Journal of Media Psychology [608890]

Journal of Media Psychology
Intensity of Facebook Use Is Associated With Lower Self-
Concept
Clarity: Cross-Sectional and Longitudinal
Evidence
Markus Appel, Constanze Schreiner, Silvana Weber, Martina Mara, and Timo Gnambs
Online First Publication, December 30, 2016. http://dx.doi.org/10.1027/1864-1105/a000192
CITATION
Appel, M., Schreiner, C., Weber, S., Mara, M., & Gnambs, T. (2016, December 30). Intensity of
Facebook Use Is Associated With Lower Self-Concept
Clarity: Cross-Sectional and Longitudinal
Evidence.
Journal of Media Psychology
. Advance online publication.
http://dx.doi.org/10.1027/1864-1105/a000192

Original Article
Intensity of Facebook Use Is
Associated With LowerSelf-Concept Clarity
Cross-Sectional and Longitudinal Evidence
Markus Appel,1Constanze Schreiner,1Silvana Weber,1Martina Mara,2and Timo Gnambs3
1University of Koblenz-Landau, Landau, Germany
2Ars Electronica Futurelab, Linz, Austria
3Leibniz Institute for Educational Trajectories, Bamberg, Germany
Abstract: Social networking sites such as Facebook provide individuals with opportunities to express and gather information relevant to their
self-concept. Previous theoretical work yielded contrasting assumptions about a potential link between individuals ’Internet use and their self-
concept clarity , that is, individuals ’perception of a clear and internally consistent self-concept content. Focusing on social networking sites,
our aim was to provide cross-sectional as well as longitudinal evidence regarding the relationship between individuals ’feelings of
connectedness to Facebook ( Facebook intensity ) and self-concept clarity. Two cross-sectional studies ( N1= 244; N2= 166) and one
longitudinal study ( N3= 101) are presented. Independent samples of adolescents, adults, and students from Austria participated. The
statistical procedures included hierarchical regression analyses (Studies 1 and 2) and a cross-lagged panel analysis (Study 3). The studies
provided consistent evidence of a negative relationship between Facebook intensity and self-concept clarity. Moreover, the longitudinal studyshowed that Facebook intensity predicted a decline in self-concept clarity over time whereas a reverse pathway was not supported. Future
research should examine the content of the self-concept and should continue searching for specific Facebook activities that might explain the
decline in self-concept clarity. Our results suggest that an intense attachment to Facebook contributes to an inconsistent and unclear self-concept.
Keywords: Facebook, self-concept clarity, cross-lagged panel analysis, unity hypothesis, fragmentation hypothesis
As of December 2015 ,o v e r 1.5billion people actively use
Facebook at least once a month, with over 1billion daily
active users on average (Facebook, 2016 ). Within a 1-min
time-span on Facebook, users request a total of 100,000
new friends, give 3,125,000 likes, send 150,000 messages,
a n du p l o a da b o u t 240,000 photos (Ahmad, 2014 ).
The popularity of social networking sites (SNSs) has fueledquestions among social scientists and the general public
regarding the antecedents, correlates, and consequences
of using these platforms. Of particular interest have beenquestions about the relationship between using SNSs and
the user ’sself,t h a ti s ,t h ek n o w l e d g e ,a t t i t u d e s ,a n d
evaluations that an individual has about him- or herself(the “me”in William James ’terms, cf. Swann & Bosson,
2010 ). Users of SNSs have ample opportunities for commu-
nicating information about themselves, getting relatedfeedback from communication partners, gathering infor-
mation about others, and providing feedback themselves
(Appel, Mara, & Weber, 2014 ). These activities appear tobe a potentially relevant source for the content of the self
(Who am I? How do I feel about myself?). Moreover, these
activities might affect the structure of the self as well
(How sure am I about my own characteristics? Are mycharacteristics consistent?). Focusing on structural aspects
of the self, the aim of the current work was to shed light
on the relationship between individuals ’feelings of
connectedness to Facebook ( Facebook intensity )a n dt h e i r
perception of a clear and internally consistent self-concept
(self-concept clarity ). After an integration of the (diverging)
theory and findings connected to this relationship, the
results of three studies with three independent samples
are reported. Our empirical approach extends previousstudies, as we focused on SNSs and used a well-established
indicator of connectedness to Facebook, the Facebook
Intensity Scale (Ellison, Steinfield, & Lampe, 2007 ).
We additionally examined particular Facebook activities
and usage patterns. Importantly, we provide longitudinal
evidence to disentangle causal influences underlying the
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing
DOI: 10.1027/1864-1105/a000192

association between the intensity of Facebook use and the
clarity of the users ’self-concept.
Self-Concept Clarity
Individuals strive for a firm sense of who they are (Fiske,
2010 ). At the same time, individuals differ with respect to
how clearly and confidently the contents of their
self-concept are defined, and to what extent the self-concept is internally consistent and temporally stable.
Individual differences in this regard have been investigated
under the term of self-concept clarity (Campbell et al., 1996 ).
Self-concept clarity (SCC) is a structural feature of the self-
concept and it is conceptually distinct from the particular
attributes people ascribe to themselves or how they feelabout themselves. SCC is conceived as a self-opinion
(Conley, 1984 ), an individual difference that is useful to
measure as a state or as a trait, because it is susceptibleto environmental influences but it also shows substantial
stability over time (Campbell et al., 1996 ). The standard
measure for assessing SCC is the Self-Concept ClarityScale, a self-report instrument with established psychomet-
ric properties (Campbell et al., 1996 ;S t u c k e , 2002 ).
In recent years, studies associated higher SCC with higherpsychological adjustment (Campbell, Assanand, &
Di Paula, 2003 ) and well-being (Church et al., 2014 ), better
adaptation to stress (Ritchie, Sedikides, Wildschut, Arndt, &Gidron, 2011), better body image (Vartanian & Dey, 2013 ),
and higher explicit self-esteem (Brandt & Vonk, 2006 ).
Communication is considered a key to achieving a clear
sense of oneself. Individuals learn about themselves by
observing their own (communicative) behavior ( self-
perception theory ;B e m , 1972). Moreover, others ’reactions
to one ’s behavior are an important source for developing a
firm self-concept ( looking glass self ; Cooley, 1902 ;perceived
appraisals ; Kenny & DePaulo, 1993 ). In their attempt at
establishing a firm self-concept, individuals can profit
from other people ’s reactions on their appearance and
behavior. A longitudinal study conducted with Dutch
adolescents (Frijns & Finkenauer, 2009 ), for example,
showed that adolescents who openly communicated withtheir parents (i.e., kept fewer secrets) indicated higher SCC
after a 6-month delay (SCC did not predict communication
with parents at a later point of time; see also van Dijket al., 2014 ). However, communication may also be related
to an unclear sense of oneself. The more students engaged
in intimate discussions with their peers and talked abouttheir evaluations of others (i.e., tended to gossip), the lower
their SCC (Watson, 2011). This is in line with research
showing that social comparison processes –both upward
and downward social comparisons –are associated with an
unclear sense of self (cf. Butzer & Kuiper, 2006 ;V a r t a n i a n
&D e y , 2013 ).Self-Concept Clarity in the Digital Age:
Competing Predictions
Today much of the communication of adolescents and
adults is conducted over the Internet, which provides ampleaccess to means of self-presentation and self-disclosure
and to getting self-relevant feedback from others.
Moreover, Internet applications provide individuals withthe opportunity to communicate more or less anonymously,
and to select which aspects of the self they wish to reveal.
Thus, it is an intriguing question how online communica-tion relates to users ’SCC. There are basically two
competing hypotheses regarding the influence of online
activities on the clarity of the user ’s self-concept –the
fragmentation hypothesis and the self-concept unity hypothesis
(cf. Valkenburg & Peter, 2011).
The fragmentation hypothesis dates back to the early
days of the Internet when chatrooms, bulletin boards, and
multi-user dungeons (MUDs) were among the most popular
applications. In these applications users were representedonly by nicknames, which facilitated taking on different
identities. MUDs typically required users to adopt a certain
non-self-identity as part of a role-playing game. Socialscientists observed that individuals indeed used the
Internet to experiment with different identities that were
only loosely tied to their identity in the offline world (e.g.,Reid, 1998 ;T u r k l e , 1995 ). The extent to which users
engaged in experimenting with different identities is a
m a t t e ro fs o m ed e b a t e( c f .S u b r a h m a n y a m& Šmahel,
2011), but it seems safe to say that even in the earlier days
of the Internet, pretending to be someone completely else
(e.g., pretending to have the opposite sex) was not verycommon (Gross, 2004 ;S u b r a h m a n y a m& Šmahel, 2011).
In contrast to the early applications, SNSs such as Face-
book require individuals to build an online representationof the user ’s true offline identity; pretending to be someone
completely else is considered a norm violation. Still, users
can and do present different facets of themselves thatmay more or less represent their true self. Michikyan and
colleagues (Michikyan, Dennis, & Subrahmanyam, 2015 ;
Michikyan, Subrahmanyam, & Dennis, 2014 )s h o w e dt h a t
users most often wish to express their real self on Facebook
(e.g., “how I am in real life ”), but other facets are commu-
nicated as well, such as the ideal self ( “to show aspects of
who I want to be ”). The authors further identified three
forms of false self, that are expressed on Facebook, thatis, for the sake of exploration ( “try out many aspects of
who I am much more than I can in real life ”), to deceive
(“try to be someone other ”), and to impress and compare
with others ( “I try to impress others with the photos I post
of myself ”). In many instances a Facebook activity arguably
reflects a mixture of these different forms of representingoneself on Facebook. Likewise, Facebook users in a study2 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

by Toma and Carlson ( 2015 ) perceived their profiles to be
more positive than their actual selves on some dimensions
(e.g., “outgoing, ”“adventurous, ”but also “relaxed ”).
On other dimensions they perceived their Facebook
profiles to be accurate (e.g., “creative, ”“friendly, ”“physi-
cally attractive ”) and they felt their profiles came across
more negative than their actual selves on even other
dimensions ( “reliable, ”“intelligent, ”or“deep ”). Negative
self-images on Facebook were attributed to postings by
friends, which are difficult to control, but add to a Facebook
profile.
According to the fragmentation hypothesis, the salience
of many possible selves and the heterogeneity of self-
expressions –and others ’feedback in response to these
different facets –impair the development of a consistent
and temporally stable self-concept (cf. Reid, 1998 ).
Consequently, the fragmentation hypothesis predictsthat more intense use of Facebook should predict lower
SCC.
The self-concept unity hypothesis emphasizes the overlap
between offline and online selves. With respect to
Facebook, a highly cited study indicates that Facebook
profiles provide valid information on the users ’personality,
as observers could gauge users ’real personality –rather
than their ideal personality –just by knowing their
Facebook page (Back et al., 2010 ). In other contexts,
however, the overlap between online and offline self might
be smaller (Gosling & Mason, 2015 ), because it is unlikely
that individuals ever meet in person, such as in onlinegaming portals (Graham & Gosling, 2013 ), or the incentive
to express one ’s ideal self is particularly large, such as on
dating websites (Ellison, Hancock, & Toma, 2012 ). In line
with the self-concept unity hypothesis, recent research
reconstructed Facebook activities as means for self-
affirmation (Toma, 2013 ; Toma & Hancock, 2013 ).
Self-affirmation theory posits that individuals have a need
for self-integrity and self-worth and that in everyone ’sl i f e
many incidents challenge this positive view of oneself(Steele, 1988
). Given these challenges and threats, individ-
uals construct the world in a way to preserve self-integrity.
Applied to Facebook activities, this theoretical frame-work suggests that users are motivated to present them-
selves in a positive, yet honest manner (Toma, 2013 ;
Toma & Hancock, 2013 ). Thus, flattering postings about
oneself are conceived as something that is an integral part
of the “true self, ”rather than a dislocated “ideal self ”or
“false self. ”From a self-affirmation perspective, engaging
in Facebook activities contributes to users ’feelings of
self-integrity (Toma & Hancock, 2013 ). According to the
self-concept unity hypothesis, Facebook users tend toc o m m u n i c a t ea s p e c t so ft h et r u es e l ft oal a r g en u m b e ro f
other individuals. These interaction partners in turn
provide information to validate one ’s self-concept, whichleads to a firm sense of oneself (cf. Calvert, 2002 ). Thus,
more intense use of Facebook should predict higher SCC.
Initial Empirical Evidence and Open
Questions
In recent years, a substantial number of studies focused
on the link between Internet and Facebook use and
self-content measures, most notably between Facebookuse and self-esteem and well-being. The findings have been
mixed, with some indicating positive and some indicating
negative relationships (e.g., Kim & Lee, 2011; Kross et al.,
2013 ; see Huang, 2010 , for a meta-analysis of early stud-
ies). Recent research suggests that for users who do not
actively engage in producing content but rather prefer toread others ’postings and comments, Facebook use is
related to lower self-esteem whereas more positive
associations were observed for more active usage patterns(Chou & Edge, 2012 ; große Deters & Mehl, 2013 ; Krasnova,
Wenninger, Widjaja, & Buxmann, 2013 ; Verduyn et al.,
2015 ). Active versus passive usage patterns might also
contribute to more rather than less SCC, because acts of
self-presentation (e.g., posting comments and photos,
commenting on things) appear to be a key for derivingself-relevant feedback that can facilitate the validation of
the self-concept. On a more cautionary note, however,
activities like changing one ’s profile picture frequently or
even actively pretending to be someone else might indicate
and perpetuate low rather than high SCC.
Empirical evidence on the relationship between Internet
use and self-concept clarity is limited. In a cross-sectional
study among Canadian undergraduates, Matsuba ( 2006 )
reported negative relationships of SCC with the time spent
online and pathological Internet use. Moreover, SCC was
associated with the motives to use the Internet for commu-nication and for entertainment. Valkenburg and Peter
(2008 ) focused on SCC as a potential consequence of
identity experiments online. A cross-sectional survey onDutch adolescents found a significant negative correlation
between SCC and engaging in such experiments, whereas
SCC was positively associated with the variety of communi-cation partners. Both relationships were small and the latter
vanished when observed as part of a larger structural
equation model. Third, a cross-sectional study was con-ducted in Israel with adolescents in seventh to ninth grade
(Israelashvili, Kim, & Bukobza, 2012 ). SCC was unrelated to
the hours surfing the Internet, but negatively related toInternet usage as indicated by the extent they engaged in
a variety of Internet-related activities such as using chats,
games, discussion groups, or exploring new websites.Moreover, SCC exhibited a negative association with the
level of Internet addiction. Finally, a cross-sectional surveyM. Appel et al., Facebook and Self-Concept Clarity 3
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing

conducted with adolescents in Barbados (Davis, 2013 )
showed that SCC was negatively linked to a self-conceived
measure of online identity expression and identity explo-ration, but positively linked to friendship quality and
mother relationship quality.
Overall, these findings have a common tenor in pointing
at a negative relationship between Internet use and SCC.
This pattern is in line with the fragmentation hypothesisand in contrast to the unity hypothesis. However, the
findings are limited in key regards: First, no prior study
focused on SNSs. At the time two of the studies wereconducted (Matsuba, 2006 ;V a l k e n b u r g&P e t e r , 2008 ),
engaging in SNSs was a much less common activity than
it is today and the measures of both other studies appliedto Internet activities as a whole. As outlined earlier, in
contrast to applications such as chatrooms, video gaming
portals, or dating websites, SNSs have been described asfacilitating a greater overlap between offline and online
selves. Thus, the association between intensive use of
Facebook and SCC could differ from earlier research thatdid not focus on SNSs. Second, the cross-sectional data of
these studies allow for the possibility of alternative explana-
tions of the results found. Possibly, the associations are dueto the tendency of individuals with low SCC to search for
self-relevant information by means of online communica-
tion and Facebook use. This is basically the reverse causal-ity of the causal path expected from the fragmentation
hypothesis. Third, the findings are based on single-item
measures or ad hoc scales of Internet use. The use ofvalidated measures would strengthen the evidence.
Overview and Predictions
The general aim of the current set of studies was to exam-ine and disentangle the relationship between the intensityof Facebook use and SCC. We used a well-established
measure of Facebook intensity, the Facebook Intensity
Scale (Ellison et al., 2007 ), which was developed, “to tap
the extent to which individuals are emotionally connected
to Facebook, and the extent to which Facebook is
integrated into individuals ’daily lives ”(p.1150). Since its
development it has been used in a large number of studies
on the antecedents, corollaries, and consequences of
Facebook use (e.g., Clayton, Osborne, Miller, & Oberle,2013 ; Pabian, De Backer, & Vandebosch, 2015 ; Valenzuela,
Park, & Kee, 2009 ). This approach adds to the prior studies
on Internet use and SCC, which frequently involvedmeasures with unknown reliability and validity.
On the basis of this Facebook intensity measure, our first
aim was to examine whether the negative relationshipbetween Internet use and SCC translates to the use of SNSs,
and Facebook in particular. To that end, data from twoindependent samples were collected (Studies 1and 2).
In Study 2we further examined particular activities and
usage patterns on Facebook, including identity shifts(pretending to be someone else). Information on the
prevalence of these activity patterns contributes to the liter-
ature on identity exploration (cf. Subrahmanyam & Šmahel,
2011). We further examined relationships between these
activities and SCC. Our second aim was to provide initialevidence on the causal patterns underlying the relationship
between Facebook intensity and SCC, based on a short-
term longitudinal design (Study 3).
Study 1
Method
Sample and Procedure
A convenience sample of 238volunteers was recruited in a
mid-sized Austrian city by research assistants. All ethicalrequirements for conducting empirical survey research
were met. Among the volunteers, 13had no Facebook
account and were therefore excluded from further analyses;one participant reported that he or she had not answered
the questionnaire seriously. The remaining sample con-
sisted of 224participants ( 62.9% women). About two thirds
of the sample accessed the questionnaire over the Internet,
one third worked on paper-and-pencil questionnaires.
The sample consisted predominantly of adolescents(n=66were between 14and18years old) and young adults
(137were between 19and26years old, 34participants were
27years or older). The participants ’age ranged from 14
to48years ( M=21.27years; SD=5.80).
Measures
The measures reported here were administered as part of a
larger survey. The means, standard deviations, and zero-
order correlations of the variables are displayed in Table 1.
Self-Concept Clarity
This construct was measured with the 12-item Self-Concept
Clarity Scale (Campbell et al., 1996 ;s a m p l ei t e m “Is p e n da
lot of time wondering what kind of person I really am ”;
reverse-coded, German adaptation, Stucke, 2002 ).
Five-point response scales were provided ( 1=not true at
allto5=completely true ). Higher mean scores indicated
higher clarity. The reliability of this scale was good, asindicated by a Cronbach ’sαof .81.
Facebook Intensity
We assessed this construct with the help of a German
language version of the Facebook Intensity Scale4 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

(Ellison et al., 2007 ).1The six items (sample item: “Face-
book is part of my everyday activity ”)w e n tw i t h 5-point
scales ranging from 1(=strongly disagree )t o5(=strongly
agree ) and showed a good Cronbach ’sαreliability of . 83.
Higher mean scores indicated a more intense relation to
Facebook.
Facebook Access Frequency
As a second indicator of Facebook usage, we assessedhow often participants went online to check Facebook.
A12-point scale was provided that ranged from 0(=less
than once a week) to11(=12times per day or more often ).
Results and Discussion
Based on the prior research that addressed Internet use
more broadly, a negative relationship between Facebook
intensity and SCC was expected. Our hypothesis wasexamined with the help of a hierarchical regression analysis
with SCC as the criterion. Age and gender were entered
first in the equation, followed by Facebook intensity enteredsecond. The demographic variables yielded a significant
effect, F(2,221)=13.05,p<.001,R
2=.11, which can be
attributed to a significant influence of age, B=0.03,SEB
=0.01,β=.27,p<.001, whereas gender was unrelated
to SCC, B=/C00.02,SEB=0.08,β=/C0.01,p=.84. Facebook
intensity turned out to be a significant predictor of SCC,B=/C00.16,SE
B=0.05,β=/C0.23,p<.001,ΔR2=.05.T h u s ,
controlled for age and gender, more intensive use of
Facebook predicted less SCC. We further inspectedhigher-order interactions to examine whether this relation-
ship varied with respect to participants ’gender or age.
The three two-way and the three-way interactionsyielded no significant result (all tvalues < | 1.43|, all
pvalues > . 15).
Extending prior research on Internet use and SCC, we
focused on the most popular SNS, Facebook, using a
reliable and valid set of items. Our findings are in line with
the fragmentation hypothesis and consistent with earlier
studies that identified a negative association betweenSCC and measures of compulsive Internet use, Internet
addiction, engaging in identity experiments online,
and popular Internet activities (Davis, 2013 ; Israelashvili
et al., 2012 ;M a t s u b a , 2006 ;V a l k e n b u r g&P e t e r , 2008 ).
Our rather diverse group of adolescents and adults allowed
us to examine whether participants ’age influenced
the results; however, neither age nor gender moderated
the core relationship.
Study 2
Questions regarding the relationship between identity and
Internet use are particularly pertinent for the group of
adolescents (cf. Arnett, 1995 ). In this age group the
development of a coherent self-concept is an important
task, and adolescents belong to the Internet ’s most avid
users (Subrahmanyam & Šmahel, 2011). The aim of our
second study was twofold. We aimed at replicating the
findings from Study 1with a different sample, in a different
setting (see, for example, Benoit & Holbert, 2008 ,o nt h e
importance of replication). This time, our particular focus
was on adolescents, as establishing a firm sense of one ’s
self is especially important for this age group (Valkenburg
& Peter, 2011). Our second aim was to extend the set of
variables examined in the first study, and we further askedfor particular Facebook activities. We were interested in the
prevalence of these activities and their relationships to SCC.
Method
Sample and Procedure
We recruited 206 students at secondary schools in a mid-
sized Austrian city, who answered the questionnaire in
class. All ethical requirements for conducting empiricalsurvey research at schools were met. Among the students,
184returned the completed questionnaire and indicated
that they had answered the questions sincerely. Eighteenstudents had no Facebook account. The remaining sample
1Originally, the Facebook Intensity Scale included two additional items with an open-ended response format (Ellison et al., 2007). Only the six
rating-scale items were used in the current studies.Table 1 .Study 1: Means (standard deviations) and zero-order correlations
M(SD)1 2 3 4 5
1 Gender 0.63 (0.48) –
2 Age 21.27 (5.80) /C0.01 –
3 Self-concept clarity 3.84 (0.62) /C0.02 .32*** –
4 Facebook intensity 2.83 (0.87) .04 /C0.23** /C0.29*** –
5 Facebook access frequency 6.51 (2.70) .04 /C0.31*** /C0.14* .64*** –
Notes. Gender was dummy-coded (0 = male, 1 = female). * p< .05. ** p< .01. *** p< .001.M. Appel et al., Facebook and Self-Concept Clarity 5
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing

consisted of 166participants ( 108female) aged 14–20years
(M=16.39years; SD=1.36).
Measures
SCC and Facebook intensity were measured with the same
scales as in Study 1and showed good reliabilities, as indi-
cated by a Cronbach ’sαof .79(self-concept clarity) and
.83(Facebook intensity). We further assessed Facebook
access frequency as in Study 1.A na d d i t i o n a l 17items were
included, which asked about the typical activities they
engage in on Facebook (e.g., changing the profile picture,
uploading pictures/videos, reading comments others wrotein response to one ’s postings, to act as if one was somebody
else; see Table 2). The items went with 5-point scales
ranging from 1(n o ta ta l lt r u e )t o5(completely true ). Among
the Facebook activity items, “playing with different
identities ”and “playing games ”were extremely uncom-
mon among our participants; 147(88.6%) had the lowest
possible score on playing with different identities (for
similar results see, e.g., Valkenburg & Peter, 2008 ), and
130 (78.3%) had the lowest possible score on playing
games. To avoid spurious results, both variables were
excluded from further analyses. Table 2provides the
complete list of activities/usage patterns along with means,standard deviations, and zero-order-correlations.
Results and Discussion
To examine our main hypothesis, we ran a hierarchicalregression analysis. SCC served as criterion and
demographic variables (age and gender) were entered first;
Facebook intensity was entered in a subsequent step. Thedemographic variables, taken together, had no significant
effect on SCC, F(2,163)=2.17,p=.12,R
2=.03,w i t ha g e
being a significant single predictor variable, B=0.07,
SEB=0.04,β=.16,p=.047. Importantly, Facebook
intensity was a significant predictor of SCC, B=/C00.12,
SEB=0.05,β=/C0.19,p=.02,ΔR2=.03.M o r ei n t e n s e
use of Facebook predicted less SCC. In order to test
whether or not this relationship was moderated by partici-
pants ’gender or age, higher-order interactions were exam-
ined. Neither one of the three two-way nor the three-way
interaction reached significance (all tvalues < | 1.14|, all
pvalues > . 26).
To examine the role of the Facebook activities on SCC,
over and above the influence of Facebook intensity, the
15Facebook activities were entered in the regression
equation as an alternative fourth step. The variables
together made a marginally significant contribution in
explaining SCC, F(15,144)=1.60,p=.08,ΔR2=.13.A m o n g
the15activities only two contributed significantly ( α=.05)
to the model. Those were “Look at others ’reactions to my
postings (e.g., status updates, links) ”,B=/C00.10,SEB=0.05,β=/C0.21,p=.03,a n d “Just browse and like,
nothing else ”,B=/C00.10,SEB=0.05,β=/C0.22,p=.03.
Of note, even if all 15activities were entered into the
equation, the predictive power of the Facebook Intensity
Scale approached significance, B=/C00.11,SEB=0.06,
β=/C0.16,p=.08.
Focusing on a sample of adolescents, we again found a
negative association between Facebook intensity andSCC, for both genders, and irrespective of participants ’
age. In line with expectations (see also Verduyn et al.,
2015 ), two rather passive specific modes of using
Facebook were negatively related to SCC. The results of
our cross-sectional studies are in contrast to the self-
concept unity hypothesis, which assumes that intense useof SNSs and related applications allows users to develop a
particularly clear sense of their selves (cf. Valkenburg &
Peter, 2011). However, owing to the cross-sectional charac-
ter of these studies we are hesitant to interpret the data as a
support for the fragmentation hypothesis, which posits that
the intense use of SNSs hinders rather than assists people ’s
strivings for a clear self-concept. Individuals with low SCC
might be more strongly attracted to Facebook and other
SNSs (as compared to individuals with higher SCC),because they find the opportunities for self-presentation
and receiving feedback from others on Facebook to be
particularly attractive. In order to disentangle these compet-ing causal pathways, a longitudinal study was conducted.
Study 3
Although both previous studies concordantly demonstratethat Facebook intensity predicted SCC, causal interpreta-tions of these results are inappropriate, because both
studies adopted cross-sectional designs. Therefore, this
study was based on a short-term longitudinal researchdesign that assessed both constructs at two measurement
occasions. Using cross-lagged panel analyses (McArdle &
Nesselroade, 2014 ), this design provides information on
causality, that is, whether Facebook intensity influences
SCC or, rather, SCC influences Facebook intensity.
Method
Sample and ProcedureThe participants were students recruited from an intro-
ductory course in consumer behavior at an Austrianuniversity. They answered the questionnaire in class and
were invited to answer the same questions 3months later.
The second questionnaire was administered online.Participants received extra credit for participation. Of the
122 students who participated at both points of time,6 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

Table 2 .Study 2: Means (standard deviations) and zero-order correlations
M(SD)12 3 4 5 6 7 8 91 01 11 21 31 41 51 6 1 7 1 8 1 9
1. Gender 0.65 (0.48) –
2. Age 16.39 (1.36) .04 –
3. Self-concept clarity 3.60 (0.63) /C0.05 .15* –
4. Facebook intensity 2.94 (0.97) .27*** .01 /C0.19* – .
5. Facebook access
frequency7.96 (2.91) .33*** /C0.02/C0.20* .51*** –
6. Change profile picture 2.17 (1.03) .26** /C0.08/C0.14 .24* .21** –
7. Upload pictures/
videos2.47 (1.17) .26** /C0.03/C0.04 .31*** .31*** .60*** –
8. Share pictures/videos 1.93 (1.81) /C0.02 /C0.04/C0.12 .22** .02 .19* .25** –
9. Share links 1.81 (1.07) /C0.04 .05 /C0.09 .14 .01 .15* .23** .61*** –
10. Look at others ’
reactions to my postings(e.g., status updates,
links)2.60 (1.28) .07 .07 /C0.20** .25** .18* .20* .32*** .08 .20** –
11. Look at others ’
reactions to photos Ihave uploaded (e.g.,
comments, likes)3.28 (1.23) .23** /C0.01/C0.08 .34*** .30*** .37*** .40*** .03 .07 .55*** –
12. Chatting/writing
personal messages4.04 (1.11) .25** /C0.14/C0.12 .33*** .33*** .25** .26** .09 .10 .17* .32*** –
13. Liking things others
have uploaded3.90 (1.10) .33*** /C0.14/C0.03 .34*** .35*** .20* .24** .15 .14 .19* .37*** .48*** –
14. Commenting on
things others have
uploaded2.89 (1.21) .20** .04 .00 .49*** .25** .28*** .30*** .16* .10 .15 .29*** .41*** .53*** –
15. Coordinating offline
activities/signalingparticipation2.30 (1.24) .03 /C0.01/C0
.04 .16* .11 .11 .20* .13 .16* .18* .11 .14 .16* .14 –
16. Being remembered
about birthdays3.14 (1.47) .30*** .11 .08 .27** .21** .10 .10 /C0.02 .07 .18* .13 .16* .29*** .25** .34*** –
17. Poking 2.01 (1.23) .14 /C0.08/C0.10 .28*** .07 .22** .20** /C0.01 .04 .21** .05 .22** .22** .34*** .19* .30*** –
18. Just browse and like,
nothing else3.27 (1.38) .06 /C0.12/C0.18* .12 .09 /C0.13 /C0.17* .13 .10 .13 .07 /C0.08 .32*** .03 .01 .10 /C0.04 –
19. Just browse 2.61 (1.43) /C0.15 .02 /C0.03 .00 /C0.02 /C0.18* /C0.18* .11 /C0.02 .04 /C0.04 /C0.17* /C0.06 /C0.10 /C0.19* /C0.08 /C0.07 .43*** –
20. Try to reveal
minimum information3.26 (1.24) /C0.10 .10 .09 /C0.23** /C0.14 /C0.29*** /C0.40*** /C0.18* /C0.15/C0.13 /C0.22** /C0.21** /C0.20* /C0.30*** /C0.04 .06 /C0.11 .17* .37***
Notes. Gender was dummy-coded (0 = male, 1 = female). The activities (rows/columns 6 –20) were introduced as follows: “When I am on Facebook, the following activities are typical for me… ”.*p< .05; ** p< .01;
***p< .001.M. Appel et al., Facebook and Self-Concept Clarity 7
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing

14had no Facebook account and five had missing values on
the relevant items. They were not included in the analyses.
Two additional participants indicated that they had notseriously answered the questions. The remaining sample
consisted of 101persons ( 62women) with an age range
of19–37years (at T1:M=22.37;SD=3.34).
Measures
Facebook intensity and SCC were assessed with thesame scales as in both previous studies. Reliabilities
(Cronbach ’sα) were satisfactory for both scales at both
points of time (Facebook intensity: α
T1=.81,αT2=.84;
self-concept clarity: αT1=.83,αT2=.89).
Statistical Analyses
The associations between Facebook intensity and SCC
across the two measurement occasions were examined
using cross-lagged panel analyses (cf. McArdle &Nesselroade, 2014 )i nM p l u s 7(Muthén & Muthén,
1998 –2012 ) with a robust maximum likelihood estimator.
In line with the previous studies, all models acknowledgedgender and age as control variables. All analyses modeled
the two constructs as latent factors. To create more
parsimonious measurement models, we did not analyzeindividual item scores, but created three item parcels
following the item-to-construct balance technique (Little,
Cunningham, Shahar, & Widaman, 2002 ). Because
meaningful interpretations of longitudinal models require
invariant measurement models (Little, 2013 ;L i t t l e ,
Preacher, Selig, & Card, 2007 ), longitudinal factorial
invariance was investigated for each construct by
comparing a model with factor loadings for a given parcel
constrained to be equal over time to a model withoutequality constraints. Following prevalent recommenda-
tions (Little et al., 2007 ), these models also included
autocorrelations among the residuals of a given parcel,which accounts for the systematic variance associated with
each parcel.
The goodness of fit of these models was evaluated
using the comparative fit index (CFI) and the root mean
square error of approximation (RMSEA). In line with
conventional standards (e.g., Hu & Bentler, 1999 ;
Schermelleh-Engel, Moosbrugger, & Müller, 2003 ), models
with a CFI > . 90and a RMSEA < . 10are interpreted as
“acceptable, ”and CFI /C21.95and RMSEA /C20.05as“good ”
fitting.
Results and Discussion
Longitudinal measurement invariance was examined in twosteps. First, we fitted an uncons trained longitudinal latent
factor model for each construct to the data that included
one latent factor at each measurement occasion.The respective models for Facebook intensity, w
2(5)=4,
CFI = 1.00,R M S E A=. 00[.00,.10], and SCC, w2(5)=2,
CFI = 1.00,R M S E A=. 00[.00,.09], showed good fits to
the data. In the next step, the factor loadings were
constrained across time. The respective models did not
fit worse than the unconstrained models, Δw2(2)=0.08,
p=.96for Facebook intensity, and Δw2(2)=5.39,p=.07
for SCC. This confirms the assumption of invariantmeasurement structures; the meaning of both constructs
did not change over time.
In light of the invariant measurement models, we fitted
the cross-lagged models presented in Figure 1to the data.
The model showed a good fit to the data, w
2(52)=62,
CFI = 1.00,R M S E A=. 00 [.00,.04]. Overall, the
demographic variables showed only rather marginal effects
on Facebook intensity and SCC: Age predicted Facebook
intensity at the first measurement occasion, B=/C00.07,
SEB=0.02,β=/C0.29,p<.001, and SCC at the second
measurement occasion, B=/C00.03,SEB=0.01,β=/C0.15,
p=.03, whereas gender predicted SCC at the first measure-
ment occasion, B=0.12,SEB=0.06,β=.22,p=.04.
All other paths of gender or age on SCC and Facebook
intensity were not significant, all pvalues > . 24.W i t hr e g a r d
to the focal constructs, both showed considerable stability
across time, B=1.02,SEB=0.06,β=.95,p<001,f o r
Facebook intensity, and B=0.82,SEB=0.14,β=.68,
p<001, for SCC. Importantly, Facebook intensity predicted
changes in SCC over time, B=/C00.18,SEB=0.09,β=/C0.22,
p=.04. By contrast, SCC was not associated with respective
changes in Facebook intensity, B=0.03,SEB=0.09,
β=.02,p=.74.
Thus, the results demonstrate that more intensive use of
Facebook predicted less SCC over time, whereas the
reverse effect could not be corroborated. This finding
supports the fragmentation hypothesis, indicating thatintensive use of Facebook contributes to a more diffuse
sense of one ’ss e l f .
Figure 1. Cross-lagged model for Facebook intensity and self-concept
clarity with standardized effects. Measurement models and control
variables are not presented. Effects in italics are significant at p<. 05.8 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

General Discussion
SNSs such as Facebook offer plenty of possibilities to
provide, share, and obtain information that is relevant for
one ’s own view of oneself. The focus of our work was on
the relationship between Facebook intensity (i.e., theemotional connection to Facebook and its relevance in
people ’s daily lives; cf. Ellison et al., 2007 )o nt h eo n eh a n d
and self-concept clarity (i.e., the extent to which one ’s
self-concept is perceived to be internally consistent and
temporally stable; cf. Campbell et al., 1996 )o nt h e
other. One theoretical perspective, the self-concept unityhypothesis, assumes that the more intense the connection to
Facebook, the better individuals know about themselves,
owing to the vast opportunities for self-presentation andreceiving self-relevant feedback through Facebook and
other means of computer-mediated communication (cf.
Valkenburg & Peter, 2011). By contrast, the fragmentation
hypothesis suggests that the more intense the connection
to Facebook, the more confused individuals are about
themselves, because the multifaceted expressions of theself and heterogeneous feedback complicate the develop-
ment of a clear self-concept. Prior results were somewhat
supportive of the fragmentation hypothesis, as negativerelationships between several Internet use indicators and
SCC were found in studies with cross-sectional designs
(Davis, 2013 ; Israelashvili et al., 2012 ;M a t s u b a , 2006 ;
Valkenburg & Peter, 2008 ). These prior studies did not
focus on SNSs. Do these findings translate to the use of
Facebook?
With our first two studies, both cross-sectional as well,
we provided evidence of a negative link between Facebook
intensity and SCC. Moreover, the contribution of specificFacebook activities was examined. Adopting a completely
different identity was very rare, which reflects the differ-
ence between the use of Facebook and the use of applica-tions that were popular in the early days of the Internet,
such as anonymous chatrooms or MUDs (cf. Turkle,
1995 ). The negative links between SCC and “Focusing on
others ’reactions to postings ”and “Just browsing and
liking ”are in line with recent evidence suggesting that
the passive use of Facebook might have particularly
deleterious effects on users ’self. However, owing to the
cross-sectional methodology, the relationships observedmight also have been the result of selective exposure, that
is, the more individuals are unclear about themselves, the
more intense their affiliation with Facebook.Our third study provides evidence on the causal
mechanisms underlying the relationship between SCC
and Facebook intensity (this is one of only two longitudinalstudies we are aware of in which any Internet-related
v a r i a b l ea n dt h ec l a r i t yo fo n e ’s self-concept was con-
nected).
2On the basis of a cross-lagged panel analysis,
we found that higher Facebook intensity predicted lower
SCC at a later point of time. The reverse causal relationshipwas not supported by our data, as higher SCC was
unrelated to Facebook intensity at a later point of time.
Thus, it appears that a strong attachment to Facebookimpedes the development of a firm sense of oneself.
Limitations and Future Research
Despite the contribution of our studies, the limitations and
open questions associated with our research need to be
noted. First, our focus was on SCC, a structural feature ofthe self-concept. Our research is silent on the content of
the self-concept, for example, on the influence of Facebook
intensity on self-ascribed attributes (e.g., thoughtful,sportive, artistic) or self-esteem (see, e.g., Johnston et al.,
2013 ; Steinfield, Ellison, & Lampe, 2008 ;T o m a , 2013 ;
Toma & Hancock, 2013 ; Verduyn et al., 2015 ). Although a
positive link between SCC and explicit (but not implicit)
self-esteem was repeatedly found (see Brandt & Vonk,
2006 , for an overview), we believe that it is important to
stick to the conceptual separation between SCC and self-
concept content variables. We believe that theoreticalmodels and empirical studies that connect patterns of
SNS use with both SCC and self-esteem or well-being could
provide intriguing insights. In a study using experiencesampling, Kross and colleagues (Kross et al., 2013 )s h o w e d
that the amount of Facebook use (item “How much have
you used Facebook since the last time we asked? ”)
predicted a decline in affective well-being. Given our find-
ings, the reduction in SCC could be a process explaining
this effect.
Second, it needs to be stressed that self-reported SCC
does not equal an accuracy of the self-concept. In fact,
SCC is positively related to tendencies of self-deceptionand the self-reported clarity might in part be due to a
positive illusion of self-concept unity (Brandt & Vonk,
2006 )–a positive illusion that intense Facebook users
might have problems to uphold.
2The only other study (Yang & Brown, 2016) was published after the present studies were conducted. In two cross-sectional mediation models
they showed that the intentional use of Facebook for self-presentation (sample item “When I posted or shared things on Facebook, I rarely
thought about its consequences, ”reverse-coded) was positively related to general self-reflection (sample item “I frequently examine my
feelings ”), which was in turn negatively related to SCC. In a longitudinal model, higher SCC was predicted by higher self-esteem at an earlier
point of time. Others ’supportive reactions to the participants ’Facebook activities (sample item “I felt supported by the feedback ”) was found to
be unrelated to SCC cross-sectionally and longitudinally.M. Appel et al., Facebook and Self-Concept Clarity 9
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing

Third, although Study 2included specific activity
measures, our research was focused on the Facebook
Intensity Scale, which allows for a reliable and validmeasurement as well as a latent factor analysis. Its psycho-
metric properties are well-established. Keeping in mind the
problem of cumulated alpha errors in significance testing,
this is preferable to multiple single-items of variables.
However, the Facebook Intensity Scale cannot illuminateparticular activities or behavioral modes that can be made
responsible for the observed decrease in SCC. Our addi-
tional findings show that pretending to be someone else –
which could contribute to an unclear sense of the offline
self –is rare on Facebook. Future research seems warranted
that further examines the exact activities and stimuli thatare responsible for decreases in SCC. Promising research
avenues include the distinction between directed communi-
cation (interactions between the focal user and a friend)and consumption (Burke, Marlo, & Lento, 2010 ), or
between active and passive use (Verduyn et al., 2015 ).
On a related note, social comparison processes (cf.Corcoran, Crusius, & Mussweiler, 2011)m i g h tb eac r u c i a l
factor: People constantly compare themselves with others
to gather information about their characteristics andabilities. These social comparisons are automatic whenever
individuals are confronted with information about how
other people behave, think, and feel –and can even occur
outside conscious awareness (Mussweiler, Rüter, &
Epstude, 2004 ). On SNSs individuals are constantly
exposed to information about others (e.g., their currentactivities or achievements) and, thus, social comparisons
are particularly likely. Indeed, Facebook intensity was
associated with the frequency of social comparisons onFacebook (Lee, 2014 ). Comparison processes, in turn, were
found to be associated with feelings of uncertainty about
oneself (e.g., Butzer & Kuiper, 2006 ; Vartanian & Dey,
2013 ). This rationale connects to prior distinctions between
directed communication and consumption (Burke et al.,
2010 ), or between active and passive use (Verduyn et al.,
2015 ). Passive activities such as reading others ’postings
or browsing others ’photos on Facebook should trigger
these comparison processes more readily than more self-centered activities such as composing new status updates.
Clearly, more research is wa rranted that explicitly
addresses this mediating mechanism of social comparisonprocesses.
Fourth, the longitudinal design of Study 3allowed us to
inspect relationships over several months, but futureresearch might profit from more than two measurement
occasions encompassing longer time spans. Facebook
intensity showed higher stability estimates than SCC acrossthe longitudinal design of Study 3. The interplay between
the two constructs could be different when focusing on
longer periods. A longer time-span, preferably several years,would likely yield larger variations in Facebook intensity.
In that sense future research might allow for a more
nuanced identification of reciprocal processes (cf. Slater,2007 ,2015 ; Stiglbauer, Gnambs, Gamsjäger, & Batinic,
2013 ).
Finally, in order to provide evidence on causality, our
third study was longitudinal, adding to the small, but
growing literature that examined antecedents and conse-quences of SNS use over time (e.g., Kross et al., 2013 ;
Saslow, Muise, Impett, & Dubin, 2013 ; Steinfield, Ellison,
&L a m p e , 2008 ;T e p p e r s ,L u y c k x ,K l i m s t r a ,&G o o s s e n s ,
2014 ;T r e p t e&R e i n e c k e , 2013 ;V e r d u y ne ta l . , 2015 ).
To date, too few of the available evidence is based on
longitudinal studies. When experimental designs areinappropriate, only longitudinal data are able to shed light
on the direction of potential causal pathways. Despite the
virtue of longitudinal studies, the unaccounted-for influence
of third variables can pose a problem. We cannot rule out
the possibility that variations of a third and unaccounted-
for variable caused both Facebook intensity to increaseand SCC to decrease. Future studies are encouraged to
include control variables, such as users ’personality or more
state-like constructs such as loneliness, which may changesubstantially even if the retest interval is short (cf. Cacioppo
et al., 2000 ;G n a m b s , 2014 ;R y a n&X e n o s , 2011).
Conclusion
Facebook is –more or less so –part of the life of many
adolescents and adults. Our research indicates that with
an increasing connection to Facebook (Facebook intensity)individuals perceive their self-concept to be less clear and
less coherent (self-concept clarity). This intriguing finding
needs additional support in the years to come, includingfurther research on its mechanisms and boundary
conditions.
Acknowledgment
We are grateful to Alexandra Preslmayr and Fabiola
Gattringer for their support in conducting the studies.
References
Ahmad, I. (2014, June 11). What happens in just ONE minute on
Facebook [Web log message] . Retrieved from http:// www.
digitalinformationworld.com/2014/06/what-happens-in-a-facebook-minute-infographic.html
Appel, M., Mara, M., & Weber, S. (2014). Media and identity. In
M. B. Oliver & A. Raney (Eds.), Media and social life (pp. 16 –28).
New York, NY: Routledge. doi: 10.1177/0267323114539432a
Arnett, J. J. (1995). Adolescents ’uses of media for self-
socialization. Journal of Youth and Adolescence, 24 , 519 –533.
doi: 10.1007/BF0153705410 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C.,
Egloff, B., & Gosling, S. D. (2010). Facebook profiles reflect
actual personality, not self-idealization. Psychological Science,
21, 372 –374. doi: 10.1177/0956797609360756
Bem, D. J. (1972). Self-perception theory. Advances in Experimen-
tal Social Psychology, 6 ,1–62. doi: 10.1016/S0065-2601(08)
60024-6
Benoit, W. L., & Holbert, R. L. (2008). Empirical intersections in
communication research: Replication, multiple quantitative
methods, and bridging the quantitative –qualitative divide.
Journal of Communication, 58 , 615 –628. doi: 10.1111/j.1460-
2466.2008.00404.x
Brandt, A. C., & Vonk, R. (2006). Who do you think you are? On the
link between self-knowledge and self-esteem. In M. H. Kernis(Ed.), Self-esteem: Issues and answers (pp. 224 –229). Hove, UK:
Psychology Press.
Burke, M., Marlow, C., & Lento, T. (2010, April). Social network
activity and social well-being. Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems
(pp.1909 –1912), ACM. doi: 10.1145/1753326.1753613
Butzer, B., & Kuiper, N. A. (2006). Relationships between the
frequency of social comparisons and self-concept clarity,intolerance of uncertainty, anxiety, and depression. Personality
and Individual Differences, 41 , 167 –176. doi: 10.1016/j.paid.
2005.12.017
Cacioppo, J. T., Ernst, J. M., Burleson, M. H., McClintock, M. K.,
Malarkey, W. B., Hawkley, L. C., …Spiegel, D. (2000). Lonely
traits and concomitant physiological processes: The MacArthur
social neuroscience studies. International Journal of
Psychophysiology, 35 , 143 –154. doi: 10.1016/S0167-8760(99)
00049-5
Calvert, S. L. (2002). Identity construction on the Internet. In S. L.
Calvert, A. B. Jordan, & R. R. Cocking (Eds.), Children in the
digital age: Influences of electronic media on development (pp.
57–70). Westport, CT: Praeger.
Campbell, J. D., Assanand, S., & Di Paula, A. (2003). The structure
of the self-concept and its relation to psychological adjust-ment. Journal of Personality, 71 , 115 –140. doi: 10.1111/1467-
6494.t01-1-00002
Campbell, J. D., Trapnell, P. D., Heine, S. J., Katz, I. M.,
Lavallee, L. F., & Lehman, D. R. (1996). Self-concept clarity:Measurement, personality correlates, and cultural boundaries.
Journal of Personality and Social Psychology, 70 , 141 –156.
doi: 10.1037/0022-3514.70.6.1114
Chou, H. T., & Edge, N. (2012). “They are happier and having better
lives than I am: ”The impact of using Facebook on perceptions
of others ’lives. Cyberpsychology, Behavior, and Social
Networking, 15 , 117 –121. doi: 10.1089/cyber.2011.0324
Church, A. T., Katigbak, M. S., Ibáñez-Reyes, J., de Jesús Vargas-
Flores, J., Curtis, G. J., Tanaka-Matsumi, J., …Simon, J. Y. R.
(2014). Relating self-concept consistency to hedonic andeudaimonic well-being in eight cultures. Journal of
Cross-Cultural Psychology, 45 , 695 –712. doi: 10.1177/
0022022114527347
Clayton, R. B., Osborne, R. E., Miller, B. K., & Oberle, C. D. (2013).
Loneliness, anxiousness, and substance use as predictors of
Facebook use. Computers in Human Behavior, 29 , 687 –693.
doi: 10.1016/j.chb.2012.12.002
Conley, J. J. (1984). The hierarchy of consistency: A review and
model of longitudinal findings on adult individual differences in
intelligence, personality and self-opinion. Personality and
Individual Differences, 5 ,1 1–25. doi: 10.1016/0191-8869(84)
90133-8
Cooley, C. H. (1902). Human nature and the social order . New York,
NY: Charles Scribner ’s Sons.Corcoran, K., Crusius, J., & Mussweiler, T. (2011). Social
comparison: Motives, standards, and mechanisms. In
D. Chadee (Ed.), Theories in social psychology (pp. 119 –139).
Oxford, UK: Wiley-Blackwell.
Davis, K. (2013). Young people ’s digital lives: The impact of
interpersonal relationships and digital media use on adoles-
cents ’sense of identity. Computers in Human Behavior, 29 ,
2281–2293. doi: 10.1016/j.chb.2013.05.022
große Deters, F., & Mehl, M. R. (2013). Does posting Facebook
status updates increase or decrease loneliness? An onlinesocial networking experiment. Social Psychology and Personal-
ity Science, 4 , 579 –586. doi: 10.1177/1948550612469233
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of
Facebook “friends: ”Social capital and college students use of
online social network sites. Journal of Computer-Mediated
Communication, 12 , 1143 –1168. doi: 10.1111/j.1083-6101.
2007.00367.x
Ellison, N. B., Hancock, J. T., & Toma, C. L. (2012). Profile as
promise: A framework for conceptualizing veracity in online
dating self-presentations. New Media and Society, 14 ,4 5–62.
doi: 10.1177/1461444811410395
Facebook. (2016). Stats . Retrieved from http://newsroom.fb.com/
company-info/
Fiske, S. T. (2010). Social beings. Core motives in social psychology .
Hoboken, NJ: Wiley.
Frijns, T., & Finkenauer, C. (2009). Longitudinal associations
between keeping a secret and psychosocial adjustment in
adolescence. International Journal of Behavioral Development,
33, 145 –154. doi: 10.1177/0165025408098020
Gnambs, T. (2014). A meta-analysis of dependability coefficients
(test-retest reliabilities) for measures of the Big Five. Journal of
Research in Personality, 52 ,2 0–28. doi: 10.1016/j.jrp.2014.
06.003
Gosling, S. D., & Mason, W. (2015). Internet research in psychol-
ogy. Annual Review of Psychology, 66 , 877 –902. doi: 10.1146/
annurev-psych-010814-015321
Graham, L. T., & Gosling, S. D. (2013). Personality profiles asso-
ciated with different motivations for playing World of Warcraft.
Cyberpsychology, Behavior, and Social Networking, 16 ,
189–193. doi: 10.1089/cyber.2012.009
Gross, E. F. (2004). Adolescent Internet use: What we expect, what
teens report. Journal of Applied Developmental Psychology, 25 ,
633–649. doi: 10.1016/j.appdev.2004.09.005
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in
covariance structure analysis: Conventional criteria versus
new alternatives. Structural Equation Modeling, 6 ,1–55.
doi: 10.1080/10705519909540118
Huang, C. (2010). Internet use and psychological well-being: a
meta-analysis. Cyberpsychology, Behavior, and Social
Networking, 13 , 241 –249. doi: 10.1089/cyber.2009.0217
Israelashvili, M., Kim, T., & Bukobza, G. (2012). Adolescents ’over-
use of the cyber world: Internet addiction or identityexploration? Journal of Adolescence, 35 , 417 –424.
doi: 10.1016/j.adolescence.2011.07.015
Johnston, K., Tanner, M., Lalla, N., & Kawalski, D. (2013). Social
capital: the benefit of Facebook ‘friends ’.Behaviour & Infor-
mation Technology, 32 ,2 4–36. doi: 10.1080/0144929X.2010.
550063
K e n n y ,D . A . ,&D e P a u l o ,B . M .( 1 9 9 3 ) .D op e o p l ek n o wh o w
others view them? An empirical and theoretical account. Psycho-
logical Bulletin, 114 , 145–161. doi: 10.1037/0033-2909.114.1.145
Kim, J., & Lee, J.-E. R. (2011). The Facebook paths to happiness:
Effects of the number of Facebook friends and self-presenta-tion on subjective well-being. Cyberpsychology, Behavior, and
Social Networking, 14 , 359 –364. doi: 10.1089/cyber.2010.0374M. Appel et al., Facebook and Self-Concept Clarity 11
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing

Krasnova, H., Wenninger, H., Widjaja, T., & Buxmann, P. (2013).
Envy on Facebook: A hidden threat to users ’life satisfaction? In
Wirtschaftsinformatik Proceedings 2013 . Paper 92. Retrieved
from http://aisel.aisnet.org/wi2013/92
Kross, E., Verduyn, P., Demiralp, E., Park, J., Lee, D. S., Lin, N., …
Ybarra, O. (2013). Facebook use predicts declines in subjective
well-being in young adults. PLoS One, 8 (8), e69841.
doi: 10.1371/journal.pone.0069841
Lee, S. Y. (2014). How do people compare themselves with
others on social network sites?: The case of Facebook.Computers in Human Behavior, 32 , 253–260. doi: 10.1016/j.chb.
2013.12.009
Little, T. D. (2013). Longitudinal Structural Equation Modeling . New
York, NY: Guilford Press.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F.
(2002). To parcel or not to parcel: Exploring the question and
weighing the merits. Structural Equation Modeling, 9 , 151–173.
doi: 10.1207/S15328007SEM0902_1
Little, T. D., Preacher, K. J., Selig, J. P., & Card, N. A. (2007). New
developments in latent variable panel analyses of longitudinal
data. International Journal of Behavioral Development, 31 ,
357–365. doi: 10.1177/0165025407077757
Matsuba, M. K. (2006). Searching for self and relationships online.
CyberPsychology & Behavior, 9 , 275 –284. doi: 10.1089/
cpb.2006.9.275
McArdle, J. J., & Nesselroade, J. R. (2014). Longitudinal data
analysis using structural equation modeling . Washington, DC: APA.
Michikyan, M., Dennis, J., & Subrahmanyam, K. (2015). Can you
guess who I am? Real, ideal, and false self-presentation onFacebook among emerging adults. Emerging Adulthood, 3 ,
55–64. doi: 10.1177/2167696814532442
Michikyan, M., Subrahmanyam, K., & Dennis, J. (2014). Can
you tell who I am? Neuroticism, extraversion, and onlineself-presentation among young adults. Computers in Human
Behavior, 33 , 179 –183. doi: 10.1016/j.chb.2014.01.010
Mussweiler, T., Rüter, K., & Epstude, K. (2004). The man who
wasn ’t there: Subliminal social comparison standards
influence self-evaluation. Journal of Experimental Social
Psychology, 40 , 689 –696. doi: 10.1016/j.jesp.2004.01.004
Muthén, L. K., & Muthén, B. O. (1998 –2012). Mplus user ’s guide
(7th ed.). Los Angeles, CA: Muthén & Muthén.
Pabian, S., De Backer, C. J., & Vandebosch, H. (2015). Dark Triad
personality traits and adolescent cyber-aggression. Personality
and Individual Differences, 75 ,4 1–46. doi: 10.1016/j.paid.
2014.11.015
Reid, E. (1998). The self and the Internet: Variations on the illusion
of one self. In J. Gackenbach (Ed.), Psychology and the Internet:
Intrapersonal, interpersonal, and transpersonal implications(pp. 29 –41). San Diego, CA: Academic Press.
Ritchie, T. D., Sedikides, C., Wildschut, T., Arndt, J., & Gidron, Y.
(2011). Self-concept clarity mediates the relation betweenstress and subjective well-being. Self and Identity, 10 ,
493–
508. doi: 10.1080/15298868.2010.493066
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 , 1658 –1664. doi: 10.1016/j.chb.2011.
02.004
Saslow, L. R., Muise, A., Impett, E. A., & Dubin, M. (2013). Can you
see how happy we are? Facebook images and relationship
satisfaction. Social Psychological and Personality Science, 4 ,
411–418. doi: 10.1177/1948550612460059
Schermelleh-Engel, K., Moosbrugger, H., & Müller, H. (2003).
Evaluating the fit of structural equation models: Test of
significance and descriptive goodness-of-fit measures.
Methods of Psychological Research Online, 8 ,2 3–74.Slater, M. D. (2007). Reinforcing spirals: The mutual influence of
media selectivity and media effects and their impact on
individual behavior and social identity. Communication Theory,
17, 281 –303. doi: 10.1111/j.1468-2885.2007.00296.x
Slater, M. D. (2015). Reinforcing spirals model: Conceptualizing
the relationship between media content exposure and the
development and maintenance of attitudes. Media Psychology,
18, 370 –395. doi: 10.1080/15213269.2014.897236
Steele, C. M. (1988). The psychology of self-affirmation: Sustain-
ing the integrity of the self. Advances in Experimental Social
Psychology, 21 , 261–302. doi: 10.1016/S0065-2601(08)60229-4
Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital,
self-esteem, and use of online social network sites: A longitu-
dinal analysis. Journal of Applied Developmental Psychology,
29, 434 –445. doi: 10.1016/j.appdev.2008.07.002
Stiglbauer, B., Gnambs, T., Gamsjäger, M., & Batinic, B. (2013).
The upward spiral of adolescents ’positive school experiences
and happiness: Investigating reciprocal effects over time.Journal of School Psychology, 51 , 231 –242. doi: 10.1016/j.jsp.
2012.12.002
Stucke, T. S. (2002). Überprüfung einer deutschen Version der
Selbstkonzeptklarheits-Skala von Campbell [Investigation of aGerman version of Campbell ’s self-concept clarity scale].
Zeitschrift für Differentielle und Diagnostische Psychologie, 23 ,
475–484. doi: 10.1024//0170-1789.23.4.475
Subrahmanyam, K., & Šmahel, D. (2011). Digital youth: The role of
media in development . New York, NY: Springer.
Swann, W. B. Jr., & Bosson, J. K. (2010). Self and identity. In
S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Handbook of
social psychology (Vol. 1, pp. 589 –628) Hoboken, NJ: Wiley.
doi: 10.1002/9780470561119.socpsy001016
Teppers, E., Luyckx, K., Klimstra, T. A., & Goossens, L. (2014).
Loneliness and Facebook motives in adolescence: A longitudi-nal inquiry into directionality of effect. Journal of Adolescence,
37, 691 –699. doi: 10.1016/j.adolescence.2013.11.003
Trepte, S., & Reinecke, L. (2013). The reciprocal effects of social
network site use and the disposition for self-disclosure:A longitudinal study. Computers in Human Behavior, 29 ,
1102–1112. doi: 10.1016/j.chb.2012.10.002
Toma, C. L. (2013). Feeling better but doing worse: Effects of
Facebook self-presentation on implicit self-esteem andcognitive task performance. Media Psychology, 16 , 199 –220.
doi: 10.1080/15213269.2012.762189
Toma, C. L., & Carlson, C. L. (2015). How do Facebook users
believe they come across in their profiles?: A meta-perception
approach to investigating Facebook self-presentation.
Communication Research Reports, 32 ,9 3–101. doi: 10.1080/
08824096.2014.990557
Toma, C. L., & Hancock, J. T. (2013). Self-affirmation underlies
Facebook use. Personality and Social Psychology Bulletin, 39 ,
321–331. doi: 10.1177/0146167212474694
Turkle, S. (1995). Life on the screen: Identity in the age of the
Internet . New York, NY: Simon & Schuster.
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital
in a social network site?: Facebook use and collegestudents ’life satisfaction, trust, and participation. Journal of
Computer-Mediated Communication, 14 , 875 –901.
doi: 10.1111/j.1083-6101.2009.01474.x
Valkenburg, P. M., & Peter, J. (2008). Adolescents ’identity
experiments on the Internet. Consequences for social compe-
tence and self-concept unity. Communication Research, 35 ,
208–231. doi: 10.1177/0093650207313164
Valkenburg, P. M., & Peter, J. (2011). Online communication
among adolescents: An integrated model of its attraction,
opportunities, and risks. Journal of Adolescent Health, 48 ,
121–127. doi: 10.1016/j.jadohealth.2010.08.02012 M. Appel et al., Facebook and Self-Concept Clarity
/C2112016 Hogrefe Publishing Journal of Media Psychology (2016)

van Dijk, M. P., Branje, S., Keijsers, L., Hawk, S. T., Hale, W. W. III,
& Meeus, W. (2014). Self-concept clarity across
adolescence: Longitudinal associations with open communica-tion with parents and internalizing symptoms. Journal
of Youth and Adolescence, 43 , 1861 –1867. doi: 10.1007/
s10964-013-0055-x
Vartanian, L. R., & Dey, S. (2013). Self-concept clarity, thin-ideal
internalization, and appearance-related social comparison as
predictors of body dissatisfaction. Body Image, 10 , 495 –500.
doi: 10.1016/j.bodyim.2013.05.004
Verduyn, P., Lee, D., Park, J., Shablack, H., Orvell, A., Bayer, J., …
Kross, E. (2015). Passive Facebook usage undermines affective
well-being: Experimental and longitudinal evidence. Journal of
Experimental Psychology: General, 144 , 480–488. doi: 10.1037/
xge0000057
Watson, D. C. (2011). Gossip and the self. Journal of Applied Social
Psychology, 41 , 1818 –1833. doi: 10.1111/j.1559-1816.2011.
00772.x
Yang, C. C., & Brown, B. B. (2016). Online self-presentation on
Facebook and self development during the college transition.
Journal of Youth and Adolescence, 45 , 402 –416. doi: 10.1007/
s10964-015-0385-y
Received December 1, 2014
Revision received March 3, 2016Accepted March 7, 2016Published online December 30, 2016
Markus Appel
Department of PsychologyUniversity of Koblenz-LandauFortstr. 7
76829 Landau
Germanyappelm@uni-landau.de
Markus Appel (PhD, University of
Cologne, Germany) is a professor ofmedia psychology at the University
of Koblenz-Landau, Germany. His
main research fields are the pro-cessing of stories, the correlates
and consequences of using social
networking sites, stereotype threat,and humanoid robots.Constanze Schreiner (MSc, Univer-
sity of Regensburg, Germany) is aPhD candidate at the Institute forCommunication Psychology and
Media Education at the University
of Koblenz-Landau, Germany. Herresearch focuses on influencingfactors, mechanisms, and boundary
conditions of narrative persuasion.
Silvana Weber (PhD, University of
Koblenz-Landau, Germany) is a
postdoc at the Institute for Com-
munication Psychology and MediaEducation, University of Koblenz-Landau, Germany. Her research
investigates effects of stereotyping
communication against immigrantsand women, social identity, health,
and well-being.
Martina Mara obtained her doctor-
ate in psychology from the Univer-
sity of Koblenz-Landau, Germany.
She is a key researcher for RoboPsychology at the Ars ElectronicaFuturelab in Linz, Austria. Her work
focuses on human –robot relation-
ships and psychological implica-tions of digital media and emerging
technologies.
Timo Gnambs (PhD, University of
Erlangen-Nuremberg) is head of
scaling and test design at the
Leibniz Institute for EducationalTrajectories, Bamberg, Germany.His research focuses on technology-
based psychological assessment
(including computer-adaptive andweb-based testing), methods of
large-scale assessment, and meta-
analytic methods.
M. Appel et al., Facebook and Self-Concept Clarity 13
Journal of Media Psychology (2016) /C2112016 Hogrefe Publishing