Content uploaded by Jeffrey T Hancock
Author content
All content in this area was uploaded by Jeffrey T Hancock
Content may be subject to copyright.
Mirror, Mirror on my Facebook Wall:
Effects of Exposure to Facebook on Self-Esteem
Amy L. Gonzales, M.A., and Jeffrey T. Hancock, Ph.D.
Abstract
Contrasting hypotheses were posed to test the effect of Facebook exposure on self-esteem. Objective Self-
Awareness (OSA) from social psychology and the Hyperpersonal Model from computer-mediated communi-
cation were used to argue that Facebook would either diminish or enhance self-esteem respectively. The results
revealed that, in contrast to previous work on OSA, becoming self-aware by viewing one’s own Facebook profile
enhances self-esteem rather than diminishes it. Participants that updated their profiles and viewed their own
profiles during the experiment also reported greater self-esteem, which lends additional support to the Hy-
perpersonal Model. These findings suggest that selective self-presentation in digital media, which leads to inten-
sified relationship formation, also influences impressions of the self.
Introduction
Over a decade ago, Internet use was thought to
promote negative psychosocial well-being, including
depression and loneliness.
1
Having attracted attention in and
out of the research community, these findings prompted re-
searchers to take a more nuanced look at the relationship
between Internet use and psychosocial health,
2,3
at times
finding evidence that Internet use could be beneficial.
3,4
The
present study extends this research by examining the effects of
the social-networking site Facebook (http://facebook.com),
which represents a popular new form of Internet communi-
cation, on self-esteem.
Previous work has addressed the role of Facebook and the
ability to socialize, and the role that socializing online plays in
supporting self-esteem and various forms of social capital.
5,6
For example, one recent study found that Facebook can en-
hance ‘‘social self-esteem,’’ measured as perceptions of one’s
physical appearance, close relationships, and romantic ap-
peal, especially when users received positive feedback from
Facebook friends.
5
Also, individuals with low self-esteem
may see particularly positive benefits from the social oppor-
tunities provided by Facebook.
6
The effect of Facebook exposure on general self-esteem has
not been explored. Yet Facebook, and other social-network
sites, have the potential to affect temporary states of self-
esteem. Social-networksites are designed to share information
about the self with others, including likes/dislikes, hobbies,
and personal musings via ‘‘wall posts,’’ and ‘‘status updates.’’
This information could make people aware of their own lim-
itations and shortcomings, which would lower self-esteem,
7
or it could be that this information represents selective and
therefore positively biased aspects of the self, which might
raise self-esteem.
8
Does Facebook operate on self-esteem in the
same way non-digital information does, by decreasing self-
esteem? Or does the opportunity to present more positive
information about the self while filtering negative informa-
tion mean that reviewing one’s own Facebook site enhances
self-esteem? The following piece examines these questions,
by exploring the theoretical predictions of Objective Self-
Awareness (OSA) theory
9
and the Hyperpersonal Model.
8
Objective self-awareness
One theoretical approach relevant to the effects of social-
networking sites on self-esteem is OSA theory, one of the first
experimentally tested psychological theories of the self. The
theory assumes that humans experience the self as both
subject and object.
9
For example, the self as subject is found in
daily experiences of life (e.g., waiting for the bus, eating
lunch, watching TV
10
). In those experiences the self is an
active participant in life and is not self-conscious. However,
people become the ‘‘object of [their] own consciousness’’
when they focus attention on the self,
9(p2)
which can have
both positive and negative effects.
In a state of objective self-awareness, Duval and Wicklund
9
claim that people are prone to self-evaluations based on
broader social standards and norms. This usually results in
a greater sense of humility, or downgraded ratings of self,
and increased pro-social behavior. For example, people
report feeling greater responsibility for social injustice,
11
or
are less likely to take an extra helping of candy without
Cornell University, Ithaca, New York.
CYBERPSYCHOLOGY,BEHAVIOR,AND SOCIAL NETWORKING
Volume 14, Number 1-2, 2011
ªMary Ann Liebert, Inc.
DOI: 10.1089/cyber.2009.0411
79
being observed.
12
On the other hand, because most people
often fall short of social standards when self-awareness is
heightened, positive affect and self-esteem typically de-
crease when people are exposed to objective self-awareness
stimuli.
13
The stimuli used to evoke objective self-awareness is most
commonly a mirror,
13
although other stimuli include images
of the self,
14
audio feedback,
15
having a video camera pointed
at participants,
16
or having participants write autobiographi-
cal information.
11
These stimuli cause people to view them-
selves as they believe others do, even if they are not
immediately under observation. Exposure to these stimuli is
what leads to pro-social behavior and decreases in self-esteem.
Given that social-networking profiles include information
about the self similar to the type of information that is used to
prompt objective self-awareness (e.g., photos, autobiograph-
ical information), viewing one’s profile should prompt a
downgrading of self-esteem according to OSA theory. That is,
viewing one’s Facebook profile should negatively affect one’s
self-esteem. Furthermore, research in computer-mediated
communication has found that information online is often
over-interpreted relative to the same information provided
offline,
17
leading to exaggerated or stereotyped impres-
sions.
18
Is it possible that this same process could occur for
impressions of the self? If Facebook acts on self-esteem in the
same way as previous OSA stimuli, only to a more extreme
degree, one prediction is:
H1: Exposure to one’s Facebook site will have a more negative
effect on self-esteem than traditional objective self-awareness
stimuli (e.g., mirror).
Selective self-presentation
A second relevant theoretical approach to understanding
effects of Facebook use is the Hyperpersonal Model.
8
Walther
posits that affordances of the Internet allow users to selectively
self-present themselves in asynchronous media. People can take
their time when posting information about themselves, care-
fully selecting what aspects they would like to emphasize.
Evidence of selective self-presentation is found in a variety of
Internet spaces, including e-mails,
19
discussion boards,
20
and
online dating Web sites.
21,22
In addition to evidence that online self-presentations are
especially positive presentations, recent research in computer-
mediated communication (CMC) suggests that online self-
presentations can become integrated into how we view
ourselves, especially when the presentations take place in a
public, digital space.
23
This phenomenon, known as identity
shift, demonstrates that self-presentations enacted in online
space can impact users’ self-concepts.
Self-presentations online can be optimized through
selective self-presentation, and online self-presentation
affects attitudes about the self. Facebook profiles may
provide sufficiently positively biased stimuli to counter
the traditional effects of objective self-awareness, and
instead prompt a positive change in self-esteem. From
this perspective, the hyperpersonal prediction of expo-
sure to Facebook is:
H2: Exposure to one’s Facebook site will have a more positive
effect on self-esteem than a control condition or traditional
self-awareness stimuli (e.g., mirror).
Furthermore, if exposure to one’s own Facebook profile
increases self-esteem due to selective self-presentation, then
behaviors associated with selective self-presentation should
correlate with changes in self-esteem. For example, because
self-stimuli are most likely to be on one’s own profile
page, we would expect that participants who only view their
own profile page would report higher self-esteem than par-
ticipants who view other profiles within Facebook. Thus:
H3: Participants who exclusively examine only their own
profile will report higher self-esteem than participants who
view other profiles in addition to their own profiles.
Finally, selective self-presentation should be reflected pri-
marily in editing of one’s online self-presentation, according
to Walther.
8
That is, the ability to edit one’s self-presentation
after the fact is a unique attribute of asynchronous, text-
based communication. Thus, according to the Hyperpersonal
Model, we predict that:
H4: Participants who make changes to their profile during the
experiment will report higher self-esteem than participants
who do not.
Each of these predictions is tested in the following study,
comparing the effect of viewing one’s Facebook site, viewing
one’s own image in a mirror, and being in a control condition
on self-reported self-esteem.
Methods
Participants
A total of 63 students (16 males, 47 females) from a large,
Northeastern university participated in this study for extra
credit. The study consisted of three conditions: exposure to a
mirror, exposure to one’s own Facebook site, and a control
condition in which participants used the same room without
any treatment. Participants were randomly assigned to one of
the three conditions, with a total of 21 participants taking part
in each of the three conditions.
Procedure
Each participant was told that the study was designed to
examine ‘‘people’s attitudes about themselves after exploring
different Internet sites.’’ People in both offline conditions
were told that they were in a control condition, and thus
would not be online. In the online condition, participants
were asked to examine their own Facebook site.
In the Facebook stimulus condition, after logging on to
Facebook, participants were instructed to click on the ‘‘Profile’’
tab after the experimenter left the room. The profile page
contains the primary source of information on an individual
user. Participants were told to look through any of the tabs on
that page (Wall posts, Photos, Info, Boxes). Participants were
given no specific instructions about making changes to their
profile during the study. In addition to the main profile photo,
the profile page has information on recent activity on Facebook
sent to and from the site owner, personal demographic infor-
mation, photos, and quizzes completed by the site owner.
After being on Facebook for 3 minutes, the experimenter re-
turned with a survey. Participants were instructed to keep the
profile page open while completing the questionnaire.
80 GONZALES AND HANCOCK
Participants in the offline conditions were taken to the
same small computer cubicle used in the online condition. In
the objective self-awareness stimulus condition, a mirror was
placed against the computer screen. To reduce suspicion of
the mirror, they were also told that the cubicle was being
used for another experiment and that they should not move
anything. Other items were laid about the room in all
conditions (e.g., intercoms, a television) in order to enhance
the perception that the room was being used for another
experiment. Participants were given a survey of questions,
which were answered while being exposed to their own
reflection in the mirror.
In the offline control condition, participants sat in the same
room as participants in the previously mentioned two con-
ditions, but without the mirror present and without the
computer screen turned on. Participants were left with the
survey and given instructions to buzz the experimenter when
they had finished completing the survey. In all conditions,
experimenters returned to collect the survey, and participants
were then debriefed and probed for suspicion or failure to
comply with instruction.
Measures
Self-esteem. Self-esteem was measured using the Ro-
senburg Self-Esteem scale,
24
in which 10 items were used to
assess self-esteem (a¼0.82). Half of the items were reverse
coded. Responses were scored on a 4-point scale, ranging
from ‘‘strongly agree’’ to ‘‘strongly disagree.’’ Although this
scale is generally used to measure trait self-esteem, as men-
tioned above, previous studies of objective self-awareness
have used this measure to capture temporary changes in self-
esteem due to awareness-enhancing stimuli.
7
Selective self-presentation. In order to examine behav-
iors predicted by the Hyperpersonal Model, we asked par-
ticipants in the Facebook condition about their behavior
while they were on Facebook. Questions included, ‘‘Did you
leave your profile at any time during the study?’’ (1 ¼‘‘yes,’’
2¼‘‘no’’), and ‘‘Did you change your profile while you were
on the Web site?’’ (1 ¼‘‘yes,’’ 2 ¼‘‘no’’).
Results
To establish that the objective self-awareness stimuli had
an effect on self-esteem, an analysis of variance (ANOVA)
was first performed. Gender was also included in the model
as a covariate, given previous research suggesting that
gender may predict differences in self-esteem.
25
The fol-
lowing analyses all reflect significant differences using two-
tailed tests of significance, unless otherwise noted. Indeed,
the stimuli did have an effect on self-esteem, F(1, 59) ¼4.47,
p¼0.02, Z
2
¼0.13.However,genderwasnotasignifi-
cant predictor of self-esteem, F(1, 60) ¼0.94, p¼0.34. This
finding reveals that self-reported self-esteem did vary by
condition.
To test the hypothesis that Facebook had a more negative
effect on self-esteem than traditional objective self-awareness
stimulus (H1), a linear contrast analysis was performed with
a weight 0 assigned to the traditional objective self-awareness
stimulus condition (i.e., mirror, M¼2.97, SD ¼0.51), a weight
of 1 assigned to the Facebook condition (M¼3.35,
SD ¼0.37), and a weight of 1 assigned to the control condition
(M¼3.23, SD ¼0.40). The results of this test were not sig-
nificant, F(1, 60) ¼0.95, p¼0.33.
To test the opposing hypothesis that Facebook has a pos-
itive impact on self-esteem (H2), a different linear contrast
analysis was performed. A contrast weight of 1 was as-
signed to the traditional objective self-awareness stimuli
condition, 0 was assigned to the control condition, and þ1
was assigned to the Facebook condition. This contrast anal-
ysis was significant, F(1, 59) ¼8.60, p<0.01, Z
2
¼0.13, dem-
onstrating support for H2 and suggesting that Facebook has a
positive effect on self-esteem relative to a traditional objective
self-awareness stimulus.
Given that viewing Facebook enhanced self-esteem, is
there additional evidence that the process of selective self-
presentation was responsible for influencing self-esteem? Our
first method of testing this question included examining
whether participants who exclusively viewed their own
profile reported having higher self-esteem than participants
who also viewed the profiles of others. An ordinary least
squares (OLS) regression of self-esteem on viewing behavior
(self-only profile vs. self and other profiles) and gender re-
vealed a significant effect on viewing behavior, b¼0.40,
p¼0.03 (one-tailed, 1 ¼‘‘yes,’’ 2 ¼‘‘no’’), indicating that par-
ticipants who left their profile during the study reported
lower self-esteem than those participants who exclusively
viewed their own profile site, supporting H3. The relation-
ship between gender and self-esteem was not significant,
b¼0.33, p¼0.12 (1 ¼female, 2 ¼male).
Finally, we expected that changes to any part of the profile
(i.e., status, photo, etc.) during the study would increase
participant self-esteem (H4), as editing is a primary means of
optimizing self-presentation, according to the Hyperpersonal
Model.
8
We tested this hypothesis using OLS regression, and
once again included gender in the analysis. In support of this
hypothesis, participants who changed their profile during the
study reported higher self-esteem than those who did not
change their profile, b¼0.53,p¼0.01, (1 ¼‘‘yes,’’ 2 ¼‘‘no ’’ ) .
These data suggest that, because asynchronous social-network
profiles allow for added time and energy to construct positive
self-presentations, profiles contain information that prompts
positive, rather than negative, effects on self-esteem. Men re-
ported having greater self-esteem than women after controlling
for the likelihood that participants changed their profile,
b¼0.45, p¼0.03. However, this result cannot be fairly inter-
preted due to the very small number of men (17 women, 4 men).
Discussion
This study was designed to test the effects of exposure to
Facebook on self-esteem relative to traditional self-awareness
enhancing stimuli, such as a mirror or photo of oneself. The
study suggests that selective self-presentation, afforded by
digitally mediated environments can have a positive influ-
ence on self-esteem.
These findings are in contrast to predictions from OSA
theory, which posits that stimuli that prompt self-awareness
(e.g., mirror, photo, autobiographical information) activate
discrepancies between oneself and social standards,
9
and
consequently lower self-esteem.
13,15
Instead, the results
demonstrate that exposure to information presented on one’s
Facebook profile enhances self-esteem, especially when
a person edits information about the self, or selectively
FACEBOOK EFFECTS ON SELF-ESTEEM 81
self-presents. These findings are consistent with Walther’s
Hyperpersonal Model
8
and suggest that the process of
selective self-presentation, which takes place in mediated
spaces due to increased time for creating a self-presentation,
makes Facebook a unique awareness-enhancing stimuli.
This study is a preliminary step toward understanding
how selective self-presentation processes, which have been
previously discussed in the context of interpersonal impres-
sion formation,
19,20,22
may also influence impressions of the
self. Whereas a non-edited view of the self (i.e., mirror) is
likely to decrease self-esteem, these findings suggest that the
extra care involved in digital self-presentations may actually
improve self-esteem. By allowing people to present preferred
or positive information about the self, Facebook is a unique
source of self-awareness stimuli in that it enhances awareness
of the optimal self. This finding is consistent with previous
work that has found that digital self-presentations can shape
self-assessments.
23
In this case, however, the findings are
striking because they contradict previous work on the neg-
ative effect of self-awareness enhancing information on
self-assessments.
Previous work examining self-esteem suggests that con-
sistency between the actual and the ideal self is an important
factor in understanding how information can affect self-
esteem.
26
Although participant perceptions between the actual
and ideal self were not measured, it is possible that Facebook
activates the ideal self. Future research on implications of self-
evaluations on self-esteem is needed to test this possibility.
Facebook may also be unique in that the public nature of
the site may contribute to objective self-awareness. In previ-
ous work, autobiographical information or photos have
prompted objective self-awareness.
11,14,15
We tested OSA in
Facebook because these features are present there. However,
Facebook is a public site, which should also remind users of
self-evaluation. In this case, the same information that is
prompting OSA is actually viewed and evaluated by others as
well. Further work is necessary to determine whether public
Internet audiences alone may stimulate OSA. In this case, we
can only speculate that the high visibility of one’s Facebook
profile further adds to a sense of objective self-awareness. The
difference is that that while Facebook may prime awareness
of an audience and self-evaluation, it is a more optimal self
that is being evaluated. Thus the effect of self-esteem is pos-
itive rather than negative.
Limitations
An important limitation of this study was our failure to
account for the effect of the number of Facebook friends on
self-esteem. As previous research has demonstrated, the so-
cial opportunities in Facebook contribute to an enhanced
feeling of social competence.
5,6
We cannot rule out the pos-
sibility that reminders of one’s social connections are partially
responsible for the increase in self-esteem. On the other hand,
social connection does not seem to be completely responsible
for this effect. Changes to one’s profile and attention to one’s
profile (vs. others’ profiles) have a positive effect on self-
esteem, which suggests that selective self-presentation is a
factor in shaping the resultant self-reports of self-esteem.
Another limitation is that we cannot know the long-term
implications of using Facebook on self-esteem from a single
study. The measure of self-esteem used in this study is gen-
erally used as a measure of stable self-esteem, but has been
used on other occasions to measure temporary shifts in self-
esteem.
7,13,15
Though difficult to perform in an experimental
setting, research that examines long-term effects of social-
network sites, such as Facebook, would be valuable. Also,
incorporating pre- and post-test measures of self-esteem and
other relevant psychological measures would be useful in
future work.
The focus in the present study is on Facebook, although we
make arguments about social-network sites in general re-
garding their effect on self-esteem. While future research will
be required to extend these findings beyond Facebook, the
Facebook interface has several advantages over other sites,
such as MySpace (http://myspace.com), including a more
uniform layout and the sheer popularity of the site. Given
that every person must view their own site, the increased
uniformity and popularity of Facebook made it a useful
starting point for examining digital self-awareness stimuli
and self-esteem.
Finally, participants in the offline conditions did not have
the same 3-minute lapse between coming into the room
and completing the questionnaire as participants in the Face-
book condition. We were concerned, however, that including
a filler task would potentially introduce an additional and
unintended manipulation into the study. It seems unlikely that
the time lapse alone was part of the reason for the different
ratings of self-esteem, but to be sure, future research will need
to account forthis effect by providing an appropriate filler task
for participants in the non-digital environments.
Conclusion
The Internet has not created new motivation for self-
presentation, but provides new tools to implement such
motives. The negative effects of objective self-awareness on
self-esteem originated from work in the early 1970s.
9,13–15
Social-networking sites, a product of the 21st century, pro-
vide new access to the self as an object. By providing multiple
opportunities for selective self-presentation—through pho-
tos, personal details, and witty comments—social-networking
sites exemplify how modern technology sometimes forces us
to reconsider previously understood psychological processes.
Theoretical development can benefit from expanding on
previous ‘‘offline’’ theories by incorporating an understand-
ing of how media may alter social processes.
Acknowledgments
We would like to thank Angela Falisi, Allison Fishler, and
Regine Mechulan for their assistance in collecting the data for
this experiment.
Disclosure Statement
No competing financial interests exist.
References
1. Kraut R, Patterson M, Lundmark V, et al. Internet paradox:
A social technology that reduces social involvement and
psychological well-being? American Psychologist 1998; 53:
1017–31.
82 GONZALES AND HANCOCK
2. Bessie
`re K, Kiesler S, Kraut R, et al. Effects of Internet use
and social resources on changes in depression. Information,
Communication & Society 2008; 11:47–70.
3. McKenna KYA, Bargh JA. Plan 9 from cyberspace: The im-
plications of the Internet for personality and social psychol-
ogy. Personality & Social Psychology Review 2000; 4:57–75.
4. Shaw LH, Gant LM. In defense of the Internet: The rela-
tionship between Internet communication and depression,
loneliness, self-esteem, and perceived social support. Cy-
berPsychology & Behavior 2002; 5:157–71.
5. Valkenburg PM, Peter J, Schouten AP. Friend networking
sites and their relationship to adolescents’ well-being and
social self-esteem. CyberPsychology & Behavior 2006; 9:
484–590.
6. Ellison NB, Steinfield C, Lampe C. The benefits of Facebook
‘‘friends’’: Social capital and college students’ use of online
social network sites. Journal of Computer-Mediated Com-
munication 2007; 12: 1. jcmc.indiana.edu/vol12/issue4/
ellison.html (Accessed Jan. 27, 2009).
7. Heine SJ, Takemoto T, Moskalenko S, et al. Mirrors in the
head: Cultural variation in objective self-awareness. Per-
sonality & Social Psychology Bulletin 2008; 34:879–87.
8. Walther JB. Computer-mediated communication: Im-
personal, interpersonal, and hyperpersonal interaction.
Communication Research 1996; 23:3–43.
9. Duval S, Wicklund RA. (1972) A theory of objective self
awareness. New York: Academic Press.
10. Moskalenko S, Heine SJ. Watching your troubles away: Tel-
evision viewing as a stimulus for a subjective self-awareness.
Personality & Social Psychology Bulletin 2003; 29:76–85.
11. Duval S, Duval VH, Neely, R. Self-focus, felt responsibility,
and helping behavior. Journal of Personality & Social Psy-
chology 1979; 37:1769–78.
12. Beaman AL, Klentz B, Diener E, et al. Self-awareness and
transgression in children: Two field studies. Journal of Per-
sonality & Social Psychology 1979; 37:1835–46.
13. Fejfar MC, Hoyle RH. Effect of private self-awareness on
negative affect and self-referent attribution: A quantitative re-
view. Personality & Social Psychology Review 2000; 4:132–42.
14. Storms MD. Videotape and the attribution process: Rever-
sing actors’ and observers’ points of view. Journal of Per-
sonality & Social Psychology 1973; 27:165–75.
15. Ickes WJ, Wicklund RA, Ferris CB. Objective self-awareness
and self-esteem. Journal of Experimental Social Psychology
1973; 9:202–19.
16. Duval T, Duval V, Mulilis J. Effects of self-focus, discrepancy
between self and standard, and outcome expectancy favor-
ability on the tendency to match self to standard and with-
draw. Journal of Personality & Social Psychology 1992;
62:340–8.
17. Hancock JT, Dunham PJ. Impression formation in computer-
mediated communication. Communication Research 2001;
28:325–47.
18. Epley N, Kruger J. When what you type isn’t what they
read: The perseverance of stereotypes and expectancies over
e-mail. Journal of Experimental Social Psychology 2005;
41:414–22.
19. Duthler KW. The politeness of requests made via email and
voicemail: Support for the hyperpersonal model. Journal of
Computer-Mediated Communication 2006; 11. jcmc.indiana
.edu/vol11/issue2/duthler.html (accessed Jan. 13, 2009).
20. Walther JB. Selective self-presentation in computer-mediated
communication: Hyperpersonal dimensions of technology,
language, and cognition. Computers in Human Behavior
2007; 23:2538–57.
21. Ellison N, Heino R, Gibbs J. Managing impressions online:
Self-presentation processes in the online dating environ-
ment. Journal of Computer-Mediated Communication 2006;
11. //jcmc.indiana.edu/vol11/issue2/ellison.html (Accessed
Sept. 12, 2007).
22. Toma CL, Hancock JT, Ellison NB. Separating fact from
fiction: An examination of deceptive self-presentation in
online dating profiles. Personality & Social Psychology
Bulletin 2008; 4:1023–36.
23. Gonzales AL, Hancock JT. Identity shift in computer-
mediated environments. Media Psychology 2008; 11:167–85.
24. Rosenberg M. (1965) Society and the adolescent self-image.
Princeton, NJ: Princeton University Press.
25. Josephs RA, Markus HR, Tafarodi RW. Gender and self-
esteem. Journal of Personality & Social Psychology 1992;
63:391–402.
Address correspondence to:
Amy L. Gonzales
327 Kennedy Hall
Cornell University
Ithaca, NY 14850
E-mail: alg49@cornell.edu
FACEBOOK EFFECTS ON SELF-ESTEEM 83
Copyright of CyberPsychology, Behavior & Social Networking is the property of Mary Ann Liebert, Inc. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.
