, who helped me in data collection and pre-processing. I am also indebted to
my friends and lab mates. Thank them for being there in every possible way. Special
thanks should go to Chen Luxi, Wu Chiao-Yi, Tan Chee-Seng, and Tan Kit Aun, for
their prompt company, definite support, and continuous encouragement during my
confusion and distress.
Last but not least, I am extremely grateful for my mother, for her unwavering
faith in me, providing me the liberty to pursue my dreams, and supporting me
unconditionally. This dissertation is dedicated to her.
ii
TABLES OF CONTENTS
CHAPTER 1: INTRODUCTION ................................................................................... 1
Emotional Disclosure and Social Networking Sites ........................................... 4
Need for Emotion Expression and Need for Impression Management .............. 7
Emotional Disclosure and Well-being .............................................................. 15
Overview of Studies.......................................................................................... 19
CHAPTER 2: STUDY 1—PATTERNS OF EMOTIONAL DISCLOSURE ON
FACEBOOK ................................................................................................................. 21
Study 1a ............................................................................................................ 21
Study 1b ............................................................................................................ 29
General Discussion ........................................................................................... 34
CHAPTER 3: STUDY 2—EMOTIONAL DISCLOSURE AND SOCIAL NETWORK
PROPERTIES ............................................................................................................... 37
A Social Network Perspective .......................................................................... 37
Network Structure and Emotional Disclosure .................................................. 39
The Present Study ............................................................................................. 43
Study 2a ............................................................................................................ 43
Study 2b ............................................................................................................ 46
General Discussion ........................................................................................... 56
CHAPTER 4: STUDY 3—EMOTIONAL DISCLOSURE AND WELL-BEING ...... 60
A Motivational Account for Well-being ........................................................... 61
Fulfillment of The Two Needs .......................................................................... 63
The Present Study ............................................................................................. 65
Method .............................................................................................................. 65
Results ............................................................................................................... 67
iii
Discussion ......................................................................................................... 71
CHAPTER 5: GENERAL DISCUSSION .................................................................... 76
Disclosure as a Contextual Behavior ................................................................ 77
Relate Emotional Disclosure to Well-being ..................................................... 80
Revisit the Need for Emotional Expression and Need for Impression
Management ...................................................................................................... 83
Implications ...................................................................................................... 87
Limitations and Future Directions .................................................................... 88
Conclusion ........................................................................................................ 89
REFERENCES ............................................................................................................. 91
NOTES........................................................................................................................ 125
TABLES ..................................................................................................................... 128
FIGURES .................................................................................................................... 136
Appendix A: Replaced text emoticons with corresponding emotional words
(Study 1a) ........................................................................................................ 142
Appendix B: Principal components analysis using varimax rotation
(Study 2b) ....................................................................................................... 144
Appendix C: Scale of need fulfillment on Facebook (Study 3) ...................... 145
iv
List of Tables
Table 1. (Study 2a) Descriptive statistics and bivariate correlation between variables
.................................................................................................................................... 128
Table 2. (Study 2a) Regress network properties on emotional words. ...................... 129
Table 3. (Study 2b) Descriptive statistics and bivariate correlation between variables.
.................................................................................................................................... 130
Table 4. (Study 2b) Indirect effects of network properties on positive emotional
disclosure through proposed mediators. ..................................................................... 131
Table 5. (Study 2b) Indirect effects of network properties on negative emotional
disclosure through proposed mediators. ..................................................................... 132
Table 6. (Study 3) Descriptive statistics and bivariate correlation between variables
.................................................................................................................................... 133
Table 7. (Study 3) Indirect effects of disclosing positive emotions on well-being
through proposed mediators........................................................................................ 134
Table 8. (Study 3) Indirect effects of disclosing negative emotions on well-being
through proposed mediators........................................................................................ 135
v
List of Figures
Figure 1. (Study 1a) Frequency of emotional words in Facebook status updates and
diaries. ......................................................................................................................... 136
Figure 2. (Study 1b) Likelihood of disclosing positive emotions vs. negative emotions
of Facebook and in real life. ....................................................................................... 137
Figure 3. (Study 2b) Coefficients representing effects of network size (Panel A) and
network density (Panel B) on mediators and disclosure of positive emotions.. ......... 138
Figure 4. (Study 2b) Coefficients representing effects of network size (Panel A) and
network density (Panel B) on mediators and disclosure of negative emotions.. ........ 139
Figure 5. (Study 3) Coefficients representing effects of disclosing positive emotions on
Life satisfaction (Panel A) and loneliness (Panel B) and proposed mediators... ........ 140
Figure 6. (Study 3) Coefficients representing effects of disclosing negative emotions
on Life satisfaction (Panel A) and loneliness (Panel B) and proposed mediators..... 141
vi
CHAPTER 1: INTRODUCTION
Emotional disclosure is prevalent in everyday life (Moreno et al., 2011; Rimé,
Finkenauer, Luminet, Zech, & Philippot, 1998; Rimé, Mehdizadeh, Philipport, &
Boca, 1991). When one encounters an event that induces emotional reaction, he or she
tends to talk with others about the emotional experience (Rimé et al., 1991). Social
networking sites (SNSs) open novel avenues for individuals to reveal themselves and
interact with others (
ler
iedl
etter
eimeister
rcmar
). Emotions
that formerly might have passed unremarked can now be easily recorded and shared.
It has been shown that emotional disclosure is ubiquitous on SNSs (Carr, Schrock, &
Dauterman, 2012; Kivran-Swaine & Naaman, 2011; Naaman, Boase, & Lai, 2010).
The overall pattern of emotional expressions is consistent with seasonal mood changes
(Golder & Macy, 2011), suggesting it is a naturally occurring behavior.
Facebook, one of the most popular SNSs, offers such an opportunity to study
emotional disclosure that may be different from that in traditional contexts.
Individuals on Facebook engage in various activities: updating status, posting photos,
clicking “likes ” giving comments, sharing videos, and as such. Among these
activities, revealing emotions is an integral and important element. On one hand,
social connections induce emotional disclosure (Rimé, 1998, 2009); on the other hand,
emotional disclosure in turn establishes greater social connections than other types of
self-disclosure (Laurenceau, Barrett, & Pietromonaco, 1998). It has shown that people
frequently disclose their emotional experiences on Facebook by updating statuses,
posting photos, and making comments (Moreno et al., 2011). The popularity of
Facebook and intensive emotional traces on it indicates that many individuals have
integrated Facebook into their communication of emotions. In recent years, there is an
emerging interest on emotional disclosure on Facebook (e.g., Kramer & Chang, 2011;
1
Schwartz et al., 2013). However, these studies studied did not examine the Facebook
context particularly.
Importantly, Facebook is a social context that differs from previous
communication channels. It is an online environment without physically presence of
audience yet have full control on the information to be shown (Walther, 2007); the
information on it is public or semi-public, visible to massive people from diverse
social circles (Boyd, 2011; Boyd & Ellison, 2007). Concerning the novelty of
Facebook context, questions arise pertaining to how the social connections on
Facebook influence individuals’ emotional disclosure and how the emotional
disclosure on Facebook relates to everyday wellness.
Barker (1968) suggested that most behaviors are not specific to the individual;
rather, they are related to the context. The presence of audience can either facilitate or
hinder an individuals’ emotional disclosure depending on the emotional content and
the interpersonal relationships between the discloser and audience (Buck, Losow, &
Murphy, 1992; Fischer, Manstead, & Zaalberg, 2003; Jakobs, Manstead, & Fischer,
1999a). To account for the relation between social context and emotional disclosure,
social motives are suggested to be the underlying processes (Fridlund, 1991a, 1994,
1997). Two key motives are highlighted in related research: the motive to let others
know (Fridlund, 1991b; Fridlund et al., 1990; Jakobs et al., 1999a) and the motive to
be perceived as good (Fischer, Manstead, Evers, Timmers, & Valk, 2004; Jakobs,
Manstead, & Fischer, 2001). In view of this, I conceptualized the two motives as the
need for emotional expression and the need for impression management.
Meanwhile, a common notion in media research holds that active engagement
in communication on Facebook is associated with greater well-being (Ahn & Shin,
2013; Burke, Marlow, & Lento, 2010; große Deters & Mehl, 2013). Some studies of
2
self-disclosure on Facebook have illustrated a potential link between emotional
disclosure and well-being (Denti et al., 2012; große Deters & Mehl, 2013). Since
fulfillment of different needs is essential in fostering well-being (e.g., Church et al.,
2013; Diener & Lucas, 2000; Ryan, 1995; Ryan & Deci, 2000; Tay & Diener, 2011), it
is plausible that emotional disclosure on Facebook is related to well-being through
fulfilling the motivational needs.
In light of above speculations, the purpose of this research is three-fold. First,
it aims to examine the distinctive pattern of emotional disclosure in Face ook’s
context relative to disclosure through other traditional channels. Second, it aims to
investigate the link between the contextual factors on Facebook and emotional
disclosure by disentangling the motivational factors underlying the relationships.
Finally, it aims to reveal the consequence of emotional disclosure in relation to the
fulfillment of corresponding motivational needs. Findings from this research are
expected to shed light on the social process of emotional disclosure on Facebook and
enrich the knowledge of media use.
In the following, I will introduce the theoretical background of the present
research. I will first review related theories and studies on emotional disclosure. In
particular, I will introduce Facebook as a vehicle for emotional disclosure by
highlighting its contextual characteristics. Then, I will propose the need for emotional
expression and the need for impression management as two crucial motives in
understanding emotional disclosure on Facebook. Following that, I will introduce
related work on the association between emotional disclosure and well-being. Finally,
I will present an overview of three studies designed to examine emotional disclosure
on Facebook in terms of its contextual contingency, motivational needs, and
psychological outcomes.
3
Emotional Disclosure and Social Networking Sites
Emotional disclosure is regarded as a form of self-disclosure (Hackenbracht &
Gasper, 2013), which essentially is “the process of making the self be known to
others” (Jourard & Lasakow, 1958, p. 91). Researchers posit that self-disclosure
consists of three layers regarding the disclosed content and depth (Altman & Taylor,
1973). The peripheral layer contains the most superficial content, such as biographical
information; the intermediate layer is about attitudes and opinions; and the core layer
is the deepest one, including information related to internal values, needs, and fears.
According to this classification, emotional disclosure should fall in the intermediate to
core layers, as it includes not only the emotional event itself but also the feelings about
the emotional experiences (Luminet, Bouts, Delie, Manstead, & Rime, 2000;
Pennebaker, 1997; Pennebaker, Zech, & Rimé, 2001). Different from descriptive selfdisclosure that includes information merely a out one’s facts in the first layer (Reis &
Shaver, 1988), emotional disclosure carries deeper revelation of self than nonaffective dimension of the disclosure (Chelune, 1975).
Facebook can be regarded as a social context where emotional disclosure is
likely to occur. For one thing, an individual are allowed to form connections with
many others on Facebook. All the contacts that one can reach on Facebook are called
“friends ” no matter whether they are known personally or not. Although invisible,
they are potential audience (Boyd, 2007). For the other thing, individuals are able to
interact with each other on Facebook. Individuals browse the content (including texts,
pictures, and videos) posted by others; meanwhile, their own posts are viewed by and
draw comments from others.
Nevertheless, Facebook context shows distinctive characteristics, compared to
traditional communicative contexts. First, an individual on Facebook often discloses
4
to a wide range of networked publics (Boyd & Ellison, 2007) without directing to a
particular one (Bazarova, Taft, Choi, & Cosley, 2013). In other words an individual’s
activities on Facebook can be observed by a large number of potential audiences.
Second, friends on Facebook are usually from different social circles, ranging
from intimate friends to mere acquaintances or even distant strangers (Boyd & Ellison,
2007; Gilbert & Karahalios, 2009). These social connections are rarely present
simultaneously in offline environments, where different social circles are separated
and individuals can manage their disclosure strategies respectively. However, these
social circles appear together on Facebook to form a collapsed context (Baym & Boyd,
2012). In a collapsed context, individuals are more likely to use a strategy called “the
lowest common denominator culture” (Hogan, 2010b). It refers to the practice
dictating that one considers not only the intended recipients but also those who are not
the intended targets, yet are able to view disclosed content. For instance, an individual
may post a Facebook status update that is not purposely for a superior. However, if
the superior is a Facebook friend of that individual, the user needs to consider the
possibility that the message might be read by the superior.
Third, as in a computer-mediated communication (CMC), direct nonverbal
cues (e.g., visual expression, auditory tone, and gestures) are lacking on Facebook.
According to social information processing theory (Walther, 1992), individuals would
like to disclose more as a compensation of the deficit in communicative cues (Hu,
Wood, Smith, & Westbrook, 2004; Joinson, 2008; Leung, 2002; Tidwell & Walther,
2002; Valkenburg & Peter, 2007). This explains why more intimate communication
are found in CMC than in face-to-face settings (Gibbs, 2006; Peter, Valkenburg, &
Schouten, 2006). Meanwhile, the invisible and asynchronous communication lacks
immediate and direct responses, which may also cause uncertainty. As individuals are
5
prone to reducing uncertainty and increasing predictability in a situation (Berger &
Calabrese, 1975), they might tend to use more direct and intimate questions and
details in order to obtain more information and reduce the uncertainty in CMC.
Indeed, intensive use of punctuation, words, emoticons, and explicit statements of
emotion were found in online text-based communication , and these were seen as an
effective way to communicate emotions (Hancock, Landrigan, & Silver, 2007).
Fourth, the communication on Facebook is hyperpersonal, because disclosers
could have full control over the information that is to be communicated and can easily
manage their online self-presentation (Walther, 1996, 2007). In face-to-face
communication, disclosers have to respond to the social situation rapidly; hence,
instinctive responses like facial expressions, body gestures, and vocal tones may not
be consciously controlled. In contrast, individuals on Facebook can freely decide what
and how to reveal information about themselves before they really send the
information off. Generally, the greater control one has in the virtual context allows
information on Facebook to be more sophisticatedly selected and refined. For
instance, college students often display pictures showing socialization with their
friends (Zhao, Grasmuck, & Martin, 2008) and choose their most attractive photos as
profile pictures (Siibak, 2009; Strano, 2008; Walther, Heide, Kim, Westerman, &
Tong, 2008). In such a manner, the hyperpersonal interaction on Facebook enables
individuals to easily establish and maintain a socially desirable self (Newman,
Lauterbach, Munson, Resnick, & Morris, 2011; Walther et al., 2008).
In sum, Facebook constitutes an environment that is rarely seen in other
context. Past research has widely acknowledged that the context can affect what and
how the emotion is disclosed (Gross & John, 1995). In an earlier experiment
(Fridlund, 1991b), participants were invited to the lab alone or with a friend. They
6
were introduced to watch a humorous film while their facial expressions were
recorded, together with their self-reported happiness. Results showed that individuals’
smiling was stronger in the condition where they were together with the friend than in
the solitary condition. Moreover, the smiling strength did not change with their selfreported happiness. The facilitative effect of presence of others has been supported by
a series of studies as well (Jakobs et al., 1999a, 1999b). Furthermore, the presence of
close friends has stronger facilitative effect than that of strangers’ (Jakobs et al.,
1999a; Wagner & Smith, 1991). Similar findings were also shown in verbal sharing of
one’s emotions with others. The presence of audience could either facilitate or inhibit
an individuals’ emotional disclosure depending on the emotional content and the
interpersonal relationships between the discloser and audience (Buck, Losow, &
Murphy, 1992).
Given the distinctive context on Facebook, it is conceivable that Facebook may
bear on a particular pattern of emotional disclosure. Therefore, an empirical study on
emotional disclosure regarding the way it fits Facebook context is called for. This
leads to my first research question:
Research Question 1: How is the pattern of emotional disclosure in Face ook’s
context different from the pattern that appears on other channels?
Chapter 2 will provide direct evidence on this issue by comparing emotional
disclosure on Facebook with disclosure in semi-private writing and daily offline
communication.
Need for Emotion Expression and Need for Impression Management
On an individual level, the context on Facebook is defined by one’s social
contacts that he or she connects with. The specific context may be endorsed with
facilitators and constraints that determine what individuals will disclose and how to
7
disclose (Saxena & Mehrotra, 2010). It is intriguing to delineate individuals’ social
context and examine how a particular context enhances or inhibits emotional
disclosure. In regard to “enhance or inhi it ” the underlying motivation could e a key
process through which the environment influences the disclosure behaviors.
Among the studies on emotional disclosure, social motives are postulated to
account for the effect of a social context (Fridlund, 1991a, 1991b, 1994). Specifically,
social motives are referred to as motivations to communicate or to predict ones’
probable behavior through emotional expression (Fridlund, 1991a, 1997). However,
the components of social motives are not clearly defined and explicitly measured
(Manstead, Fischer, & Jakobs, 1999); it covers almost every motive that is related to
communicative inclinations, such as the urge to talk, the motive to be helped, and to
seek comfort (Fischer et al., 2003; Parkinson, 2005). The theory of social sharing of
emotion (Rimé, 2009; Rimé et al., 1998; Rimé et al., 1991) also conceives that desire
for social interaction drives emotional disclosure. However, it did not explicitly
measure such a motivation either.
Meanwhile, there are instances that individuals suppress their emotions. For
example, in Jakobs et al.’s (2001) study, sadness was induced by watching film clips,
and individuals expressed less sadness in social conditions than in a solitary context.
The concept of “display rules ” therefore, is highlighted to explain the general
tendency for inhibiting negative emotions when others are present. Display rules are
first proposed by Ekman and Friesen (1975) as the appropriateness in a society that
individuals need to manage emotions to align them in accordance with. Individuals
are assumed to learn to display or not to display particular emotions in a given social
setting. According to the view of social motives, display rules are supposed to activate
certain social motives that one expects to establish or maintain certain impression in a
8
given circumstance. Jacobs and colleagues (1999) attempted to interpret such a social
motive as the need to adapt one’s emotional expression to adhere to the norms in the
given context.
To assume an appropriate role in a particular context and display emotions
accordingly is an important facet of social motives. In a situation where a particular
emotion is considered favorable, social motives lead to greater amount of emotion
expression; in contrast, in a situation where the same emotion is not considered
appropriate, social motives lead to suppression. For instance, individuals may express
negative emotions such as sadness and anger, when friends are around and the friends
are expected to provide social support (Zeman & Garber, 1996). Supposing when a
student speaks in front of a faculty member, he or she would be likely to be displaying
his or her competence. Whereas later, when the student meets up with friends, he or
she would be more likely to talk about the anxiety during the former conversation
(Omarzu, 2000). A study showed that the display rule is an antecedent of social
motives (Zaalberg, Manstead, & Fischer, 2004). In other words, social motives entail
enhancing or suppressing one’s expression, depending on the display rules endorsed in
the specific setting. In either case, one of the motives that could determine emotional
disclosure is to maintain favorable impression in front of others. Fischer et al. (2004)
also identified the motive of impression management as one of the crucial social
motives along with other motives of sharing.
As such, the motive to express and the motive to play good both occur at an
equivalent dimension. Congruence takes place when the disclosers’ expressive need is
in agreement with the need to maintain desirable impression; whereas conflicts
probably arise when the two needs lead to different tendencies of disclosure
(Parkinson, 2005). Take an example to illustrate the coexistence of the two motives.
9
Buck et al. (1992) contended that the presence of a distant friend, compared to a close
friend, would lessen an individual’s emotional expression. One possible reason is the
presence of close friend could elicit the desire to share with the friend, while distant
friend could not elicit such a desire. An alternative reason can be distant friends may
be more sensitive to inappropriate behaviors than close friends, so their presence may
evoke the discloser’s concern about impression. To avoid the risk of being
inappropriate, the discloser may opt to disclose less.
From above review, it is evident that there are two key factors of social
motives. In fact, Fridlund (1994) mentioned that revealing sadness may be driven by
the need for help in some circumstances but suppressed by other motives (e.g., need
for impression management) in other situations. In Jakobs et al.’s (2001) studies,
social motives are further specified as the concern about others’ evaluation and the
desire to communicate. The two different pathways are more explicitly addressed by
Parkinson (2005): The expression of positive emotions is mediated by the motive to
share on one hand, and on the other hand, inhibition of negative emotions is mediated
by the motive to conform to the inhibitory display rules.
It is also suggested that individuals on Facebook engage in both self-expressed
and self-censored disclosure. Individuals attempt to balance between what they want
to express and what is desirable for the public; sometimes one need might outweigh
the other. An interview with Facebook users found that individuals posted status
updates after experiencing frustration and felt relief from the posting behaviors (Zhao
et al., 2013). However, later on they worried that the messages might be
misunderstood or criticized, so they may delete the messages to maintain self-image.
In another study, individuals in a health community on Facebook expressed a strong
need to share their health-related information in hopes of getting emotional support
10
and advice from others. Nevertheless, they hesitated to do so because they wanted to
avoid giving a negative impression to others (Newman et al., 2011). It demonstrates
that two motivational needs may jointly influence individuals’ expression on
Facebook.
Therefore, I conceptualize the two factors of social motives as the need for
emotional expression and the need for impression management. Specifically, the need
for emotional expression refers to the tendency of sharing and letting others know,
which is similar to the motive that is mostly highlighted in Fridlund’s (1991a, 1991b,
1994) theory and Rimé et al.’s (1991) social sharing of emotion. The need for
impression management refers to the motivation to convey a favorable self-image in
the presence of others (Leary, Allen, & Terry, 2011; Leary & Kowalski, 1990; Martin,
Leary, & Rejeski, 2000). In the following, I will elaborate on the processes of the two
needs.
First, both needs have an innate readiness. It is suggested that self-disclosure
could be driven by a basic and unconscious impulse for validation, and people do so
by maintaining conversations with others (Giles, Coupland, & Coupland, 1991). A
recent study (Tamir & Mitchell, 2012) found that individuals were more willing to
forgo money for self-disclosure than for talking about others. Neuroimaging results
further showed that participants’ rain regions that are related to rewards exhibit
stronger activation in the self-disclosure condition than in the gaining money
condition. It is concluded that human beings have an inherent tendency to express
themselves and this behavior is self-rewarding. At the same time, research has long
identified that humans by nature are prone to engaging in impression management to
maintain a positive image for others (Goffman, 1959; Leary & Kowalski, 1990;
Schlenker & Weigold, 1992). The emphasis on impression management is considered
11
a critical need for social survival (Hogan, Jones, & Cheek, 1985), and it takes place in
daily life through dressing, expression, gestures, and as such (Goffman, 1959). Even
under stress, individuals attempt to avoid additional rejection and disapproval by
presenting themselves in a positive manner; they tend to engage in more pleasant
expression in order to entertain the audience (Renner & Laux, 2004).
The second and the most important feature of the two needs is their
contingency on the context. The need for emotional expression is enhanced in
situations where people can acquire affirmation and comfort (Zeman & Garber, 1996).
It is found that inducing an individual’s anxiety in a la oratory can elicit the desire to
communicate with others, suggesting that social interaction is an important motive of
emotional disclosure (Moscovici, 1984). A recent study also demonstrated that the
degree to which an individual is free to form or terminate relationships with others
(i.e., relational mobility) significantly influences the tendency of the individual to
engage in self-disclosure with close friends. Specifically, individuals who perceive a
higher degree of relational mobility in their environment are motivated to show social
commitment and to reinforce relationships through self-disclosure, compared to
individuals perceiving a lower relational mobility (Schug, Yuki, & Maddux, 2010).
These findings suggest that the need for emotional expression is determined by
interpersonal relations in the context. A high level of need for emotional expression is
supposed to promote sharing of genuine emotions
Meanwhile, how strong the need for impression management is also contingent
on the social settings. For example, in service-oriented industry, positivity is highly
encouraged while negative emotions are strictly prohibited (Parkinson, 1991). Some
other settings, such as family or friendship, may have a greater tolerance of negativity,
leading to a low need for impression management. Also, the impression motivation is
12
greatly affected by the publicity, and how likely one’s ehavior can e o served and
how many others might observe (Leary & Kowalski, 1990). The more public the
environment is, the more motivated the individuals will be to manage impression. In
contrast, the need for impression management is low in a solitary setting (Arkin,
Appelman, & Burger, 1980). As the presence of audience plays a crucial role in
triggering one’s impression management, greater positive impression is observed in
face-to-face interview than in private (Schlenker, Dlugolecki, & Doherty, 1994). In
short, the audience in a context poses a need for impression management that may
facilitate sharing of favorable emotions while inhibiting expression of unfavorable
ones.
To summarize, both the need for emotional expression and the need for
impression management come from deep impulse within, but they are manifest in
facilitative or suppressive emotional disclosure depending on the socially constructed
context. The two needs are divergent at the instrumental purpose. The need for
emotional expression points to social attachment that could console emotions.
Differently, the need for impression management focuses on social acceptance that is
to construct self-competence and avoid social sanctions. The need for emotional
expression is always a facilitative drive for emotional disclosure, whereas the need for
impression management would have different impact depending on the favorability of
the emotion in the given context.
As aforementioned, the two motivational needs are manifest on Facebook, too.
Expression on Facebook is ubiquitous and one can share experiences with others,
allowing the need for emotional expression to be met. Indeed, expressing oneself has
been addressed as one of the important motivations for using online services (Jung,
Youn, & McClung, 2007; Papacharissi & Rubin, 2000). In Andre’s (
13
) study an
application was developed for users to report emotional status on an SNS by simply
clicking certain icons. User evaluation showed that people considered this application
useful for self-reflection and subtle communication, suggesting that people have a
propensity to disclose their emotional state during online networking activities.
Studies also suggested that those who actively express their emotional experience
could improve their social connectedness (Kobler et al., 2010; Sheldon, Abad, &
Hinsch, 2011), which reinforces one’s motivation to express. In contrast, there are
also individuals who lurk on Facebook without disclosing themselves. Research
indicates that lurking is related to low relational intimacy between friends in the
networks (Rau, Gao, & Ding, 2008). Henceforth, it is conceivable that the social
relations one has could influence the need for express oneself on Facebook, with
strong social connections enhancing the need while week ones decreasing it.
Regarding the need for impression management, it is also a major motivation
to use Facebook (Nadkarni & Hofmann, 2012). Publicity is high on Facebook, as each
post is visible to multiple friends from various social circles. In order to maintain a
positive self-image in public, individuals present themselves in a selective and
desirable manner on Facebook (Barash, Ducheneaut, Isaacs, & Bellotti, 2010;
Bazarova et al., 2013; Papacharissi, 2011). Further, the complex publicity on
Facebook makes impression management more complicated (Leary et al., 2011).
However, while the content is not catered for every single audience, it is expected to
be appropriate to all the potential viewers while remaining appealing to important
ones according to “the lowest common denominator culture” (Hogan, 2010b). As a
result, the coexistence of different audiences decontextualizes the communication and
requires individuals to therefore display one single and desirable image to all friends.
It is thus speculated that the strength of the impression motivation depends on the
14
possi le oundary of one’s network. If the network includes audience with lower
requirement of impression, the motivation would not be as strong as in a network
where audience has higher requirement of impression.
All in all, the two needs are social and communicative, varying with the
particular context. As individuals on Facebook differ in their network compositions,
individuals’ motivations would e strengthened or weakened differently, leading to
different disclosure tendency. Thus, the context effect can be understood on the
individual level by taking into motivational needs into account. Therefore, my second
research question is:
Research Question 2: How do the need for emotional expression and the need
for impression management play a role in the association between personal social
network context and emotional disclosure on Facebook?
Chapter 3 will elaborate on the association between personal context on
Facebook and emotional disclosure from a perspective of social network analysis. A
study will examine how the two proposed needs mediate this association.
Emotional Disclosure and Well-being
In addition to the concern about the contextual and motivational account for
emotional disclosure on Facebook, there is a growing attention on well-being in
Facebook research. Ellison, Steinfield, and Lampe (2007) conducted a survey among
college students and found the intensity of Facebook use is correlated with increasing
social capital. In another study, positive self-disclosure is found to correlate with
greater life satisfaction, because the disclosure increases perceived social support (Kim
& Lee, 2011). In a field experiment, participants who were encouraged to post more
status updates than usual showed decreasing loneliness (große Deters & Mehl, 2013).
It appears that using Facebook benefits well-being. The findings are inconclusive,
15
however, due to the diverse facets of Face ook use. Some studies focused on users’
interaction pattern (Burke et al., 2010), while some concerned only the length of time
spent on Facebook (Kalpidou, Costin, & Morris, 2011). Some studies looked into
overall usage (Ellison et al., 2007; Kalpidou et al., 2011; Valenzuela et al., 2009),
while some examined specific behaviors, such as commenting or browsing (Chen &
Lee, 2013; große Deters & Mehl, 2013;Qiu, Lin, & Leung, 2010). It is unclear which
aspect of Facebook activities is really related to well-being and how the benefit has
been established.
A common notion concerning Facebook usage and well-being is that only
active engagement in communication on Facebook is associated with greater wellbeing (Ahn & Shin, 2013; Burke et al., 2010; große Deters & Mehl, 2013). For
example, direct communication with friends on Facebook was found to correlate with
greater life satisfaction and less loneliness (Burke et al., 2010), while browsing
Facebook showed no such effect. In addition, the amount of self-disclosure in SNSs
was positively related to life satisfaction (Lee, Lee, & Kwon, 2011). Given that
emotional disclosure is an active behavior, it is plausible that emotional disclosure is
associated with greater well-being.
Few studies have been done on the relationship between emotional disclosure
on Facebook and well-being, but insight can be gained from other lines of research.
Studies on expressive writing have suggested that writing about emotions is predictive
of positive health outcomes (Greenberg & Stone, 1992; Pennebaker & Seagal, 1999).
On the contrary, suppressing emotions may be detrimental to psychological health
(King & Pennebaker, 1998). Individuals who tend to withhold their emotion are more
likely to experience low well-being (DeNeve & Cooper, 1998).
Particularly, sharing positive events with others is considered capitalization
16
(Langston, 1994), because the process produces greater positive affect beyond the
pleasure induced by the events. Individuals who share positive emotions with others
can experience prolonged hedonic feelings (Mauss et al., 2011; Tugade & Fredrickson,
2007). Psychophysical experiments illustrate that a higher level of somatic activity
and amusement follows the action of disclosing positive emotions, while a lower level
of psychophysical reactions is associated with suppressing the positive emotions
(Gross & Levenson, 1997). In the long run, capitalization on positive emotions is
positively associated with everyday positive affect and life satisfaction (Gable & Reis,
2010; Gable et al., 2004). With regards to interpersonal relationship, it is found that
disclosing positive emotions increases a person’s lika leness (Collins & Miller, 1994)
and strengthens the person’s social bond (Peters & Kashima, 2007). Other studies also
demonstrate that sharing positive emotions enhances the satisfaction of interpersonal
relationship and social connectedness (Gable & Reis, 2010; Gable et al., 2004; Mauss
et al., 2011).
Concerning the consequence of disclosing negative emotions, however, the
results are complicated. When individuals experience negative events, disclosing the
distress is perceived as a way to gain relief (Rimé, 2009). Writing about negative
emotions is used as an effective way to cope with negative affect (Pennebaker, 1997),
such as fear (Langens, 2005), traumatic stress (Greenberg & Stone, 1992), and
depression (Radcliffe, Lumley, Kendall, Stevenson, & Beltran, 2010). Pennebaker
and Beall (1986) found profound effects on participants who wrote about both events
and feelings, with physiological arousal increasing in a short time and health problems
decreasing in four months. Describing traumatic experience by email was also found
to associate with improvement in physical and psychological health in the following
weeks (Sheese, Brown, & Graziano, 2004). Refusing talking about negative emotional
17
events is associated with recurrent unwanted thoughts and several psychophysical
symptoms including anxiety, depression and insomnia (Pennebaker, 1997). The
reason for the therapeutic effect is that expressing negative emotions reduces the stress
of hiding negative feelings and allows reappraisal on the negative experience
(Pennebaker, 1997). Nevertheless, the beneficial effect is mainly observed in
personally writing practice which focuses on distant past (Rimé, Paez, Kanyangara, &
Yzerbyt, 2011). In a series of experiments (Rimé et al., 1998; Zech & Rimé, 2005),
participants were asked to talk about their recent emotions to an experimenter, results
failed to support the association between disclosing negative emotions and emotional
recovery. Rimé (2009) attributed the effect of disclosing negative emotions to
cognitive dissonance. If the disclosers stick to initial appraisal of the emotional
situation and fail to resolve the cognitive dissonance, emotional recovery would not be
obtained. In line with this argument, disclosing negative emotions in certain
circumstances is detrimental because the disclosure behavior replays the memory of
negative episodes and leads to rumination over the unpleasant experience (Rimé,
Philippot, Boca, & Mesquita, 1992).
In terms of interpersonal consequence, participants who expressed their anxiety
were more likely to receive help and perceived as more likable than those who did not
express their anxiety (Graham et al., 2008). Interpersonal intimacy and trust is found
to increase with disclosing negative emotion when it occurs necessarily but not
continuously (Collins & Miller, 1994; Gable, Reis, Impett, & Asher, 2004; Graham,
Huang, Clark, & Helgeson, 2008; Mauss et al., 2011). Drawing on these findings,
disclosing negative emotions seems beneficial to individuals and interpersonal
relationships. However, such a beneficial effect mostly emerges in close relationships,
because intimate friends can provide empathy, consolidation, and support (Rimé et al.,
18
2011). As for general audience, negative emotions may not be so desirable, and
audience may intrinsically defend against the uncomfortable feelings that are evoked
by the unpleasant disclosure (Gross, Richards, & John, 2006). Moreover, disclosing
own negativity may expose ones’ weak points to others resulting in eing disliked y
the audience (Hatfield, 1984).
Taken together, emotional disclosure has its indisputable benefit on well-being,
but there are also instances when emotional disclosure does not pay. For one thing,
some contents are perceived as socially inappropriate (Chaikin & Derlega, 1974). The
contents may look awkward, leaving the audience embarrassed and incapable of
responding (Collins & Miller, 1994). For the other thing, self-disclosure in general is
at the risk of being misunderstood and criticized (Sermat & Smyth, 1973), which may
hamper interpersonal relationships. It is found that disclosing deep aspects of self may
cause loneliness and instability of personal relationship, due to the anxiety about the
responses from recipients (Matsushima & Shiomi, 2001). Given Facebook is a unique
context, how emotional disclosure afforded by Facebook is related to well-being
remains an open question. This drives me to ask the third research question:
Research Question 3: What is the relationship between emotional disclosure on
Facebook and well-being?
Chapter 4 will look into this issue. In particular, the association between
emotional disclosure on Facebook and well-being will be discussed in light of the
fulfillment of motivational needs.
Overview of Studies
Given that Facebook has a unique communication context, this research aims
to understand the pattern, motivation, and consequence of emotional disclosure in this
context. Study 1 demonstrates that emotional disclosure on Facebook is different from
19
those in other communication media. Study 2 further examines how the Facebook
context influences emotional disclosure, by investigating the relationships between
two variables that measure individuals’ social context on Facebook (i.e., social
network size and density) and emotional disclosure pattern. Study 3 then examines
how emotional disclosure on Facebook is related to subjective well-being.
Specifically, Study 1 is designed to investigate the pattern of emotional
disclosure on Facebook, in comparison with semi-private diary writing and face-toface communication. Drawing on the effect of social context on emotional disclosure,
Study 2 focuses on Facebook context itself and look into how emotional disclosure is
particularly related to individuals’ social network structure on Facebook. Two
motivational pathways are proposed to explain the relationship between structural
properties and emotional disclosure. One pathway emphasizes the need for emotional
expression, and the other one concerns the need for impression management.
The primary assumption throughout Study 1 and Study 2 is that people can
actively utilize SNSs in the service of both need for emotional expression and need for
impression management. Study 3 extended this contention by examining whether the
motivational needs are really fulfilled on Facebook and how the fulfillment implies on
individuals’ well-being.
Overall, this thesis aims to fill up several gaps in previous research on
emotional disclosure. It can demonstrate how Facebook context shapes emotional
disclosure as compared to other traditional contexts. The results will shed light on the
social context effect on emotional disclosure by identifying the underlying motives—
the need for emotional expression and the need for impression management.
Additionally, the findings will deepen the understanding of emotional disclosure in
relation to individuals’ well-being.
20
CHAPTER 2: STUDY 1
PATTERNS OF EMOTIONAL DISCLOSURE ON FACEBOOK
As reviewed in Chapter 1, Facebook provides a context that is rarely
encountered in other settings; it is conceivable that Facebook context bears on
particular patterns of emotional disclosure. Thus, a portrait of emotional disclosure on
Facebook and an explanation fitting the Facebook’s context is necessary. Two substudies are conducted to directly and quantitatively examine emotional disclosure on
Facebook. In Study 1a, emotional disclosure is compared with semi-private writing so
as to capture the influence of publicness of Facebook context. In Study 1b, disclosure
on Facebook is compared with everyday face-to-face communication to address the
collapsed context on Facebook. Together, the two sub-studies are expected to
demonstrate Facebook as a distinctive context for emotional disclosure.
Study 1a
One of the prominent features of Facebook is status updates, which enables
individuals to disclose their feelings and experiences through short texts (Wang,
Burke, & Kraut, 2013). It is reported that more than 300,000 status updates are
published every minute (Pring, 2012). Among these status updates, 80% are simply
about users’ own immediate experiences (Naaman et al., 2010), reflecting people’s
psychological state (Kramer & Chung, 2011). A substantial number of studies have
suggested that word use indicates psychological processes, such as emotional states,
biological states, personality, thinking styles, and social relationships (Chung &
Pennebaker, 2012, in press; Pennebaker, Mehl, & Niederhoffer, 2003; Tausczik &
Pennebaker, 2010). Thus, in this study, emotional disclosure on Facebook is
represented by word use in the status updates.
21
To understand the effect of Facebook context on emotional disclosure, I choose
to compare it with a less sociable context. This is an important comparison to make,
because some of the purported social functions of emotional disclosure (e.g.,
reinforcing self-affirmation and interpersonal intimacy) imply the necessity for an
audience that is familiar and responsive, whereas other functions (e.g., reliving
positive experiences and relief from suppressing negative feelings) do not. Thus, one
could argue that emotional disclosure is purely hedonic and nonsocial—in which case,
emotional disclosures on Facebook would be no different from disclosures to strangers
or even those made privately as in a journal.
For this reason, I provide such a comparison in the present paper by using data
from a previous study (Tov, Ng, Lin, & Qiu, 2013) in which participants reported their
daily experiences in the form of short entries akin to Facebook status updates. This
allows emotional disclosure to be examined in a less sociable context in which
participants shared their experiences with an unfamiliar audience (i.e., a small team of
researchers), with no expectation of a response or reaction. If the theory about social
context effect on emotional expression is applicable to emotional disclosure on
Facebook, one would expect more emotional disclosure (both positive and negative)
on Facebook than in the diary writing samples. However, if the motivation to disclose
emotions is purely to express emotional state, one would expect no difference between
the two datasets. Therefore, I hypothesized that Facebook status updates would show
a greater tendency to disclose emotions than diaries do.
Linguistic Inquiry and Work Count (LIWC)
Computerized text analysis can handle large volumes of open-ended responses
at greatly enhanced speed, without sacrificing consistency in coding. One of the text
analysis program, Linguistic Inquiry and Word Count (LIWC; Pennebaker, Booth, &
22
Francis, 2007), was initially developed to understand contents in people’s expressive
writing and has been intensively validated (Pennebaker, Chung, Ireland, Gonzales, &
Booth, 2007). Nowadays, it is widely used to capture a wide variety of psychological
constructs (Tausczik & Pennebaker, 2010). The core function of LIWC is based on its
internal default dictionary, which defines approximately 70 word categories of a total
of about 4500 words and word stems. The word categories include four major
language dimensions: linguistic components (e.g., pronouns, articles, adverbs),
psychological constructs (e.g., affect, social, cognitive processes), personal concerns
(e.g., work, family, leisure) and paralinguistic features (e.g., assents, fillers, nonfluencies). The dictionary can be customized, such as adding words into certain
categories or defining new categories. Given a writing sample, LIWC searches each
word in the text and counts the frequencies of words that are defined in the dictionary,
and word frequency in each category is taken as a percentage of the total word count
in a given text sample. Usually, texts from an individual are compiled and treated as
one writing sample; multiple individuals’ texts were analyzed as separate samples. In
this way, percentage of each category could be generated for each individual.
According to Penne aker et al.’s (
7) psychometric evaluations, the default
dictionary LIWC2007 has desirable internal reliability (correlations between each
word’s occurrence and the sum of other words in the same category) and external
validity (the correlation etween judges’ evaluation and results from IWC analysis).
With regards to emotional disclosure in particular, LIWC contains two word
categories that predominantly signify different expressions of emotions: positive
emotions (e.g., love, nice, sweet) and negative emotions (e.g., hurt, ugly, nasty). The
two categories have been used to detect emotion words that people use when writing
about positive and negative events in daily life (e.g., Kahn, Tobin, Massey, &
23
Anderson, 2007; Pennebaker & Chung, 2011; Slatcher & Pennebaker, 2006). These
two categories of emotion words have also been applied to datasets from social media
to indicate temporal pattern of individual’s happiness on Facebook and Twitter
(Dodds, Harris, Kloumann, Bliss, & Danforth, 2011), to demonstrate emotional
fluctuations over time on Twitter (Golder & Macy, 2011), to generate index of
national happiness based on Facebook status updates (Kramer, 2010), and to
understand the relationship between religion and happiness (Ritter, Preston, &
Hernandez, 2014). In general, the two LIWC categories have been shown to be
reliable indicators of emotional disclosure in textual samples.
Method
Participants. Participants included 441 undergraduate students. The diary
sample consisted of 211 students (88 males; mean age = 21.60, SD = 1.68). The
Facebook sample consisted of 230 students (68 males; mean age = 20.96, SD = 1.64)
who consented to share their Facebook status updates.
Diary entries. Participants were asked to report two events (one positive and
one negative) that occurred during the day via a website each night for 21 days. This
requirement was implemented to ensure that participants would talk about their
emotional experiences as those on Facebook. Eventually, each participant had an
average of 348.79 words in total (i.e., 12.92 words per day). While participants
understood that their confidentiality was assured, they knew that they were
contributing to a research study. For this reason, their diary entries may best be
described as semi-private. Nevertheless, as is the case with truly private personal
diaries, participants were free to discuss a range of daily experiences and thoughts
without concern for how others—particularly those in their social network—might
react. Thus, a comparison between semi-private diary entries and Facebook status
24
updates can demonstrate the effect of context on communication. It can offer insight
on the patterns of emotional experiences that individuals choose to reveal to those
within their social networks.
Status updates. Participants’ status updates were retrieved through Facebook
application programming interface (API). To ensure that the Facebook sample had a
word count comparable to the diary sample, I randomly selected 27 status updates
from each Facebook user’s status updates.1 Eventually, the Facebook sample had an
average of 344.20 words in total with 11.47 words per status update on average.2
Text analysis. I used LIWC to analyze Facebook status updates and diaries.
Since the major concern of current study is emotional disclosure, I focused only on the
two word categories: positive emotions and negative emotions. Common emoticons
(see Appendix A) were replaced by corresponding emotional words beforehand.
In addition, to ensure the consistency of language use in the status updates
sample, I applied the LIWC analysis both to the selected sample (27 status updates
from each Facebook user) and to the complete set, and correlated the corresponding
word frequencies in each category. It showed that the selected sample shared high
consistency with the entire set, r(230) = .84 and .82 for positive emotions and negative
emotions respectively, in both instances p < .05. This suggests that the pattern of
emotional words in status updates was stable with minimal variation.
Results
Positive emotion words and negative emotional words were positively
correlated (r = .26, p < .001), suggesting individuals who disclosed more positive
emotions were also more likely to disclose negative emotions. A repeated-measured
analysis was carried out with emotion valence (positive vs. negative) as a withinsubject factor and the disclosure channel (Facebook vs. diary) as a between-subject
25
factor. As Figure 1 illustrates, emotion valence had a significant main effect on word
frequency, F(1, 439) = 342.28, p < .001, p2 = .44, with participants disclosing more
positive emotions (M = 5.82%, SD = 2.90%) than negative emotions in general (M =
3.13%, SD = 1.96%). The main effect of the disclosure channel showed that the
emotional disclosure significantly differed between Facebook status updates (M =
5.34%, SD = 1.73%) and diary (M = 3.53%, SD = 1.73%), F(1, 439) = 122.14, p <
.001, p2 = .22. The interaction effect between emotion valence and disclosure
channel was not statistically significant, F(1, 439) = 1.58, p = .21, p2 = .004.
Demographic differences may confound the results, as prior studies on daily
expression suggests that women express emotions more often than men do (LaFrance
& Banaji, 1992). To rule out this possibility, gender was added into the analysis as a
between-group factor. The results did not change. The effect of emotion valence
remained significant, F(1, 437) = 289.78, p < .001, p2 = .40. Likewise, the difference
between two channels remained significant as well, F(1, 437) = 96.98, p < .001, p2 =
.18. The interaction between emotional valence and channel was not significant, F(1,
437) = .20, p = .65, p2 = .001. Therefore, overall emotional disclosure was greater on
Facebook than in the diary entries. This effect did not vary by the valence of emotion
and were independent of gender.
An inherent limitation of computerized text analysis is its insensitivity to
context. For example, a status stating “I am in a good mood” and another status
saying “I am not in a good mood” would oth e regarded as containing positive
emotions. To eliminate this confusion, I removed all the instances of words in the
positive and negative emotions categories occurring with “no” or “not.” Analyses of
these filtered data did not change the results from original coding. Specifically,
emotion valence still had a significant main effect on word frequency, F(1, 438) =
26
351.73, p < .001, p 2 = .45, with participants disclosing more positive emotions (M =
5.68%, SD = 2.95%) than negative emotions in general (M = 2.95%, SD = 1.89%).
The main effect of the disclosure channel remained with Facebook status updates
containing more (M = 5.12%, SD = 1.77%) than diary (M = 3.44%, SD = 1.76%), F(1,
438) = 99.19, p < .001, p2 = .19. In fact, past research also suggested that removing
instances of negative use of emotional words do not alter the results significantly
(Golder & Macy, 2011; Ritter et al., 2014).
Discussion
The current study reveals that emotional words, both positive and negative,
appear more often in Facebook status updates than in semi-private diaries. It supports
the hypothesis that there is generally more emotional disclosure on Facebook than in
diaries. It demonstrates that the social context of Facebook promotes explicit
articulation of emotional experience.
Importantly, it contributes to existing research by uncovering how emotional
disclosure is manifested in text format and in an SNS setting. According to Baron’s
(1998) model of language use, Facebook status updates and diaries are similar,
because both are text-based, durable, and editable. They also share a common goal
that is to express and reflect on one’s recent experiences. However the two writing
forms have different social contexts. Status updates are publicized to a large and
mixed audience, whereas diaries are primarily restricted to the author oneself and the
researchers in the study who are outside participants’ social network. Therefore, the
comparison between the two writing samples sheds light on how the Facebook context
influences users’ emotional disclosure.
Note that these differences were found even though diary participants were
explicitly instructed to report positive and negative experiences—which could have
27
encouraged them to express more emotions than usual. That they still expressed less
emotion than the Facebook sample supports the notion that emotional disclosure is
socially motivated and not purely hedonic. This provides strong evidence that
Facebook users have a greater tendency to disclose emotional experiences than diary
writers do. Because individuals on Facebook face a potential audience, the messages
they posted are supposed to serve for communication with audience. By articulating
emotions on Facebook, individuals anticipate to receive supportive responses from
their social network (Ellison et al., 2007). The anticipated feedback from audience
may reward the behavior of disclosing emotional experiences in status updates. The
necessity for communication and the interpersonal rewards are lacking in diary
writing, which involves no third party (besides the researcher). As a result, emotional
disclosure appears in a less intensive extent. The results clearly support the theories
about social context effect on emotional expression, which states that individuals in a
social context are more likely to express their emotion (Fridlund 1991a, 1991b, 1994).
The result also aligns with uncertainty reduction theory (Berger & Calabrese, 1975)
and social information theory (Walther, 1992), which believe individuals in a
communicative situation tend to provide more information for reducing uncertainty
and enriching information.
Since the diary dataset was initially collected for other purposes, it leaves some
limitations in this study. The diary used a forced instruction method to ensure that
participants disclose emotional experience, which is a common technique in studies on
emotional expression (Pennebaker, 1997; Slatcher & Pennebaker, 2006). For
example, Lavallee and Campbell’s (1995) participants described a negative event
twice a day for two weeks. These written descriptions were then coded for the degree
of self-focus and showed that negative events were more likely to induce self-focus if
28
they were related to important goals. Although the method is often used, it makes the
writing process not as natural as in the Facebook condition. Disclosure on Facebook is
often spontaneous and voluntary, whereas the disclosure in a diary study is
retrospective and forced. It is possible that individuals in the diary study were less
motivated than those on Facebook, because they were forced; or they wrote
meaningless things as they had nothing special to write about yet they were required to
do so. As such, the results may be attributed to these artifacts caused by the task itself
other than the social process in the context.
Meanwhile, the study might be challenged for its between-subject design. In
fact, a similar approach has been employed in previous research to compare
communication styles between specific contexts. For example, text messages on
mobile phones were compared with instant messages on computers from two samples
of participants, demonstrating that text messages were longer and contained more
sentences than instant messages (Ling & Baron, 2007). Two large bodies of spoken
and written materials in computer conferences were compared to identify differences
between spoken and written utterances (Yates, 1996). While between-subject design
is preferred as to avoid carry-over effect, there is nonetheless a caveat to these
findings. It may leave some of the variations unexplainable because of individual
differences. A within-subject design can directly show how an individual shifts from
one situation to another, and can further understand how the same individual adapts
his or her communication style to different social platforms.
Because of above limitations, I conducted the following study, using a more
straightforward and controlled design.
Study 1b
Strictly speaking, the semi-private diary writing is not a communication
29
medium. To better understand how Facebook is different from other social settings,
another setting that really affords interpersonal communication is desired. As
reviewed in Chapter 1, Facebook context is featured by its massive and mixed
audience, which is rarely encountered in offline settings. Therefore, this study aims to
directly compare Facebook with offline context in real life.
According to “the lowest common denominator culture” (Hogan, 2010b),
maintaining a positive impression in a collapse context is implemented by presenting
contents that are supposed to be accepted by all possible audience. In daily
circumstances, where one deals with only a particular context at a time, a minimal
level of impression management is required. However, on Facebook, the maximum
level of impression management is required. Furthermore, the technology itself
enables selective self-presentation to be easily completed. According to the
hyperpersonal interaction model, impression management is enhanced when
information is more manageable, especially in the online context. Drawing on the fact
that positive emotions are often more favorable than negative emotions (Ekman &
Friesen, 1975) and frequent display of negative emotions may lead to the impression
that one is incapable of self-control and self-regulation (Gross et al., 2006), one of the
strategies for impression management would be to display positive emotions while
hiding the negative one. Thus, I hypothesized that the preference for positive
emotions over negative emotions will be more salient on Facebook than in the offline
context.
In respect of overall disclosure intensity, the difference between Facebook and
real life can be inferred from the information richness theory (Daft & Lengel, 1984) as
reviewed in Chapter 1. Since real life contains richer communicative information
(e.g., immediate feedback, verbal and nonverbal cues, and linguistic and paralinguistic
30
exchange of thoughts) than Facebook, it is hypothesized that real life will afford more
emotional disclosure than Facebook.
To test the two hypotheses, a within-subject comparison is carried out to
identify differences between Facebook and real life communication. Because the
actual disclosure in real life is so trivial and complicated to track, emotional disclosure
is measured by self-report in this study. The baseline of opportunities to disclose
emotions on Facebook and real life might differ, because Facebook does not require
the physical presence of an audience whereas face-to-face disclosure requires so. It is
plausible that individuals used Facebook more often have greater opportunities to
reveal their emotional experience on Facebook than in the real life. To eliminate this
possibility, the intensity of Facebook use was included as a covariate.
Method
Participants. Participants were 196 undergraduate students (66 males; mean
age = 21.06 years, SD = 1.61). They participated in exchange for course credit. They
had been using Facebook for at least one year, each with more than 50 friends on
Facebook.
Procedure and Measures. Emotional disclosure on Facebook was measured
by asking participants how likely they would be to disclose on Facebook if they
encounter positive and negative emotional experience, respectively. Participants were
asked the same questions about disclosure likelihood in real life. All the questions
were on a 7-point Likert scale (1=very unlikely, 7=very likely). The intensity of
Facebook use was measured by the Facebook Intensity Scale (Ellison et al., 2007),
including six statements on a 5-point Likert scale ranging from 1 (strongly disagree) to
5 (strongly agree). Examples of questions are “Facebook is part of my everyday
activity” and “I feel out of touch when I haven’t logged onto Face ook for a while.”
31
The scale showed good relia ility (Cron ach’s α = .86; M = 3.57, SD = .81).
Results
The intensity of Facebook use was positively correlated with the likelihood of
disclosing positive emotions (r = .29, p < .001) and negative emotions (r = .23, p <
.001) on Facebook. The intensity of Facebook use was not correlated with the
likelihood of disclosing either positive emotions (r = –.06, p = .41) or negative
emotions (r = –.05, p = .47) in real life. It suggests that users who use Facebook more
intensively are more likely to disclose their emotions on Facebook.
A repeated-measures fully factorial analysis of covariance (ANCOVA) was
carried out with 2 within-subject factors—emotion valence (positive vs. negative) and
disclosure setting (Facebook vs. real life). There was a main effect of emotion
valence, F(1, 195) = 135.13, p <.001, ηp2 = .41, with a greater likelihood to disclose
positive emotions (M = 5.51, SD = 1.04) than to disclose negative emotions (M = 4.42,
SD = 1.16). There is also a main effect of disclosure setting, F(1, 195) = 150.43, p <
.001, ηp2 = .44, suggesting that participants were less likely to disclose emotions on
Facebook (M = 4.25, SD = 1.22) than in real life (M = 5.68, SD = 1.18). There was an
interaction effect between valence and disclosure setting, F(1, 195) = 45.87, p <
.001, ηp2 = .19.
However, when controlling for the intensity of Facebook use, the main effects
were unchanged while the interaction between valence and disclosure settings became
not significant, F(1, 194) = 1.76, p = .19, ηp2 = .01. It suggests that the effect of
valence on Facebook was the same as in real life when controlling for the intensity of
Facebook use. The result was unchanged when gender was included as a betweensubject factor: The main effect of valence remained significant, F(1, 193) = 6.10, p =
.014, ηp2 = .03. The main effect of setting was significant as well, F(1, 193) = 48.41,
32
p < .001, ηp2 = .20. There was no significant interaction between valence and setting.
Discussion
Using a within-subject comparison, the current study sheds light on how
individuals accommodate their emotional disclosure in the massive and collapsed
context of Facebook. Consistent with my hypothesis, emotional disclosure on
Facebook is less frequent than disclosure in real life. Unexpectedly, the preference for
positive emotions over negative emotions is not more evident in the context of
Facebook than real life.
Previous studies found greater disclosure in online platforms than face-to-face
(Antheunis, Valkenburg, & Peter, 2007; Tidwell & Walther, 2002). However, those
results were often obtained in controlled laboratories, where the communication
settings were manipulated, the communication targets were strangers, and the
conversation took place in one single discourse. In the present study, emotional
disclosure is investigated on the basis of daily experiences. Individuals may have
more channels to talk about emotions in real life. They can speak with different
friends whenever necessary. Moreover, individuals in real life interact with each other
in physical presence and through both verbal and nonverbal channels. The
information richness theory suggests that communication media are ranked on a
continuum of richness in terms of information that can be processed (Daft & Lengel,
1984). In this sense, real life is a richer environment than Facebook, because the
former includes more immediate feedback, verbal and nonverbal cues, as well as
linguistic and paralinguistic exchange of thoughts than the latter (Nguyen, Bin, &
Campbell, 2012). Individuals may prefer a rich environment to share their emotions.
Another possible reason is that disclosing emotion without a particular target on
Facebook may not be able to get sufficient attention as it is expected. The diffused
33
focus may lessen individuals’ desire to express. Therefore, emotional disclosure on
Facebook may take place in a less frequent manner than that in real life.
Additionally, the lack of interaction effect between valence and disclosure
setting suggests that the preference of positive emotions over negative disclosure is the
same across settings. It is noteworthy that the intensity of Facebook usage is a
confounding variable in determining the frequency of emotional disclosure. The more
intensively the person uses Facebook, the more likely that the individual discloses.
Greater disclosure may enlarge the difference between the observed positive
disclosure and negative disclosure. However, it may be just because the increment of
positive emotions and negative emotions are not in the same amount. After all, people
reported more positive emotions than negative emotions in general.
General Discussion
Taken together, the two studies demonstrate the pattern of emotional disclosure
on Facebook versus semi-private diaries and real life communication. It reveals how
emotional disclosure is related to the publicness and collapsed context.
As shown in the two studies, Facebook affords emotional disclosure as diaries
and real life do. Nevertheless, Facebook stimulates a different likelihood of emotional
disclosure from diary and real life. Study 1a shows that individuals indeed express
their emotions on Facebook, and the tendency to express is more explicitly represented
than in semi-private context. I argue that the presence of the public audience and
probable interaction contributes to such a tendency. Study 1b presents results that the
likelihood of emotional disclosure on Facebook decreased, with emotional disclosure
in real life as the counterpart. I argue that the less communicative cues and collapsed
context on Facebook accounts for the results, by attenuating the motivation to express
emotions. In other words, being involved in a public context, individuals might be
34
driven by the need for emotional expression; however, if the context is too diverse and
diffused, the need might decline. In both cases, results suggest that the context
influences emotional disclosure.
However, across the two studies, results consistently show that positive
emotions are more frequently disclosed than negative emotions. Such a discrepancy
between positive and negative emotional disclosure demonstrates the necessity to
investigate the two types of emotional disclosure separately. However, the greater
impression management tendency on Facebook as suggested by the hyperpersonal
interaction model seems not supported here. One possible reason is that individuals
always have a stronger motivation to express positive emotions. It has been found that
individuals in a day share 70.8% of positive events they encounter (Gable et al., 2004)
but only 59% of negative events. Cunningham (1988) also found that when
individuals were induced to experience positive emotions, they were more active in
initiating conversations with others than those who were induced to feel neutral or
negative emotions. It seems that a common display rule in the society is to favor
positive emotions more (Ekman & Friesen, 1975). It is likely that impression
management may become a habit that one gets used to carry over across settings
(Leary & Kowalski, 1990). Alternatively, it is also possible that impression
management on Facebook is not as higher as expected. In fact, the hyperpersonal
interaction model is mainly drawn on comparisons of one-to-one communication
between CMC and face-to-face in laboratory settings. When the comparison is made
on naturalistic and complex settings, the enhanced effect of CMC may be cancelled
out by the variations within the setting. It indicates that Facebook may not be a
uniform context; the emotional disclosure may vary with the Facebook network of a
particular individual. This makes the follow-up study necessary.
35
As the first step in this thesis, Study 1 substantiates the theoretical basis that
emotional disclosure is a social act. I make the point that the Facebook context is
closely related to emotional disclosure on it. It supports the social account of
emotional disclosure on one hand; and on the other hand, it also leaves the underlying
motives and within Facebook variations open to question. As I will elaborate in the
following chapter, need for emotional expression and need for impression
management would be important mechanisms in accounting for the emotional
disclosure under a particular social network of Facebook.
36
CHAPTER 3: STUDY 2
EMOTIONAL DISCLOSURE AND SOCIAL NETWORK PROPERTIES
Study 1 found that emotional disclosure in some situations occurs more
frequently than in others; even the same individuals would modify their emotional
disclosure accordingly. It supports previous theories about the effect of social context
on emotional disclosure. More importantly, it demonstrates that the Facebook context
is different from other settings in terms of affording emotional disclosure. Meanwhile,
an individual’s social context can be defined by his or her network, through which the
individual relates to others (Burt, 1976). On Facebook, every individual has his or her
own social context, which is constituted by the particular connections with others.
Hence, one’s social context within Face ook could e related to his or her emotional
disclosure online. Therefore, the current study aims to focus on the Facebook and
explore emotional disclosure on a proximal and individual level.
A Social Network Perspective
In order to examine an individual’s social context I took a social network
perspective. Instead of looking solely at individuals’ own propensity social network
perspective focuses on the relationships (also known as ties) that exist between
individuals, the interactions, and outcomes of these relationships (Wasserman & Faust,
1994). Specifically, an ego-centered social network analysis refers to examining a
local network from a viewpoint of a person at the center. It focuses on the focal
individual (also known as “ego”) and defines the local network y links to the person.
On this level of analysis, specific structural indices can be generated to characterize an
ego’s network structure and contextual differences are there y compara le among
egos (Hogan, 2008a). Therefore individual’s context on Face ook can e quantified
and explicitly recognized. Individuals who have structurally equivalent networks are
37
supposed to exhibit similar responses by virtue of the similar facilitators and
constrains in the context (Borgatti, Mehra, Brass, & Labianca, 2009). Social network
perspective has been suggested to be especially valuable in understanding individual
behaviors (Garton, Haythornthwaite, & Wellman, 1997).
Recently, media studies have showed that social network properties could
explain the variability in media usage (e.g., Alison et al., 2006; Dimmick, Ramirez,
Wang, & Lin, 2007; Park et al., 2012). For example, Dimmick et al. (2007) found that
number of friends in a network was associated with the value that they can obtain from
using the technology, such as telephone, email, or text messages. Park et al. (2012)
showed that similarity of races among friends was positively corrected with the time
spent on Facebook. Such a study offers preliminary evidence that structural properties
of online networks are associated with individual behaviors in media. On Facebook,
the list of friends establishes a personal network, centering at “me” (Chua, Madej, &
Wellman, 2011). It is conceivable that emotional disclosure on Facebook is also
related to individuals’ personal network structure on it.
Among various metrics that can characterize one’s social network structure,
network size and network density are two basic and core properties (Garton et al.,
1997; Hipp, 2008; Izquierdo & Hanneman, 2006). A network’s size is defined as the
number of nodes in a network, and network density is the ratio of existing ties between
nodes over all possible ties in the network. In an undirected network such as an
individual’s network on Face ook the num er of friends is the size N; the number of
possible ties is N (N –1) / 2; density is the number of actual ties L divided by the
number of possible ties, i.e., 2L/N (N –1). The value of density ranges from 0 to 1,
ranking networks from completely sparse where none of the members knows each
other, to completely interconnected so that everybody is linked to everyone else. In
38
other words, low density suggests a fragmented network whereas high density
suggests a closed network. The number of possible ties increases as the number of
actors grows within the network; however, it is less likely that the number of actual
ties will increase to the same extent (Hogan, 2009). As a result, network density is
inherently correlated with network size in a negative direction. The larger the network
is, the sparser the interconnections are. The two measurements per se, however,
capture distinct dimensions of a social network. Network size determines how many
resources the one can gain from the network, reflecting the quantity of connections
(Hanneman & Riddle, 2005), while network density is usually interpreted as
cohesiveness and closure of the network, indicating the quality of interpersonal
relations (Hogan, 2009).
Network Structure and Emotional Disclosure
There is emerging interest in the relation between emotional disclosure and
online network structure. One study analyzing Facebook status updates showed a
negative correlation between network size and both positive and negative emotion
words (Facebook, 2010). However, another study on Twitter found that emotional
words increased as the network became larger and sparser (Kivran-Swaine & Naaman,
2011). In spite of the inconsistency, the two studies have demonstrated the attempt to
capture structural correlates of emotional disclosure. It stands to reason that emotional
disclosure is related to the social network structure.
In Study 1a, the publicity to large audience on Facebook is suggested to induce
emotional disclosure. It indicates that network size might be positively correlated with
emotional disclosure. A recent study on Facebook revealed that a larger network is
associated with stronger motivation to use Facebook for social connection
(Spiliotopoulos & Oakley, 2013). It is also found that the size of a person’s network
39
on Facebook predicts the person’s popularity (Tong, Van Der Heide, Langwell, &
Walther, 2008). This leads to the inference that that a large network may afford for
active and widely diffused behaviors. Thus, it is plausible to expect that overall
emotional disclosure is promoted in large networks.
Meanwhile, Study 1b showed decreasing emotional disclosure is caused by
collapsed context on Facebook, which is supposed to be at low density. According to
the principle of triadic closure (Granovetter, 1973), for three individuals, A, B, and C,
if A is closely connected to B and C respectively, B and C are likely to be closely
connected as well. This suggests that in a dense network, members are more likely to
be close friends and they would know each other, creating a socially coherent
community (Marsden, 1990; Reagans & McEvily, 2003). Close connections within a
network are thought to improve communication between members (Burt, 2001).
Individuals who have denser networks are more engaged in Facebook, marked by
more time and messages posted on Facebook (Park, Lee, & Kim, 2012). Sparse
networks, on the contrary, are comprised of diverse social circles with distinct social
contacts and roles (Burt, 1992), where individuals would be confronted with people
whom they are not interested in sharing information with. From this point of view, it
is reasonable for individuals to disclose more in a dense network while withholding
the disclosure in a sparse network.
To recap, emotional disclosure occurs because social motives are evoked in a
given context (Jakobs et al., 1999a, 1999b; Zaalberg et al., 2004). The same rationale
leads to the speculation that the size and density of one’s egocentric network would
heighten certain motives that eventually lead to adaptive disclosure behaviors. Thus, it
is necessary to identify how the social network properties are related to specific
motives. Based on the review in Chapter 1, I propose the dual pathways—(a) the need
40
for emotional expression and (b) need for impression management—through which
network properties are related to emotional disclosure on Facebook.
A large network is supposed to afford higher level of social constraints than a
small one (Burt, 2001; Izquierdo & Hanneman, 2006; Wasserman & Faust, 1994).
The more audiences are, the more likely that individuals need to concern about the
impression (Riley & Eckenrode, 1986; Leary & Kowalski, 1990). On one hand, a
large network is likely to contain a variety of members ranging from close to distant
friends, and distant friends are more sensitive to inappropriate behaviors than close
friends (Buck et al., 1992). Stokes (1983) suggested that the overall satisfaction with
social networking declined as one’s network ecame too large ecause of the
increasing stresses of managing self-images. On the other hand, the nature of
broadcasting in a wide range of audience drives the disclosers to focus on selfpresentation (Barasch & Berger, 2014). In a series of controlled experiments, Barasch
and Berger (2014) found that large group of audience, compared to only one audience,
would promote individuals to focus on self so as to keep a positive impression in front
of the audience. In contrast, narrow communication to only one listener would draw
the discloser’s focus to others and lessen the concern about self-image (Baumeister,
Bratslavsky, Finkenauer, & Vohs, 2001). Drawing on the two possibilities, network
size is expected to positively correlate with the need for impression management, and
increased need for impression management is thereby expected to promote positive
disclosure and hinder negative disclosure.
Network density, serving as another structural property, is supposed to carry a
different psychological meaning. Dense networks are usually thought to produce high
social capital (Burt, 2001), that is, people receive more social support in a dense
network (Wellman, Carrington, & Hall, 1997). Not only might members feel more
41
comfortable initiating disclosures, they might also be induced to reciprocate the
disclosures they receive from other members (Cozby, 1973). In a dense network,
individuals are expected to have a stronger need to share emotions to maintain
interpersonal connection, and facilitate emotional support and trust (Rimé et al., 2011).
Meanwhile, individuals in dense networks can establish bonding social capital in the
form of social support and trust from each other (Kilburn, 2011; Lin, 1999). Indeed,
dense networks are also found to be associated with high emotional well-being
(Fischer, 1982; Liem & Liem, 1978), strong sense of belonging, and little loneliness
(Stokes, 1985). A more nuanced view suggests that high density is supportive only in
emergencies or for long-term assistance, while low density is conducive to providing
companionship (Walker, Wasserman, & Wellman, 1994). Since social connections on
Facebook are long-term based and the communication is spontaneous, a high density is
supposed to conducive to social support. As the need for emotional expression
concerns about reciprocal interaction and social attachment, it is likely to be induced
in dense networks. With increasing need for emotional expression, disclosure of both
positive emotions and negative emotions are expected.
To sum, the size of a network implies the width of broadcasting and density
signifies social attachment; thus, the need for impression management is more likely to
be elicited by network size while the need for emotional expression is more sensitive
to network density. As an evidence to support such a speculation, a recent study
shows that the amount of self-promotional contents is not related to individuals’ desire
for social attachment on SNSs (Winter et al., 2014). As such, the need for impression
management is less likely to derive from the connectedness within the audience;
rather, it is closely related to the range of broadcasting (Barasch & Berger, 2014). As
for the need for emotional expression, it concerns about the strength of social
42
connections rather than the broadcasting range (Rimé et al., 2011). Hence, it is more
relevant to network density.
The Present Study
Despite of the theoretical construct of social motives in literature, clear
operationalization and measures are still lacking. Researchers advocate to refine the
concepts and specify how the concepts vary with social context (Jakobs et al., 2001;
Timmers, Fischer, & Manstead, 1998). As reviewed above, a general association
between network properties and emotional disclosure is expected, and the association
is hypothesized to be variously mediated by the need for emotional expression and the
need for impression management. Specifically, it is hypothesized that network size is
positively associated with positive emotional disclosure and negatively associated with
negative emotional disclosure because of the need for impression management. In
contrast, network density is hypothesized to be positively associated with negative
emotional disclosure and positive emotional disclosure because of the need for
emotional expression.
Therefore, Study 2a is designed to establish the association between network
properties and emotional disclosure. Study 2b investigates the dual-pathway
hypothesis by testing the mediating roles of need for emotional expression and need
for impression management in the association between network properties and
emotional disclosure.
Study 2a
Method
Participants and Procedure. Participants were 101 undergraduate students
(32 males; mean age = 20.93, SD = 1.77). To ensure participants included in the study
had sufficient numbers of status updates and network connections, only those who had
43
been Facebook users for more than one year and had more than 50 Facebook friends
were eligible to participate.
Measures.
Emotional words. With participants’ consent, their 100 most recent status
updates were retrieved from Facebook API as in Study 1a. The frequency of positive
emotions and negative emotional words were computed by LIWC (Pennebaker, Booth,
& Francis, 2007).
Network properties. Participants granted permission to experimenters to install
an application “NameWe Gen” (Hogan, 2010a) that would generate a text file listing
all of one’s Facebook contacts and the connections among these contacts. The
application was removed right after the list was downloaded. Then, these text files
were imported into a social network analysis software UCINET 6 (Borgatti, Everett, &
Freeman, 2002). The software computes the egocentric network size and density based
on the connections within a given social network.
Results
Table 1 presents the descriptive statistics and bivariate correlations between all
the variables. Network size was positively correlated with use of positive emotional
words (r = .29, p = .007), while network density was positively correlated with use of
negative emotional words (r = .22, p = .03). Since size and density were moderately
and negatively correlated (r = –.54, p < .001), the two variables were mean-centered
firstly and then entered into the regression model to predict positive emotional words
and negative emotions, respectively. The interaction term between mean-centered size
and mean-centered density was also included in the regression models. As shown in
Table 2, the frequency of positive emotional words was significantly and positively
correlated with both network size (β = .37 p = .
44
) and network density (β = . 8 p =
.02). No significant interaction effect was found etween network size and density (β
= .22, p = .06). The use of negative emotional words was significantly correlated only
with network density (β = .3
interaction term (β = .
p=.
). Neither network size (β = . 9 p = .11) nor the
p = .97) was significantly correlated with negative emotional
words.
When gender was considered in predicting the disclosure of positive emotion,
no significant effect was found on gender (β = .03, p = .81). The interaction between
gender and network size (β = -.22, p = .38), gender and density (β = -.18, p = .52), or
three-way interaction between gender, network size and density (β = -.27, p = .21)
were not significant either. The results were similar when considering the gender
effect on the disclosure of negative emotion: No significant effect was shown for
gender (β = .27, p = .06), the interaction between gender and network size (β = .20, p =
.42), gender and density (β = .24, p = .40), or three-way interaction between gender,
network size, and density (β = -.11, p = .63).
Discussion
The results support the hypotheses that network size and density are related to
emotional disclosure. When network density is controlled for, the larger a network is,
the more positive emotional words are used. On the contrary, when controlling for
network size, the denser a network is, the more likely people will be to use both
positive emotional words and negative emotional words, suggesting a generally
enhanced disclosure.
The interesting finding is that positive emotions are correlated with both
network size and density. Notably, that network size has a larger effect than network
density. It appears that the breadth of the broadcasting is an essential factor in
promoting positive disclosure. This may occur because impression concern is mostly
45
heightened in large networks, causing individuals to display a more desirable image.
Unexpectedly, negative emotions are not related to network size. It looks
contradictory to the prediction of high impression concern in large networks.
However, it is possible that the need for impression management is merely represented
by enhancing positive disclosure. A follow-up study is desired to examine the
underlying mechanisms. Nevertheless, network density accounts for the variation of
disclosing negative emotions. It suggests that a dense network may offer a facilitative
and supportive context for emotional expression, regardless of the emotion valence.
Social barriers seem lesser in such a network, where individuals concern more about
the expressive need.
A recent study on microblogging (i.e., Twitter) found a positive correlation
between network size and emotional disclosure but a negative correlation between
network density and emotional disclosure (Kivran-Swaine & Naaman, 2011). The
different results are probably due to different contextual nature of social networking
platforms. Social networks on Twitter include both unidirectional and directional
connections (Ellison et al., 2007); the majority of connections are strangers
(Huberman, Romero, & Wu, 2008), which may produce an extreme low density and
much large network size. Moreover, there are a lot of direct conversations taking
place on Twitter, targeting specific audience while visible to the public (Marwick &
Boyd, 2011). Thus, it is not surprising that the emotional disclosure on Twitter shows
a different pattern.
Study 2b
The present study is expected to test the roles of two proposed needs—need for
expression and need for impression management—underlying the association between
network properties and emotional disclosure. In Study 2a, the use of emotional words
46
was assessed to indicate emotional disclosure. However, the limitation of the text
analysis is that it only captures emotional expression in status updates, which
neglected disclosures that might have occurred on other features, such as comments
and photos. Thus, in the current study, I used self-reported experiences to assess the
intensity of emotional disclosure.
Method
Participants. Participants were from the same sample in Study 1b. Only
those who provided completed social network data were included. This resulted in
164 undergraduate students (56 males; mean age = 21.09, SD = 1.65).
Measures.
Emotional Disclosure. The questions were identical to those in Study 1b, to
measure the likelihood of disclosing positive or negative emotions on Facebook. The
same questions with reference to real life were also included as a control for
individual’s daily expressiveness.
Network Properties. The procedure used to assess network properties was
identical to Study a. Participants’ friend list and connections among friends were
obtained from their Facebook account upon their full consent.
Need for Emotional Expression and Impression Management. Scales were
modified based on previous research, with reference to Facebook context.3 Seven
items of need for emotional expression were adopted from ring et al.’s (1994) survey
on emotional expression (Cron ach’s α = .82; M = 2.60, SD = .68). Example
questions were “I want friends on Face ook to be able to read my emotions” and
“Even if I am feeling very emotional I don’t want to let others on Face ook know my
feelings” (reversely scored). Ten items for impression management were adopted
from Rioux and Penner’s (2001) scale on the motive of impression management
47
(Cron ach’s α = .85 ; M = 2.79, SD = .69). Example items included questions such as
“I want to avoid looking ad on Face ook” and “I want compliments from others on
Face ook”. Participants were required to indicate to what extent that they agreed with
each item on a 5-point Likert Scale (1 = Strongly disagree, 5 = Strongly agree).
Intensity of Facebook Use. Considering that the intensity of emotional
disclosure may also depend on the level of engagement on Facebook, the intensity of
Facebook use was included as a control variable. It was measured as the same as in
Study 1b (Cron ach’s α = .80; M = 4.83, SD = 3.60).
Results
Table 3 shows the descriptive statistics and bivariate correlation of all the
variables in the current study. Although the correlations between network properties
and emotional disclosure were not statistically significant, I proceeded to test the
mediation models. Researchers have suggested that mediation effect could occur
without a significant direct effect in the cases of low statistic power or competing
mediators (Hayes, 2009; Rucker, Derek, Preacher, & Kristopher, 2011). They
contended that the size of the indirect effect, in lieu of the correlation between
independent variable and dependent variable, is the key in evaluating a mediation
effect. To estimate the indirect effect, Preacher and Hayes (2008) proposed a
bootstrapping procedure. This procedure employs a nonparametric resampling
approach using bias-corrected and accelerated (BCa) percentile confidence intervals.
It is superior to traditional product-of-coefficients strategy (e.g., Sobel test), because
no assumption of sampling distribution is required. More important, it enables the
unique indirect effect to be tested while controlling for other mediators
simultaneously. Therefore, the current analysis was performed using Preacher and
Hayes’ INDI ECT.spd custom dialog (availa le at http://afhayes.com/spss-sas-and-
48
mplus-macros-and-code.html) on SPSS 21.0. Results were obtained by using 95%
BCa confidence intervals (CIs) with 5,000 bootstrapping resamples.
To test the meditational hypotheses, multiple mediator analysis was conducted
in which the two proposed mediators were entered simultaneously, being controlled
for one another. First, for each type of emotional disclosure (positive vs. negative),
two models were tested: one for the effect of network size and one for the effect of
network density. Then, for each of the network properties, the other property was
included as a covariate. Meanwhile, Facebook intensity and the corresponding
emotional disclosure in offline contexts were also included in the model as covariates.
No gender differences were found on disclosure of positive emotion on
Facebook, t(162) = .64, p = .52, or on disclosure of negative emotion on Facebook,
t(162) = .94, p = .35. In addition, given study 2a did not show any gender effect,
gender is not considered as a factor in the following analysis.
Network properties and disclosing positive emotions. Regarding disclosure
of positive emotions, results showed that the total effect of network size (β = .15, p =
.05) was marginally significant. As Figure 3 (Panel A) illustrates, the direct effect of
network size on positive emotions was not significant (β = .12, p = .13). However, the
indirect effect of network size on positive emotions through need for impression
management was significant (indirect effect = .0002, BCa 95% CI [.0001, .0005],
excluding zero), while the indirect effect through need for emotional expression was
not statistically significant (indirect effect = 0, BCa 95% CI [–.0002, .0003],
containing zero). The results qualified for a significant mediation effect (See Table 4
Panel A). The overall model was significant, F(3, 160) = 7.74, p < .001, R2 = .13,
adjusted R2 = .11. Results remained when network density, Facebook intensity and
positive emotional disclosure in real life were included as covariates (p’s > . 5).
49
Concerning the relationship between network density and positive emotions, as
Figure 3 (Panel B) shows, the total effect of network density (β = .
4 p = .96) was
not significant; neither was the direct effect (β = -.03, p = .65). Table 4 (Panel B)
shows only the indirect effect through need for emotional expression was significant
(indirect effect = 1.00, BCa 95% CI [.30, 2.09], excluding zero), while the effect
through impression motive was not statistically significant (indirect effect = -.25, BCa
95% CI [–1.13, .21], containing zero). The overall model was significant, F(3, 160) =
6.94, p < .001, R2 = .12, adjusted R2 = .10. Results were unchanged when network
size, Facebook intensity and positive emotional disclosure in real life were included as
covariates (p’s > . 5).
Consistent with Study 2a, the findings above showed that both network size
and density were related to positive emotions. More important, the effect works
through increasing need for impression management and increasing need for emotion
expression respectively.
Network properties and disclosing negative emotions. As Figure 4 (Panel
A) shows, the total effect of network size on negative emotional disclosure was not
statistically significant (β = .
p = .76), and direct effect was not statistically
significant either (β = .04, p = .54). Moreover, Table 5 (Panel A) demonstrates the
indirect effect through either need for emotional expression (indirect = .0001, BCa
95% CI [–.0005, .0007], containing zero) or need for impression management (indirect
= -.0002, BCa 95% CI [–.0005, 0], containing zero) was not statistically significant,
although the mediation model was significant, F(3, 160) = 21.38, p < .001, R2 = .29,
adjusted R2 = .27. The results were unchanged when network density, Facebook
intensity, and negative emotional disclosure in real life were added as covariates (p’s >
.05).
50
Concerning the association between network density and negative emotions, as
Figure 4 (Panel B) illustrates, the total effect of network density on negative emotional
disclosure was not statistically significant (β = .01, p = .95), and direct effect was not
statistically significant either (β = -.09, p = .20). However, as Table 5 (Panel B)
shows, the indirect effect of network density was significant through need for
emotional expression (indirect = 2.36, BCa 95% CI [.23, 4.38], excluding zero). The
indirect effect of network density was not significant through need for impression
management (indirect = .18, BCa 95% CI [-.13, .89], containing zero). Overall, the
mediation model was significant, F(3, 160) = 21.98, p < .001, R2 = .29, adjusted R2 =
.28. The patterns remained when network size, Facebook intensity, and negative
emotional disclosure in real life were added as covariates (p’s > . 5).
In short, the findings showed that only network density contributed to
disclosure of negative emotions, which is consistent with the result in Study 2a.
Moreover, the effect could be explained by increased need for emotional expression.
Discussion
The current study examines how the need for emotional expression and the
need for impression management mediate the association between social network
properties and emotional disclosure. Results revealed that the context influences
disclosure behavior by shifting the relative force of two motivational needs.
Specifically, the need for emotional expression mediates the association between
network density and both disclosure of positive emotions and negative emotions,
whereas the need for impression management only mediates the association between
network size and disclosure of positive emotions.
As shown in Study 2a, network size is associated with positive emotional
disclosure. Importantly, the current study found that the association is mediated by the
51
need for impression management. The larger the network is, the more likely that the
individual will be to disclose emotion in support of a positive image, leading to
increasing disclosure of positive emotions. It indicates that positive emotions are
shared for the purpose of image construction and maintenance. It thus offers evidence
for previous studies that claim that SNSs users selectively present favorable
information as a strategy of impression management (e.g., Ellison, Heino, & Gibbs,
2006; Gibbs, 2006; Jung et al., 2007; Ong et al., 2011). On the other hand, the finding
supports previous speculation that large networks enhance the concern about positive
impressions (Barasch & Berger, 2014; Riley & Eckenrode, 1986). Need for
impression management appears to derive from the width of the influence in a large
environment, rather than the density of the network. This is consistent with previous
study (Barasch & Berger, 2014), which shows that the motivation of impression
management occurs when broadcasting to a large group of audience rather than when
narrowcasting to only one person. The reason is that large audience size arouses the
focus on self, so individuals put more weight on own interests. Although the influence
of density is not stated in previous studies, the current research indicates that the
connection among the audience is not a key factor in triggering impression
management. Following the logic above, it seems that how interconnected the social
network is does not enhance self-focus. Perhaps the density only signals social
attachment and support, which is not the foundation for getting impression; or the
density only concerns about deep relationships such as trust and commitment, which
are not related to the focus of self-competence. In contrast, the large network size,
which can make the disclosure achieve broad influence, is more likely to meet the
need for impression management.
Concerning the coherence of audience, the focus on impression is replaced by
52
the expressive need. This is manifest in the association between network density and
emotional disclosure, mediated by the need for emotional expression. No matter
whether it is positive or negative emotions, the desire to express is enhanced in a dense
network. It is likely that individuals are more personally attached to a dense network
and more frequently engage in interactions with this group of friends. A possible
reason is the mutual interaction in such a network is more likely to be anticipated, and
the social attachment is more likely to be developed. Indeed, disclosure of emotion
often takes place within intimates in real life (Rimé, 2009). It appears that the same
rule is applicable to Facebook context. When the need for emotional expression is
enhanced by the supportive context, the behavior of emotional disclosure is facilitated.
This also supports previous assumption that social interaction is a strong motivation
for emotional disclosure (Jakobs et al., 2001,1999a, 1999b; Rimé, 2009; Rimé et al.,
1998; Rimé et al., 2011).
Results also confirm an earlier speculation that the need for impression
management is irrelevant to network density, whereas the need for emotional
expression is not associated with network size. It illustrates that network size and
network density are distinctive metrics that carry different psychological meaning.
While network size is related to self-promotion, network density is important to social
attachment.
Interestingly, the need for impression management is not related to less
disclosure of negative emotions. It may indicate that impression motivation on
Facebook is manifested in a positively enhancing rather than a negatively inhibiting
manner. Literature has identified two distinct types of impression management in selfpresentation, acquisitive and protective (Arkin, Lake, & Baumgardner, 1986).
Acquisitive impression management refers to the action of gaining social approval by
53
presenting oneself in a favorable fashion (Brown, Collins, & Schmidt, 1988). In
contrast, the protective type involves efforts to avoid social disapproval by distancing
themselves from negative events and creating merely “safe” images (Arkin, Lake, &
Baumgardner, 1986). The current result implies that impression management on
Facebook is mainly adopted in an acquisitive manner (to approach positivity) instead
of a protective manner (to avoid negativity). It reflects that Facebook users are willing
to acquisitively seek the approval of others rather than protectively avoid disapproval
of others. Supplementary analysis supports this explanation. Based on the scale of
need for impression management, acquisitive statements (e.g., “I want compliment
from others on Facebook”) and protective statements (e.g., I want to avoid a reprimand
from others on Facebook) were summed up respectively. Mediation analysis showed
that the mediation path from network size to positive emotions is only significant
through acquisitive impression management (indirect effect = .0004, 95% CI [.0001,
.0008]), confirming the role of acquisitive impression management.
The null effect of impression management motive on negative emotions needs
a few more words. Past research suggests that individuals in everyday life engage in
protective impression management more frequently than in acquisitive impression
management (Baumeister et al., 2001). For example, the need for impression
management leads to less negative expression because negative emotions such as
crying is not an appropriate act in the presence of others (Fischer et al., 2003; Jakobs
et al., 2001). Nevertheless, this seems not the case on Facebook. It is possible that the
display rule on Facebook is somewhat different. While previous research presumes
negative emotions are unacceptable in public, this assumption may not be necessary on
Facebook. While positive emotions are still favorable, negative emotions are not as
disliked as in real life. Probably, Facebook has fostered an environment that
54
encourages active engagement rather than conservative conversation, which can be
inferred from the social information theory and uncertainty reduction theories as
reviewed in Chapter 1. The tactic to maintain impression on Facebook may mainly
rely on boosting positive side.
Although the current study suggests that social network structure is associated
with psychological needs and the emotional disclosure pattern, no definite causal
relationship is established. It is possible that network structure is an outcome of
individuals’ psychological needs through individuals’ intentional selections. People
who have greater need for emotional expression may develop denser network, and
those who have greater need for impression management may tend to enlarge their
networks. To test this alternative explanation, I compare the two casual models. I first
tested the hypothesized model with the two network properties as independent
variables, the two needs as mediators, and both disclosure of positive emotions and
negative emotions as dependent variables. A path analysis was conducted with IBM
SPSS AMOS 21.0. The paths were drawn according to the hypotheses. Results of the
analysis revealed a satisfactory model fit to the data: χ2 (8) = 6.42, p = .6 and χ2/df =
.80, less than the criteria of 3 (Kline, 2011). Further, the comparative fit index (CFI) =
1.00, the Goodness-of-Fit Index (GFI) = 0.99, adjusted goodness of fit index (AGFI) =
0.97, and normed fit index (NFI) = 0.94, all exceeded the critical value of 0.9 (Kline,
2011). The non-normed fit index (NNFI) = 1.03 was greater than 0.95, larger than the
cutoff for a good model fit (Hu & Bentler, 1999). Root mean square error of
approximation (RMSEA) = 0.00 meets the guideline of smaller than 0.05 (McDonald
& Ho, 2002). Then, I tested the alternative model by reversing the original paths
between the psychological needs and network properties. Results showed a poor
model fit: χ2 (8) = 75.5
χ2/df = 9.44 > 3; CFI = 0.32, GFI = 0.87, NFI = 0.34, AGFI =
55
0.67, NNFI = -.28, all are smaller than 0.9 and RMSEA = 0.24 > 0.05. In addition, the
AIC for the alternative model (102.58) is much larger than that of the original model
(AIC = 33.97), suggesting the alternative model is poorer. Statistically, the results rule
out the possibility of reversed causal relationship and confirm the hypothesized path
from network properties to psychological needs.
A caveat of this study is that the participants were a subsample of those in
Study 1b. However, the two studies are supposed to address different research
questions, with Study 1b aiming at comparing emotional disclosure between Facebook
and real-life, and Study 2b focusing on the emotional disclosure on Facebook as a
function of network structure. The data for the two studied were collected at the same
time, but each study’s data was analyzed for a different purpose. The results of Study
1b do not intervene with that of Study 2b.
General Discussion
Taking a social network perspective, the current study examines the association
between social network structure and emotional disclosure. It also disentangles the
underlying motivations in terms of need for emotional expression and need for
impression management. Suffice it to say, the two types of needs account for the
association between social network properties and emotional disclosure in divergent
ways.
First, network size and density are suggested to have differential associations
with emotional disclosure. Although the bivariate correlations found in Study 2a are
not replicated in Study 2b, the link between social network structure and emotional
disclosure is confirmed by the indirect effect. The deficit in total effect could have
occurred because of different measurements of emotional disclosure. In Study 2a,
emotional disclosure was measured by linguistic markers, whereas it was measured by
56
self-reported likelihood in Study 2b. While linguistic analyses focus only on status
updates, self-reports cover a wider range of activities on Facebook. Nonetheless, the
mediation model reveals the indirect relationship between network structure and
disclosure behaviors, which aligns with results of Study 2a. It is still reasonable to
believe the association between network structure and emotional disclosure does exist.
More important, Study 2b implies that the likelihood of emotional disclosure
on Facebook could be regarded as a function of motivational needs that are contingent
on personal network properties. Note that the results should be interpreted with one
property in consideration while the other is controlled. When network density is
controlled and network size is concerned, the need for impression management
becomes salient. As the audience size grows, the need to be perceived as popular and
competent increases. To foster a positive image, individuals are inclined to display
positive emotions. In a divergent aspect, network density contributes to the likelihood
of disclosing negative emotions through enhancing the need for emotional expression.
A denser network, in this sense, may form a safe harbor for less socially desirable
emotions. The meaning of density is also translated to positive emotional disclosure
by enhancing the need for emotional expression. Probably, a dense network can offer
not only caring and understanding for distress expression, but also focused social
attention and affirmation that are anticipated when sharing pleasure.
Some studies suggest that the larger the network is, the greater the likelihood is
that the person can obtain resources and accumulate social capital (Burt, 2000).
Seeman and Berkman (1988) measured the elderly’s social networks in daily life and
found that network size was positively correlated with availability of both instrumental
and emotional aid. It appears that large networks are more supportive than small
networks. In this regard, network size is also related to need for emotional expression.
57
However, the current study does not support such a claim. In fact, previous work
focused on a specific and regional social network in real life, where the density is
supposed to be relatively high. It is possible that the observed social capital and social
support are rooted in the interconnection within the network rather than the size of
community.
A dense network usually means a closed network, where people are likely to
interconnect. Such networks tend to involve high surveillance, as the impression can
be easily spread within the network (Coleman, 1988) and sanctions can be easily
implemented (Burt, 2001, 2012). Dense networks are suggested to be stable because
of coherent norms and strict social monitors (Granovetter, 1985). It is also shown that
dense networks enhance prevention-focused regulation, which is supposed to relate to
higher impression management (Zou, 2009). Although the current study could not
exclude such a possibility, the seeming contradiction might be explained by the
characteristics of Facebook. Previous studies are mostly based on offline network in a
particular domain. For example, in Zou’s (2009) study, networks were measured in a
professional organization which included only connections at one’s workplace. In the
case of Facebook, a network usually includes a more diverse composition, containing
connections from not only workplace but also other circles, such as alumni, social
communities, and remote acquaintances. This may result in a much lower density
versus the workplace.4 Because of this, the social sanction in terms of closure of
network may be too weak to observe.
While Study 1 shows that emotional disclosure differs across various
disclosure channels, the current study offers an alternative perspective for
understanding the results. Comparing Facebook with diary writing, the large audience
size increases the need for impression management, which is supposed to lead to more
58
positive emotional disclosure. However, there is another variable differentiating
Facebook from diary writing: Friends on Facebook have established social
connections with the individual, but the “readers” of diary are unacquainted
researchers. Therefore, the need for emotional expression is supposed to be greater on
Facebook too, leading to more disclosure of both positive and negative emotions.
Comparing Facebook with real-life one’s social network on Facebook is larger and
sparser than that in real life. Thus, the need for impression management is higher on
Facebook than in real life, yielding more emotions that are positive. In contrast, the
need for emotional expression is evoked more intensively in real life, leading to more
positive and negative emotional disclosure. The overall outcome is that greater
emotional disclosure occurs in real life than on Facebook.
To summarize, Study 2a and 2b discuss the effects of social context in the light
of a theoretical framework of social motive processes. The effects are straightforward
for disclosing positive emotions, with network size triggering need for impression
management and network density inducing need for emotional expression. In the case
of disclosing negative emotions, the effect of network density is mediated by the need
for emotional expression. The results also imply that disclosing positive emotions and
disclosing negative emotions may serve different motivational needs. Disclosure of
positive emotions may be conducive to both image enhancement and emotional
expression. As for negative emotions, disclosure may be beneficial for the expressive
need only. This speculation leads to the follow-up study on the relationship between
emotional disclosure and need fulfillment.
59
CHAPTER 4: STUDY 3
EMOTIONAL DISCLOSURE AND WELL-BEING
Study 1 and 2 found that emotional disclosure is closely related to the context,
and emotional disclosure was driven by the need for emotional expression and need
for impression management in a particular social network. A related question is:
What would be the consequence of emotional disclosure on Facebook? Although
previous studies have demonstrated the associations between emotional disclosure and
well-being, can emotional disclosure on Facebook yield the same association with
well-being given the unique context of Facebook?
Recap the literature reviewed Chapter 1, disclosing positive emotions always
brings along positive effect (Gable & Reis, 2010; Gable et al., 2004). The massive and
collapsed context of Facebook allows individuals to broadcast their joy, thus
capitalization might be enhanced in interpersonal interactions. Hence, it would be
expected that disclosing positive emotions on Facebook could benefit well-being.
With regards to negative emotional disclosure and well-being, there are conflicting
findings in the literature. On one hand, disclosing negative emotions may have a
cathartic effect (Pennebaker, 1997; Radcliffe et al., 2010; Sheese et al., 2004). It is
possible that disclosure on Facebook can also be beneficial, with Facebook being an
outlet for freely expressing negative emotions. On the other hand, sharing negative
emotions is supposed to gain benefits from social support (Graham et al., 2008; Gross
& John, 2003) and the effect of disclosing negative emotions requires cognitive
reappraisal of the experience (Rimé, 2009)However, the brief and asynchronous
communication on Facebook may not be able to allow for this process to occur.
Meanwhile, replaying the memory of negative episodes could cause unpleasant
experiences (Rimé et al., 1992). Especially on Facebook, as the disclosed content will
60
not disappear, it may repeatedly remind individuals of the unpleasant experiences. A
negative effect on well-being may also likely accompany the disclosure of negative
emotions on Facebook.
If emotional disclosure on Facebook is associated with well-being, how can the
motivational needs identified in Study 2 shed light on this issue? As the results
indicate that individuals expect that disclosing emotions would serve their needs, is it
true that the behaviors do meet the needs? A relevant and important speculation
would be this: Does the effect of emotional disclosure on well-being, if any, work by
fulfilling the need for emotional expression, the need for impression management, or
both?
The current study is conducted on above rationale. It aims to attest whether
well-being is related to emotional disclosure on Facebook; furthermore, to examine the
explainability of fulfillment of need for emotional expression and fulfillment of need
for impression management in the relationship. To elaborate on the logic leading up to
the hypothesis, I shall first review related theories of well-being from the perspective
of needs fulfillment.
A Motivational Account for Well-being
As well-being is a multidimensional construct (Ryan & Deci, 2001), various
indicators of well-being may lead to different findings. In the current study, I
concentrate on two indicators that are most concerned in media research. One is life
satisfaction, which comprises the cognitive processes concerning an overall life
experiences (Diener, Suh, Lucas, & Smith, 1999; Pavot & Diener, 2009). It is a
su jective evaluation of one’s quality of life ased on self-defined criteria (Diener,
Emmons, Larsen, & Griffin, 1985). A number of studies on media effect have referred
to it as a central indicator of well-being (e.g., Burke et al., 2010; Kim, Chung, & Ahn,
61
2013; Kramer, 2010; Lee, Lee, & Kwon, 2011). For instance, Burke et al. (2010)
correlated behavioral data on Facebook with life satisfaction and indicated how the
activities on Face ook enefit individuals’ well-being. Kramer (2010) examined the
association between emotional words and life satisfaction to illustrate the possibility of
using Facebook status updates to predict national well-being. Lee et al. (2011)
discussed how the number of Facebook friends affected individuals’ well-being by
measuring life satisfaction.
In addition to psychological well-being that emphasizes personal functioning,
social scientists also advocate investigating social well-being (Keyes, 1998). It is
critical for a person to function well while involving others in the social contexts. As
emotional disclosure is conceptualized as a social act, social well-being appears
extremely relevant. For this reason, the second indicator—loneliness—is considered in
the current study too. Loneliness refers to a feeling of unhappiness caused by being
isolated from interpersonal relationships (Cacioppo & Patrick, 2008). It has received
intensive attention in media research pertaining to well-being (e.g., Burke et al., 2010;
große Deters & Mehl, 2013; Kraut et al., 2002; Kraut et al., 1998; Peter et al., 2006;
Ryan & Xenos, 2011). For example, Kraut et al. (1998 & 2002) studied how using
Internet was related to well-being, with loneliness as one of the indicators; Ryan and
Xenos (20 ) studied how Face ook users’ personality and loneliness related to their
usage of Facebook; Burke et al. (2010) showed some individuals had low well-being
on Facebook because the passive use pattern caused loneliness.
In the exploration of factors that could contribute to well-being, need
fulfillment is considered a key reason that contributes to well-being (e.g., Church et
al., 2013; Diener & Lucas, 2000; Ryan, 1995; Ryan & Deci, 2000; Tay & Diener,
2011). For instance, identification with a community can facilitate individuals’ well-
62
being by satisfying basic psychological needs (Molix & Nichols, 2013). Another
study reveals that fulfillment of psychological needs could explain the effect of
materialism on well-being (Chen, Yao, & Yan, 2013). A review of past studies
proposed a mediation model to explain the relation between leisure and well-being
(Newman, Tay, & Diener, 2013). In this model, a set of psychological mechanisms,
including feelings of autonomy, affiliation, mastery, detachment from work, and
understanding the meaning of life are posited as the psychological pathways through
which engagement in leisure activities influences well-being. These studies have
demonstrated the role of need fulfillment in accounting for the association between
several antecedents and well-being. In a similar vein, it is plausible to expect the
association between emotional disclosure and well-being to be mediated by fulfillment
of specific motivational needs.
Fulfillment of The Two Needs
In Study 2b, the need for emotional expression drives both disclosure of
positive emotions and disclosure of negative emotions. Therefore, it is conceivable
that the need for emotional expression would be satisfied by emotional disclosure
regardless of the valance. The need for emotional expression may be satisfied in two
possible ways. First, disclosure of emotion, either positive or negative, is rewarded by
the expression process itself (Tamir & Mitchell, 2012). Disclosing positive emotions
often occurs with re-experiencing happiness brought by the positive experience (Gable
& Reis, 2010; Maisel, Gable, & Strachman, 2008). Disclosing negative emotions can
facilitate a healing process through the release of emotional distress (McCance, Nye,
Wang, Jones, & Chiu, 2010). It suggests that the need for emotional expression may
be met by the nature of the disclosure process itself. Second, the sense of sharing
emotional experience with others enables the personal experience to be cared for and
63
understood (Moreno et al., 2011; Stone & Pennebaker, 2002). In Graham et al.’s
(2008) experiment, participants showed higher tendency to help another person who
explicitly expressed his or her negative emotions than to help a person who did not
express negative emotions. In one way or another, disclosure of either positive
emotions or negative emotions is suggested to satisfy the need for emotional
expression.
Despite any relief or support that is expected, disclosure of emotion may not be
a le to alter the person’s overall state. Previous studies demonstrated that people
generally reported emotional benefits in emotional disclosure, but actual recovery of
emotion was rarely achieved (Rimé, 1998; Zech & Rimé, 2005). A meta-analysis
found that disclosing emotion in a single session is insufficient to alleviate emotional
burden (Van Emmerik, Kamphuis, Hulsbosch, & Emmelkamp, 2002). Emotional
disclosure on Facebook is often spontaneous and brief, rather than systematic and
continuous as in writing therapy. In addition, the feedback from others on Facebook is
always brief and delayed, rather than elaborative and immediate. Thus, the satisfaction
of expression may not bring actual benefit on the overall life satisfaction and
interpersonal relationship. Therefore, it seems that need for emotional expression
might be effectively met on the present pleasure or relief, but it may not contribute to
long-term well-being.
Regarding the need for impression management, it may be fulfilled by the
desirable images established on Facebook. Given results in Study 2b that need for
impression management only accounted for disclosure of positive emotions, it is
expected that the need for impression management would be mainly fulfilled by
positive emotional disclosure. Indeed, disclosers can gain a sense of self-enhancement
through audience’s responses to positive disclosure (Beals, Peplau, & Gable, 2009;
64
Gable, Gonzaga, & Strachman, 2006; Gable & Reis, 2010; Maisel, Gable, &
Strachman, 2008). Thus, the need for impression management is likely to be fulfilled
as the self-image is positively constructed in the sharing process. On Facebook, this
need is even fulfilled by simply looking at own profiles, from which the established
positive impression affirms the sense of self-competence (Toma, 2010; Toma &
Hancock, 2013). As the perception of self-competence is a basic need for producing
well-being (Ryan & Deci, 2001), fulfillment of need for impression management is
likely to produce increased well-being.
The Present Study
As reviewed above, the relationship between emotional disclosure and wellbeing depends on the valance of emotion. Taking the need for emotional expression
and need for impression management into account, the relationship may be partially
accounted for by fulfillment of different needs. Given the features of Facebook, need
for emotional expression is expected to be satisfied by disclosing positive emotions
and negative emotion; however, it will not account for the association between
emotional disclosure and well-being. Need for impression management is expected to
be met by disclosing positive emotions, which will account for the positive association
between disclosure of positive emotions and well-being.
Method
Participants and procedure
One hundred and thirty-six undergraduate students (53 males; mean age =
22.28, SD = 1.59) participated in the present study in exchange for course credit. They
were required to answer the questionnaires about emotional disclosure on Facebook,
fulfillment of needs by emotional disclosure on Facebook, life satisfaction, loneliness,
and demographic information.
65
Measures
Emotional disclosure. The questions were identical to those used in Study 1b
and 2b, that is about the likelihood to disclose positive and negative emotions on
Facebook, on a 7-point Likert scale (1=very unlikely, 7=very likely).
Need fulfillment on Facebook. It was assessed by the extent to which
emotional disclosure on Facebook fulfill the needs of emotional expression and
impression management respectively (see Appendix C). Eight questions used in Study
2b were selected and modified to address the fulfillment of need for emotional
expression (e.g., “ y disclosing my emotion on Facebook, friends on Facebook can
understand my feelings”) and fulfillment of need for impression management (e.g.,
“ y disclosing my emotion on Facebook, I build an ideal self-image on Facebook”).
Participants were asked to indicate to what extent they agreed with each statement on a
5-point Likert Scale with 1 as strongly disagree and 5 as strongly agree. The two
scales were reliable (Fulfillment of need for emotional expression: M = 2.51, SD = .85,
Cron ach’s α = .87; Fulfillment of need for impression management: M = 2.67, SD =
.8
Cron ach’s α = .84).
Life satisfaction. Life satisfaction was assessed by the Satisfaction with Life
Scale (SWLS; Diener et al., 1985). The scale contains five items about global life
satisfaction such as “the conditions of my life are excellent” and “so far I have got the
important things I want in life,” on a 7-point Likert scale (1 = strongly disagree, 7 =
strongly agree; M = 4.54, SD = . 9; Cron ach’s α =.85).
Loneliness. The UCLA Loneliness Scale (Russell, 1996) was used to assess
the feelings of loneliness in past three months. Examples of questions include “how
often do you feel that you are ‘in tune’ with the people around you” and “how often do
you feel part of a group of friends,” with responses on a 4-point Likert Scale (1=
66
never, 4 = always; M = 2.33, SD = .4
Cron ach’s α = .91).
Results
Descriptive statistics and bivariate correlation are presented in Table 6.
Disclosure of positive emotions was positively correlated with fulfillment of need for
emotional expression (r = .26, p = .002) and fulfillment of need for impression
management (r = .36, p < .001). Disclosure of negative emotions was positively
correlated with fulfillment of need for emotional expression (r = .26, p = .002), and
not significantly correlated with fulfillment of need for impression management (r =
.16, p = .06). It suggested that need for emotional expression was satisfied by both
positive and negative emotional disclosure, while need for impression management
was satisfied by disclosure of positive emotions.
Regarding the relationship with well-being, disclosing positive emotions was
not significantly correlated with either life satisfaction (r = –.03, p = .76) or loneliness
(r = –.002, p = .98). Contrarily, disclosing negative emotions was positively
correlated with loneliness (r = .23, p = .007) and negatively correlated with life
satisfaction (r = –.20, p = .02). Next, multiple mediator models were performed by
using Preacher and Hayes’ ootstrapping method (Preacher & Hayes, 2008) as in
Study 2b. For each type of emotional disclosure (positive vs. negative), two separate
models were tested on life satisfaction and loneliness, with fulfillment of need for
emotional expression and fulfillment of need for impression management entered
simultaneously as mediators. Because disclosing positive emotions was positively
correlated with disclosing negative emotions (r = .56, p < .001), it is most likely that
individuals who like disclosing positive emotions were also those who like disclosing
negative emotions. To differentiate the effects between positive emotional disclosure
and negative emotional disclosure, it is necessary to test the effect while controlling
67
for one another. Thus in the second step of each model testing, the other type of
emotional disclosure was included as a covariate.
Effect of disclosing positive emotion
The total effect of disclosing positive emotions was not statistically significant
on life satisfaction (β = –.03, p = .76). The direct effect of disclosing positive emotions
was not statistically significant either (β = –.07, p = .46). Nevertheless, as Table 7
(Panel A) shows, the indirect effect through fulfillment of need for emotional
expression (indirect effect = –.04, BCa 95% CI [–.11, –.007], excluding zero) and
satisfaction of the need for impression management (indirect effect = .03, BCa 95% CI
[.02, .16], excluding zero) were significant. The whole model was significant, F(3,
132) = 2.91, p = .04, R2 = .06., adjusted R2 = .04. However, when disclosing negative
emotions was included as a covariate, only the pathway through fulfillment of need for
impression management remained (indirect effect = .03, BCa 95% CI [.02, .17],
excluding zero). As shown in Figure 5 (Panel A), the indirect effect through
fulfillment of need for emotional expression disappeared (indirect effect = –.02, BCa
95% CI [–.09, .001], containing zero).
Concerning the effect of disclosing positive emotions on loneliness, the
mediation effect appeared despite the non-significant total effect (β = –.002, p = .98)
and non-significant direct effect of disclosing positive emotions (β = .07, p = .41). The
indirect effect showed the effect of disclosing positive emotions on loneliness was
taken through the satisfaction of the need for emotional expression (indirect effect =
.02, BCa 95% CI [.01, .05], excluding zero) and the fulfillment of need for impression
management (indirect effect = –.04, BCa 95% CI [–.08, –.02], excluding zero), as
shown in Table 7 (Panel B). The whole model was significant, F(3, 132) = 5.65, p <
.01., R2 = .11, adjusted R2 = .09. However, only the pathway through fulfillment of
68
need for impression management remained when disclosing negative emotions was
controlled as a covariate (indirect effect = –.04, BCa 95% CI [–.08, –.02], excluding
zero). Fulfillment of need for emotional expression did not contribute to the
association between disclosing positive emotions and loneliness (indirect effect = .01,
BCa 95% CI [–.001, .03], containing zero) any more (see Figure 5 Panel B).
When including gender as a moderator in the meditational process, results
showed neither significant effect of gender nor significant interaction effect between
gender and disclosing positive emotion on the fulfillment of the two needs.
Henceforth, the two mediation models revealed that disclosure of positive
emotions, when disclosure of negative emotions was controlled, was connected with
increasing life satisfaction and decreasing loneliness because of satisfied need for
impression management. The relationship between emotional disclosure and wellbeing could not be accounted for by the fulfillment of need for emotional expression.
Effect of Disclosing Negative Emotion
The total effect of disclosing negative emotions on life satisfaction was
negative and significant (β = –.20, p = .02), and the direct effect showed the similar
pattern (β = –.19, p = .03). Table 8 (Panel A) exhibits that the indirect effect was
significant through both the fulfillment of need for emotional expression (indirect
effect = –.04, BCa 95% CI [–.11, –.001], excluding zero) and the fulfillment of need
for impression management (indirect effect = .34, BCa 95% CI [.004, .09], excluding
zero). The whole model was significant, F(3, 132) = 4.48, p = .005, R2 = .09, adjusted
R2 = .07. However, when including disclosing positive emotions as a covariate, the
indirect effect through fulfillment of need for emotional expression (indirect effect =
.01, BCa 95% CI [–.001, .03], containing zero) and the effect through fulfillment of
need for impression management (indirect effect = .007, BCa 95% CI [–.001, .03],
69
containing zero) became non-significant (see Figure 6 Panel A).
With regard to the relationship between disclosing negative emotions and
loneliness, the total effect was positive and significant (β = .23, p = .007), and the
direct effect was significant too (β = .24, p = .004). The indirect effect was significant
through fulfillment of need for emotional expression (indirect effect = .02, BCa 95%
CI [.001, .04], excluding zero) and through fulfillment of need for impression
management was significant (indirect effect = –.02, BCa 95% CI [–.05, –.0009],
excluding zero), as shown in Table 8 (Panel B). The whole model was significant,
F(3, 132) = 8.58, p < .001, R2 = .16, adjusted R2 = .14. As shown in Figure 6 (Panel
B), when including disclosing positive emotions as a covariate, the indirect effect
became non-significant through either fulfillment of need for emotional expression
(indirect effect = –.01, BCa 95% CI [–.09, .002], containing zero) or fulfillment of
need for emotional expression (indirect effect = –.01, BCa 95% CI [–.07, .02],
containing zero).
Again, considering gender as a moderator on the meditational process did
change the findings: No significant effect of gender was found, neither were
significant interaction effect between gender and disclosing negative emotion on the
fulfillment of the two needs.
In short, the disclosure of negative emotions is negatively associated with wellbeing. However, the relationship could not be accounted for by either satisfaction of
the need for emotional expression or need for impression management, when
controlling for the likelihood of disclosing positive emotions. Despite a lack of
significant mediation effect, it must be noted that the negative relationship between
disclosing negative emotions and life satisfaction (β = –.26, p = .01) and the positive
relationship with loneliness (β = .34 p = .001) remained, suggesting that disclosing
70
negative emotions was related to poorer well-being as expected. These findings
suggested that disclosure of negative emotions could satisfy the need for emotional
expression, but the need fulfillment is not related to well-being.
Discussion
Study 3 adds to Study 2 in two aspects. First, it extends the study on the two
motivational needs—need for emotional expression and need for impression
management—by examining the extent to which the actual behavior really fulfills the
needs. Results indicate that emotional disclosure fulfills the need in both expressional
and presentational aspects. Second, it moves forward from targeting the antecedents
of behavior toward the consequences, in terms of subjective life satisfaction and
loneliness. Although initial analyses show significant mediation effects, not all the
effects hold as the covariate is controlled. The results support only the meditational
hypothesis regarding fulfillment of need for impression management in the association
between disclosure of positive emotions and well-being. Results show no mediation
effect of need for impression management on negative emotions. The null mediation
effect of fulfillment of need for emotional expression is as expected. Results suggest
that the relationship between emotional disclosure on Facebook and well-being is
explained by need fulfillment only in a limited way.
Regarding the bivariate correlation between emotional disclosure and need
fulfillment, the result echoes the findings in Study 2b. As the need for emotional
disclosure motivates disclosure of positive emotions and negative emotions, the actual
disclosure fulfills the need in return. This supports the claim that the function of
emotional disclosure is to gain social attention (Moreno et al., 2011; Stone &
Pennebaker, 2002) and emotional pleasure (Gable & Reis, 2010; Maisel et al., 2008).
In a similar vein, the need for impression management drives the action of disclosing
71
positive emotions; in return, the behavior of disclosure seems to meet the need. This
aligns with previous finding that positive disclosure is an effective strategy of image
construction (Newman et al., 2011; Walther, Heide et al., 2008). In addition,
disclosing negative emotions yields no negative effect on impression management,
which is consistent with findings in Study 2b that need for impression management
was irrelevant to negative emotional disclosure. It again suggests that negative
emotions may not be so undesirable on Facebook. Alternatively, the null effect of
negative emotions may be attributed to the variations among specific emotions.
Previous research shows that anger expression is perceived as an indicator of power
and competence, leading to compliance and positive impression; sadness is perceived
as weakness, resulting in poor impression (Tiedens, 2001). As the perception of
negative emotions may vary, a significant pattern may not be found from the aggregate
results.
Although the need for emotional expression is fulfilled by disclosing positive
emotions and negative emotions, it renders no greater well-being. In fact, empirical
studies showed that talking about negative emotions did not actually achieve
emotional recovery, even though people always hold such a belief (Rimé et al., 2011).
The disclosure may fulfill the need for emotional expression at once, but its
subsequent impact on well-being may be detrimental and persistent. Usually, the
therapeutic effect of disclosing negative emotions requires cognitive reappraisal of the
experience (Rimé, 2009). Since the disclosure on Facebook is always on the fly and
brief (Naaman et al., 2010), it may not be able to help disclosers with cognitive
reappraisal. Meanwhile, replaying the memory of negative episodes could cause
unpleasant experiences (Rimé et al., 1992). Especially on Facebook, as the disclosed
content will not disappear, it may repeatedly remind individuals of the unpleasant
72
experiences. This may cause persistent and negative feelings. From an interpersonal
perspective, the audience may be unwilling to hear about negative experience; or, they
may provide helpless comments such as criticism and disagreement that could hurt the
disclosers more (House, 1981). Considering the invisible audience and asynchronous
communication on Facebook in particular, immediate and detailed responses may be
limited. More support may be needed for lifting well-being, such as physical
company, instrumental support, and in-depth conversation. For either possibility, the
association between emotional disclosure and individuals’ well-being may rely on the
responses from the audience. After all, emotional disclosure on SNSs is a practice
based on interpersonal communication. The result reveals the discrepancies between
individuals’ expectation and the actual outcome.
With respect to the role of need for impression management, results partially
support the meditational hypothesis. While disclosure is not directly related to life
satisfaction, greater disclosure of positive emotions nonetheless is associated with
greater life satisfaction by way of fulfilling the need for impression management.
Similarly, disclosing positive emotions connects to less loneliness through the same
route. As the need for impression management is fulfilled, greater life satisfaction and
less loneliness are anticipated. Conceptually, this type of need is very closed to one of
the fundamental needs—competence—that are suggested to be critical for well-being
Ryan & Deci, 2001). In eary and owalski’s (1990) theory, need for impression
management fosters well-being by maximizing rewards from social relations,
enhancing self-esteem and developing desired identities. Since Facebook affords rich
social interactions, it is very possible for disclosers to gain these rewards from the
public. Moreover, this occurs only in cases of disclosing positive emotions, as
negative disclosure is not related to image construction. Compared to the null effect of
73
fulfillment of need for emotional expression, this fact highlights that one of the critical
functions on Facebook is to gain positive impression from others. Moreover, such a
benefit could be manifest in a long-term well-being.
It is noteworthy that the association between disclosing negative emotions and
well-being was weaken once the disclosure of positive emotions is controlled. Since
positive disclosure is significantly correlated with negative disclosure, it suggests that
individuals who disclose positive emotions more are also likely to disclose negative
emotions. The original effect of negative emotions through need fulfillment may be
attributed to the overall intensity of emotional disclosure. That is why the effect
disappeared when positive disclosure was taken into account. The pure effect of
negative emotions disclosure on well-being cannot be explained by the need
fulfillment.
Considering the significant total effect and direct effect of negative emotional
disclosure on decreasing life satisfaction and increasing loneliness, the disclosure of
negative emotions itself does not necessarily alleviate loneliness. The re-experiencing
of negative emotions after posting negative experiences is one possible reason. An
alternative explanation may be that disclosers who disclose negative emotions are poor
at adjusting emotions (Lyubomirsky & Nolen-Hoeksema, 1995). Although the
disclosure may bring emotional benefit at that moment, the disclosers may not be able
to change the overall state. A similar paradox is presented in a recent study (Sheldon
et al., 2011), which showed that people feeling disconnected with others used
Facebook in the same fashion as people feeling connected. The underlying reason for
disconnected people to engage in Facebook is to use Facebook for coping with the
disconnectedness rather than for improving connectedness. In a similar vein, lonely
persons may just tend to cope with loneliness by disclosing negative emotions rather
74
than to enhance connectedness.
Taken together, emotional disclosure on Facebook serves for the need for
emotional expression and impression management; however, the function could not
fully account for the association between emotional disclosure and well-being. While
fulfillment of impression management by positive disclosure is an important part of
fostering well-being, the fulfillment of emotional expression shows lack of lasting
effect on well-being. Paradoxically, disclosing negative emotions is associated with
poor well-being, suggesting Facebook might be an emotional outlet for people who are
experiencing low well-being, but it cannot influence the overall well-being.
75
CHAPTER 5: GENERAL DISCUSSION
The overarching goal of this thesis is to understand emotional disclosure in the
social context of Facebook. I started the attempt by asking how the novel context of
Facebook makes it different from traditional disclosure channels, and then examined
the underlying motivational processes that connected social network structure with
emotional disclosure, and finally investigated the function of emotional disclosure in
light of its association with well-being.
In Study 1, I found self-disclosure on Facebook contains more emotional
words than semi-private writing, but fewer than real-life communication. The findings
supported the notion that emotional disclosure is related to the social context. In
addition, from the results it can be inferred that emotional disclosure on Facebook
reflects both the motivation to let others know as well as the motivation to maintain
positive impression. In Study 2, I testified this speculation with focus on the
contextual variations within Facebook context. Specifically, I formulated the two
motives as need for emotional expression and need for impression management. I
investigated how they mediated the association between network structure and
emotional disclosure. As expected, network size is associated with disclosure of
positive emotions through need for impression management, while network density is
associated with both disclosure of positive emotions and negative emotions because of
the need for emotional expression. The results indicate that the two motivational
needs separately account for the relationship between network properties and
emotional disclosure. As an extension of this finding, Study 3 tested the fulfillment of
need for emotional expression and need for impression management in explaining the
association between emotional disclosure behavior and well-being. It turns out that
emotional disclosure is positively associated with need fulfillment. However, only the
76
fulfillment of need for impression management accounts for the positive association
between positive emotional disclosure and well-being.
Altogether, this research demonstrates that impression management is an
important motivation in a large network context, and its fulfillment from disclosing
positive emotions is related to greater well-being. In contrast, a dense network
facilitates the need for emotional expression, but the fulfillment of need for emotional
expression cannot contribute to well-being.
Disclosure as a Contextual Behavior
Facebook, a novel communication environment, offers an opportunity to study
emotional disclosure that may be different from disclosure in traditional contexts.
Results from Study 1 confirmed that Face ook context influences individuals’
emotional disclosure. Study 2 further identified the contextual correlates with one’s
pattern of emotional disclosure.
Despite the evident link between social context and emotional disclosure, it is
unlikely that individuals have a concrete map of their entire social network in mind.
In reality, how exactly individuals conform with display rules depends on how they
interpret the contexts (Hogan, 2010b). It is possible that they react in light of a group
of salient individuals, which is merely a portion of the network that described in the
present research. In other words, discrepancy might exist etween an individuals’
actual network and his or her perceived audience. The discrepancy derives from the
invisibility of the audience and asynchronicity of communication. There might be a
silent audience that disclosers are unaware of and an audience that disclosers target but
that overlooks the posts. Nevertheless, it is still reasonable to assume that the overall
network structure reflects yet is not identical to the mental model group. When an
individual has a dense network on Facebook, there is a low chance for the post to be
77
seen by very different social circles; likewise, for an individual have a large network,
the number of viewers should be still relatively greater than that in a small network.5
Despite significant associations between social network properties and
emotional disclosure are found in this research, a definitive causal conclusion could
not be drawn. Social network structure and behaviors are intertwined, each shaping
the other (Golubović
9). Social network structure determines how individuals
behave; it can also go the other way around; that is, individuals’ ehaviors reinforce or
modify the structure. It is still possible that an individual in a large network may
disclose frequently in support of self-image; the frequent disclosure may attract more
audience, and that further expands the person’s network. However, as shown by the
model comparison in Study2b, the reversed cause-effect is not supported by the data in
current study. Comparing the two possibilities, the effect from social network on
behaviors might be stronger. Boyd (2007) contended that Face ook’s social etiquette
enjoins that refusing a friend request as socially inappropriate and so such refusals are
uncommon. Thus, a friend list is unlikely to be an intentional outcome; the formation
of network on Facebook is more likely to be passive. Instead of customizing the
audience, users are more likely to use a self-censorship strategy to decide things to
share or not. In this regard, it appears more logical for a social environment to
influence the motivations and behaviors of individuals inside the environment.
Although the reversed relationship cannot be completely excluded, current research
demonstrates the probable influence of social context on emotional disclosure. In
particular, need for emotional expression and need for impression management are
considered motivational processes that are elicited by a particular context and lead to
following behaviors of emotional disclosure. Although the present research cannot
completely exclude the reversed causal link, it mainly demonstrates the influence of
78
social context on emotional disclosure.
Another challenge to the social context account of emotional disclosure is the
results might reflect attributes that are more fundamental rather than the social context.
Individuals having a large social network are likely to be those who are inherently
prone to disclose. A construct related to this concern is emotional expressivity, which
refers to the trait of behaviorally (both verbally and nonverbally) reacting emotional
experience (Gross & John, 1995; Kring et al., 1994). Indeed, there are individual
differences in the innate propensity to express emotions (Trierweiler, Eid, &
Lischetzke, 2002). For example, in the event of an accident, there could be some
people crying out loudly while some others remaining calm. However, expressivity is
a measure of generalized expressivity, without concerning the communication channel
or valence of the emotion. It is an implicit propensity that may be independent of selfcontrol. Expressivity is different from emotional disclosure, which is conceptualized
as an act responding to the social environment in current research. In addition, past
research suggests that the use of emotional words is related to personality traits, such
as extraversion (e.g., Pennebaker & King, 1999; Qiu, Lin, Ramsay, & Yang, 2012).
However, data in this research showed no significant correlations between personality
traits and emotional disclosure, and only extraversion was correlated with network
size. Additionally, the need for emotional expression and need for impression
management were weakly related to personality, suggesting that personality would not
be a strong factor in this research.6 This is consistent with imé’s (1998) finding that
the tendency of sharing emotion is unrelated to the big five trait dimensions.
Altogether, it suggests that the relationship between social context and emotional
disclosure is a robust finding and is not attributed to personality.
Gender is another factor that may cause individual differences. A recent text
79
analysis on Facebook status updates found that women used more emotional words
than men did (Schwartz et al., 2013). However, in the present research, analyses
across all the three studies did not show significant contribution of gender to the
results. To date, the gender effect on emotional disclosure is still in controversy.
Women are always thought to be more emotional than men (LaFrance & Banaji, 1992)
and they express their emotion more frequently than men (Kring et la., 1994). An
analysis on a large corpus that combined various studies showed woman generally
used more affect words than men did (Newman, Groom, Handelman, & Pennebaker,
2008). However, LaFrance and Banaji (1992) suggested that in some other
circumstances, such as in private self-report or considering specific emotions, gender
effects were nonexistent or inconsistent,. For instance, men in management jobs may
be more likely to express negative emotions such as anger, compared to women
(Simpson & Stroh, 2004). A study on an online discussion platform found no gender
difference in emoticon usage and suggested that males would adapt to the females
style by using more emoticons to display emotions (Wolf, 2000). Another study about
comments on MySpace found that woman expressed more positive emotions than men
did, yet woman and men did not differ in expressing negative emotions (Thelwall,
Wilkinson, & Uppal, 2010). Given these inconsistent findings, future studies are
needed to better understand the gender differences on Facebook.
Relate Emotional Disclosure to Well-being
Recently, there is a widespread concern in social media research pertaining to
the relationship between using SNSs and well-being (e.g., Burke et al., 2010; Chen &
Lee, 2013; Ellison et al., 2007; große Deters & Mehl, 2013; Kalpidou et al., 2011;
Kim & Lee, 2011; Lampe, Vitak, & Ellison, 2013). The present research contributes
to this stream of research by focusing on how the emotional disclosure on Facebook is
80
related to well-being. Study 3 shows that disclosure of positive emotions is positively
associated with well-being via fulfilling the need of impression management; whereas
disclosure of negative emotions is negatively associated with well-being and the need
fulfillment does not contribute to the relationship. It implies that emotional disclosure
on Facebook is a double-edged sword and the Facebook context is not always
supportive for well-being.
As shown in the present research, need fulfillment is not always predictive of
greater well-being, especially with regards to the fulfillment of the need for emotional
expression. A possible explanation is that fulfillment of need for emotional expression
is temporal, taking place merely at the moment of disclosing. From the negative
correlation between this fulfillment and well-being, it could be inferred that people
who gain emotional relief on Facebook are those who persistently experience low
well-being. The need may be fulfilled by the support from the audience, but this
support cannot contribute to well-being in general. Moreover, past research suggests
that some motivations are negatively associated with well-being, especially if the
motivation points to external rewards (Sheldon, Ryan, Deci, & Kasser, 2004). In
pursuit of the same goal, such as accomplishing a project, people who value internal
rewards (e.g., personal growth, societal contribution) are more likely to attain greater
well-being. On the contrary, people who value external benefits (e.g., wealth, fame)
are likely to end up with poorer well-being. The underlying argument is that people
with extrinsic motives always gravitate towards superficial benefit that is contingent
and vulnerable. In dyadic communication a “reciprocity norm” (Gouldner
96 ) is
always expected. Individuals are likely to perform self-disclosure in response to the
other who have disclosed. In an online setting such as Facebook, however, the
relationship between recipients and disclosers becomes inequitable. Individuals may
81
put in more inputs (disclose a lot) than the audience’ inputs; the audience may respond
y the means of mere clicking “likes” instead of actually revealing their thoughts.
This may raise interpersonal tensions, which is harmful for well-being. As fulfillment
of emotional expression might mainly rely on the responses from the audience, it
would make individuals vulnerable and have a negative impact on well-being.
On the contrary, Facebook activity supports impression management in a more
lasting and positive way that is beneficial to well-being. This is in congruence with
previous research which suggests that Face ook increase individuals’ self-affirmation
(Toma & Hancock, 2013), without concerning about real feedback from others. It is
possible that the positive discourse itself provides a sense of self-competence to the
discloser. Certainly, how exactly the two needs rely on extrinsic rewards require
further investigation.
Besides the puzzle of negative influence on well-being, the concern about
causality arises in the association between emotional disclosure and well-being as
well. The relationship can be reciprocal too. It could be that lonely people are more
likely to disclose negative emotions and less likely to engage in positive disclosure,
whereas individuals with greater well-being are more likely to share positive emotions
and have less negative emotions to disclose (Jin, 2013). Likewise, individuals disclose
positive emotions that eventually support their self-images; but meanwhile, it could be
that people are more likely to present positive emotions when they feel they are
desirable. Nevertheless, well-being is always studied as the a consequence of using
media (e.g., Mckenna & Bargh, 2009; Valkenburg & Peter, 2009). Specifically, the
motivational theory emphasizes that the motivation is a proximal account for
psychological outcome (Vallerand, 1997). Social factors as causes trigger need
fulfillment that subsequently leads to psychological outcomes including affect,
82
cognition, and behavior. Hence, it is logically appropriate to regard well-being as an
outcome in this research. Future studies using a longitudinal approach can shed more
light on this debate.
Further, results in the present research showed that need fulfillment cannot
fully explain the relationship between emotional disclosure and well-being. Other
mechanisms may have not yet been covered in the current study. For example, coping
stress or reliving past experience may be the mediators between disclosing negative
emotions and well-being. In light of the facilitative mechanisms, close intimacy may
increase and stress may decrease in the process of emotional disclosure (Collins &
Miller, 1994; Gable et al., 2004; Graham et al., 2008; Mauss et al., 2011). Other
psychological processes, such as trust and cognitive reappraisal, may also serve for the
association between emotional disclosure and well-being and are worthy of further
exploration.
Revisit the Need for Emotional Expression and Need for Impression Management
Throughout the three studies, a core concern is the roles of need for emotional
expression and need for impression management, in terms of how their contingency on
context is related to the disclosure behaviors and how their fulfillment is related to
well-being. The two needs are proposed based on social motives that have been
implied in previous research. Study 2 demonstrates that the need for emotional
expression and the need for impression management are independent and divergent,
and Study 3 shows the fulfilled need for impression management accounts for the
positive correlation between positive emotional disclosure and greater well-being.
The two needs tap into different aspects of social motives in the process of
emotional disclosure. They both root in the need for socialization, because only
through the connection with others can the two needs be manifest. Put differently,
83
social connectedness is a potent source for emotional expression and impression
management. From another perspective, the two needs may comprise both
components of the self-regulatory systems—to approach rewards and to avoid threats
(Elliot & Covington, 2001), yet do so with different targets. The need for emotional
expression could contain a desire for seeking hedonism or an inclination to escape
from negative feelings; the need for impression management may contain the
propensity to gain positive evaluation as well as to avoid sanctions. Nevertheless, the
two needs are distinct from each other. The need for emotional expression aims to
signal an emotional attachment and trust; it is enhanced in dense networks, and it leads
to greater emotional disclosure regardless of valence.
Results in current research suggest that the need for emotional expression
motivates disclosure of both positive and negative emotions, suggesting the
importance of need for emotional expression in disclosure behaviors. After all, selfdisclosure is the behavior of broadcasting personal information (Forgas, 2011; Jourard
& Lasakow, 1958). In return, disclosing emotion indeed meets the need. However,
fulfillment of the need for emotional expression has no bearing on the relationship
between the behavior of emotional disclosure and the psychological outcome on wellbeing. The astonishing paradox raises an important issue of the boundary of emotional
disclosure on SNSs. Disclosure of negative emotions on Facebook appears not helpful
for people to alter lasting state and lift well-being; it may serve only as a buffer for
emotional expression in a short time frame.
The present research also demonstrates how impression management is
manifest by emotional disclosure. As need for impression management has been
frequently addressed in Facebook research as the reason for selective self-presentation
(e.g., Siibak, 2009; Strano, 2008; Tong et al., 2008; Zhao et al., 2008), the results from
84
Study 2b provide direct evidence that positive disclosure is driven by need for
impression management. Moreover, disclosing positive emotions indeed fulfills the
need for impression management. The convergent results echo a previous claim that
SNSs are ideal media for impression management (Ellison et al., 2006). In a social
platform such as Face ook the pu lic creates constraints on individuals’ disclosure
but meanwhile, it offers affirmation for individuals. It implies that using Facebook for
emotional disclosure may have instrumental effects on interpersonal impression,
which in general benefits well-being.
Comparison between the two needs indicates that they did not conflict with
each other. Both needs motivate greater emotional disclosure; there is no interaction
effect between the two needs. The only difference is that need for impression
management is not related to disclosure of negative emotions whereas need for
emotional expression promotes it. Although the results should be interpreted with one
network property taken into account and the other controlled, both needs may play
their roles jointly in a given context. A question is which need plays a predominant
role? Results from Study 2 showed that the need for impression management would
increase with an expanding network, leading to greater disclosure of positive
emotions. At the same time, as density decreases, the need for emotional expression is
expected to reduce. As a result, negative emotions disclosure will decrease, but the
change of positive emotions depends on the magnitude of density change. Similarly,
as a network shrinks, density increases with size decreasing, the emotional disclosure
of negative emotions and positive emotions would both increase as a result. Whether
the overall disclosure of positive emotions decline or not depends on how the size
changes the need for impression management. It is possible that in a very large
network, emotional disclosure is mainly motivated by the need for impression
85
management. The context as such is less likely to provide attachment and care as
anticipated by the need for emotional expression. In contrast, emotional disclosure in
a very small and tight network may be mainly driven by the need for emotional
expression, as the close relationship makes impression management less important,
with the provision of social support being more available. Since a personal network
on Facebook is often relatively large with diverse social connections, it is most likely
that the impression management is more salient than the need for emotional
expression.
According to Bargh and Chartrand’s (1999) theoretical model, the contextual
cues could activate the pursuit of certain goals in an automatic fashion without
conscious deliberation. If this is true, the decision of emotional disclosure in the
context of SNSs appears to be made unconsciously. This might explain the low
ratings on need for emotional expression and need for impression management, which
were just around the midpoint. Fulfillment of the two needs showed similarly low
ratings on average. It is possible that individuals are not fully aware of their needs.
Probably, the dual pathways, in terms of need for emotion expression and need for
impression management, may vary in the degree of conscious effort. The need for
emotional expression could be more spontaneous and automatic, while the need for
impression management is more deliberate, as it needs efforts to craft the information.
Concerning the finding that need for expressing emotion increases with network
density, it is reasonable for people to employ less conscious effort in a closed network
and to be more expressive. It also makes sense that need for impression management
increases with network size, because people may need more deliberation in a massive
environment.
86
Implications
The findings have several theoretical implications. First, they contribute to
existing literature on emotional disclosure by placing the issue in a novel and complex
social environment. The results shed light on how emotional disclosure is governed
by the massive and collapsed context on Facebook. Second, the study fills up some
gaps in the research on emotional disclosure by explicitly and empirically addressing
the underlying motives. In particular, need for emotional expression and need for
impression management are highlighted and distinguished. It demonstrates which
motivational needs promote emotional disclosure is contingent on the social context.
Third, the study examines how motivational needs are translated into social
networking practice and how they contribute to individuals’ well-being.
From a practical perspective, this research enriches the understanding of
emotional disclosure as a strategy of emotional regulation. Gross (1998) emphasizes
that emotional regulation, which consists of sustaining positive emotions and coping
with negative emotions, is associated with social context. By examining how
emotional disclosure is related to social network structure, this research indicates how
individuals could harness Facebook for emotional regulation. Clinicians might use
Facebook as a convenient social setting for group sessions, in combination with
updating status as a tool for expression. According to the results, it is essential to
carefully manage the size and density of such a group. However, the effect of
leveraging SNSs for well-being may be limited and demands cautions.
For practitioners in the industry of social media, the findings offer insight in
designing an engaging social network. Individuals have different needs under
different social contexts. As an individual’s network expands, it appears that
impression management would play a major role in hampering in-depth self-
87
disclosure. If individuals could freely switch between different social contexts and
choose an appropriate context that supports a particular motivation, they would be
more likely to engage in the platform.
Limitations and Future Directions
The present research has several limitations. First, it is limited by the samples.
Participants in this research include only Singaporean college students. It is unclear to
what extent that current findings can be generalized to other populations, especially
because college students’ social networks might not be as diverse as those of adults.
Another issue open for exploration is the social network metrics. On one hand,
although network size and density represent important properties of a network, there
are other sophisticated metrics possibly playing a role. For example, a person’s
centrality in the network and the homophily of the network may affect how and what
an individual to say. Those properties may have additional implications about the
social processes. Thus, much effort is needed to clarify psychological meanings of
these network properties. On the other hand, to attain a more valid measurement of
the actual audience is important in future study. The network can be assessed on a
post-by-post basis, with disclosed contents corresponding to specific visibility
settings.7 There is no doubt that a person’s social network would evolve with the use
of Facebook, and the users’ communication strategy might change as well. Certainly,
future study should investigate the dynamic changes of both network and emotional
disclosure to uncover the causal relationship. Controlled experiments are also
desirable to manipulate specific network structures, to clarify the causal relationship.
The measure of emotional disclosure is also a complicated issue. Choosing
self-report or linguistic data entails a trade-off. Although social media provide a
convenient database for studying natural language, the validity of the results is
88
restricted by the analytical approach. As for the self-report, it may be biased by
memory, introspection, and social desirability. The different correlations between
emotional disclosure and network structure as shown in Study 2a and 2b reveal the
differences of the two approaches. Thus, to develop a more valid measure is crucial in
the future.
Another issue concerns the categorization of emotion. The present research
categorizes emotion under two headings––positive and negative. This leaves some
particular emotions undistinguishable. For example, some instances of positive
emotions may e less desira le ecause it relies on another one’s pain. Meanwhile,
some negative emotions may be acceptable, such as guilty, that indicates a person’s
reflection and intention to be good. In addition, the targets involved in the emotional
experience are not taken into account. Emotional events involving interpersonal
relationship may be judged differently from those concerning only a single person’s
private issues. In the future, a finer categorization could refine the understanding of
emotional disclosure.
Even though this research is subject to above limitations, it opens a venue for
further exploration. It would be intriguing to test other psychological mechanisms for
a better understanding of the contextual contingency and well-being consequence of
emotional disclosure. Future studies could also consider potential moderators such as
the individuals’ interpersonal goals and dispositional expressivity, or the
responsiveness from the audience. Explorations on different media should help
researchers to establish a comprehensive framework for understanding emotional
disclosure.
Conclusion
The present research reveals that emotional disclosure on Facebook is
89
distinctive from disclosure under other settings. By considering emotional disclosure
in a social network context, results indicate that emotional disclosure on Facebook is
contingent on the structure of personal networks, with need for emotional expression
and need for impression management as the underlying processes. The actual
disclosure behavior meets the two types of needs, but only the fulfillment of need for
impression management contributes to the association between positive emotional
disclosure and greater well-being. Through three studies, the present research sheds
light on the contextual antecedents and possible consequences of emotional disclosure
on Facebook. It highlights the critical roles of motivational needs in the associations
between emotional disclosure, social context, and well-being.
90
REFERENCES
Ahn, D., & Shin, D.-H. (2013). Is the social use of media for seeking connectedness or
for avoiding social isolation? Mechanisms underlying media use and subjective
well-being. Computers in Human Behavior, 29(6), 2453-2462.
doi:10.1016/j.chb.2012.12.022
Alison B., J., Sanders-Jackson, A., & Smallwood, A. M. (2006). IMing, text
messaging, and adolescent social networks. Journal of Computer-Mediated
Communication, 11(2), 577-592.
Altman, I., & Taylor, D. (1973). Social penetration:The development of interpersonal
relationships. New York: Holt, Rinehart &Winston.
Andre, P., Dix, A., & White, R. (2011, February). Expressing well-being online:
Towards self-reflection and social awareness. Paper presented at the
iConference, Seattle, WA.
Antheunis, M. L., Valkenburg, P. M., & Peter, J. (2007). Computer-mediated
communication and interpersonal attraction: An experimental test of two
explanatory hypotheses. Cyberpsychology & Behavior, 10(6), 831-836.
doi:10.1089/cpb.2007.9945
Arkin, R., Lake, E., & Baumgardner, A. (1986). Shyness and self-presentation. In W.
Jones, J. Cheek & S. Briggs (Eds.), Shyness (pp. 189-203): Springer US.
Arkin, R. M., Appelman, A. J., & Burger, J. M. (1980). Social anxiety, selfpresentation, and the self-serving bias in causal attribution. Journal of
Personality and Social Psychology, 38(1), 23-35. doi:10.1037/00223514.38.1.23
Barasch, A., & Berger, J. (2014). Broadcasting and narrowcasting: How audience size
impacts what people share. Journal of Marketing Research. Advance online
91
publication. doi:10.1509/jmr.13.0238
Barash, V., Ducheneaut, N., Isaacs, E., & Bellotti, V. (2010). Faceplant: Impression
(mis)management in Facebook status update. Proceedings of the Fourth
International AAAI Conference on Weblogs and Social Media, 207-210.
Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being.
American Psychologist, 54(7), 462-479. doi:10.1037/0003-066X.54.7.462
Bargh, J. A., & McKenna, K. Y. A. (2004). The internet and social life. Annual Review
of Psychology, 55(1), 573-590. doi:10.1146/annurev.psych.55.090902.141922
Barker, R. G. (1968). Ecological psychology: Concepts and methods for studying the
environment of human behavior. Stanford, CA: Stanford University Press.
Baron, N. S. (1998). Letters by phone or speech by other means-the linguistics of
email. Language & Communication, 18(2), 133-170. doi:10.1016/S02715309(98)00005-6
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is
stronger than good. Review of General Psychology, 5(4), 323-370.
doi:10.1037/1089-2680.5.4.323
Baym, N. K., & Boyd, d. (2012). Socially mediated publicness: An introduction.
Journal of Broadcasting & Electronic Media, 56(3), 320-329.
doi:10.1080/08838151.2012.705200
Bazarova, N. N., Taft, J. G., Choi, Y. H., & Cosley, D. (2013). Managing impressions
and relationships on Facebook: Self-presentational and relational concerns
revealed through the analysis of language style. Journal of Language and
Social Psychology, 32(2), 121-141. doi: 10.1177/0261927x12456384.
Beals, K. P., Peplau, L. A., & Gable, S. L. (2009). Stigma management and wellbeing: The role of perceived social support, emotional processing, and
92
suppression. Personality and Social Psychology Bulletin, 35(7), 867-879.
doi:10.1177/0146167209334783
Berger, C. R., & Calabrese, R. J. (1975). Some explorations in initial interaction and
beyond: Toward a developmental theory of interpersonal communication.
Human Communication Theory, 1, 99-112. doi:10.1111/j.14682958.1975.tb00258.x
Bernstein, M. S., Bakshy, E., Burke, M., & Karrer, B. (2013). Quantifying the
invisible audience in social networks. Proceedings of the 31th SIGCHI
Conference on Human Factors in Computing Systems, 21-30.
doi:10.1145/2470654.2470658
Borgatti, S. P., Everett, M. G., & Freeman, L. C. (2002). Ucinet for Windows:
Software for Social Network Analysis. Harvard, MA: Analytic Technologies.
Borgatti, S. P., Mehra, A., Brass, D. J., & Labianca, G. (2009). Network analysis in
the social sciences. Science, 323(5916), 892-895. doi:10.1126/science.1165821
Boyd, D. (2007). Social network sites: Public, private, or what? Knowledge Tree, 13,
1-7.
Boyd, D. (2011). Social network sites as networked publics: Affordances, dynamics,
and implications. In Z. Papacharissi (Ed.), A networked self: Indentiy,
community, and culture on social netowrk sites (pp. 39-58). New York, NY:
Routledge.
Boyd, D., & Ellison, N. (2007). Social network sites: Definition, history, and
scholarship. Journal of Computer-Mediated Communication, 13(1), 210–230.
doi:10.1111/j.1083-6101.2007.00393.x
Boyd, D. M. (2008). Taken out of context: American teen sociality in networked
publics (Doctoral disseration). Available from ProQuest disseration and theses
93
databse. (UMI No. 3353116)
Brown, J. D., Collins, R. L., & Schmidt, G. W. (1988). Self-esteem and direct versus
indirect forms of self-enhancement. Journal of Personality and Social
Psychology, 55(3), 445-453. doi:10.1037/0022-3514.55.3.445
Buck, R., Losow, J. I., & Murphy, M. M. (1992). Social facilitation and inhibition of
emotional expression and communication. Journal of Personality and Social
Psychology, 63(6), 962-968.
Burke, M., Marlow, C., & Lento, T. (2010). Social network activity and social wellbeing. Proceedings of the 28th SIGCHI Conference on Human Factors in
Computing Systems, 1909-1912. doi:10.1145/1753326.1753613
Burt, R. S. (1976). Positions in networks. Social Forces, 55(1), 93-122.
Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge,
MA: Harvard University Press.
Burt, R. S. (2000). The network structure of social capital. Reseasrch in
Organizational Behavior, 22, 345-423. doi:10.1016/S0191-3085(00)22009-1
Burt, R. S. (2001). Structural holes versus network closure as social capital. In N. Lin,
K. Cook & R. S. Burt (Eds.), Social capital: Theory and research. Sociology
and economics: Controversy and integration series (pp. 31-56). New York,
NY: Aldine de Gruyter.
Burt, R. S. (2012). Network-related personality and the agency question: Multirole
evidence from a virtual world1. American Journal of Sociology, 118(3), 543591. doi:0002-9602/2012/11803-0001
Cacioppo, J. T., & Patrick, W. (2008). Loneliness: Human nature and the need for
social connection. New York: W.W. Norton.
94
Cameron, C. (2012). Millions have not reviewed Facebook privacy settings: Here's
how [web log message]. Retrived from
http://www.welivesecurity.com/2012/05/15/millions-have-not-reviewedFacebook-privacy-settings-heres-how/
Carr, C. T., Schrock, D. B., & Dauterman, P. (2012). Speech acts within Facebook
status messages. Journal of Language and Social Psychology, 31(2), 176-196.
doi: 10.1177/0261927X12438535
Chaikin, A. L., & Derlega, V. J. (1974). Liking for the norm=breaker in selfdisclosure. Journal of Personality, 42(1), 117-129. doi:10.1111/j.14676494.1974.tb00560.x
Chelune, G. J. (1975). Self-disclosure: An elaboration of its basic dimensions.
Psychological reports, 36(1), 79-85. doi:10.2466/pr0.1975.36.1.79
Chen, W., & Lee, K.-H. (2013). Sharing, liking, commenting, and distressed? The
pathway between Facebook interaction and psychological distress.
CyberPsychology, Behavior & Social Networking, 16(10), 728-734. doi:
10.1089/cyber.2012.0272
Chen, Y., Yao, M., & Yan, W. (2013). Materialism and well-being among chinese
college students: The mediating role of basic psychological need satisfaction.
Journal of Health Psychology, Advance online publication. doi:
10.1177/1359105313488973
Chua, V., Madej, J., & Wellman, B. (2011). Personal communities: The world
according to me. In P. Carrington & J. Scott (Eds.), The sage handbook of
social network analysis (pp. 101-115). London: Sage.
Chung, C. K., & Pennebaker, J. W. (2012). Linguistic inquiry and word count (liwc):
95
Pronounced "luke" and other useful facts. In P. McCarthy & C. Boonthum
(Eds.), Applied natural language processing and content analysis:
Identification, investigation, and resolution (pp. 206-229). Hershey, PA: IGI
Global.
Chung, C. K., & Pennebaker, J. W. (in press). Counting little words in big data: The
psychology of communities, cultures, and history. In J. László, J. Forgas & O.
Vincze (Eds.), Social cognition and communication. New York: Psychology
Press. Retrieved from http://www.sydneysymposium.unsw.edu.au/2012/
chapters/PennebakerEASP2012.pdf
Church, A. T., Katigbak, M. S., Locke, K. D., Zhang, H., Shen, J., Vargas-Flores, J. d.
J., . . . Ching, C. M. (2013). Need satisfaction and well-being: Testing selfdetermination theory in eight cultures. Journal of Cross-Cultural Psychology,
44(4), 507-534. doi:10.1177/0022022112466590
Coleman, J. S. (1988). Social capital in the creation of human capital. American
Journal of Sociology, 94, 95-120. doi:0002-9602/89/9407-0010
Collins, N. L., & Miller, C. (1994). Self-disclosure and liking: A meta-analytic review.
Psychological Bulletin, 116(3), 457-475. doi:0033-2909/94
Cozby, P. C. (1973). Self-disclosure: A literature review. Psychological bulletin,
79(2), 73-91. doi:10.1037/h0033950
Cunningham, M. R. (1988). Does happiness mean friendliness? Induced mood and
heterosexual self-disclosure. Personality and Social Psychology Bulletin,
14(2), 283-297. doi: 10.1177/0146167288142007
Daft, R. L., & Lengel, R. H. (1984). Information richness. A new approach to
managerial behavior and organization design. Research in Organizational
Behaviour, 6, 191-233.
96
DeNeve, K. M., & Cooper, H. (1998). The happy personality: A meta-analysis of 137
personality traits and subjective well-being. Psychological Bulletin, 124(2),
197-229. doi:0033-2909/98
Denti, L., Barbopuolos, I., Nilsson, I., Holmberg, L., Thulin, M., Wendeblad, M., . . .
Davidsson, E. (2012). Sweden's largest Facebook study: A survey of 1000
swedish Facebook users (gri-rapport 2012:3). Retrieved from University of
Gothenburg website: https://gupea.ub.gu.se/bitstream/2077/28893/1/
gupea_2077_28893_1.pdf
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with
life scale. Journal of personality assessment, 49(1), 71-75.
doi:10.1207/s15327752jpa4901_13
Diener, E., & Lucas, R. E. (2000). Subjective emotional well-being. In M. Lewis & J.
M. Haviland (Eds.), Handbook of emotions (2nd ed., pp. 325-337). New York,
NY: Guilford.
Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being:
Three decades of progress. Psychological bulletin, 125(2), 276-302.
doi:10.1037/0033-2909.125.2.276
Dimmick, J., Ramirez, A., Wang, T., & Lin, S.-F. (2007). Extending society': The role
of personal networks and gratification-utilities in the use of interactive
communication media. New Media & Society, 9(5), 795-810.
doi:10.1177/1461444807081225
Dodds, P. S., Harris, K. D., Kloumann, I. M., Bliss, C. A., & Danforth, C. M. (2011).
Temporal patterns of happiness and information in a global social network:
Hedonometrics and Twitter. PLoS One, 6(12), e26752. doi:
10.1371/journal.pone.0026752
97
Ekman, P., & Friesen, W. V. (1975). Unmasking the face: A guide to recognizing
emotions from facial clues. Oxford, England: Prentice-Hall.
Elliot, A. J., & Covington, M. V. (2001). Approach and avoidance motivation.
Educational Psychology Review, 13(2), 73-92.
Ellison, N., Steinfield, C., & Lampe, C. (2007). The benefit of Facebook " friends":
Social capital and college students' use of online social network sites. Journal
of Computer-Mediated Communication, 12(4), 1143-1168. doi:10.1111/j.10836101.2007.00367.x
Ellison, N. B., Heino, R. D., & Gibbs, J. L. (2006). Self-presentation processes in the
online dating environment. Journal of Computer-Mediated Communication,
11(2), 415–441. doi:10.1111/j.1083-6101.2006.00020.x
Facebook. (2010). What's on your mind. Retrived from
http://www.Facebook.com/notes/Facebook-data-team/whats-on-yourmind/477517358858
Fischer, A., Manstead, A., Evers, C., Timmers, M., & Valk, G. (2004). Motives and
norms undelrying emotion regulation. In P. Philipport & R. S. Feldman (Eds.),
The regulation of emotion. New York: Lawrence Erlbaum Associates.
Fischer, A. H., Manstead, A. S. R., & Zaalberg, R. (2003). Social influences on the
emotion process. European Review of Social Psychology, 14(1), 171-201.
doi:10.1080/10463280340000054
Fischer, C. S. (1982). To dwell among friends: Personal networks in town and city.
Chicago, IL: University of Chicago Press.
Fitzgerald, B. (2012, Sepetember 7). Facebook has more female users but much more
male journalists covering it. The Huffington Post. Retrieved from
http://www.huffingtonpost.com/2012/07/09/women-Facebook-twitter-
98
pinterest_n_1655164.html
Forgas, J. P. (2011). Affective influences on self-disclosure: Mood effects on the
intimacy and reciprocity of disclosing personal information. Journal of
Personality and Social Psychology, 100(3), 449-461. doi:10.1037/a0021129
Fridlund, A. J. (1991a). Evolution and facial action in reflex, social motive, and
paralanguage. Biological Psychology, 32(1), 3-100. doi:10.1016/03010511(91)90003-Y
Fridlund, A. J. (1991b). Sociality of solitary smiling: Potentiation by an implicit
audience. Journal of personality and social psychology, 60(2), 229-240.
doi:10.1037/0022-3514.60.2.229
Fridlund, A. J. (1994). Human facial expression: An evolutionary view. San Diego:
Academic Press.
Fridlund, A. J. (1997). The new ethology of human facial expressions. In J. A. Russell
& Fernandez-Dols (Eds.), The psychology of facial expression (pp. 103-129).
Cambridge: Cambridge Universit Press.
Fridlund, A. J., Sabini, J. P., Hedlund, L. E., Schaut, J. A., Shenker, J. I., & Knauer,
M. J. (1990). Audience effects on solitary faces during imagery: Displaying to
the people in your head. Journal of Nonverbal Behavior, 14, 113-137.
Gable, S. L., Gonzaga, G. C., & Strachman, A. (2006). Will you be there for me when
things go right? Supportive responses to positive event disclosures. Journal of
Personality and Social Psychology, 91(5), 904-917. doi:10.1037/00223514.91.5.904
Gable, S. L., & Reis, H. T. (2010). Good news! Capitalizing on positive events in an
interpersonal context. Advances in Experimental Social Psychology, 42, 195257. doi:10.1016/s0065-2601(10)42004-3
99
Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when
things go right? The intrapersonal and interpersonal benefits of sharing positive
events. Journal of Personality and Social Psychology, 87(2), 228-245.
doi:10.1037/0022-3514.87.2.228
Garton, L., Haythornthwaite, C., & Wellman, B. (1997). Studying online social
networks. Journal of Computer－Mediated Communication, 3(1).
doi:10.1111/j.1083-6101.1997.tb00062.x/full
Gibbs, J. L. (2006). Self-presentation in online personals: The role of anticipated
future interaction, self-disclosure, and perceived success in internet dating.
Communication Research, 33(2), 152-177. doi:10.1177/0093650205285368
Gilbert, E., & Karahalios, K. (2009). Predicting tie strength with social media.
Proceedings of the 27th SIGCHI Conference on Human Factors in Computing
Systems, 211-220. doi: 10.1145/1518701.1518736
Giles, H., Coupland, N., & Coupland, J. (1991). Accommodation theory:
Communication context and consequence. In H. Giles, J. Coupland & N.
Coupland (Eds.), Contexts of accommodation: Developments in applied
sociolinguistics (pp. 1-68). Cambridge, UK: Cambridge University Press.
Goffman, E. (1959). The presentation of self in everyday life. New York: Anchor.
Golder, S. A., & Macy, M. W. (2011). Diurnal and seasonal mood vary with work,
sleep, and daylength across diverse cultures. Science, 333(6051), 1878-1881.
doi:10.1126/science.1202775
Golu ović N. (
9). Network structure of social capital. Economics and
Organization, 6(3), 211-219.
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American
sociological review, 25(2), 161-178. doi:10.2307/2092623
100
Graham, S. M., Huang, J. Y., Clark, M. S., & Helgeson, V. S. (2008). The positives of
negative emotions: Willingness to express negative emotions promotes
relationships. Personality and Social Psychology Bulletin, 34(3), 394-406.
doi:10.1177/0146167207311281
Granovetter, M. (1985). Economic action and social structure: The problem of
embeddedness. American Journal of Sociology, 91(3), 481-510.
doi:10.1086/228311
Granovetter, M. S. (1973). The strength of weak ties. American journal of sociology,
1360-1380.
Greenberg, M. A., & Stone, A. A. (1992). Emotional disclosure about traumas and its
relation to health: Effects of previous disclosure and trauma severity. Journal
of Personality and Social Psychology, 63(1), 75-84. doi:10.1037/00223514.63.1.75
Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review.
Review of General psychology, 2(3), 271-299. doi:10.1037/1089-2680.2.3.271
Gross, J. J., & John, O. P. (1995). Facets of emotional expressivity: Three self-report
factors and their correlates. Personality and Individual Differences, 19(4), 555568. doi:10.1016/0191-8869(95)00055-B
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects of inhibiting
negative and positive emotion. Journal of Abnormal Psychology, 106(1), 95103. doi:10.1037/0021-843X.106.1.95
Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation
processes: Implications for affect, relationships, and well-being. Journal of
Personality and Social Psychology, 85(2), 348-362. doi: 10.1037/00223514.85.2.348
101
Gross, J. J., Richards, J. M., & John, O. P. (2006). Emotion regulation in everyday life.
In D. K. Snyder, J. Simpson & J. N. Hughes (Eds.), Emotion regulation in
couples and families: Pathways to dysfunction and health (pp. 13–35).
Washington, DC: American Psychological Association.
große Deters, F., & Mehl, M. R. (2013). Does posting Facebook status updates
increase or decrease loneliness? An online social networking experiment.
Social Psychological and Personality Science, 4(5), 579-586.
doi:10.1177/1948550612469233
Hackenbracht, J., & Gasper, K. (2013). I'm all ears: The need to belong motivates
listening to emotional disclosure. Journal of Experimental Social Psychology,
49(5), 915-921. doi:10.1016/j.jesp.2013.03.014
Hancock, J. T., Landrigan, C., & Silver, C. (2007). Expressing emotion in text-based
communication. Proceedings of the 25th SIGCHI Conference on Human
Factors in Computing Systems, 929-932. doi:10.1145/1240624.1240764
Hanneman, R. A., & Riddle, M. (2005). Introduction to social network methods.
Retrieved from http:/faculty.ucr.edu/~hanneman/nettext/
Hatfield, E. (1984). The dangers of intimacy. In V. J. Derlega (Ed.), Communication,
intimacy, and close relationships (pp. 207-220). New York: Academic Press.
Hayes, A. F. (2009). Beyong Baron and Kenney: Statistical mediation analysis in the
new milennium. Communication Monographs, 76(4), 408-420. doi:
10.1080/03637750903310360
Henderson, S., & Gilding, M. (2004). I've never clicked this much with anyone in my
life: Trust and hyperpersonal communication in online friendships. New Media
Society, 6(4), 487-506. doi:10.1177/146144804044331
Hipp, J. R. (2008). Network-based approaches for measuring social capital. In I.
102
Kawachi, S. V. Subramanian & D. Kim (Eds.), Social capital and health (pp.
63-81). New York: Springer.
Hogan, B. (2008a). Analysing social networks via the internet. In N. Fielding, R. M.
Lee & G. Blank (Eds.), Handbook of online research methods (pp. 141–160).
London: Sage Publications.
Hogan, B. (2008b). A comparison of on and offline networks through the Facebook
API. Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1331029
Hogan, B. (2010a). Name gen web. Retrieved from http://apps.Facebook.com/
namegenweb/
Hogan, B. (2010b). The presentation of self in the age of social media: Distinguishing
performances and exhibitions online. Bulletin of Science, Technology &
Society, 30(6), 377-386. doi:10.1177/0270467610385893
Hogan, B. J. (2009). Networking in everyday life (Doctoral dissertation). Available
from ProQuest Dissertation and Theses Database. (UMI No. NR59091)
Hogan, R., Jones, W. H., & Cheek. (1985). Socioanalytic theory: An alternative to
armadilo psychology. In B. R. Schlenker (Ed.), The self and social life (pp.
175-198). New York: Mcgraw-hill.
House, J. S. (1981). Work stress and social support. Reading, MA: Addison-Wesley
Hu, L. t., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling: A Multidisciplinary Journal, 6(1), 1-55.
Hu, Y., Wood, J. F., Smith, V., & Westbrook, N. (2004). Friendships through im:
Examining the relationship between instant messaging and intimacy. Journal
103
of Computer-Mediated Communication, 10(1). doi:10.1111/j.10836101.2004.tb00231.x
Huberman, B., Romero, D., & Wu, F. (2008). Social networks that matter: Twitter
under the microscope. First Monday, 14(1).
Izquierdo, L. R., & Hanneman, R. A. (2006). Introduction to the formal analysis of
social networks using mathematica. Retrieved from
http://luis.izqui.org/papers/Izquierdo_Hanneman_2006-version2.pdf
Jakobs, E., Manstead, A. S. R., & Fischer, A. (2001). Social context effects on facial
activity in a negative emotional sharing. Emotion, 1(1), 51-69.
doi:10.1037/1528-3542.1.1.51
Jakobs, E., Manstead, A. S. R., & Fischer, A. H. (1999a). Social motives and
emotional feelings as determinants of facial displays: The case of smiling.
Personality and Social Psychology Bulletin, 25(4), 424-435.
doi:10.1177/0146167299025004003
Jakobs, E., Manstead, A. S. R., & Fischer, A. H. (1999b). Social motives, emotional
feelings, and smiling. Cognition & Emotion, 13(4), 321-345.
doi:10.1080/026999399379212
Jin, B. (2013). How lonely people use and perceive Facebook. Computers in Human
Behavior, 29(6), 2463-2470.
Joinson, A. N. (2008). Looking at, looking up or keeping up with people?: Motives
and use of Facebook. Proceedings of the 26th SIGCHI Conference on Human
Factors in Computing Systems, 1027-1036. doi:10.1145/1357054.1357213
Jourard, S. M., & Lasakow, P. (1958). Some factors in self-disclosure. Journal of
Abnormal and Social Psychology, 56(1), 91-98.
104
Jung, T., Youn, H., & McClung, S. (2007). Motivations and self-presentation
strategies on korean-based "cyworld" weblog format personal homepages.
Cyberpsychology & Behavior, 10(1), 24-31. doi:10.1089/cpb.2006.9996
Kahn, J. H., Tobin, R. M., Massey, A. E., & Anderson, J. A. (2007). Measuring
emotional expression with the linguistic inquiry and word count. The American
Journal of Psychology, 120(2), 263-286. doi: 10.2307/20445398
Kalpidou, M., Costin, D., & Morris, J. (2011). The relationship between Facebook and
the well-being of undergraduate college students. CyberPsychology, Behavior
& Social Networking, 14(4), 183-189. doi:10.1089/cyber.2010.0061
Keltner, D., & Haidt, J. (1999). Social functions of emotions at four levels of analysis.
Cognition & Emotion, 13(5), 505-521. doi:10.1080/026999399379168
Keyes, C. L. M. (1998). Social well-being. Social Psychology Quarterly, 61(2), 121140. doi:10.2307/2787065
Kilburn, M. (2011). Why individuals choose to post incriminating information on
social networking sites: Social control and social disorganization theories in
context. International Review of Information Ethics, 16, 55-59.
Kim, J., & Lee, J.-E. R. (2011). The Facebook paths to happiness: Effects of the
number of Facebook friends and self-presentation on subjective well-being.
Cyberpsychology, Behavior, and Social Networking, 14(6), 359-364.
doi:10.1089/cyber.2010.0374
Kim, J. Y., Chung, N., & Ahn, K. M. (2013). Why people use social networking
services in korea the mediating role of self-disclosure on subjective well-being.
Information Development, Advanced online publication. doi:
10.1177/0266666913489894
King, L. A., & Pennebaker, J. W. (1998). What's so great about feeling good?
105
Psychological Inquiry, 9(1), 53-56. doi:10.1207/s15327965pli0901_8
Kivran-Swaine, F., & Naaman, M. (2011). Network properties and social sharing of
emotions in social awareness streams. Proceedings of the ACM 2011
Conference on Computer Supported Cooperative Work, 379-382.
doi:10.1145/1958824.1958882
Kline, R. B. (2011). Principles and practice of structural equation modeling. New
York: Guilford Press.
ler F.
iedl C.
etter C. Leimeister, J. M., & Krcmar, H. (2010). Social
connectedness on Facebook: An explorative study on status message usage.
Proceedings of the Sxteen Americas Conference on Information Systems, Paper
247. Retrieved from http://aisel.aisnet.org/amcis2010/2247
Kramer, A. D. I. (2010). An unobtrusive behavioral model of "gross national
happiness". Proceedings of the 28th SIGCHI Conference on Human Factors in
Computing Systems, 287-290. doi:10.1145/1753326.1753369
Kramer, A. D. I., & Chung, K. (2011). Dimensions of self-expression in Facebook
status updates. Proceedings of the 5th International AAAI Conference on
Weblogs and Social Media, 169-176.
Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford, A.
(2002). Internet paradox revisited. Journal of Social Issues, 58(1), 49-74.
doi:10.1111/1540-4560.00248
Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukophadhyay, T., & Scherlis,
W. (1998). Internet paradox: A social technology that reduces social
involvement and psychological well-being? American Psychologist, 53(9),
1017-1031. doi:10.1037/0003-066X.53.9.1017
Kring, A. M., Smith, D. A., & Neale, J. M. (1994). Individual differences in
106
dispositional expressiveness: Development and validation of the emotional
expressivity scale. Journal of Personality and Social Psychology, 66(5), 934949. doi:10.1037/0022-3514.66.5.934
LaFrance, M., & Banaji, M. (1992). Toward a reconsideration of the gender-emotion
relationship. In M. S. Clark (Ed.), Emotion and social behavior (Vol. 14, pp.
178-201). Newbury Park, CA: Sage Publications.
Lampe, C., Vitak, J., & Ellison, N. (2013). Users and nonusers: Interactions between
levels of adoption and social capital. Proceedings of the 2013 Conference on
Computer Supported Cooperative Work, 809-820.
doi:10.1145/2441776.2441867
Langens, T. A. (2005). Written emotional expression and emotional well-being: The
moderating role of fear of rejection. Personality and Social Psychology
Bulletin, 31(6), 818-830. doi:10.1177/0146167204271556
Langston, C. (1994). Capitalizing on and coping with daily-life events: Expressive
resonses to positive events. Journal of Personality and Social Psychology,
67(6), 1112-1125. doi:10.1037/0022-3514.67.6.1112
LaFrance, M., & Banaji, M. (1992). Toward a reconsideration of the gender-emotion
relationship. In M. S. Clark (Ed.), Emotion and social behavior (Vol. 14, pp.
178-201). Newbury Park, CA: Sage Publications.
Laurenceau, J.-P., Barrett, L. F., & Pietromonaco, P. R. (1998). Intimacy as an
interpersonal process: The importance of self-disclosure, partner disclosure,
and perceived partner responsiveness in interpersonal exchanges. Journal of
Personality and Social Psychology, 74(5), 1238-1315. doi: 10.1037/00223514.74.5.1238
107
Lavallee, L. F., & Campbell, J. D. (1995). Impact of personal goals on self-regulation
processes elicited by daily negative events. Journal of personality and social
psychology, 69(2), 341-352. doi:10.1037/0022-3514.69.2.341
Leary, M. R., Allen, A. B., & Terry, M. L. (2011). Managing social images in
naturalistic versus laboratory settings: Implications for understanding and
studying self-presentation. European Journal of Social Psychology, 41(4), 411421. doi:10.1002/ejsp.813
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review
and two-component model. Psychological Bulletin, 107(1), 34-47.
doi:10.1037/0033-2909.107.1.34
Lee, G., Lee, J., & Kwon, S. (2011). Use of social-networking sites and subjective
well-being: A study in south korea. Cyberpsychology, Behavior, and Social
Networking, 14(3), 151-155. doi:10.1089/cyber.2009.0382
Leung, L. (2002). Loneliness, self-disclosure, and ICQ ("I seek you") use.
Cyberpsychology & Behavior, 5(3), 241-251.
doi:10.1089/109493102760147240
Liem, R., & Liem, J. (1978). Social class and mental illness reconsidered: The role of
economic stress and social support. Journal of Health and Social Behavior,
139-156. doi:10.2307/2136530
Lin, N. (1999). Building a network theory of social capital. Connections, 22(1), 28-51.
Ling, R., & Baron, N. S. (2007). Text messaging and im: Linguistic comparison of
american college data. Journal of Language and Social Psychology, 26(3),
291-298. doi:10.1177/0261927X06303480
Liu, Y., Gummadi, K. P., Krishnamurthy, B., & Mislove, A. (2011). Analyzing
Facebook privacy settings: User expectations vs. Reality. Proceedings of the
108
2011 ACM SIGCOMM conference on Internet Measurement Conference, 6170. doi:10.1145/2068816.2068823
Luminet, O., Bouts, P., Delie, F., Manstead, A. S. R., & Rime, B. (2000). Social
sharing of emotion following exposure to a negatively valenced situation.
Cognition and Emotion, 14(5), 661-688. doi:10.1080/02699930050117666
Lyubomirsky, S., & Nolen-Hoeksema, S. (1995). Effects of self-focused rumination
on negative thinking and interpersonal problem solving. Journal of Personality
and Social Psychology, 69(1), 176-176. doi:10.1037/0022-3514.69.1.176
Maisel, N. C., Gable, S. L., & Strachman, A. (2008). Responsive behaviors in good
times and in bad. Personal Relationships, 15(3), 317-338 doi:10.1111/j.14756811.2008.00201.x
Manstead, A., Fischer, A., & Jakobs, E. (1999). The social and emotional functions of
facial displays. In P. Philipport, S. Feldstein & E. Coats (Eds.), The social
context of nonverbal behavior (pp. 287-313). New York: Cambridge
University Press.
Marsden, P. V. (1990). Network data and measurement. Annual Review of Sociology,
16(1), 435-463. doi:doi:10.1146/annurev.so.16.080190.002251
Martin, K. A., Leary, M. R., & Rejeski, W. J. (2000). Self-presentational concerns in
older adults: Implications for health and well-being. Basic and Applied Social
Psychology, 22(3), 169-179. doi:10.1207/15324830051036072
Marwick, A. E. & Boyd, D. (2011). I tweet honestly, I tweet passionately: Twitter
users, context collapse, and the imagined audience. New Media & Society,
13(1), 114-133. doi:10.1177/1461444810365313
Matsushima, R., & Shiomi, K. (2001). The effect of hesitancy toward and the
motivation for self-disclosure on loneliness among japanese junior high school
109
students. Social Behavior and Personality, 29(7), 661-670.
doi:10.2224/sbp.2001.29.7.661
Mauss, I. B., Shallcross, A. J., Troy, A. S., John, O. P., Ferrer, E., Wilhelm, F. H., &
Gross, J. J. (2011). Don't hide your happiness! Positive emotions dissociation,
social connectedness, and psychological functioning. Journal of Personality
and Social Psychology, 100(4), 738-748. doi:10.1037/a0022410
McCance, A. S., Nye, C. D., Wang, L., Jones, K. S., & Chiu, C.-y. (2010). Alleviating
the burden of emotional labor: The role of social sharing. Journal of
Management, 39(2), 392-415. doi:10.1177/0149206310383909
McDonald, R. P., & Ho, M.-H. R. (2002). Principles and practice in reporting
structural equation analyses. Psychological methods, 7(1), 64-82. doi:
10.1037/1082-989X.7.1.64
Mckenna, K., & Bargh, J. A. (2009). Causes and consequences of social interaction on
the internet: A cconceptual framework. Media Psychology, 1(3), 249-269.
doi:10.1207/s1532785xmep0103_4
Mesch, G. S., & Beker, G. (2010). Are norms of disclosure of online and offline
personal information associated with the disclosure of personal information
online? Human Communication Research, 36(4), 570-592. doi:10.1111/j.14682958.2010.01389.x
Molix, L. A., & Nichols, C. P. (2013). Satisfaction of basic psychological needs as a
mediator of the relationship between community esteem and wellbeing.
International Journal of Wellbeing, 3(1), 23-34. doi:10.5502/ijw.v3i1.2
Moreno, M. A., Jelenchick, L. A., Egan, K. G., Cox, E., Young, H., Gannon, K. E., &
Becker, T. (2011). Feeling bad on Facebook: Depression disclosures by college
110
students on a social networking site. Depression and Anxiety, 28(6), 447-455.
doi:10.1002/da.20805
Moscovici, S. (1984). The phenomenon of social representations. In R. M. Farr & S.
Moscovici (Eds.), Social representations (pp. 3-69). Cambridge, UK:
Cambridge University Press.
Naaman, M., Boase, J., & Lai, C.-H. (2010). Is it really about me? Message content in
social awareness streams. Proceedings of the 2010 ACM conference on
Computer supported cooperative work, 189-192.
doi:10.1145/1718918.1718953
Nadkarni, A., & Hofmann, S. G. (2012). Why do people use Facebook? Personality
and Individual Differences, 52(3), 243-249. doi:10.1016/j.paid.2011.11.007
Newman, D., Tay, L., & Diener, E. (2013). Leisure and subjective well-being: A
model of psychological mechanisms as mediating factors. Journal of
Happiness Studies, 1-24. doi:10.1007/s10902-013-9435-x
Newman, M. L., Groom, C. J., Handelman, L. D., & Pennebaker, J. W. (2008). Gender
differences in language use: An analysis of 14,000 text samples. Discourse
Processes, 45: 211-236. doi: 10.1080/01638530802073712
Newman, M. W., Lauterbach, D., Munson, S. A., Resnick, P., & Morris, M. E. (2011).
“It's not that I don't have problems, I'm just not putting them on Facebook":
Challenges and opportunities in using online social networks for health.
Proceedings of the 2011 ACM Conference on Computer Supported
Cooperative Work, 341-350. doi:10.1145/1958824.1958876
Nguyen, M., Bin, Y. S., & Campbell, A. (2012). Comparing online and offline selfdisclosure: A systematic review. CyberPsychology, Behavior & Social
Networking, 15(2), 103-111. doi:10.1089/cyber.2011.0277
111
Omarzu, J. (2000). A disclosure decision model: Determining how and when
individuals will self-disclose. Personality and Social Psychology Review, 4(2),
174-185. doi:10.1207/S15327957PSPR0402_05
Ong, E. Y. L., Ang, R. P., Ho, J. C. M., Lim, J. C. Y., Goh, D. H., Lee, C. S., & Chua,
A. Y. K. (2011). Narcissism, extraversion and adolescents’ self-presentation on
Facebook. Personality and Individual Differences, 50(2), 180-185.
doi:10.1016/j.paid.2010.09.022
Papacharissi, Z. (2011). A networked self: Identity, community, and culture on social
netowork New York, NY: Routledge.
Papacharissi, Z., & Rubin, A. M. (2000). Predictors of internet use. Journal of
Broadcasting & Electronic Media, 44(2), 175-196.
doi:10.1207/s15506878jobem4402_2
Park N.
ee S.
im J. H. (
). Individuals’ personal network characteristics
and patterns of Facebook use: A social network approach. Computers in
Human Behavior, 28(5), 1700-1707. doi:10.1016/j.chb.2012.04.009
Parkinson, B. (1991). Emotional stylists: Strategies of expressive management among
trainee hairdressers. Cognition & Emotion, 5(5-6), 419-434.
doi:10.1080/02699939108411051
Parkinson, B. (2005). Do facial movements express emotions or communicate
motives? Personality and Social Psychology Review, 9(4), 278-311.
doi:10.1207/s15327957pspr0904_1
Pavot, W., & Diener, E. (2009). Happiness experienced: The science of subjective
well-being. In S. David, I. Boniwell & A. C. Ayers (Eds.), Oxford handbook of
happiness (pp. 134-151). Oxford, UK: Oxford University Press.
112
Pennebaker, J. W. (1997). Writing about emotional experiences as a therapeutic
process. Psychological Science, 8(3), 162-166. doi:10.1111/j.14679280.1997.tb00403.x
Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an
understanding of inhibition and disease. Journal of Abnormal Psychology,
95(3), 274-281. doi:10.1037/0021-843X.95.3.274
Pennebaker, J. W., Booth, R., & Francis, M. (2007). Linguistic inquiry and word count
[Computer software]. Mahwah, NJ.: Erlbaum.
Pennebaker, J. W., & Chung, C. K. (2011). Expressive writing: Connections to
physical and mental health. In H. S. Friedman (Ed.), Oxford handbook of
health psychology (pp. 417-437). New York, NY: Oxford University Press.
Pennebaker, J. W., Chung, C. K., Ireland, M., Gonzales, A., & Booth, R. J. (2007).
The development and psychometric properties of LIWC2007. Austin, TX,
LIWC. Net.
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psychological aspects
of natural language. Use: Our words, our selves. Annual Review of Psychology,
54, 547-577. doi:10.1146/annurev.psych.54.101601.145041
Pennebaker, J. W., & Seagal, J. D. (1999). Forming a story: The health benefits of
narrative. Journal of Clinical Psychology, 55(10), 1243-1254.
doi:10.1002/(SICI)1097-4679(199910)55:10<1243::AID-JCLP6>3.0.CO;2-N
Pennebaker, J. W., Zech, E., & Rimé, B. (2001). Disclosing and sharing emotion:
Psychological, social, and health consequences. In M. S. Stroebe, W. Stroebe
& R. O. Hansson (Eds.), Handbook of bereavement research: Consequences,
coping, and care (pp. 517–543). Washington, DC: American Psychological
Association.
113
Peter, J., Valkenburg, P. M., & Schouten, A. P. (2006). Characteristics and motives of
adolescents talking with strangers on the internet. CyberPsychology &
Behavior, 9(5), 526-530. doi:10.1089/cpb.2006.9.526
Peters, P., & Kashima, Y. (2007). From social talk to social action: Shaping the social
triad with emotion sharing. Journal of Personality and Social Psychology,
93(5), 780-797. d oi: 10.1037/0022-3514.93.5.780
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(3), 879-891. doi:10.3758/brm.40.3.879
Pring, C. (2012). 100 social media statistics for 2012. Retrieved from
http://thesocialskinny.com/100-social-media-statistics-for-2012/
Qiu, L., Lin, H., Leung, A. K.-y. (2010). How does Facebook browsing affect selfawareness and social well-being: The role of narcissism. In M. Haller & Y.
Kitamura (Eds.), In Proceedings of the 7th International Conference on
Advances in Computer Entertainment Technology (pp. 100-101). New York,
NY: Association for Computing Machinery.
Qiu, L., Lin, H., Ramsay, J., & Yang, F. (2012). You are what you tweet: Personality
expression and perception on Twitter. Journal of Research in Personality,
46(6),710-718. dx.doi.org/10.1016/j.jrp.2012.08.008
Radcliffe, A. M., Lumley, M. A., Kendall, J., Stevenson, J. K., & Beltran, J. (2010).
Written emotional disclosure: Testing whether social disclosure matters.
Journal of Social and Clinical Psychology, 26(3), 362-384.
doi:10.1521/jscp.2007.26.3.362
Rau, P.-L. P., Gao, Q., & Ding, Y. (2008). Relationship between the level of intimacy
and lurking in online social network services. Computers in Human Behavior,
114
24(6), 2757-2770. doi:10.1016/j.chb.2008.04.001
Reagans, R., & McEvily, B. (2003). Network structure and knowledge transfer: The
effects of cohesion and range. Administrative science quarterly, 48(2), 240267.
Reis, H. T., & Shaver, P. (1988). Intimacy as an interpersonal process. In S. Duck
(Ed.), Handbook of personal relationships (pp. 367-389). Chichester, England:
Wiley.
Renner, K.-H., & Laux, L. (2004). The relationship between self-presentaton styles
and coping with social stress. Anxiety, Stress & Coping: An International
Journal, 17(1), 1-22. doi:10.1080/10615800310001601449
Riley, D., & Eckenrode, J. (1986). Social ties: Subgroup differences in costs and
benefits. Journal of Personality and Social Psychology, 51(4), 770-778.
doi:10.1037/0022-3514.51.4.770
Rimé, B. (2009). Emotion elicits the social sharing of emotion: Theory and empirical
review. Emotion Review, 1(1), 60-85. doi: 10.1177/1754073908097189
Rimé, B., Finkenauer, C., Luminet, O., Zech, E., & Philippot, P. (1998). Social sharing
of emotion: New evidence and new questions. European Review of Social
Psychology, 9(1), 145-190. doi:10.1080/14792779843000072
Rimé, B., Mehdizadeh, S., Philipport, P., & Boca, S. (1991). Beyond the emotional
event: Six studies on the social sharing of emotion. Cognition and Emotion,
5(5-6), 435-465. doi:10.1080/02699939108411052
Rimé, B., Paez, D., Kanyangara, P., & Yzerbyt, V. (2011). The social sharing of
emotions in interpersonal and in collective situations: Common psychosocial
consequences. In I. Nyklíček A. ingerhoets
M. Zeelen erg (Eds.),
Emotion regulation and well-being (pp. 147-163).
115
Rimé, B., Philippot, P., Boca, S., & Mesquita, B. (1992). Long-lasting cognitive and
social consequences of emotion: Social sharing and rumination. European
Review of Social Psychology, 3(1), 225 - 258.
doi:10.1080/14792779243000078
Rioux, S. M., & Penner, L. A. (2001). The causes of organizational citizenship
behavior: A motivational analysis. Journal of Applied Psychology, 86(6),
1306-1314. doi:10.1037/0021-9010.86.6.1306
Ritter, R. S., Preston, J. L., & Hernandez, I. (2014). Happy tweets: Christians are
happier, more socially connected, and less analytical than Atheists on Twitter.
Social Psychological and Personality Science, 5(2), 243-249. doi:
10.1177/1948550613492345
Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011). Mediation
analysis in social psychology: Current practices and new recommendations.
Social and Personality Psychology Compass, 5(6), 359-371. doi:
10.1111/j/1751-9004.2011.00355.x
Russell, D. W. (1996). Ucla loneliness scale (version 3): Reliability, validity, and
factor structure. Journal of Personality Assessment, 66(1), 20-40.
doi:10.1207/s15327752jpa6601_2
Ryan, R. M. (1995). Psychological needs and the facilitation of integrative processes.
Journal of Personality, 63(3), 397-427. doi:10.1111/1467-6494.ep9510042298
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. American
Psychologist, 55(1), 68-78. doi:10.1037/0003-066X.55.1.68
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of
research on hedonic and eudaimonic well-being. Annual Review of Psychology,
116
52(1), 141-166. doi:0066-4308/01/0201-0141
Ryan, T., & Xenos, S. (2011). Who uses Facebook? An investigation into the
relationship between the big five, shyness, narcissism, loneliness, and
Facebook usage. Computers in Human Behavior, 27(5), 1658-1664.
doi:10.1016/j.chb.2011.02.004
Saxena, P., & Mehrotra, S. (2010). Emotional disclosure in day-to-day living and
subjective well being. Psychological Studies, 55(3), 208-218.
doi:10.1007/s12646-010-0034-1
Schwartz, H. A., Eichstaedt, J. C., Kern, M. L., Dziurzynski, L., Ramones, S. M.,
Agrawal, M., . . . Ungar, L. H. (2013). Personality, gender, and age in the
language of social media: The open-vocabulary approach. PLoS ONE, 8(9),
e73791. doi:10.1371/journal.pone.0073791Schlenker, B. R., Dlugolecki, D.
W., & Doherty, K. (1994). The impact of self-presentations on self-appraisals
and behavior: The power of public commitment. Personality & Social
Psychology Bulletin, 20(1), 20-33
Schlenker, B. R., & Weigold, M. F. (1992). Interpersonal processes involving
impression regulation and management. Annual Review of Psychology, 43,
133-168. doi:0066-4308/92/0201-0133
Schug, J., Yuki, M., & Maddux, W. (2010). Relational mobility explains between-and
within-culture differences in self-disclosure to close friends. Psychological
Science, 21(10), 1471-1478. doi: 10.1177/0956797610382786
Seeman, T. E., & Berkman, L. F. (1988). Structural characteristics of social networks
and their relationship with social support in the elderly: Who provides support.
Social Science & Medicine, 26(7), 737-749. doi:10.1016/0277-9536(88)900652
117
Sermat, V., & Smyth, M. (1973). Content analysis of verbal communication in the
development of a relationship: Conditions influencing self-disclosure. Journal
of personality and social psychology, 26(3), 332-346. doi:10.1037/h0034473
Sheese, B. E., Brown, E. L., & Graziano, W. G. (2004). Emotional expression in
cyberspace: Searching for moderators of the pennebaker disclosure effect via
e-mail. Health Psychology, 23(5), 457. doi:10.1037/0278-6133.23.5.457
Sheldon, K. M., Abad, N., & Hinsch, C. (2011). A two-process view of Facebook use
and relatedness need-satisfaction: Disconnection drives use, and connection
rewards it. Journal of Personality and Social Psychology, 100(4), 766-775.
doi:10.1037/a0022407
Sheldon, K. M., Ryan, R. M., Deci, E. L., & Kasser, T. (2004). The independent
effects of goal contents and motives on well-being: It's both what you pursue
and why you pursue it. Personality and Social Psychology Bulletin, 30(4), 475486. doi:10.1177/0146167203261883
Siibak, A. (2009). Constructing the self through the photo selection-visual impression
management on social networking websites. Cyberpsychology: Journal of
Psychosocial Research on Cyberspace, 3(1), article 1. Retrieved from
http://cyberpsychology.eu/view.php?cisloclanku=2009061501
Simpson, P. A. & Stroh, L. K. (2004). Gender differences: Emotional expression and
feelings of personal inauthenticity. Journal of Applied Psychology, 89(4), 715721. doi: 10.1037/0021-9010.89.4.715
Slatcher, R. B., & Pennebaker, J. W. (2006). How do I love Thee? Let me count the
word: The social effects of expressive writing. Psychological Science, 17(8),
660-664. doi: 10.1111/j.1467-9280.2006.01762.x
118
Sleeper, M., Balebako, R., Das, S., McConahy, A. L., Wiese, J., & Cranor, L. F.
(2013). The post that wasn't: Exploring self-censorship on Facebook.
Proceedings of the 2013 Conference on Computer Supported Cooperative
Work, 793-802. doi:10.1145/2441776.2441865
Spiliotopoulos, T., & Oakley, I. (2013). Understanding motivations for Facebook use:
Usage metrics, network structure, and privacy. Proceedings of the 31th
SIGCHI Conference on Human Factors in Computing Systems, 3287-3296.
doi:10.1145/2470654.2466449
Steinfield, C., Ellison, N., & Lampe, C. (2008). Social capital, self-esteem, and use of
online social network sites: A longitudinal analysis. Journal of Applied
Developmental Psychology, 29(6), 434-445. doi:10.1016/j.appdev.2008.07.002
Stokes, J. P. (1983). Predicting satisfaction with social support from social network
structure. American Journal of Community Psychology, 11(2), 141-152.
doi:10.1007/BF00894363
Stokes, J. P. (1985). The relation of social network and individual difference variables
to loneliness. Journal of Personality and Social Psychology, 48(4), 981-990.
doi:10.1037/0022-3514.48.4.981
Stone, L. D., & Pennebaker, J. W. (2002). Trauma in real time: Talking and avoiding
online conversations about the death of princess diana. Basic and Applied
Social Psychology, 24(3), 173-183. doi:10.1207/153248302760179101
Strano, M. M. (2008). User descriptions and interpretations of self-presentation
through Facebook profile images. Cyberpsychology: Journal of Psychosocial
Research on Cyberspace, 2(2), article 5. Retrieved from
http://cyberpsychology.eu/view.php?cisloclanku=2008110402&article=5
Tamir, D. I., & Mitchell, J. P. (2012). Disclosing information about the self is
119
intrinsically rewarding. Proceedings of the National Academy of Sciences,
109(21), 8038-8034. doi:10.1073/pnas.1202129109
Tausczik, Y. R., & Pennebaker, J. W. (2010). The psychological meaning of words:
Liwc and computerized text analysis methods. Journal of Language and Social
Psychology, 29(1), 24-54. doi:10.1177/0261927x09351676
Tay, L., & Diener, E. (2011). Needs and subjective well-being around the world.
Journal of Personality and Social Psychology, 101(2), 354-365.
doi:10.1037/a0023779
Thelwall, M., Wilkinson, D., & Uppal, S. (2010). Data mining emotion in social
network communication: Gender differences in myspace. Journal of the
American Society for Information Science and Technology, 61(1), 190-199.
doi:10.1002/asi.v61:1
Tidwell, L. C., & Walther, J. B. (2002). Computer-mediated communication effects on
disclosure, impressions, and interpersonal evaluations: Getting to know one
another a bit at a time. Human Communication Research, 28(3), 317-348.
doi:10.1111/j.1468-2958.2002.tb00811.x
Tiedens, L. Z. (2001). Anger and advancement versus sadness and subjugation: The
effect of negative emotions expressions on social status conferral. Journal Of
Personality And Social Psychology, 80(1), 86-94. doi:10.1037/00223514.80.1.86
Timmers, M., Fischer, A. H., & Manstead, A. S. R. (1998). Gender differences in
motives for regulating emotions. Personality and Social Psychology Bulletin,
24(9), 974-985. doi:10.1177/0146167298249005
Toma, C. L. (2010). Affirming the self through online profiles: Beneficial effects of
social networking sites. Proceedings of the 28th SIGCHI Conference on
120
Human Factors in Computing Systems, 1749-1752.
doi:10.1145/1753326.1753588
Toma, C. L., & Hancock, J. T. (2013). Self-affirmation underlies Facebook use.
Personality and Social Psychology Bulletin, 39(3), 321-331.
doi:10.1177/0146167212474694
Tong, S. T., Van Der Heide, B., Langwell, L., & Walther, J. B. (2008). Too much of a
good thing? The relationship between number of friends and interpersonal
impressions on Facebook. Journal of Computer-Mediated Communication,
13(3), 531-549. doi:10.1111/j.1083-6101.2008.00409.x
Tov, W., Ng, K. L., Lin, H., & Qiu, L. (2013). Detecting well-being via computerized
content analysis of brief diary entries. Psychological assessment, 25(4), 10691078. doi:10.1037/a0033007
Trierweiler, L. I., Eid, M., & Lischetzke, T. (2002). The structure of emotional
expressivity: Each emotion counts. Journal of personality and social
psychology, 82(6), 1023-1040. doi:10.1037/0022-3514.82.6.1023
Tugade, M. M., & Fredrickson, B. L. (2007). Regulation of positive emotions:
Emotion regulation strategies that promote resilience. Journal of Happiness
Studies, 8(3), 311-333. doi:10.1007/s10902-006-9015-4
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social
network site?: Facebook use and college students' life satisfaction, trust, and
participation. Journal of Computer-Mediated Communication, 14(4), 875-901.
doi:10.1111/j.1083-6101.2009.01474.x
Valkenburg, P. M., & Peter, J. (2007). Online communication and adolescent well‐
being: Testing the stimulation versus the displacement hypothesis. Journal of
Computer-Mediated Communication, 12(4), 1169-1182. doi:10.1111/j.1083-
121
6101.2007.00368.x
Valkenburg, P. M., & Peter, J. (2009). Social consequences of the internet for
adolescents a decade of research. Current Directions in Psychological Science,
18(1), 1-5. doi:10.1111/j.1467-8721.2009.01595.x
Vallerand, R. J. (1997). Toward a hierarchical model of intrinsic and extrinsic
motivation. Advances in Experimental Social Psychology, 29, 271-360.
doi:10.1016/S0065-2601(08)60019-2
Van Emmerik, A. A., Kamphuis, J. H., Hulsbosch, A. M. & Emmelkamp, P. M.
(2002). Single seession debriefing after psychological trauma: A meta-analysis.
The Lancet, 360(9935), 766-771. doi:10.1016/S0140-6736(02)09897-5
Wagner, H. L., & Smith, J. (1991). Facial expression in the presence of friends and
strangers. Journal of Nonverbal Behavior, 15(4), 201-214.
doi:10.1007/BF00986922
Walker, M. E., Wasserman, S., & Wellman, B. (1994). Statistical models for social
support networks. In S. Wasserman & J. Galaskiewicz (Eds.), Advances in
social network analysis (Vol. 171, pp. 53-78). Thousand Oaks, CA: Sage
Publications.
Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A
relational perspective. Communication Research, 19(1), 52-90.
doi:10.1177/009365092019001003
Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal,
and hyperpersonal interaction. Communication Research, 23(1), 3-43.
doi:10.1177/009365096023001001
Walther, J. B. (2007). Selective self-presentation in computer-mediated
communication: Hyperpersonal dimensions of technology, language, and
122
cognition. Computers in Human Behavior, 23(5), 2538-2557.
doi:10.1016/j.chb.2006.05.002
Walther, J. B., Heide, B. V. D., Kim, S.-Y., Westerman, D., & Tong, S. T. (2008). The
role of friends' appearance and behavior on evaluations of individuals on
Facebook: Are we known by the company we keep? Human Communication
Research, 34(1), 28-49. doi:10.1111/j.1468-2958.2007.00312.x
Wang, Y.-C., Burke, M., & Kraut, R. E. (2013). Gender, topic, and audience response:
An analysis of user-generated content on Facebook. Proceedings of the 31th
SIGCHI Conference on Human Factors in Computing Systems, 31-34.
doi:10.1145/2470654.2470659
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and
applications (Vol. 8). Cambridge, UK: Cambridge University Press.
Wellman, B., Carrington, P. J., & Hall, A. (1997). Networks as personal communities.
In B. Wellman & S. D. Berkowitz (Eds.), Social structures: A network
approach (Vol. 15, pp. 130-184). Cambridge, UK: Cambridge University
Press.
Winter, S., Neubaum, G., Eimler, S. C., Gordon, V., Theil, J., Herrmann, J., . . .
Krämer, N. C. (2014). Another brick in the Facebook wall–how personality
traits relate to the content of status updates. Computers in Human Behavior,
34, 194-202. doi:10.1016/j.chb.2014.01.048
Wolf, A. (2000). Emotional expression online: Gender differences in emoticon use.
Cyberpsychology & Behavior, 3(5), 827-833. doi:
10.1089/10949310050191809
Yates, S. J. (1996). Oral and written linguistic aspects of computer conferencing. In S.
C. Herring (Ed.), Computer-mediated communication: Linguistic, social and
123
cross-cultural perspectives (pp. 29-46). Amsterdam and Philadelphia: John
Benjamins.
Zaalberg, R., Manstead, A., & Fischer, A. (2004). Relations between emotions, display
rules, social motives, and facial behaviour. Cognition & Emotion, 18(2), 183207. doi:10.1080/02699930341000040
Zech, E., & Rimé, B. (2005). Is talking about an emotional experience helpful? Effects
on emotional recovery and perceived benefits. Clinical Psychology &
Psychotherapy, 12(4), 270-287. doi:10.1002/cpp.460
Zeman, J., & Garber, J. (1996). Display rules for anger, sadness, and pain: It depends
on who is watching. Child development, 67(3), 957-973. doi:10.2307/1131873
Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook:
Digital empowerment in anchored relationships. Computers in Human
Behavior, 24(5), 1816-1836. doi: 10.1016/j.chb.2008.02.012
Zhao, X., Salehi, N., Naranjit, S., Alwaalan, S., Voida, S., & Cosley, D. (2013). The
many faces of Facebook: Experiencing social media as performance,
exhibition, and personal archive. Proceedings of the 31th SIGCHI Conference
on Human Factors in Computing Systems, 1-10.
doi:10.1145/2470654.2470656
Zou, X. (2009). Social networks and subjective well-being: Regulatory fit between
self-regulation and network structure (Doctoral disseration). Available from
ProQuest Dissertations and Theses database. (UMI No. 3374089)
124
NOTES
1
The reason to make word count comparable is because LIWC takes the percentage of
the word occurrence over the total word count. If the two samples have very different
word count, the chance for the same word occurred in each sample will be quite
different, resulting in biased results. Alternatively, it is also reasonable to control for
the number of events to make it the same as diaries (2 events × 21 days = 42 entries).
As I selected 42 messages, the sample size of Facebook reduced to 210, the average
word count increased to 536, while the results remained: emotion valence still showed
a significant main effect on word frequency, F(1, 419) = 463.27, p < .001, 2 = .53,
with participants disclosing more positive emotions (M = 5.59, SD = 2.53) than
negative emotions in general (M = 2.91, SD = 1.56). The main effect of the disclosure
channel showed that the emotional disclosure significantly differed between Facebook
status updates (M = 4.97, SD = 1.51) and diary (M = 3.53, SD = 1.73), F(1, 419) =
96.64, p < .001, 2 = .19. The interaction effect between emotion valence and
disclosure channel was not statistically significant, F(1, 419) = 1.86, p = .16, 2 =
.005.
2
An independent-samples t-test showed no significant difference between the
Facebook and diary samples regarding total word count, t (439) = .28, p = .78.
3
To first validate the two dimensions of motivational needs on Facebook, questions
were tested in a pilot study with 91 participants (37 males; mean age = 21.98, SD =
1.54). A principal components analysis using Varimax rotation on the 18 items
revealed two factors that together explained 50% of the variance (see Appendix B for
items and loadings). The items for each factor were averaged to create two sub-scale
scores: need for emotional expression (M = 2.56, SD = .71, α = .84) and need for
impression management (M = 3.05, SD = .7
125
α = .86).
4
Concerning network density, past studies suggested that users have a looser network
on Facebook than in real life. In a case study, Hogan (2008b) reported the personal
Facebook network density of .05, which is much sparser than network density ranges
from .13 in real life (Chua et al., 2011). In current research, it is about .09 in both
Study 2a and 2b. It suggests Facebook network in a student sample is relatively
denser, yet still looser than network in real life.
5
It is found that each post is reached to 35% of Facebook friends, and 61% over a one-
month period (Bernstein, Bakshy, Burke, & Karrer, 2013).
6
In Study 1a, only weak correlation between conscientiousness and negative emotion
words was found in the Facebook sample (r = -.19, p = .01) and agreeableness with
positive emotion words in the diary sample (r = .14, p = .03). In Study 2a, among Big
Five traits, only extraversion was found to be positively correlated with size (r = .31, p
= .002) and negatively correlated with density (r = -.23, p = .22); there was no
significant correlation between emotional words and traits. In Study 2b, extraversion
was correlated with disclosure of positive emotions (r = .29, p < .001) and negative
emotions (r = .19, p =.01) in real life. Neuroticism was correlated with negative
emotional disclosure (r = .20. p = .01) in real life. No significant correlations between
traits and emotional disclosure on Facebook were found. Personality was not
measured in Study 3.
7
Data in Study 1 and Study 2 were collected between June 2011 and May 2012, while
the feature of customizing visibility on Facebook was launched in the middle of study.
I have reasons to believe that the majority of the users did not use the new settings to
customize the visibility at that time. A study in 2011 revealed that 36% of contents on
Facebook were open with the default setting, that is public (Liu, Gummadi,
Krishnamurthy, & Mislove, 2011). A recent report based on 900 million Facebook
126
users (Cameron, 2012) showed that only 37% of users have customized the visibility
of information to third parties. A similar pattern is identified in a recent study, too
(Sleeper et al., 2013).
127
TABLES
Table 1.
(Study 2a) Descriptive statistics and bivariate correlation between variables.
1
1. Network size
2
3
4
519.25 (192.64)
2. Network density
–.54***
3. Positive emotions
.29**
.09 (.03)
–.03
.06 (.02)
4. Negative emotion
.02
.22*
.17
.03 (.01)
Note. Means are indicated in the main diagonal with standard deviations indicated in
parentheses. *p < .05, **p < .01, ***p < .001.
128
Table 2.
(Study 2a) Regress network properties on emotional words.
β
Predictor
t
Positive emotion
Network size
.37
3.29**
Network density
.28
2.21*
Network size × network density
.22
1.94
Negative emotion
Network size
.19
1.63
Network density
.32
2.44*
Network size × network density
.01
.04
Note. Network size and density are mean-centered. *p < .05, **p < .01.
129
Table 3.
(Study 2b) Descriptive statistics and bivariate correlation between variables.
130
1.
2.
3.
4.
5.
6.
7.
8.
9.
Network size
Network density
Need for emotional expression
Need for impression management
Positive emotions (Facebook)
Negative emotions (Facebook)
Positive emotions (real life)
Negative emotions (real life)
Facebook Intensity
1
2
499.14 (202.2)
–.36***
.09 (.05)
.01
.16*
.20**
–.03
.15
.00
.02
.01
.16*
–.09
.03
–.17*
.27***
–.18*
3
2.59 (.68)
.08
.28***
.52***
–.18*
–.06
.27**
4
2.79 (.68)
.21**
–.07
–.10
–.16*
.38***
5
5.12 (1.41)
.17*
.09
.01
.25**
6
7
3.4 (1.67)
–.06
5.96 (1.22)
–.01
.38***
.21** –.05
8
9
5.45 (1.56)
–.07
3.54 (.79)
Note. Means are indicated in the main diagonal with standard deviations indicated in parentheses. *p < .05, **p < .01, ***p < .01.
Table 4.
(Study 2b) Indirect effects of network properties on positive emotional disclosure through proposed mediators.
Bias-corrected and
accelerated CIs
Normal theory tests
Mediator
Bootstrap
effect
Normal
effect
SE
Z
p
Lower
Upper
(A) Mediated effect of network size
131
Total effect
.0003
.0003
.0002
1.22
.22
–.0001
.0007
Need for emotional expression
.0000
.0000
.0001
.15
.88
–.0002
.0003
Need for impression management
.0002
.0002
.0001
1.67
.09
.0001
.0005
(B) Mediated effect of network density
Total effect
.76
.95
.76
1.30
.20
.40
1.94
Need for emotional expression
1.00
1.11
.61
1.81
.07
.30
2.09
Need for impression management
Note. CI = confidence interval.
-.25
–.16
.36
–.43
.66
-1.13
.21
Table 5.
(Study 2b) Indirect effects of network properties on negative emotional disclosure through proposed mediators.
Bias-corrected and
accelerated CIs
Normal theory tests
Mediator
Bootstrap
effect
Normal
effect
SE
Z
p
Lower
Upper
(A) Mediated effect of network size
Total effect
132
Need for emotional expression
Need for impression Management
–.0001
–.0002
.0004
–.41
.68
–.0007
.0005
.0001
.0001
.0003
.15
.88
–.0005
.0007
–.0002
–.0002
.0001
–1.47
.14
–.0005
.0000
(B) Mediated effect of network density
Total effect
2.55
2.73
1.30
2.10
.04
.52
4.59
Need for emotional expression
2.36
2.61
1.29
2.02
.04
.23
4.38
.18
.12
.28
.42
.67
–.13
.89
Need for impression Management
Note. CI = confidence interval.
Table 6.
(Study 3) Descriptive statistics and bivariate correlation between variables.
1
2
3
4
5
6
1. Disclosure of positive emotion
2.82 (1.36)
2. Disclosure of negative emotion
.56***
2.13 (1.27)
**
3. Fulfillment of need for emotional expression
.26
.26**
2.51 (.84)
***
4. Fulfillment of need for impression management
.36
.16
.48**
5. Loneliness
–.002
.23**
.08
*
6. Life satisfaction
–.03
–.20
–.10
Note. Means are indicated in the main diagonal with standard deviations indicated in parentheses.
2.67 (.82)
–.25**
2.33 (.42)
.15
–.61**
4.54 (1.09)
*p < .05, **p < .01, ***p < .001
133
Table 7.
(Study 3) Indirect effects of disclosing positive emotions on well-being through proposed mediators.
Bias-corrected and
accelerated CIs
Normal theory tests
Mediator
Bootstrap
effect
Normal
effect
Z
p
Lower
Total effect
(A) Mediated effect on life satisfaction
–.001
–.03
.03
–0.13
.89
–.04
.11
Satisfaction of emotional expression
–.04
–.04
.02
–1.57
.12
–.11
–.007
.03
.04
.02
1.55
.12
.02
.16
–1.69
.09
–.06
.004
Satisfaction of impression management
134
Total effect
Satisfaction of emotional expression
Satisfaction of impression management
Note. CI = confidence interval
SE
(B) Mediated effect on loneliness
–.02
–.02
.01
Upper
.02
.02
.01
2.02
.04
.01
.05
–.04
–.04
.01
–2.98
.003
–.08
–.02
Table 8.
(Study 3) Indirect effects of disclosing negative emotions on well-being through proposed mediators.
Bias-corrected and
accelerated CIs
Normal theory tests
Mediator
Bootstrap
effect
Normal
effect
SE
Upper
p
–.13
.89
–.06
.05
–.11
–.001
135
Total effect
(A) Mediated effect on life satisfaction
–.01
–.003
.03
Satisfaction of emotional expression
–.04
–.04
.02
–1.57
.12
Satisfaction of impression management
–.03
.04
.02
1.55
.12
Total effect
(B) Mediated effect on loneliness
.003
–.003
.01
–.24
.81
Satisfaction of emotional expression
.02
.02
.01
1.77
.08
–.02
–.02
.01
–1.71
.09
Satisfaction of impression management
Note. CI = confidence interval.
Lower
Z
.004
–.03
.001
–.05
.09
.02
.04
–.0009
FIGURES
Figure 1. (Study 1a) Frequency of emotional words in Facebook status updates and
diaries.
136
Figure 2. (Study 1b) Likelihood of disclosing positive emotions vs. negative emotions
of Facebook and in real life.
137
Need for emotional
expression
Size
.12 (.15)
Positive
Emotion
Need for impression
management
(A)
138
Need for emotional
expression
Density
-.03 (.004)
Positive
Emotion
Need for impression
management
(B)
Figure 3. (Study 2b) Coefficients representing effects of network size (Panel A) and network density (Panel B) on mediators and disclosure of
positive emotions. * p < .05, ** p < .01, *** p < .001.
Need for emotional
expression
Size
.04 (.02)
Negative
Emotion
Need for impression
management
(A)
139
Need for emotional
expression
Density
-.09 (.01)
Negative
Emotion
Need for impression
management
(B)
Figure 4. (Study 2b) Coefficients representing effects of network size (Panel A) and network density (Panel B) on mediators and disclosure of
negative emotions. * p < .05, ** p < .01, *** p < .001.
Fulfillment of
need for emotional
expression
Positive
Emotion
.06 (.12)
Life
Satisfaction
Fulfillment of
need for impression
management
(A)
140
Fulfillment of
need for emotional
expression
Positive
Emotion
-.09 (-.12)
Loneliness
Fulfillment of
need for impression
management
(B)
Figure 5. (Study 3) Coefficients representing effects of disclosing positive emotions on Life satisfaction (Panel A) and loneliness (Panel B) and
proposed mediators. Disclosure of negative emotions was controlled. * p < .05, ** p < .01, *** p < .001.
Fulfillment of
need for emotional
expression
Negative
Emotion
-.22* (-.26*)
Life
Satisfaction
Fulfillment of
need for impression
management
(A)
141
Fulfillment of
need for emotional
expression
Negative
Emotion
.29** (.34**)
Loneliness
Fulfillment of
need for impression
management
(B)
Figure 6. (Study 3) Coefficients representing effects of disclosing negative emotions on Life satisfaction (Panel A) and loneliness (Panel B) and
proposed mediators. Disclosure of positive emotions was controlled. * p < .05, ** p < .01, *** p < .001.
APPENDICES
Appendix A
Replaced text emoticons with corresponding emotional words (Study 1a)
Text
Expression Meaning
Replaced with
:(
>:(
>.<
;/
O.o
:(((((((((
:'(
T_T
x_x
D:
-.:|
&_&
:X
Angry
Angry
angry, annoying
skeptical, annoyed
Confused
Crying
Crying
crying
dead or unconscious
horror/disgust
flipping off
straight face/disgusted, grim
Eye roll
embarrassed
evil
evil
frown
Frown
frown
frown
frown
frown
Big Grin
blissful
Happy
happy face
angry
angry
angry
annoy
confuse
cry
cry
cry
dead
disgust
disgust
disgust
dizzy
embarrass
evil
evil
frown
frown
frown
frown
frown
frown
grin
happy
happy
happy
unamused
unamused
kiss
laugh
like
love
playful
sad
sleep
smile
smile
>:)
>;)
:-(
>:C
:'c
C:
:[[
:<
:D
^^
=)
:3
:/
-.-''
(^3^)
^o^
:^D
<3
:p
=(
~,~
:)
:]
Not Amused
sweating, not amused
kiss
singing, laughing
Great! I like it
heart/love
cheeky/playful
sad
sleeping
smile/happy face
smiley
(table continues)
142
(continued)
Text
Expression Meaning
Replaced with
=))
^_^
:Q
*_*
:O
>:O
(:
((((:
smiley
smiley
Smoking while talking
surprise, disbelief
surprised
surprise
Happy
Tired
upset
Angry
annoying
smile
smile
smoke
surprise
surprise
surprise
happy
tired
upset
angry
annoy
-_):
:S
143
Appendix B
Principal components analysis using varimax rotation (Study 2b)
Standardized estimates
Factor 1
Factor 2
Factor 1: need for impression management
I want compliment from others on Facebook.
I want to avoid a reprimand from others on
Facebook.
I want to impress my classmates/colleagues on
Facebook.
I want to look better than my cohort on Facebook.
I fear appearing irresponsible on Facebook.
I want to avoid looking lazy on Facebook.
I want to look like I am busy on Facebook.
I want to avoid looking bad on Facebook.
Rewards from others on Facebook are important
to me.
I want to stay out of trouble on Facebook.
.730
.727
.727
.721
.673
.664
.630
.619
.599
.449
Factor 2: need for emotional expression
I don’t like to let friends on Face ook see how I’m
feeling.
I want to display my emotions to others on
Facebook.
I don’t want friends on Face ook to easily o serve
what I’m feelings.
Even if I am feeling very emotional I don’t want
to let others on Facebook know my feelings.
-.804
.786
-.771
-.723
I want friends on Facebook can read my emotions.
.722
I don’t want to hide the way I’m feeling on
Facebook.
I want to hold my feelings in on Facebook.
.692
Eigenvalue
% of variance
Cronbach's α
Note. Loadings below .40 are not shown
144
-.443
4.82
28.34
3.69
21.70
.85
.82
Appendix C
Scale of need fulfillment on Facebook (Study 3)
Fulfillment of emotional expression
1. Friends on Facebook can understand my feelings.
2. Friends on Facebook know me better.
3. Friends on Facebook provide me emotional support.
4. I feel emotionally relieved.
Fulfillment of impression management
1. I appear popular on Facebook.
2. I get compliment from friends on Facebook.
3. I impress my friends on Facebook.
4. I build an ideal self-image on Facebook.
145