Computers in Human Behavior 66 (2017) 138e148

Contents lists available at ScienceDirect

Computers in Human Behavior

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

Full length article

A study of the relationship between narcissism, extraversion, drive for

entertainment, and narcissistic behavior on social networking sites
Di Wang
C202, Macau University of Science and Technology, Avenida Wai Long, Taipa, Macau, China

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

Article history: As social networking sites are gaining popularity, the rise in narcissistic culture on theses sites are also
Received 1 November 2015 documented. Literature suggests users' personality traits may be important factors leading people to
Received in revised form engage in narcissistic behaviors. Using a national sample of Chinese adults, this study investigated the
22 March 2016
relationship between narcissism, extraversion, drive for entertainment and narcissistic behaviors on a
Accepted 21 September 2016
Chinese social networking site, Wechat Moment. We also examined whether demographic variables and
online time played a role in that dynamic. Results revealed that narcissism predicted exhibitionistic
behaviors on Wechat Moment, ratings of one's profile picture and using an image of oneself as profile
Keywords:
Social networking site
picture over and above extraversion and drive for entertainment controlling for online time and de-
Wechat moment mographic variables. Contrary to our hypothesis, narcissism did not predict higher frequency of Wechat
Narcissism Moment posting over and above drive for entertainment and extraversion. Different sub-factors of
Extraversion narcissism also contributed differently to the variance of narcissistic behavior on SNS.
Drive for entertainment © 2016 Elsevier Ltd. All rights reserved.
Narcissistic behavior

1. Introduction face-to-face basis? Most previous studies used narcissism as a sole

The widespread use of social networking sites (SNS) and the rise
of a narcissistic online culture have been documented in recent
years (Buffardi & Campbell, 2008; Hughes, Rowe, Batey, & Lee,
2012; Ong et al., 2011; Ryan & Xenos, 2011). Showing off has
replaced the original goals of some activities. For some the aim of
travel has become the taking and sharing of pictures on SNS, rather
than having a memorable experience or exploring. Instead of eating
food in a restaurant as soon as it is brought to us, some of us first
serve the food to our SNS audience. Taking pictures of ourselves,
then modifying those pictures and sharing them on SNS has
become a routine social networking activity for some people (Fox &
Rooney, 2015). Checking whether people have “liked” our posts or
commented on our posts are also routines for some. How should we
interpret narcissistic behavior on SNS?
An often-cited datum about narcissism is that although 30%e
40% of our face-to-face communication relates to our own experi-
ences (Landis & Burtt, 1924), around 80% of social media commu-
nication (such as Twitter) is self-related (Naaman, Boase, & Lai,
2010). Why is narcissistic behavior more frequent on SNS than in
face-to-face interactions? Are we more narcissistic on SNS than on a
E-mail address: dwang@must.edu.mo.
predictor of narcissistic behavior on SNS (Bergman, Fearrington,
Davenport, & Bergman, 2011; Buffardi & Campbell, 2008), but
narcissism only explained a small proportion of the variance in
narcissistic behavior (e.g., Ong et al., 2011; Panek, Nardis, &
Konrath, 2013). One explanation for this is that there are other
variables, besides narcissistic personality, which also contribute to
the variance in the narcissistic behavior on SNS. Narcissistic
behavior on SNS may be motivated by other reasons other than
narcissism. Extraversion has been found to predict narcissistic
behavior on Facebook (Ong et al., 2011). This study hypothesizes
that drive for entertainment also predicts narcissistic behavior on
SNS.
We investigated this hypothesis by examining narcissistic
behavior on one of China's largest social networking sites, Wechat
Moment. Wechat is an instant messaging application used on over
90% of smart phones used in China and it had 549 million active
users per month in the first quarter of 2015 (Curiosity China, 2015;
June, 8). Wechat Moment is a social networking site built on the
Wechat software. Despite its popularity, few studies have studied
narcissistic behavior on this Chinese social media. This study shall
fill in the gap and explore what predicts narcissistic behavior on
this Chinese SNS. Specifically, this study explores the relationship
between narcissism, extraversion, drive for entertainment and
narcissistic behavior on SNS.

http://dx.doi.org/10.1016/j.chb.2016.09.036
0747-5632/© 2016 Elsevier Ltd. All rights reserved.
 D. Wang / Computers in Human Behavior 66 (2017) 138e148
1.1. Narcissism and narcissistic behavior on SNS
Social networking sites by design are great venues for exhibi-
tionism (Bumgarner, 2007). Users of SNS have full control over the
information they disclose and can be strategic about their verbal
and pictorial self-presentation than in face-to-face communication
(Buffardi & Campbell, 2008; Kramer & Winter 2008). This makes
SNS the perfect platform for narcissists.
Narcissists are typically characterized as individuals with a
highly inflated, unrealistic, positive self-concept, who constantly
seek veneration from others and whose thoughts and behavior are
self-centered (Campbell, Rudich, & Sedikides, 2002; Morf &
Rhodewalt, 2001). As is generally the case for personality con-
structs, narcissism tends to occur along a continuum in the general
population.
Narcissism is also multi-dimensional in nature. Several studies
have demonstrated that narcissistic behavior and thought consist
of multiple discrete components (Ackerman et al., 2011; Emmons,
1984; Kansi, 2003). Raskin and Terry (1988) developed a 40-item
index called narcissism personality index (NPI) that measured
seven components of narcissism: authority, self-sufficiency, supe-
riority, exhibitionism, exploitativeness, vanity, and entitlement.
Other versions involve four components (Emmons, 1984; Kansi,
2003) or three components (Ackerman et al., 2011). For this study
we adopted Kansi (2003)’s four-component index because it was
developed using samples from a general population similar to that
from which our sample was drawn, whereas other indexes are
based on student samples. Kansi (2003)'s 29 -item scale included
four components: Leadership/Power (i.e., feeling like a leader),
Exhibitionism/Self-admiration (i.e., showing off), Superiority/
Arrogance (i.e., feeling better than others), and Uniqueness/Enti-
tlement (i.e., believing that one deserves the best).
Most previous studies of narcissism and SNS use treated
narcissism as a one-dimensional construct (Bergman et al., 2011;
Ong et al., 2011), although there have been some exceptions
(Carpenter, 2012; Panek et al., 2013). Panek et al. (2013) found that
posting on Twitter was correlated with the superiority component
of narcissism whereas posting on Facebook was correlated with the
exhibitionism component of narcissism among college students.
Using Ackerman et al. (2011)’s three-component conceptualization
of narcissism: leadership/authority, grandiose exhibitionism, and
entitlement/exploitativeness, Carpenter (2012) found the gran-
diose exhibitionism component strongly predicted self-promoting
behaviors including posting status updates and posting photos of
one's self on Facebook, while the entitlement/exploitativeness
component partially predicted anti-social behaviors, such as
responding to negative comments.
These findings suggest that the exhibitionism component of
narcissism is most closely related to the self-enhancing presenta-
tion on SNS. Narcissists are attention seeking. They are very con-
cerned about their physical appearances (Vazire, Naumann,
Rentfrow, & Gosling, 2008) and overestimate their physical
attractiveness (Bleske-Rechek, Remiker, & Baker, 2008; Gabriel,
Critelli, & Ee, 1994). It is not surprising that narcissism predicted
higher self-ratings of Facebook profile picture over and above ex-
traversion (Ong et al., 2011). Based on this finding, we predict that
narcissism predicts ratings of one's profile picture on Wechat
Moment as well.
Using an image of oneself as one's profile picture on SNS might
be an indication of inflated self-confidence. Narcissists love to look
at themselves in the mirror (Kansi, 2003) and using an image of
oneself as a profile picture serves the same function as looking at
oneself in the mirror; it may also be considered as a form of
showing off. Therefore we hypothesize that narcissism predicts
using an image of oneself as a profile picture.
139
Narcissists put a lot of effort into their clothes and makeup in
order to appear attractive (Vazire et al., 2008). This effort helps to
confirm them in their illusions of their own attractiveness (Bleske-
Rechek et al., 2008; Gabriel et al., 1994). It is easier to enhance one's
self-image on SNS than in real life. Modern picture-editing software
has made it easy to alter online images. The popularity of picture
editing can be gauged from the number of downloads of one of the
most popular photo-editing applications in China, Meitu. Meitu app
has been downloaded over 1.2 billion times on 750 million mobile
devices by 2015 (Huang, 2015; July, 22). It has been shown that in
the USA narcissism is positively related to the frequency of photo-
editing on SNS (Fox & Rooney, 2015) and we predict that the same
associations would be observed with respect to editing of pictures
in the Chinese social networking site, Wechat Moment.
The exhibitionism component of narcissism encompasses an
element of showing off one's life. Narcissists believe that everybody
is interested in their lives and likes hearing their stories (Raskin &
Terry, 1988). Sharing every bit of one's life, especially the positive
bits, is now common on SNS - having a fancy dinner, getting a new
haircut, going for a trip, etc. - almost anything to do with one's life is
worth sharing with friends on SNS. In many cases showing off has
replaced the original goals of the activity. We predict that narcis-
sism is positively related to uploading pictures of one's dinner and
travel experiences.
Taking selfies and displaying them on SNS is another popular
form of narcissism. The prevalence of smartphones and other small,
affordable photographic devices has enabled many people to take
pictures of themselves from one arm's length; these pictures are
called selfies, or ego shots (Lobinger & Brantner, 2015). Most studies
have found a positive relationship between narcissism and posting
selfies on SNS (Fox & Rooney, 2015; Mullen-Cooper, 2014; Weiser,
2015) with a few exceptions. Sorokowski et al. (2015) found that
whilst women posted more selfies on SNS than men, narcissism
was only associated with selfie-posting on SNS in men. Reed (2015)
found no relationship between narcissism and the posting of selfies
on Instagram. These discrepant findings notwithstanding we pre-
dict that narcissism would be positively related to selfie posting
frequency, based on the majority finding.
Narcissists need constant external self-affirmation as they are
unable to regulate their own self-esteem (Morf & Rhodewalt,
2001). Self-affirmation can easily be obtained on SNS by posting
and then waiting for the post to be liked or commented on. Ford's
2014 trend report stated that 62% of American adults said they felt
better about themselves when others reacted positively to their
posts on social media (Gara, 2013; Dec, 17). Based on this evidence
we predicted that narcissism is positively related to the frequency
with which individuals check the comments on their social media
posts.
In summary, we predict that narcissism is positively related to
exhibitionistic behavior on SNS, including selfies posting, upload-
ing pictures of one's meals and travel experiences, editing pictures
before uploading them to SNS and the checking comments on one's
posts. Because all these behaviors were measured in terms of fre-
quency, they can be combined into a single variable, exhibitionistic
behavior. Other narcissistic behaviors, including high rating of one's
profile picture and using an image of oneself as a profile picture,
were also predicted to be positively related to narcissism.
Frequency of SNS posting is another indicator of showing off.
Narcissism has been found to be positively related to Facebook
posting frequency (Ong et al., 2011; Panek et al., 2013) and based on
this we predicted that narcissism would be positively related to
Wechat Moment posting frequency.
Besides narcissism, extraversion and drive for entertainment
have also been found to be related to SNS use, which may explain
why narcissistic behavior is common on SNS. Below I shall review
140 D. Wang / Computers in Human Behavior 66 (2017) 138e148

these variables and their relationship with narcissistic behavior on
SNS.
1.2. Extraversion and narcissistic behavior on SNS
Extraversion refers to a preference to seek social interaction
(Costa & McCrae, 1992). There are two competing hypotheses about
the relationship between extraversion and use of computer-
mediated-communication (CMC). The “rich get richer” hypothesis
posits that extraverts gain more from CMC usage as they transfer
offline sociability to CMC platforms (Kraut et al., 2002); in contrast
the “social compensation” hypothesis posits that introverts gain
more from CMC usage as they are able to compensate for their
weaker social skills through use of CMC platforms (Valkenburg,
Schouten, & Peter, 2005).
Consistent with the “rich get richer” hypothesis, studies found
that extraverts were more likely to use social media than introverts
(Correa, Hinsley, & de Zuniga, 2010; Kim, Sin, & Tsai, 2014; Ryan &
Xenos, 2011), have more Facebook friends (Amichai-Hamburger &
Vinitzky, 2010; Zywica & Danowski, 2008) and belong to more
Facebook groups (Ross et al., 2009) than introverts do.
The “social compensation” hypothesis also received some sup-
port. Valkenburg and Peter (2007) found that although introverts
communicated online less often than extraverts, they rated the
Internet as more valuable for intimate self-disclosure than did
extroverts.
The findings suggest that although extraverts are more likely
than introverts to disclose information online, they do not perceive
the Internet as a better medium for self-disclosure than face-to-face
communication. Given that extraverts are more likely to disclose
information online than introverts, are they also more likely to
engage in narcissistic behavior on SNS? Research has shown that
the answer is “yes”. Ong et al. (2011) found that extraversion was
positively associated with rating of one's own Facebook profile
picture, number of Facebook friends, number of Facebook photos,
and frequency of Facebook status updates. These findings indicate
that extraversion may also be a predictor of narcissistic behavior on
SNS.
1.3. Drive for entertainment and SNS use
Entertainment has been found to be one of the most important
motives for use of SNS (Kim, Sohn, & Choi, 2011). Drive for enter-
tainment could also be seen as a trait that varies between in-
dividuals (Brock & Livingston, 2004). Some people are willing to
spend a lot of time and money on entertainment, while others
believe entertainment is a waste of time and money. Showing off or
exhibitionist behavior on SNS may reflect a high drive for enter-
tainment. Individuals with a high drive for entertainment might
seek to communicate with others to alleviate their boredom.
Sharing details of one's life with others is the easiest way of starting
a conversation on SNS, as other users can “like” or comment on
these posts and those comments can be commented. Thus drive for
H3. Narcissism will predict higher frequency of Wechat Moment
posting over and above drive for entertainment and extraversion.
H4. Narcissism will predict using an image of oneself as a profile
picture on Wechat Moment over and above drive for entertainment
and extraversion.
2. Method
2.1. Participants
A total of 810 participants aged 18 or above were referred by
undergraduate students from a university in Macau, China, to fill in
an online survey. The students who referred participants received
extra course credits. As the focus of this study was on people's
narcissistic behavior on Wechat Moment, only those who have used
Wechat Moment were remained in the final sample of 778 (502
females, 276 males). Based on IP addresses, participants were
Chinese readers from all over China and across the globe. Several
participants filled in the survey from America, Australia, Italy, U.K.,
Canada, Singapore and Japan. Participants ranged in age from 18 to
61 years old (M ¼ 28.88, SD ¼ 9.31). Participants' years of education
were 15.45 years on average (SD ¼ 16.00), which meant they were
mostly college graduates. They spent 2.19 h per day online with
their smartphone on average (SD ¼ 0.88). The descriptive statistics
were listed in Table 1.
2.2. Measures
2.2.1. Independent variables
2.2.1.1. Narcissism. The personality trait of narcissism was assessed
using the Narcissistic Personality Inventory e 29-item version
(NPIe29; Kansi, 2003). The NPI-29 is a forced-choice dichotomous
measure of narcissistic tendencies. Each item pairs a narcissistic
response with a non-narcissistic response (e.g., 1 ¼ ‘‘I always know
what I'm doing’’, and 0 ¼ “Sometimes I am not sure of what I am
doing’’) and asks respondents to rate the extent to which these
statements reflect themselves. It provides a total narcissism scale,
as well as scores for four sub-factors: leadership/power (8 items;
referred as “leadership” in the following for short), exhibitionism/
self-admiration (7 items; “exhibitionism” for short), superiority/
arrogance (6 items; “superiority” for short), and uniqueness/enti-
tlement (8 items; “uniqueness” for short). The NPI-29 was chosen
for the current study as it was developed using samples from a
general population similar to that from which our sample was
drawn (Kansi, 2003), whereas other versions of the NPI were
developed from student samples.
The obtained Cronbach's alpha reliability estimate for the
overall scale was at 0.85. Internal consistencies for four subscales
were: leadership (8 items: a ¼ 0.76), exhibitionism (7 items:
Table 1
Descriptive statistics.

entertainment may be associated with narcissistic behavior on SNS. N % M SD Mdn

Based on the literature reviewed above, we made the following Age 778 28.88 9.31 26.00
predictions: Gender 778
Female 502 64.5%
H1. Narcissism will predict higher exhibitionistic behavior (selfie Male 276 35.5%
posting, uploading pictures of one's meals and travel experiences, Years of Education 778 15.45 1.81 16.00
editing pictures before uploading them to SNS and checking com- Income (Thousands yuan per month) 778 5.64 2.9 4.25
ments on one's posts) over and above drive for entertainment and Marriage status 778
Married 282 36.2%
extraversion.
Single 483 62.1%
H2. Narcissism will predict higher ratings of one's Wechat profile Divorced 13 1.7%
Online time per day (hour) 778 2.19 0.88 2.50
picture over and above drive for entertainment and extraversion.
 D. Wang / Computers in Human Behavior 66 (2017) 138e148 141

a ¼ 0.61), superiority (6 items: a ¼ 0.57) and uniqueness (8 items:
a ¼ 0.61). Inter-factor correlations ranged from r ¼ 0.32 between
exhibitionism and leadership to r ¼ 0.65 between uniqueness and
leadership.
2.2.1.2. Extraversion. The personality trait of extraversion was
measured using the 12- item extraversion subscale of the NEO Five-
Factor Inventory (Costa & McCrae, 1992). Participants were asked to
rate their agreement to each of the 12 items on a five-point Likert
scale (1 ¼ ‘Strongly Disagree’ to 5 ¼ ‘Strongly Agree’). Higher scores
indicate higher levels of extraversion as a trait. Cronbach's alpha
reliability estimate was adequate at 0.76.
2.2.1.3. Drive for entertainment. Drive for entertainment was
measured using the 12- item Drive for entertainment subscale of
the need for entertainment scale (Brock & Livingston, 2004). Need
for entertainment scale consists of three factors, drive for enter-
tainment (12 items), entertainment utility (4 items) and enter-
tainment passivity (3 items). While trying to validate their scale,
Brock and Livingston (2004) found that not everyone believed
that entertainment is passive and usefulness of entertainment is
correlated with drive for entertainment. Thus only drive for
entertainment was used to measure people's entertainment need
in the current study. Participants were asked to rate each of the 12
items (e.g. “I do not spend much time during the week on enter-
taining activities.” “I spend most of my free time seeking out
entertainment.”) on a five-point Likert scale (1 ¼ ‘Extremely unlike
me’ to 5 ¼ ‘Extremely like me’). Higher scores indicate higher levels
of drive for entertainment. Cronbach's alpha reliability estimate
was high at 0.87.
2.2.2. Dependent variables
2.2.2.1. Exhibitionistic behavior. Bergman et al. (2011) suggested
that single-item criterion measures may attenuate the relationship
between predictor variables and narcissistic behavior variables on
SNS due to their focus on very specific behavior and they recom-
mended using composite criterion variable. Following this recom-
mendation, this study asked participants to indicate their
frequency of doing the following behaviors, “All of the pictures and
videos I posted at a time are selfies.” “Before having dinner, I will
take pictures of the food and post them on Wechat Moment.”
“When I'm traveling, I will take pictures and post on Wechat (both
posting while traveling and posting afterwards are counted).”
“Before I post pictures on Wechat, I will modify pictures that are not
ideal.” “After posting on Wechat Moment, I will check whether the
post has been commented or liked.” Frequencies were measured on
a 5-point scale, never (1), seldom (2), sometimes (3), often (4),
every time (5). An exhibitionistic act index was created by taking
the average of the answers to these five questions. Cronbach's alpha
reliability estimate was adequate at 0.77.
picture will make me look attractive to others.” “My Wechat profile
picture makes me look very cool.” Similarities ranged from
extremely unlike me (1) to extremely like me (5). An index of
profile picture attractiveness was created by taking the average of
the scores of the four questions.
The Inter-item reliability of the scales for extraversion, drive for
entertainment, narcissism, leadership, exhibitionism, superiority,
uniqueness, narcissistic behavior, profile attractiveness were listed
in Table 2.
2.2.3. Control variables
Previous studies have found that gender was related to narcis-
sism behavior on SNS (Bergman et al., 2011; Ong et al., 2011). Age
and education were also found to be associated with narcissism
behavior on SNS (Ong et al., 2011) and therefore they were
controlled in this study. Online time could also be related to
narcissistic behavior on SNS. Those who have more free time to
browse online are more likely to have time to engage in narcissistic
behavior than those who do not have time to browse online and it
was controlled in the analysis as well.
2.2.3.1. Demographic variables. The demographic variables
included gender (female ¼ 1, male ¼ 0), age (years) and education
(years of schooling).
2.2.3.2. Online time per day. Participants were asked to estimate on
average how much time they spent online with their smartphone
on a daily basis. Answers were recoded into a continuous variable,
online time per day (0 min ¼ 0, 30 min or less ¼ 0.5, 31 min to 1 h
(including 1 h) ¼ 0.75, 1e2 h (including 2 h) ¼ 1.5, 2e3 h (including
3 h) ¼ 2.5, above 3 h ¼ 3).
2.2.4. Descriptive variables
To understand participants' SNS activities in details, two other
variables were asked in the survey, checking frequency and the
content of their posts. Though they were not the focus of the cur-
rent study, they provided some background information for the
study.
2.2.4.1. Checking frequency. Participants were asked to report
whether or not they have used Wechat Moment. If they had, they
were asked how often they checked Wechat Moment. This variable
was later recoded as Wechat Moment checking times per day
(every half hour ¼ 24, every hour ¼ 12, every 2e3 h ¼ 4.8, every
4e5 h ¼ 3, every half day ¼ 2, once a day ¼ 1, once every few
days ¼ 0.25).
2.2.4.2. Content of posts. Participants were asked which kind of
message they posted most often on Wechat Moment,‘‘selfies’’,
‘‘pictures about friends or family’’, ‘‘original posts about my life’’,

2.2.2.2. Posting frequency. Participants were asked how often they

posted on Wechat Moment. Answers were recoded as posting Table 2
frequency per day (never ¼ 0, once a month ¼ 0.033, once a Inter-item reliability of the scales for extraversion, drive for entertainment, narcis-
sism, leadership, exhibitionism, superiority, uniqueness, exhibitionistic behavior
week ¼ 0.143, several times a week ¼ 0.5, once a day ¼ 1, two or
and profile attractiveness.
three times a day ¼ 2.5, four or more times a day ¼ 4).
N Item Crobach's alpha

2.2.2.3. Real person profile picture. Participants were asked Extraversion 778 12 0.76
whether their profile picture was the image of themselves (yes ¼ 1, Drive for entertainment 778 12 0.87
Narcissism 778 29 0.85
no ¼ 0).
Leadership 778 8 0.76
Exhibitionism 778 7 0.61
2.2.2.4. Profile picture attractiveness. Participants were asked to Superiority 778 6 0.57
indicate their similarity to the following four statements, “My Uniqueness 778 8 0.61
Wechat profile picture look attractive to others.” “I look very Exhibitionistic behavior 755 5 0.77
Profile attractiveness 308 4 0.84
fashionable in my Wechat profile picture.” “My Wechat profile
142 D. Wang / Computers in Human Behavior 66 (2017) 138e148

‘‘original posts about my work/study’’, ‘‘original posts about news’’, Table 4

‘‘original posts about entertainment (e.g. humor, travel)’’, ‘‘original T-test results of gender differences of predictor and criterion variables.

posts about life (e.g. health, food)’’ ‘‘forwarded posts about my life’’, Variables Women Men df t
‘‘forwarded posts about my work/study’’, ‘‘forwarded posts about M SD M SD
news’’, ‘‘forwarded posts about entertainment (e.g. humor, travel)’’,
Online time per day (hour) 2.29 0.84 1.99 0.91 530a 4.57***
‘‘forwarded posts about life (e.g. health, food)’’ and other (please
Posting per day 0.42 0.62 0.40 0.64 771b 0.32
indicate__). Answers were recoded into a dichotomous variable, Extraversion (scale 1 to 5) 3.28 0.47 3.36 0.50 776 2.23*
posting about oneself (‘‘selfies’’ ¼ 1, ‘‘original posts about my Entertainment (1e5) 2.79 0.65 2.87 0.69 776 1.53
life’’ ¼ 1, all other answers ¼ 0). Narcissism (0e1) 0.37 0.19 0.43 0.21 776 4.07***
Leadership 0.33 0.28 0.47 0.30 776 6.33***
Exhibitionism 0.50 0.23 0.46 0.27 508 2.33*
Superiority 0.32 0.25 0.37 0.27 523 2.76*
3. Results Uniqueness 0.33 0.22 0.41 0.25 776 4.83***
Exhibitionistic behavior (1e5) 2.99 0.78 2.63 0.81 753 6.06***
Close to 96% of the participants used Wechat Moment. Partici- Real profile picture 0.39 0.49 0.42 0.49 774 0.78
Profile picture attractiveness (1e5) 3.45 0.68 3.67 0.86 198c 2.36*
pants spent on average 2.19 h online using smartphone every day
*
(SD ¼ 0.88). On average, they checked Wechat Moment 9.31 times a p < 0.05, **p < 0.01, ***p < 0.001.
a
day (SD ¼ 8.20) and post 0.41 times a day on Wechat Moment For the variables “online time per day”, “exhibitionism”, and “superiority”,
Levene's Test for Equality of Variances were significant. Equal variances were not
(SD ¼ 0.63). This finding was very similar to Wechat's official assumed, and therefore what was reported was the adjusted df and t.
finding, which examined Wechat users across 31 provinces of China b
For the variable “posting per day,” “exhibitionistic behavior” and “real profile
and found that on average users check Wechat over 10 times a day picture,” the df was not 776, this was because of missing data.
c
in 2015 (http://tencent.com/Tencent.com, 2015). When asked what The question “profile picture attractiveness” was answered by those who used
their own pictures as their profile picture, n ¼ 308. Levene's Test for Equality of
contents were posted most frequently, 38.6% of the participants
Variances were significant. Equal variances were not assumed, and therefore what
chose posts about one's picture or one's life. was reported was the adjusted df and t.
The means, standard deviations and the inter-correlations of the
three predictor variables, four criterion variables and two control
variables were presented in Table 3. The three predictor variables exhibitionistic behavior, r (755) ¼ 0.23, p < 0.001 and using an
were moderately correlated, indicating an acceptable degree of image of oneself as a profile picture, r (776) ¼ 0.15, p < 0.001 and it
multicollinearity among predictors. was controlled in the analysis.
Gender was examined to see whether women and men differ The first three hypotheses were tested with three separate hi-
across key independent variables and dependent variables using t- erarchical regression analyses and the fourth hypothesis was tested
tests. As can be seen in Table 4, women (M ¼ 2.99, SD ¼ 0. 78) with a logistic regression. To investigate whether narcissism pre-
behaved more exhibitionistically on SNS than men (M ¼ 2.63, SD ¼ . dicted each of the three dependent variables (exhibitionistic
81), t (753) ¼ 6.06, p < 0.001; they spent more time browsing behavior, ratings of one's profile picture and posting frequency)
online with their smartphone than men (women: M ¼ 2.29, over and above drive for entertainment and extraversion while
SD ¼ 0.84; men: M ¼ 1.99, SD ¼ 0.91), t (530) ¼ 4.57, p < 0.001; controlling for online time and demographic variables, all three
scored lower on extraversion than men (women: M ¼ 3.28, hierarchical regression analyses were conducted by entering de-
SD ¼ 0.47; men: M ¼ 3.36, SD ¼ 0.50), t (776) ¼ 2.23, p < 0.05; mographic variables in the first step, online time in the second step,
scored lower on narcissism than men (women: M ¼ 0.37, SD ¼ 0.19; drive for entertainment in the third step, extraversion in the fourth
men: M ¼ 0.43, SD ¼ 0.21), t (776) ¼ 4.07, p < 0.001 and their step, followed by narcissism in the fifth step. Similarly, to examine
rating of their profile picture was lower than men's rating (women: whether narcissism predicted using an image of oneself as a profile
M ¼ 3.45, SD ¼ 0.68; men: M ¼ 3.67, SD ¼ 0.86), t (198) ¼ 2.36, picture over and above drive for entertainment and extraversion
p < 0.05. As gender indeed affect many of the predictor and crite- while controlling for online time and demographic variables, a hi-
rion variables, it was controlled in the following analysis. erarchical binary logistic regression was run by entering de-
Age was found to be significantly correlated with drive for mographic variables in the first step, online time in the second step,
entertainment, r (778) ¼ 0.28, p < 0.001, exhibitionistic behavior, drive for entertainment in the third step, extraversion in the fourth
r (755) ¼ 0.16, p < 0.001, profile picture attractiveness, r step, followed by narcissism in the fifth step.
(308) ¼ 0.23, p < 0.001 and using an image of oneself as a profile To test how the four narcissism sub-factors predict narcissistic
picture, r (776) ¼ 0.08, p < 0.05 and it was controlled in the behavior on SNS, another three hierarchical regressions and
analysis. another logistic regression were run with the narcissism variable
In addition to gender and age, online time was found to be replaced with four narcissism sub-factors while keeping all the
significantly correlated with posting frequency, r (773) ¼ 0.13, other variables in the regressions the same.
p < 0.001, drive for entertainment, r (778) ¼ 0.19, p < 0.001, H1 hypothesized that narcissism predicts exhibitionistic

Table 3
Descriptive statistics and correlations.

1 2 3 4 5 6 7 8 M SD
** ** ** **
1. Online time per day (hour) _ 0.13 0.03 0.19 0.01 0.25 0.15 0.05 2.19 0.88
2. Posting frequency per day _ 0.10** 0.07 0.09* 0.26** 0.07* 0.17** 0.41 0.63
3.Extraversion (scale 1 to 5) _ 0.27** 0.28** 0.22** 0.12** 0.27** 3.31 0.48
4.Entertainment (1e5) _ 0.26** 0.26** 0.12** 0.16** 2.82 0.67
5. Narcissism (0e1) _ 0.16** 0.11* 0.33* 0.39 0.20
6. Exhibitionistic behavior (1e5) _ 0.24** 0.24** 2.86 0.81
7. Real profile picture _ _ 0.40 0.49
8. Profile picture attractiveness (1e5) _ 3.53 0.75
* ** ***
p < 0.05, p < 0.01, p < 0.001.
 D. Wang / Computers in Human Behavior 66 (2017) 138e148 143

behavior over and above drive for entertainment and extraversion. Table 5.2
Age and education were not significantly related to the criterion Hierarchical multiple regression analyses for gender, online time per day, drive for
entertainment, extraversion, exhibitionism, superiority, leadership and uniqueness
variable in the full regression model and accordingly they were not predicting exhibitionistic behavior on Wechat Moment.
included in the analysis. Consistent with past research, gender was
significantly related to exhibitionistic behavior (Bergman et al., Predictor and step b R2 D R2 DF
2011; Ong et al., 2011) and thus it was controlled in the multiple Step 1
regression predicting exhibitionistic behavior. Results were pre- Female (1 ¼ yes, 0 ¼ no) 0.22*** 0.05 0.05 36.77***
Step 2
sented in Table 5.1, after controlling for gender and online time,
Female (1 ¼ yes, 0 ¼ no) 0.18***
narcissism significantly predicted exhibitionistic behavior over and Online time per day 0.22*** 0.09 0.05 37.43***
above drive for entertainment and extraversion (DR2 ¼ 0.01, DF (1, Step 3
749) ¼ 7.58, p < 0.01, b ¼ 0.10). H1 was supported. Female (1 ¼ yes, 0 ¼ no) 0.20***
Online time per day 0.17***
Weighing the relative impact of the predictor variables, we can
Entertainment 0.24*** 0.15 0.05 46.67***
refer to the R2 change of each multiple regression. For the indicators Step 4
predicting exhibitionistic behavior on Wechat Moment, the largest Female (1 ¼ yes, 0 ¼ no) 0.21***
R2 change was caused by drive for entertainment (DR2 ¼ 0.05), Online time per day 0.17***
online time (DR2 ¼ 0.05) and gender (DR2 ¼ 0.05), followed by Entertainment 0.19***
Extraversion 0.17*** 0.17 0.03 24.68***
extraversion (DR2 ¼ 0.03). The smallest R2 change was caused by
Step 5
narcissism (DR2 ¼ 0.01). Female (1 ¼ yes, 0 ¼ no) 0.19***
Table 5.2 showed that only the exhibitionism component of Online time per day 0.16***
narcissism contributed to the prediction of exhibitionist behaviors Entertainment 0.15***
on SNS (b ¼ 0.22, p < 0.001) while other components were not Extraversion 0.15***
Exhibitionism 0.22***
significantly related to exhibitionistic behaviors on SNS. The R2 Superiority 0.05
change was larger when we use the four sub-factors of narcissism Leadership 0.01
to predict exhibitionistic behavior (DR2 ¼ 0.04) than when we use Uniqueness 0.01 0.21 0.04 9.57***
the overall narcissism scale to predict exhibitionistic behavior N 755
(DR2 ¼ 0.01). p < 0.05, **
p < 0.01, ***
p < 0.001.
H2 hypothesized that narcissism predicts higher ratings of one's
Wechat profile picture over and above drive for entertainment and
extraversion. Age was significantly related to profile picture ratings Table 6.2 showed that both the superiority component (b ¼ 0.15,
and was controlled in the regression. Gender and online time were p < 0.05) and the uniqueness component (b ¼ 0.18, p < 0.05) of
not significantly related to profile picture ratings and were narcissism contributed to the prediction of rating of one's Wechat
excluded from the analysis. Results were shown in Table 6.1, after profile picture while other components were not significantly
controlling for age, narcissism significantly predicted higher ratings related to rating of one's Wechat profile picture. R2 change was
of one's Wechat profile picture over and above drive for enter- larger when we use the four sub-factors of narcissism to predict
tainment and extraversion (DR2 ¼ 0.07, DF (1, 303) ¼ 25.98, ratings of one's Wechat profile picture (DR2 ¼ 0.08) than using
p < 0.001, b ¼ 0.28). H2 was supported. overall narcissism scale to predict ratings of one's Wechat profile
Comparing the indicators predicting ratings of one's Wechat picture (DR2 ¼ 0.07).
profile picture, the largest R2 change was caused by narcissism H3 hypothesized that narcissism predicts higher frequency of
(DR2 ¼ 0.07), followed by drive for entertainment (DR2 ¼ 0.05), Wechat Moment posting over and above drive for entertainment
extraversion (DR2 ¼ 0.05) and age (DR2 ¼ 0.05). and extraversion. Age and online time were significantly associated
with Wechat Moment posting frequency and they were controlled
in the analysis. Results were shown in Table 7.1, after controlling for
Table 5.1
Hierarchical multiple regression analyses for gender, online time per day, drive for
age and online time, narcissism did not significantly predict the
entertainment, extraversion and narcissism predicting exhibitionistic behavior on frequency of Wechat Moment posting over and above drive for
Wechat Moment.

Predictor and step b R2 D R2 DF

Step 1 Table 6.1
Female (1 ¼ yes, 0 ¼ no) 0.22*** 0.05 0.05 36.77*** Hierarchical multiple regression analyses for online time per day, drive for enter-
Step 2 tainment, extraversion and narcissism predicting ratings of one's Wechat profile
Female (1 ¼ yes, 0 ¼ no) 0.18*** picture.
Online time per day 0.22*** 0.09 0.05 37.43***
Predictor and step b R2 D R2 DF
Step 3
***
Female (1 ¼ yes, 0 ¼ no) 0.20 Step 1
Online time per day 0.17*** Age 0.23*** 0.05 0.05 17.11***
Entertainment 0.24*** 0.15 0.05 46.67***
Step 2
Step 4 Age 0.28***
Female (1 ¼ yes, 0 ¼ no) 0.21*** Drive for entertainment 0.22*** 0.10 0.05 15.95***
Online time per day 0.17*** Step 3
Entertainment 0.19*** Age 0.27***
Extraversion 0.17*** 0.17 0.03 24.68*** Drive for entertainment 0.16**
Step 5 Extraversion 0.23*** 0.15 0.05 17.74***
Female (1 ¼ yes, 0 ¼ no) 0.22*** Step 4
Online time per day 0.17*** Age 0.27***
Entertainment 0.17*** Drive for entertainment 0.12*
Extraversion 0.15*** Extraversion 0.16**
Narcissism 0.10** 0.18 0.01 7.58** Narcissism 0.28*** 0.22 0.07 25.98***
N 755 N 308
* ** *** * ** ***
p < 0.05, p < 0.01, p < 0.001. p < 0.05, p < 0.01, p < 0.001.
144 D. Wang / Computers in Human Behavior 66 (2017) 138e148

Table 6.2 Table 7.2

Hierarchical multiple regression analyses for online time per day, drive for enter- Hierarchical multiple regression analyses for online time per day, drive for enter-
tainment, extraversion, exhibitionism, superiority, leadership and uniqueness pre- tainment, extraversion, exhibitionism, superiority, leadership and uniqueness pre-
dicting ratings of one's Wechat profile picture. dicting frequency of posting on Wechat Moment.

Predictor and step b R2 D R2 DF Predictor and step b R2 D R2 DF

Step 1 Step 1
Age 0.23*** 0.05 0.05 17.11*** Age 0.11**
Step 2 Online time per day 0.16*** 0.03 0.03 10.49***
Age 0.28*** Step 2
Drive for entertainment 0.22*** 0.10 0.05 15.95*** Age 0.13***
Step 3 Online time per day 0.15***
Age 0.27*** Drive for entertainment 0.08* 0.03 0.01 4.16*
Drive for entertainment 0.16** Step 3
Extraversion 0.23*** 0.15 0.05 17.74*** Age 0.12**
Step 4 Online time per day 0.15***
Age 0.28*** Drive for entertainment 0.05
Drive for entertainment 0.10 Extraversion 0.08* 0.04 0.01 4.29*
Extraversion 0.16** Step 4
Exhibitionism 0.09 Age 0.13***
Superiority 0.15* Online time per day 0.15***
Leadership 0.04 Drive for entertainment 0.04
Uniqueness 0.18* 0.23 0.08 7.6*** Extraversion 0.07
N 308 Exhibitionism 0.02
* ** *** Superiority 0.03
p < 0.05, p < 0.01, p < 0.001.
Leadership 0.03
Uniqueness 0.07 0.04 0.00 1.04
N 773
entertainment and extraversion (DR2 ¼ 0.00, DF (1, 767) ¼ 2.82, * ** ***
p < 0.05, p < 0.01, p < 0.001.
p > 0.05, b ¼ 0.06). H3 was rejected. Age was significantly associ-
ated with Wechat Moment posting frequency (b ¼ 0.11, p < 0.001).
Online time per day significantly predicted Wechat Moment post- oneself as a profile picture with online time, drive for entertain-
ing frequency (b ¼ 0.16, p < 0.001). The largest R2 change was ment, extraversion and narcissism as predictors. Age and gender
caused by age and online time (DR2 ¼ 0.03), followed by drive for were not significantly related to using an image of oneself as a
entertainment (DR2 ¼ 0.01) and extraversion (DR2 ¼ 0.01). profile picture and were excluded from the analysis. Results were
Table 7.2 showed that none of components of narcissism were shown in Table 8.1.
significantly related to rating of one's Wechat profile picture. As A test of the full model against a constant only model was sta-
neither overall narcissism scale nor the sub-factors of narcissism tistically significant, indicating that the predictors as a set reliably
was significantly related to rating of one's Wechat profile picture, R2 distinguished between using an image of oneself as a profile picture
change was the same when we use the four sub-factors of narcis- or otherwise (Chi square ¼ 36.02, p < 0.001, df ¼ 4). The difference
sism (DR2 ¼ 0.00) as when we use the overall narcissism scale to between the 2 log likelihood at each step was significant (from
predict frequency of posting on Wechat Moment (DR2 ¼ 0.00). step 1 to step 2: Chi-square ¼ 6.72, p < 0.01, from step 2 to step 3:
H4 hypothesized that narcissism predicts using an image of Chi-square ¼ 7.03, p < 0.01, from step 3 to step 4: Chi-square ¼ 4.16,
oneself as a profile picture on Wechat over and above drive for p < 0.05).
entertainment and extraversion. A hierarchical binary logistic Nagelkerke's R2 of 0.06 indicated a weak relationship between
regression analysis was conducted to predict using an image of predictors and criterion variable. Prediction success overall was
62.6% (87.0% for not using an image of oneself and 26.5% for using
an image of oneself as a profile picture). The Wald criterion
Table 7.1 demonstrated that online time (B ¼ 0.34, S.E. ¼ 0.09, p < 0.001),
Hierarchical multiple regression analyses for online time per day, drive for enter- extraversion (B ¼ 0.35, S.E. ¼ 0.17, p < 0.05) and narcissism
tainment, extraversion and narcissism predicting frequency of posting on Wechat (B ¼ 0.81, S.E. ¼ 0.40, p < 0.05) made a significant contribution to
Moment.
prediction. Drive for entertainment was not a significant predictor.
Predictor and step b R2 D R2 DF Exp (B) values indicate that for every one-unit increase in online
Step 1 time per (one hour), one is 1.40 times more likely to use an image of
Age 0.11** oneself as a profile picture; for every one-unit increase in extra-
Online time per day 0.16*** 0.03 0.03 10.49*** version, one is 1.42 times more likely to use an image of oneself as a
Step 2
profile picture; for every one-unit increase in narcissism, one is 2.25
Age 0.13***
Online time per day 0.15*** times more likely to use an image of oneself as a profile picture.
Drive for entertainment 0.08* 0.03 0.01 4.16* In summary, after controlling for online time, drive for enter-
Step 3 tainment and extraversion, narcissism significantly predicted using
Age 0.12**
an image of oneself as a profile picture. H4 was supported.
Online time per day 0.15***
Drive for entertainment 0.05
The relationship between narcissism and using an image of
Extraversion 0.08* 0.04 0.01 4.29* oneself as a profile picture seemed to have disappeared when we
Step 4 use the four sub-factors of narcissism to replace the overall
Age 0.12*** narcissism scale. Table 8.2 showed that none of the sub-factors of
Online time per day 0.16***
narcissism significantly predicted using an image of oneself as a
Drive for entertainment 0.04
Extraversion 0.06 profile picture. The pseudo R2 significantly increased by 0.01, Chi
Narcissism 0.06 0.04 0.00 2.82 square ¼ 36.02, p < 0.001, which showed that the overall narcissism
N 773 scale still significantly predicted using an image of oneself as a
*
p < 0.05, **
p < 0.01, ***
p < 0.001. profile picture. The pseudo R2 change was larger when we use the
 D. Wang / Computers in Human Behavior 66 (2017) 138e148 145

Table 8.1
Hierarchical binary logistic regression analyses for online time per day, drive for entertainment, extraversion and narcissism predicting using an image of oneself as a profile
picture.

Predictor and steps B S.E. Exp(B) Model chi square df Pseudo R2 (Nagelkerke's R2)

Step 1
Online time per day 0.36*** 0.09 1.44*** 18.12*** 1 0.03
Step 2
***
Online time per day 0.32 0.09 1.38
Entertainment 0.09* 0.03 6.72** 24.84*** 2 0.04
Step 3
Online time per day 0.33*** 0.09 1.39
Entertainment 0.22 0.12 1.24
Extraversion 0.43** 0.16 1.54 31.87*** 3 0.05
Step 4
***
Online time per day 0.339 0.089 1.404
Entertainment 0.166 0.122 1.181
Extraversion 0.350* 0.168 1.419
Narcissism 0.813* 0.399 2.254 36.02*** 4 0.06
N 776

Note: The dependent variable in the analysis is using one's picture as a profile picture coded so that 0 ¼ not using an image of oneself as a profile picture and 1 ¼ using an image
of oneself as a profile picture.
*
p < 0.05, **p < 0.01, ***p < 0.001.

four sub-factors of narcissism to predict using an image of oneself
as a profile picture (DR2 ¼ 0.07) than using overall narcissism scale
(DR2 ¼ 0.06).
4. Discussion
This study aimed to examine the predictors of narcissistic
behavior on SNS. Besides trait narcissism, the study also found that
drive for entertainment, extraversion, online time and some de-
mographic variables all significantly predicted narcissistic behavior
on SNS. This finding can broaden our understanding of the
narcissistic culture on SNS. In contradiction to the conventional
wisdom that only narcissists behave narcissistically on SNS we
found that narcissistic behavior on SNS was predicted by other
traits and habits in addition to narcissism.
First, narcissistic behavior was predicted by trait narcissism.
Individuals who were high in narcissism were more likely to
engage in exhibitionistic behavior on SNS (H1), were more likely
rate their own profile picture as attractive (H2) and more likely to
use an image of themselves as a profile picture (H4) than those who
were low on narcissism. These results were consistent with previ-
ous studies showing that narcissism was positively associated with
rating of one's Facebook profile picture (Ong et al., 2011).
H3 was the only hypothesis that was rejected. After accounting
for the effects of online time per day, drive for entertainment and
extraversion, narcissism did not predict Wechat Moment posting
frequency. This result conflicts with previous studies that found
that narcissism predicted the frequency of Facebook status updates
(Ong et al., 2011; Panek et al., 2013). The difference between this
study and previous research is that this study investigated Wechat
Moment posting, whereas the previous research was related to
Facebook or Twitter status updates. Naaman et al. (2010) found that
around 80% of Twitter posting was self-related. However, our data
showed that only 38.6% of users chose self-rated items (posting
picture of oneself, posting about one's life) as the most frequently
posted Wechat Moment posts. The rest were posts about other is-
sues, such as news, entertainment, life, etc. As the majority posts on
Wechat Moment were not self-related, narcissism did not predict
the frequency of Wechat Moment updates.
Second, narcissistic behavior was predicted by extraversion.

Table 8.2
Hierarchical binary logistic regression analyses for online time per day, drive for entertainment, extraversion, exhibitionism, superiority, leadership and uniqueness predicting
using an image of oneself as a profile picture.

Predictor and steps B S.E. Exp(B) Model chi square df Pseudo R2 (Nagelkerke's R2)

Step 1
Online time per day 0.36*** 0.09 1.44*** 18.12*** 1 0.03
Step 2
Online time per day 0.32*** 0.09 1.38
Entertainment 0.09* 0.03 6.72** 24.84*** 2 0.04
Step 3
***
Online time per day 0.33 0.09 1.39
Entertainment 0.22 0.12 1.24
Extraversion 0.43** 0.16 1.54 31.87*** 3 0.05
Step 4
Online time per day 0.33*** 0.09 1.39
Entertainment 0.15 0.12 1.49
Extraversion 0.36* 0.17 4.56
Exhibitionism 0.52 0.34 2.39
Superiority 0.49 0.38 1.70
Leadership 0.16 0.38 0.18
Uniqueness 0.15 0.45 0.11 38.46*** 7 0.07
N 776

Note: The dependent variable in the analysis is using an image of oneself as a profile picture coded so that 0 ¼ not using an image of oneself as a profile picture and 1 ¼ using an
image of oneself as a profile picture.
*
p < 0.05, **p < 0.01, ***p < 0.001.
146 D. Wang / Computers in Human Behavior 66 (2017) 138e148
Individuals who were high on extraversion were more likely to
engage in exhibitionistic behavior on SNS (H1), more likely to rate
their own profile picture as attractive (H2), posted more frequently
on Wechat Moment (H3) and were more likely to use an image of
themselves as a profile picture (H4) than those who were low on
extraversion. This finding was consistent with Ong et al. (2011)
study, which found that extraversion was positively associated
with ratings of Facebook profile picture and the frequency of
Facebook status updates. Extraverts treat SNS as another social
interaction medium and post more frequently on SNS than in-
troverts. These findings support the “rich get richer” hypothesis.
Extraversion was positively associated with narcissistic behavior,
presumably because sharing or disclosing a picture of oneself or
information about one's life is a good way to start a conversation.
Comments on posts about one's picture or one's life lead to more
social interaction such as replying to one's comments. Extraversion
was also positively associated with narcissism, which may explain
why extraverts were more likely to rate their profile picture as more
attractive than introverts.
Third, narcissistic behavior was predicted by drive for enter-
tainment. This study is the first to use the drive for entertainment
variable in research on narcissistic behavior in SNS and hence these
results represent an important contribution to the field. Individuals
with a high drive for entertainment were more likely to engage in
narcissistic behavior on SNS (H1), more likely to rate their own
profile picture as attractive (H2) and more likely to use an image of
themselves as a profile picture (H4) than those with a low drive for
entertainment. Drive for entertainment was found to contribute
similarly, or in some cases more strongly, to variance in narcissistic
behavior than trait narcissism. In fact, much of what appeared to be
narcissistic behavior seems to have been motivated by a drive for
entertainment.
Fourth, narcissistic behavior on SNS was predicted by online
time. The probability of behaving narcissistically on SNS increased
with daily time online. This is the first study to investigate time
online as a predictor of narcissistic behavior on SNS. Previous
studies have not controlled for the effects of this variable and may
thus have confounded the relationship between trait narcissism
and narcissistic behavior (e.g. Bergman et al., 2011; Ong et al., 2011).
For instance, a person who is very narcissistic may be too busy to
spend time online and hence may be less likely to behave narcis-
sistically online. On the other hand, a person who is not very
narcissistic may have a lot of time to spend online and may have
more time to behave narcissistically online than those who do not
spend much time online.
We found that people who spent a lot of time online via a
smartphone were more likely to engage in exhibitionistic behavior
on SNS (H1), posted more frequently on SNS (H3) and were more
likely to use pictures of themselves as a profile picture (H4) than
those who spent rather little time online. Almost all narcissistic
behavior on SNS takes time and people who spend a lot of time
online necessarily have more opportunity to edit pictures (H1),
check other people's comments (H1), post pictures or status up-
dates (H3), or upload profile picture (H4) than people who spend
rather little time online. Free time did not predict rating of one's
own profile picture (H2), which is unsurprising as evaluation is not
time-consuming; personality variables are sufficient to account for
variance in rating of one's profile picture.
This study also contributes to the literature because it provides
evidence on the relationship between certain demographic vari-
ables and narcissistic behavior. We found that women were more
likely than men to engage in narcissistic behavior on Wechat
Moment. This finding is consistent with a study that found that
female Chinese university students were more likely to upload
selfies and update their status on SNS than their male counterparts
(Wang, Jackson, Zhang, & Su, 2012). This finding is interesting as the
women in this study scored lower on narcissism than men. Detailed
inspection of the scores revealed that although men scored higher
than women on overall narcissism, they scored lower than women
on the exhibitionism component; it is therefore not surprising that
women were more likely to behave exhibitionistically.
Older adults were more likely to rate their Wechat profile pic-
ture as attractive and posted more frequently on Wechat Moment
than younger adults. This finding was not consistent with Ong et al.
(2011) study in which age was not related to ratings of one's profile
picture in their full regression model. The reason for this discrep-
ancy may be that Ong et al. (2011) used an adolescent Singaporean
sample, which was rather homogeneous with respect to age. Future
research should be carried out to resolve this discrepancy.
We turn now to the relative impact of the three predictor vari-
ables (drive for entertainment, extraversion and narcissism) on the
three criterion variables (exhibitionistic behavior, ratings of one's
Wechat profile picture and using an image of oneself as a profile
picture1). All three predictor variables were significantly related to
all three criterion variables, the differences were in the relative
contribution of the predictor variables to variance in each criterion
variable. Drive for entertainment was the strongest predictor of
exhibitionistic behavior on Wechat Moment and narcissism was
the strongest predictor of ratings of one's Wechat profile picture.
Drive for entertainment, extraversion and narcissism contributed
similarly to variance in use of an image of oneself as a profile pic-
ture. As there was no consistent pattern of relative contribution to
the three criterion variables, we conclude that all the predictor
variables predict narcissistic behavior and the strength of their
contribution to variance in narcissistic behavior depends on which
aspect of narcissism is measured.
This study has demonstrated that the different types of narcis-
sistic behavior on SNS are motivated by different sub-factors of
narcissism. For instance, only the exhibitionism component of
narcissism contributed to variance in exhibitionistic behavior on
SNS, whereas both the superiority component and the uniqueness
component contributed to variance in ratings of one's Wechat
profile picture. As posting selfies, sharing details of one's travel
experiences and meals, editing pictures and checking comments on
one's posts are ways of showing off, it is unsurprising that exhibi-
tionism was the only narcissism component to predict these be-
haviors. Individuals who rated their Wechat profile picture highly
also believed that they were better than other people (superiority)
and that they deserved the best (uniqueness/entitlement).
We found that narcissism sub-factors were better predictors of
the narcissistic behavior on SNS than overall narcissism scale.
Future research should take advantage of the sub-factors of
narcissism in the study of narcissism online rather than treating
narcissism as a single variable.
Although we found many significant effects, we have to
acknowledge that the R2 values and R2 change in all of the regres-
sion analyses were quite small. For instance, gender, online time
per day, drive for entertainment, extraversion and sub-factors of
narcissism explained only 21% of the variance in exhibitionistic
behavior. Only 4% of the total variance was contributed by narcis-
sism sub-factors. Age, drive for entertainment, extraversion and
sub-factors of narcissism explained 23% of the variance in ratings of
one's Wechat profile picture. Narcissism sub-factors contributed 8%
of the total variance. In the logistic regression predicting using an
image of oneself as a profile picture, the pseudo R2 value was only
0.06. Although we cannot interpret the pseudo R2 like the R2 values
1
Posting frequency was not discussed here as this hypothesis was not supported
by the data.
 D. Wang / Computers in Human Behavior 66 (2017) 138e148
in linear regression, it was still rather small. All these findings
showed that personality, demographic variables and online time
were not the major determinant of one's narcissistic behavior on
SNS. The conventional wisdom is that only narcissists behave nar-
cissistically; however our data suggest that narcissistic behavior
may be motivated by drive for entertainment or other motivations.
Extraversion also explained a proportion of variance in narcissistic
behavior.
This study extended research on narcissism on SNS beyond
Facebook and Twitter to an Asian SNS, Wechat Moment. China, the
country with the most Internet users (649 million in 2014, McKirdy,
2015), was understudied in terms of SNS use in the English litera-
ture. As Facebook and Twitter are blocked in the mainland China,
one has to study a popular Chinese SNS to understand Chinese
netizens' behavior on SNS. This paper represents an important
contribution to knowledge of SNS use by Chinese netizens and
provides a basis for cross-country comparisons.
Although this study provides evidence on the relationship
among personality variables, demographic variables, time online
and narcissistic behavior on SNS it has several limitations that
should be addressed in future studies:
First, as stated above, although we found significant relation-
ships, the value of R2 was still rather small when all three predictor
variables and two control variables (age and gender) had been
entered. In other words, these variables account for only a small
proportion of the variance in narcissistic behavior on SNS. Future
research could explore other predictors of narcissistic behavior on
SNS, such as privacy concerns, need for social support, need for
communication and attitudes towards the Internet.
Second, although we used an adult sample in which age was
normally distributed, our sample was not gender balanced (64.5%
women; 35.5% men) and this may have biased our results as many
of the criterion variables were associated with gender. Future
research should be based on a more gender-balanced sample.
Third, the self-report data on which this study relies is limited
by the reliability of the respondents' recollections. Furthermore, in
the Chinese culture, narcissism is socially undesirable and so it is
likely that narcissistic behaviors underreported. Future research
could address this problem by asking participants to log on to their
SNS account in the lab and report their behaviors on SNS objec-
tively based on observation of their own account activities in the
past.
Finally, we cannot make causal implications based on a cross-
sectional survey. Future research could use longitudinal studies to
study the relationships.
Despite these limitations, we believe this study contributes to
understanding of narcissistic behavior on SNS among Chinese
adults. In addition to narcissism, extraversion, drive for entertain-
ment, online time, age and gender are all predictors of narcissistic
behavior on SNS.
Disclosure statement
No competing financial interests exist.
References
Ackerman, R. A., Witt, E. A., Donnellan, M. B., Trzesniewski, K. H., Robins, R. W., &
Kashy, D. A. (2011). What does the Narcissistic Personality Inventory really
measure? Assessment, 18, 67e87.
Amichai-Hamburger, Y., & Vinitzky, G. (2010). Social network use and personality.
Computers in Human Behavior, 26, 1289e1295.
Bergman, S. M., Fearrington, M. E., Davenport, S. W., & Bergman, J. Z. (2011). Mil-
lennials, narcissism, and social networking: What narcissists do on social
networking sites and why. Personality and Individual Differences, 50, 706e711.
Bleske-Rechek, A., Remiker, M. W., & Baker, J. P. (2008). Narcissistic men and women
think they are so hot e But they are not. Personality and Individual Differences,
147
45, 420e424.
Brock, T. C., & Livingston, S. D. (2004). The drive for entertainment scale. In
L. J. Shrum (Ed.), The psychology of entertainment media: Blurring the lines be-
tween entertainment and Persuasion (pp. 255e274). Mahwah, NJ: Erlbaum.
Buffardi, L. E., & Campbell, W. K. (2008). Narcissism and social networking web sites.
Personality and Social Psychology Bulletin, 34, 1303e1314. http://dx.doi.org/
10.1177/0146167208320061.
Bumgarner, B. A. (2007). You have been poked: Exploring the uses and gratifications of
Facebook among emerging adults. First Monday, 12. Retrieved September 24,
2015, from http://www.firstmonday.dk/ojs/index.php/fm/article/view/2026/
1897#author.
Campbell, W. K., Rudich, E. A., & Sedikides, C. (2002). Narcissism, self-esteem, and
the positivity of self-views: Two portraits of self-love. Personality and Social
Psychology Bulletin, 28, 358e368.
Carpenter, C. (2012). Narcissism on facebook: Self-promotional and anti-social
behavior. Personality and Individual Differences, 52, 482e486. http://dx.doi.org/
10.1016/j.paid.2011.11.011.
Correa, T., Hinsley, A. W., & de Zuniga, H. G. (2010). Who interbehavior on the web?:
the intersection of users' personality and social media use. Computers in Human
Behavior, 26, 247e253.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory and NEO five-
factor inventory. Odessa, FL: Psychological Assessment Resources.
Curiosity China. (2015, June, 8). WeChat user global report 2015. Retrieved
September 24, 2015, from https://www.linkedin.com/pulse/wechat-user-
global-report-2015-alexis-bonhomme.
Emmons, R. A. (1984). Factor analysis and construct validity of the Narcissistic.
Personality Inventory. Journal of Personality Assessment, 48, 291e300.
Fox, J., & Rooney, M. C. (2015). The Dark Triad and trait self-objectification as pre-
dictors of men's use and self-presentation behaviors on social networking sites.
Personality & Individual Differences, 76, 161e165. http://dx.doi.org/10.1016/
j.paid.2014.12.017.
Gabriel, M. T., Critelli, J. W., & Ee, J. S. (1994). Narcissistic illusions in self-evaluations
of intelligence and attractiveness. Journal of Personality, 62, 143e155.
Gara, T. (2013, Dec, 17). Why our online persona is needier than our real one. Wall
Street Journal. Retrieved Feb., 2, 2016 from http://blogs.wsj.com/corporate-
intelligence/2013/12/17/cell-phones-help-us-be-alone-together/.
Huang, Q. (2015, July, 22). Meitu mobile terminal has been downloaded over 1.2 billion
times, will enter the mobile games market. Retrieved from http://tech.ifeng.com/
a/20150722/41355711_0.shtml.
Hughes, D. J., Rowe, M., Batey, M., & Lee, A. (2012). A tale of two sites: Twitter vs.
Facebook and the personality predictors of social media usage. Computers in
Human Behavior, 28, 561e569.
Kansi, J. (2003). The narcissistic personality Inventory: Applicability in a swedish
population sample. Scandinavian Journal of Psychology, 44, 441e448.
Kim, K. S., Sin, S. C. J., & Tsai, T. I. (2014). Individual differences in social media use
for information seeking. The Journal of Academic Librarianship, 40, 171e178.
Kim, Y., Sohn, D., & Choi, S. M. (2011). Cultural difference in motivations for using
social network sites: A comparative study of American and Korean college
students. Computers in Human Behavior, 27, 365e372.
Kramer, N. C., & Winter, S. (2008). Impression management 2.0: The relationship of
self-esteem, extraversion, self-efficacy, and self-presentation within social
networking sites. Journal of Media Psychology, 20, 106e116.
Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford, A. (2002).
Internet paradox revisited. Journal of Social Issues, 58, 49e74. http://dx.doi.org/
10.1111/1540-4560.00248.
Landis, M. H., & Burtt, H. E. (1924). A study of conversations. Journal of Comparative
Psychology, 4, 81e89.
Lobinger, K., & Brantner, C. (2015). In the eye of the Beholder: Subjective views on
the authenticity of selfies. International Journal of Communication, 9, 1848e1860.
McKirdy, E. (2015). China's online users more than double entire U.S. population.
Retrieved Feb. 1, 2016 from http://www.cnn.com/2015/02/03/world/china-
internet-growth-2014/.
Morf, C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A dynamic
self-regulatory processing model. Psychological Inquiry, 12, 177e196.
Mullen-Cooper, X. (2014). ‘Are selfies selfish?’: a study of selfie-engaging behaviors and
personality factors as potential predictors of these behaviors. Unpublished
doctoral thesis. UK: Manchester Metropolitan University.
Naaman, M., Boase, J., & Lai, C. H. (2010). Is it really about me?: Message content in
social awareness streams. In Proceedings of the 2010 ACM conference on com-
puter supported cooperative work (Savannah, Georgia, USA, February 06-10, 2010)
(pp. 189e192). New York, NY: CSCW '10. ACM. http://dx.doi.org/10.1145/
1718918.1718953. Retrieved Feb, 2, 2016 from.
Ong, E. Y. L., Ang, R. P., Ho, J. C. M., Lim, J. C. Y., Goh, D. H., Lee, C. S., et al. (2011).
Narcissism, extraversion and adolescents' self-presentation on Facebook. Per-
sonality and Individual Differences, 50, 180e185.
Panek, E. T., Nardis, Y., & Konrath, S. (2013). Mirror or Megaphone?: How re-
lationships between narcissism and social networking site use differ on Face-
book and Twitter. Computers in Human Behavior, 29, 2004e2012.
Raskin, R., & Terry, H. (1988). A principal-components analysis of the Narcissistic
Personality Inventory and further evidence of its construct validity. Journal of
Personality and Social Psychology, 54, 890e902.
Reed, M. (2015). Narcissism and the selfie: Investigating millennial narcissism on
Instagram. Unpublished undergraduate's thesis. United States: Andrews
University.
Ross, C., Orr, E. S., Sisic, M., Arseneault, J. M., Simmering, M. G., & Orr, R. R. (2009).
148 D. Wang / Computers in Human Behavior 66 (2017) 138e148
Personality and motivations associated with Facebook use. Computers in Human
Behavior, 25, 578e586.
Ryan, T., & Xenos, S. (2011). Who uses Facebook? An investigation into the rela-
tionship between the Big Five, shyness, narcissism, loneliness, and Facebook
usage. Computers in Human Behavior, 27, 1658e1664.
Sorokowski, P., Sorokowska, A., Oleszkiewicz, A., Frackowiak, T., Huk, A., &
Pisanski, K. (2015). Psychometric examination and normative data for the
Narcissistic Personality Inventory 29 item version. Personality and Individual
Differences, 85, 123e127.
Tencent.com. (2015). Wechat platform data analysis report. Retrieved October, 27,
2015, from http://tech.qq.com/a/20150127/018482.htm#p¼5.
Valkenburg, P. M., & Peter, J. (2007). Preadolescents' and adolescents' online
communication and their closeness to friends. Developmental Psychology, 43,
267e277.
Valkenburg, P. M., Schouten, A. P., & Peter, J. (2005). Adolescents' identity
experiments on the Internet. New Media & Society, 7, 383e402.
Vazire, S., Naumann, L. P., Rentfrow, P. J., & Gosling, S. D. (2008). Portrait of a
narcissist: Manifestations of narcissism in physical appearance. Journal of
Research in Personality, 42, 1439e1447.
Wang, J., Jackson, L. A., Zhang, D., & Su, Z. (2012). The relationships among the Big
Five Personality factors, self-esteem, narcissism, and sensation-seeking to
Chinese University students' uses of social networking sites (SNSs). Computers
in Human Behavior, 28, 2313e2319.
Weiser, E. B. (2015). #Me: Narcissism and its facets as predictors of selfie-posting
frequency. Personality and Individual Differences, 86, 477e481.
Zywica, J., & Danowski, J. (2008). The faces of Facebookers: Investigating social
enhancement and social compensation hypotheses; predicting Facebook™ and
offline popularity from sociability and self-esteem, and mapping the meanings
of popularity with semantic networks. Journal of Computer-Mediated Commu-
nication, 14, 1e34. http://dx.doi.org/10.1111/j.1083-6101.2008.01429.x.