“Selfie-ists” or “Narci-selfiers”?: A cross-lagged panel analysis of selfie taking and narcissism

Abstract

We examine the widely popular social phenomenon of “selfies” (self-portraits uploaded and shared in social media) in terms of the observed positive relationship between this individualistic form of social media usage and narcissism. We conducted a cross-lagged analysis of a two-wave, representative panel survey to understand whether narcissists take selfies as an outlet for maintaining their positive self-views (the self-selection hypothesis), or if by taking selfies' users would increase their level of narcissism (the media effect hypothesis). The findings, however, are consistent with both hypotheses, suggesting a self-reinforcement effect: whereas narcissist individuals take selfies more frequently over time, this increase in selfie production raises subsequent levels of narcissism.

Full text

Selfie-ists” or “Narci-selfiers 1
“Selfie-ists” or “Narci-selfiers”?: A cross-lagged panel analysis of selfie taking and
narcissism
Daniel Halpern and Sebastián Valenzuela
Pontificia Universidad Católica de Chile
James E. Katz
Boston University
[Published in Personality and Individual Differences, 97, 98-101.
doi:10.1016/j.paid.2016.03.019. Available at:
http://www.sciencedirect.com/science/article/pii/S0191886916301635]
Abstract: We examine the contemporary, widely popular social phenomenon of ‘‘selfies’’
(self-portraits uploaded and shared in social media) in terms of the observed positive
relationship between this individualistic form of social media usage and narcissism. We
conducted a cross-lagged analysis of a two-wave, representative panel survey to understand
whether narcissists would take selfies as an outlet for maintaining their positive self-views
(the self-selection hypothesis), or if by taking selfies users would increase their level of
narcissism (the media effect hypothesis). Results, however, are consistent with both
hypotheses, suggesting a self-reinforcement effect: whereas narcissist individuals take
selfies more frequently over time, this increase in selfie production raises subsequent levels
of narcissism.
Keywords: Selfie; Narcissism, Social Media, Self-reinforcement effect; cross-lagged
analysis

Selfie-ists” or “Narci-selfiers 2
Introduction
Found in many people to a degree, subclinical narcissism traits include grandiosity,
dominance and superiority (Paulhus & Williams, 2002). Narcissism may be a response to
parental insensitivity, leading to a defensive grandiosity accompanied by a sense of
inferiority (Kernberg, 1975). Narcissists often believe they are special and superior to
others, but respond to self-esteem threats with rage, defiance, shame, and humiliation (Morf
& Rhodewalt, 2001). They present a success-oriented personality (Campbell & Foster,
2007), and expect to be treated accordingly while giving little back (Bergman et al., 2011).
Narcissists may not seek long-term relationships preferring to use relationships to augment
their popularity and satisfy short-term status needs (Campbell, 1999). Relevant for our
research is the fact that they generally need social contacts in order to obtain the attention
they require for building their self-esteem (Buffardi & Campbel, 2008). This, instead, leads
them to maintain several attention-seeking and exhibitionist strategies (Bergman et al.,
2011). Narcissists also tend to use their interpersonal relationships to make themselves
appear more socially desirable and to seek out external sources of admiration to help
maintain their self-esteem (Brown & Zeigler-Hill, 2004).
Narcissism and social media
Social network sites (SNS) offer affordances possibly appealing narcissists. First,
narcissists tend to develop less intimate -as opposed to emotionally committed-
relationships (Bergman et al., 2011), and SNS are places where social relationships with
weak ties can be cultivated (Ellison, Steinfield & Lampe, 2007). Thus, these platforms have
become fertile ground for narcissists to form ‘‘weak tie’’ connections that do not require

Selfie-ists” or “Narci-selfiers 3
emotional investments (Bergman et al., 2011), but also satisfy their need to get the attention
of more people (Morf & Rhodewalt, 2001). Second, because the quantity of social media
contacts are counted and displayed, these platforms are a plausible indicator of status and
popularity (Sundar, 2008). Consequently, narcissists can maintain their inflated self-views
by gaining higher numbers of SNS friends, which affirms their grandiosity and ensures a
large online audience for their posts, videos or messages (Bergman et al., 2011). In fact,
research has shown a positive relationship between the number of contacts, activity in SNS
and levels of narcissism (Carpenter, 2011). Third, SNS present several advantages for
strategic self-presentation where users can promote themselves in highly tailored fashion
via self-descriptions (e.g., profile picture) and satisfy their vanity via photos. Ong et al.
(2011) found that narcissistic individuals rated their Facebook profile pictures as more cool
and fashionable, suggesting that they select profile photos that are self-perceived as more
physically appealing. This is consistent with strategies adopted by narcissists in affirming
their positive illusions of their physical appearance (Morf & Rhodewalt, 2001).
An argument then can be that narcissism is positively related to SNS use. However,
relatively little research has focused on the causality aspect: most of the studies have
assumed that this personality trait is a predictor for SNS use. Nevertheless, a positive
relationship between narcissism and SNS could be explained by two quite different
perspectives: (1) Narcissists utilize more social media (the self-selection hypothesis), or (2)
Use of SNS produce higher levels of narcissism (the media effect hypothesis). Whereas
both views predict a positive association between SNS use and narcissism, the implications
are quite the opposite. In the first case, use of social media would be perceived as a
consequence of a character trait. In the second case, a specific SNS use would be capable of
augmenting or stimulating the same personality trait. In fact, research has shown that being

Selfie-ists” or “Narci-selfiers 4
exposed on SNS can have different effects on individuals, such as elevating users’ self-
disclosure (Trepte & Reinecke, 2013), their level of bridging social capital (Burke, Kraut &
Marlow, 2011), and empower users by acquiring the role of helpers (Barak, Boniel-Nissim
& Suler, 2008). It is likely, too, that these two types of effects co-exist, reinforcing each
other in positive feedback fashion.
Thus, narcissistic users would use SNS more, and by using SNS more, they become
more narcissistic over time. Thus, as communication researchers, it is worth examining
whether and how specific uses of SNS, such as selfies, are related to narcissism.
Selfie-ssism or Narci-selfiers?
Popular in social media use, “selfies” (self-portraits uploaded and shared in social media)
are being used transversally across SNS platforms, and so provide a gateway to investigate
this topic. Selfies allow users novel opportunities to customize their self-presentation by
giving picture-takers a vast array of ways to portray themselves (Katz & Crocker, 2015).
Interestingly, recent research has shown that this individualistic form of SNS usage, where
users focus on how they want to be seen by others from their own perspective, is highly
related to narcissism. Both Sorowsky et al. (2015) and Fox and Rooney (2015) found that
men’s overall narcissism scores positively predict posting selfies, and Fox and Rooney also
determined that they edit their SNS-posted selfies more frequently. Overall, these studies
provide evidence for a strong relationship between this personality trait and selfie
production. Nevertheless, they do not address the causality quandary, leaving open the
question of whether the selfies-narcissism relationship is explained by selectivity, media
effects, or both. Thus, we posit the following hypotheses, which will be examined using
longitudinal survey data:

Selfie-ists” or “Narci-selfiers 5
H1: Higher levels of narcissism will predict increasing frequency of taking selfies
for sharing.
H2: Higher frequency of taking selfies for sharing will predict increasing levels of
narcissism.
Method
To examine the hypotheses, a two-wave panel survey was conducted one year apart.
To assure a more accurate representation, a national panel was employed following the
Chilean National Socioeconomic Characterization Survey (CASEN). For this purpose,
three variables were considered: gender (male: 48.7%; female 51.3%); age (18–34: 55%;
35–44: 20%; 45-64: 22%; 65+: 3%) and geography. The selected panel members received
the survey’s URL through an e-mail invitation. 8,840 participants received the first email,
1,225 responded the questionnaire, yielding a 13.9% response rate. A follow-up email was
sent to participants after one year, inviting them to participate in a follow-up survey. In
total, 314 of the participants completed both questionnaires.
Measures
Narcissism. Narcissism was assessed using 10 items of the Narcissism Personality
Inventory (NPI), with statements responding to the authority/superiority dimension (e.g., “I
am more capable than other people”), exhibitionism (e.g., “I like to be the center of the
attention”), exploitativeness (e.g., “I will never be satisfied until I get all that I deserve”)
and vanity (e.g., “I like to look at myself in the mirror”). For each item, respondents used a
5-point Likert scale from strongly disagree to strongly agree. Subsequently, responses were
added and averaged (wave 1: M = 2.75, SD = 0.55, Cronbach’s α = 0.77; wave 2: M = 2.84,
SD = 0.57, Cronbach’s α = 0.74). A paired t-test showed a significant increase in the
average of narcissism levels between waves, t(276) = 3.27, p < 0.001.

Selfie-ists” or “Narci-selfiers 6
Selfies. Respondents were asked how often, in the past year, they had taken a picture of
themselves for sharing on a 7-point scale, from never to everyday (wave 1: M = 2.58, SD =
1.24; wave 2: M = 2.97, SD = 1.43). As was the case with narcissism, selfie taking
increased significantly over time, t(301) = 5.63, p < 0.001.
Control variables. A host of control variables were measured, all in wave 1 (with one
exception, explained below). Given that extroversion and openness to new experiences
have been important personality predictors of SNS usage (Correa, Hinsley & De Zuniga,
2010; Gosling et al., 2011), we asked participants to respond from 1 to 10, how much
extroverted (M = 6.04, SD = 2.14) and how much open to experience new situations were
they (M = 6.98, SD = 2.15). We also controlled for age (M= 31.3, SD = 11.3) and gender
(65% female). Due to changes between waves, frequency of use of seven online media that
support sharing selfies (including Facebook, Twitter, Instagram, Snapchat and WhatsApp)
was measured in both waves on a 7-point scale, from never to all the time (wave 1: M =
3.20, SD = 0.80, Cronbach’s α = 0.66; wave 2: M = 2.93, SD = 0.71, Cronbach’s α = 0.66;
t(278) = -6.54, p < 0.001).
Statistical Analysis
To analyze over-time relationships between selfie behavior and narcissism, we estimated a
cross-lagged panel model (Campbell, 1963; Kessler & Greenberg, 1981), in which both
synchronous and asynchronous associations between the key variables were calculated
using structural equation modeling. The key advantage of this design over other common
panel data models (e.g., change scores, lagged dependent variable) is that it takes into
account both inertial effects (i.e., the influence of a variable at time 1 on the same variable
at time 2) and possible reciprocal effects (i.e., the influence of a variable at time 1 on
another variable at time 2 and vice versa). Hence, it is well suited for testing for selective,

Selfie-ists” or “Narci-selfiers 7
causal and/or reinforcing effects. Because 7% of the sample had missing values in the
variables of interest, full information maximum likelihood estimation was used (Holbert &
Stephenson, 2002) on the rather small sample size (N = 303). Considering the fact that we
were predicting directionality in the hypotheses, the α level of statistical significance was
set at p < 0.10. To facilitate interpretation, all results employ standardized (z-score)
coefficients. The model was estimated using Stata 14.1.
Results
Table 1 shows the zero-order, bivariate correlations between all variables that were
subsequently included in the cross-lagged model. The results of this preliminary analysis
show that the correlations between frequency of taking selfies and narcissism levels at time
1, time 2 and across times 1 and 2 are all positive and statistically significant. However,
these results may be spurious, as they do not control for inertial effects and other
individual-level differences, such as age, gender, personality traits and general online and
social media use.
[INSERT TABLE 1]
Thus, we estimated the cross-lagged structural model depicted in Figure 1 using
structural equation modeling. This model included the key variables as well as the five
control variables (not shown in the figure for ease of presentation). As shown, there was
strong evidence of a positive reinforcement between selfie behavior and narcissism. On the
one hand, after controlling inertial effects and contemporaneous relationships, the cross-
lagged effect of narcissism at time 1 on selfie behavior at time 2 was positive and
significant (standardized β = .10 [90% confidence interval = .03, .18], p < 0.05). On the
other hand, the significant effect from frequency of selfie taking at time 1 on narcissism
levels at time 2 was also positive and significant (standardized β = .07 [90% C. I. = .01, .

Selfie-ists” or “Narci-selfiers 8
16], p < 0.10). And, considering the confidence intervals of the standardized coefficients,
the reciprocal relationship between selfies and narcissism was rather symmetrical.
[INSERT FIGURE 1]
Discussion
This study aims to understand whether narcissists take selfies as an outlet for
maintaining their positive self-views (the self-selection hypothesis), or if by taking selfies
users increase their level of narcissism (the media effect hypothesis). The results of a two-
wave panel study demonstrated that this character trait influences selfie production (self-
selection effect) and that, in turn, increases the levels of narcissism reported by users over
time (media effect). Thus, the data supports a self-reinforcement effect (Slater, 2007) that
can be understood as a reciprocal process that is reinforced by higher levels of narcissism
that users gain through the practice of selfies. This means that users who engage in this
behavior probably feel rewarded by sharing their own images with other users, augmenting
their levels of narcissism and consequently their use of SNS for selfie production.
The present study extends prior research in selfies and narcissism (Fox & Rooney,
2015; Sorowsky et al., 2015) by offering two new insights. First, it presents what we
believe is the first longitudinal study of causal effects between these variables,
demonstrating that they reinforce over time. Secondly, it validates the application of a new
theoretical approach to understand reciprocal processes that drive computer-mediated
communication. In our study, people with high levels of narcissism engage in frequent use
of selfies. According to previous research, it could be argued that this trend is attributable to
the fact that posting and editing photos encourages users to engage in self-promoting and
superficial behaviors. Thus, the prevalence of narcissistic individuals practicing selfies may
lead to a rise of this trait among users who have some initial degree of narcissism, or

Selfie-ists” or “Narci-selfiers 9
possibly (and here we speculate beyond what our data support) even inducing it in those
who do not manifest it.

Selfie-ists” or “Narci-selfiers 10
REFERENCES
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A dynamic
self-regulatory processing model. Psychological inquiry, 12(4), 177-196.
Barak, A., Boniel-Nissim, M., & Suler, J. (2008). Fostering empowerment in online support
groups. Computers in Human Behavior, 24(5), 1867-1883.
Bergman, S. M., Fearrington, M. E., Davenport, S. W., & Bergman, J. Z. (2011).
Millennials, narcissism, and social networking: What narcissists do on social networking
sites and why. Personality and Individual Differences,50(5), 706-711.
Brown, R. P., & Zeigler-Hill, V. (2004). Narcissism and the non-equivalence of self-esteem
measures: A matter of dominance?. Journal of Research in Personality, 38(6), 585-592.
Buffardi, L. E., & Campbell, W. K. (2008). Narcissism and social networking web
sites. Personality and social psychology bulletin, 34(10), 1303-1314.
Burke, M., Kraut, R., & Marlow, C. (2011, May). Social capital on Facebook:
Differentiating uses and users. In Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems (pp. 571-580). ACM.
Campbell DT. (1963). From description to experimentation: Interpreting trends as quasi-
experiments. In CW. Harris Ed., Problems in measuring change. Madison: University of
Wisconsin Press.
Campbell, W. K. (1999). Narcissism and romantic attraction. Journal of Personality and
Social Psychology, 77(6), 1254.
Campbell WK, Foster, JD. (2007). The narcissistic self: Background, an extended agency
model, and ongoing controversies. In: Sedikides C, Spencer SJ, eds. The self: Frontiers of
social psychology, Psychology Press: New York, pp. 115–138

Selfie-ists” or “Narci-selfiers 11
Carpenter, C. J. (2012). Narcissism on Facebook: Self-promotional and anti-social
behavior. Personality and individual differences, 52(4), 482-486.
Correa, T., Hinsley, A. W., & De Zuniga, H. G. (2010). Who interacts on the Web?: The
intersection of users’ personality and social media use.Computers in Human
Behavior, 26(2), 247-253.
Ellison Nicole, B., & Charles, S. (2007). The benefits of Facebook» friends «: Exploring
the relationship between college students’ use of online social networks and social
capital. Journal of Computer-Mediated Communication,12(3).
Fox, J., & Rooney, M. C. (2015). The Dark Triad and trait self-objectification as predictors
of men’s use and self-presentation behaviors on social networking sites. Personality and
Individual Differences, 76, 161-165.
Gosling, S. D., Augustine, A. A., Vazire, S., Holtzman, N., & Gaddis, S. (2011).
Manifestations of personality in online social networks: Self-reported Facebook-related
behaviors and observable profile information.Cyberpsychology, Behavior, and Social
Networking, 14(9), 483-488.
Holbert, R. L., & Stephenson, M. T. (2002). Structural equation modeling in the
communication sciences, 1995–2000. Human Communication Research,28(4), 531-551.
Katz, J. E., & Crocker, E. T. (2015). Selfies| Selfies and Photo Messaging as Visual
Conversation: Reports from the United States, United Kingdom and China. International
Journal of Communication, 9, 12.
Kernberg O. (1975). Borderline conditions and pathological narcissism. New York: Jason
Aronson.
Kessler RC, Greenberg DF. (1981). Linear panel analysis: Models of quantitative change.
London: Academic Press.

Selfie-ists” or “Narci-selfiers 12
Ong, E. Y., Ang, R. P., Ho, J. C., Lim, J. C., Goh, D. H., Lee, C. S., & Chua, A. Y. (2011).
Narcissism, extraversion and adolescents’ self-presentation on Facebook. Personality and
Individual Differences, 50(2), 180-185.
Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personality: Narcissism,
Machiavellianism, and psychopathy. Journal of research in personality, 36(6), 556-563.
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(3), 281-303.
Sorokowski, P., Sorokowska, A., Oleszkiewicz, A., Frackowiak, T., Huk, A., & Pisanski, K.
(2015). Selfie posting behaviors are associated with narcissism among men. Personality
and Individual Differences, 85, 123-127.
Sundar SS. (2008). The MAIN model: A heuristic approach to understanding technology
effects on credibility. In: Metzger MJ, Flanagin AJ, eds. Digital Media, Youth, and
Credibility. Cambridge, MA: The MIT Press, pp.72-100.
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(3),
1102-1112.