Content uploaded by Yi-Ching Liu
Author content
All content in this area was uploaded by Yi-Ching Liu on Jan 15, 2018
Content may be subject to copyright.
When your Second Life comes knocking: Effects of personality on
changes to real life from virtual world experiences
q
Poppy Lauretta McLeod
⇑
, Yi-Ching Liu, Jill Elizabeth Axline
Department of Communication, Cornell University, Ithaca, NY 14853, United States
article info
Article history:
Keywords:
Avatar
Big Five
Personality
Trait activation
Virtual world
Virtual environment
abstract
A survey study (N= 223) of participants in the social virtual world, Second Life, examined the relationship
between Big Five personality factors, experiences in the virtual world and reports of changes to real life
resulting from the virtual world experiences. Hypotheses about direct and indirect effects of personality
on real life changes were tested with structural equation modeling. Results showed that the strength of
users’ relationship to the virtual environment, identification with and similarity to their avatars
positively predicted reports of changes to real life, and that these three factors mediated effects of Agree-
ableness, Extraversion, Intellect, Conscientiousness, and Emotional Stability, on real life changes. Consci-
entiousness also had a direct negative relationship with real life changes. Implications are discussed for
the potential of virtual social media features for activating facets of personality traits.
Ó2014 Elsevier Ltd. All rights reserved.
1. Introduction
The relationship between people’s online and offline experi-
ences has been of considerable interest since the beginning of
the Internet era, with researchers and practitioners alike concerned
with how computer-mediated interactions affect people’s lives. As
the sophistication of virtual social media increases, questions to
consider include how the perceived distinction between the real
and the virtual may be changing, and how that perception affects
both online and offline behaviors (Guitton, 2012). Multiple factors,
such as technological features, social context, and individual differ-
ences between users, shape perceptions of how virtual environ-
ments relate to the material environment. The current research
builds on previous findings that individual differences predict
effects of Internet activities on offline behaviors (e.g., Bagozzi,
Dholakia, & Pearo, 2007), and addresses the question of how per-
sonality predicts the effects of social virtual worlds on the real lives
of people involved in those environments. We develop and test a
conceptual model of how users’ experiences with virtual social
media, through activation of certain personality traits (Tett &
Guterman, 2000), mediate the relationship between personality
and the perception of real life changes.
Social virtual worlds are massive multi-user online applications
that simulate 3D environments through which people navigate and
interact with others by means of customizable avatars (Messinger
et al., 2009). The setting for the current research is the social virtual
world Second Life. In contrast to the gaming environment of other
virtual worlds such as World of Warcraft or Everquest II, the Sec-
ond Life (SL) environment does not impose any specific interaction
framework. Instead users are provided with the tools and compo-
nents to create any kind of content, activities and communities
that they desire (Linares, Subrahmanyam, Cheng, & Guan, 2013).
2. Theoretical framework
Much research has demonstrated that people’s experiences
with virtual environment media affect their offline attitudes, per-
ceptions and behaviors, and vice versa (Guitton, 2012;
Wiederhold, 2013). For example, people’s self-concept can be
affected by how others respond to their online self-presentation
(cf., Bargh & McKenna, 2004); online social activities can affect off-
line relationships with family, friends and community organiza-
tions (e.g., Bagozzi et al., 2007; Gilbert, Murphy, & Ávalos, 2011;
Kolotkin, Williams, Lloyd, & Hallford, 2012); people re-create real
life experiences in virtual environments (e.g., Lomanowska &
Guitton, 2014); and people can experience physiological,
emotional, and behavioral changes – both positive and negative –
following certain online experiences, (e.g., Persky & Blascovich,
http://dx.doi.org/10.1016/j.chb.2014.06.025
0747-5632/Ó2014 Elsevier Ltd. All rights reserved.
q
This research was supported in part by the Cornell Agricultural Experiment
Station federal formula funds, Project No. NYC-131442 received from Cooperative
State Research, Education, and Extension Service, U.S. Department of Agriculture.
Any opinions, findings, conclusions, or recommendations expressed in this publi-
cation are those of the author(s) and do not necessarily reflect the view of the U.S.
Department of Agriculture.
⇑
Corresponding author. Tel.: +1 6072548896; fax: +1 6072541322.
E-mail address: plm29@cornell.edu (P.L. McLeod).
Computers in Human Behavior 39 (2014) 59–70
Contents lists available at ScienceDirect
Computers in Human Behavior
journal homepage: www.elsevier.com/locate/comphumbeh
2007). Personality has emerged as an important factor that can
explain individual differences in how people are affected by Inter-
net usage in general (e.g., Correa, Hinsley, & de Zúñiga, 2010), and
immersive virtual worlds in particular (e.g., Dunn & Guadagno,
2012).
Our research model, depicted in Fig. 1, is based on three
propositions: first, due in part to personality, people differ in
how predisposed they are to be affected in their real lives by their
virtual world experiences; second, different environmental fea-
tures of virtual worlds activate different personality traits, which
leads to variance in how people experience the virtual world;
finally people’s virtual world experiences mediate the effects of
personality on likelihood to experience offline changes. Three vir-
tual world experience factors are at the center of the model: (a)
relationship to the environment, which incorporates the concepts
of presence, absorption, and immersion (Biocca, Harms, &
Burgoon, 2003; Lombard & Ditton, 1997); (b) identification with
the avatar, which includes dimensions of perceived closeness,
attachment, liking, and attraction to the avatar (Messinger et al.,
2008; Van Looy, Courtois, De Vocht, & De Marez, 2012); and (c)
user–avatar similarity, which refers to the degree of perceived
self-resemblance along dimensions of behavior, attitudes, person-
ality, and physical appearance (Evans, 2012). In the sections that
follow we describe these three virtual world experience factors,
review empirical evidence relevant to the relationships proposed
in our model, and develop the rationale for the study hypotheses.
2.1. Facets of virtual environment experience
2.1.1. Relationship to the environment
A user’s relationship to a virtual environment is multi-faceted,
comprised of feeling transported into, present in and immersed
within the environment, (Biocca et al., 2003; Lee, 2004; Lombard
& Ditton, 1997), of being absorbed with activities and interactions
in the environment (e.g., Weibel, Wissmath, & Mast, 2010), and
being attached to and emotionally involved with the environment
(e.g., Wirth et al., 2007; Witmer & Singer, 1998). These facets are
interrelated and some of the terms, such as presence and immer-
sion, are often used interchangeably (e.g., Witmer & Singer,
1998). There are nevertheless distinctions to be taken into account.
The most important of these distinctions for our research focuses
on the definition of presence as an experience limited to the time
a user is connected to a virtual environment which then ends once
the user goes offline (Lombard & Ditton, 1997; Slater, 1999). In
contrast, we argue that the feelings of attachment and emotional
involvement and thoughts of the environment can remain with a
user long after a person quits the application, facilitated in part
by the fact that virtual environments are persistent (Bell, 2008).
The immediate experience of presence, immersion, and absorption
experienced while online will affect the likelihood that the virtual
environment leaves psychological traces with a person subsequent
to exposure to the environment. Our concept of relationship to the
environment includes both proximal effects experienced while
online (e.g., presence, immersion, absorption) and distal effects
that remain following exposure (e.g., attachment and emotional
involvement).
1
In short, we define a strong relationship to the envi-
ronment as characterized by high experience of presence, immersion
and absorption, and strong emotional involvement and attachment.
People’s relationship to a virtual environment has been shown
to influence the extent to which virtual world experiences affect
them outside of the virtual world. For example, Behm-Morawitz
(2013) found that presence strengthened the effects of avatars on
users’ offline appearance and health-related behaviors, and Fox,
Bailenson, and Binney (2009) reported that people were more
likely to imitate the eating behaviors of their avatars as the expe-
rience of presence increased. Tamborini and Skalski (2006)
reviewed research showing that presence-inducing electronic
games could help to reduce offline anxiety and phobias. The rela-
tionship to a virtual environment may act on a user’s sense of self
and social encounters to the extent that the line between the vir-
tual and physical world is blurred, and virtual world experiences
may be more likely to carry over to offline experiences (McLeod
& Leshed, 2011).
2.1.2. Avatars
The model includes two factors related to avatars: identification
with the avatar and degree of similarity between the user and the
avatar. Identification with virtual avatars includes the notions of
assimilating characteristics of the avatar into the self-perception
(e.g., Van Looy et al., 2012), moral and emotional attachment to
the avatar (e.g., Wolfendale, 2007), and adopting the avatar’s goals,
abilities, resources and impediments as one’s own (e.g., Steen,
Control Variables:
1. Hours per Week in SL
2. SL Age
3. SL Activities
H7a
H7c
H8
H9a
H7b
H10
H4a
H4b
H6a
H6b
H5
H1
H2
H3
Extraversion
Conscientiousness
Emotional
Stability
Intellect
Identification
with Avatar
Similarity to
Avatar
Real Life
Changes
Agreeableness
H9c
H9b
Relationship to
Environment +
−
+
+
+
+
+
+
+
−
−
+
−
−
+−
Fig. 1. Hypothesized conceptual model.
1
We recognize that distal effects such as attachment and emotional involvement
do also occur proximately; we argue that these effects do not require the proximate
online connection.
60 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
Davies, Tynes, & Greenfield, 2006). The extent to which people
identify with their avatars has been found to affect their enjoyment
of virtual world experiences (e.g., Steen et al., 2006; Trepte &
Reinecke, 2010), but there is only limited empirical evidence about
how identification with avatars affects offline behaviors. Bargh and
McKenna (2004) reported data suggesting that parts of identities
formed from virtual group membership may sometimes be assim-
ilated into the offline self-concept, especially when these identity
factors are deemed important to the user. Behm-Morawitz
(2013) found evidence among SL users for the positive influence
of identifying with avatars on offline health behaviors. Based on
these limited findings, it seems reasonable to hypothesize that
avatar identification in the virtual world may lead to a variety of
alterations in the real world.
Though related to avatar identification, similarity to the avatar
is a distinct aspect of users’ virtual world experiences. For instance,
someone may be highly identified with an avatar that is very dis-
similar – such an avatar with a non-human form. The research
shows that there is general consistency between users and their
avatars in both personality and behavior, and that exceptions to
this trend tend to be related to specific goals or motivations. For
example, Koles and Nagy (2012) reported that divergence between
real self and avatar identities was important for many Second Life
users, and that the social aspects of identity were more relevant for
their avatars than for their real selves. Sung and Moon (2011)
found that people’s avatars were significantly more extraverted
than their real selves in a social-networking context, but no differ-
ences in extraversion were seen in an online gaming or a virtual
classroom context. With respect to the relationship between
user–avatar similarity and real life changes, Trepte and Reinecke
(2010) reported that people who attributed their avatars with
similar personality profiles as their own personalities showed posi-
tive effects on their real life satisfaction. Behm-Morawitz (2013)
suggested that people who perceived their avatar as being more
attractive than themselves and closer to their idealized self-image
were more likely to report that their avatar influenced their offline
health and appearance. It thus seems reasonable to expect that
avatar–user similarity will have real world effects. We offer the
following three hypotheses with respect to the virtual world
experience factors:
H1. Strength of relationship to virtual environment will be positively
related to perceptions of real life changes of virtual world experiences.
H2. Identification with one’s avatar will be positively related to
perceptions of real life changes of virtual world experiences.
H3. Similarity to the avatar will be positively related to perceptions of
real life changes of virtual world experiences.
2.2. Personality
The Five Factor Model of personality (Goldberg, 1993; McCrae
et al., 2008) is commonly applied in research related to Internet
use (e.g., Hamburger & Ben-Artzi, 2000). The ‘‘Big Five’’ factors
are: Extraversion – the extent to which a person is predisposed
toward the external world. Talkative, assertive, and active are com-
mon words that are used to describe extraverts; Agreeableness –
the extent to which an individual is kind, trustworthy, and warm;
Conscientiousness – the extent to which a person is perceived as
organized, thorough, and reliable; Emotional Stability (also labeled
Neuroticism) – the extent to which a person is relaxed, not easily
upset, and able to withstand unexpected occurrences with equa-
nimity; and Intellect (also labeled Openness to Experience) – the
extent to which a person has imagination, curiosity, and creativity.
Research has shown that personality predicts patterns of Inter-
net usage in general (Landers & Lounsbury, 2006) and people’s
motivations for using virtual world applications in particular
(e.g., Graham & Gosling, 2013). Personality has also been found
to correlate with specific behaviors that users exhibit online. For
example McCreery, Krach, Schrader, and Boone (2012) found that
Agreeableness predicted agreeable behaviors in a virtual environ-
ment. The evidence about the influence of personality on the
relationship between people’s online and offline experiences
comes primarily from the literature on violent online games. For
example, Bartholow, Sestir, and Davis (2005) reported that trait
hostility partially accounted for increases in offline aggressive
behavior following online game play. Hartmann, Toz, and
Brandon (2010) found that trait empathy increased feelings of guilt
following the commission of unjustified virtual violence. Outside of
the gaming context one study found that Introversion and Neurot-
icism predicted the extent to which people perceived their ‘‘real
selves’’ to be more fully expressed on the Internet than in their real
lives (Amichai-Hamburger, Wainapel, & Fox, 2002), and another
reported that among a sample of college students, trait Extraver-
sion affected Internet users’ perceptions of school life satisfaction
(Liu & LaRose, 2008).
Based on the research reviewed thus far, it seems clear that per-
sonality can predict various aspects of Internet usage and that
Internet usage can cause changes in real life beliefs, emotions
and behaviors. Very few studies however, have examined the
entire causal chain from how personality predisposes people to
use the Internet in different ways, to the experiences resulting
from their usage, to the effects of their usage and experience on life
offline. One example is the Bartholow et al. (2005) study of violent
video games in which they showed that even after controlling for
personality factors that predispose people to seek out such games,
the violence within the game stimulated cognitive and emotional
changes resulting in aggressive behavior and perhaps long term
tendency to develop aggressive personality. To our knowledge,
however, such a model has not been tested in the context of social
virtual world media, where a wider array of personality factors and
offline effects come into play.
The model presented in Fig. 1 proposes that the three virtual
world experience factors described earlier (i.e., relationship to
the environment, identification with the avatar and similarity
between the user and avatar) mediate the effects of personality
on likelihood that virtual world use will lead to real life changes.
The theoretical justification for this proposition can be derived
from trait activation theory (Tett & Guterman, 2000), which is
based on the fundamental premise that environments differ in
their relevance to the expression of different personality traits.
Our model indicates further that we predict Conscientiousness
and Emotional Stability also to have direct effects on real life
changes. In the following section, we discuss how the differential
activating potential of various features of social virtual worlds
leads to specific predictions of how different personality factors
affect users’ online experiences and offline changes.
3. Activation of personality traits in Second Life
Although personality is relatively stable across time and con-
texts, it is nevertheless widely acknowledged that the expression
of personality traits will vary across situations (cf., Mischel,
1977). According to trait activation theory people are inherently
motivated to express their traits and they will likely do so when
situational cues offer the opportunity (Tett & Burnett, 2003; Tett
& Guterman, 2000). Tett and Burnett give the example that to a
person high in Conscientiousness an untidy office provides a cue
that would likely induce cleaning behavior. Moreover, the same
P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70 61
situational characteristic could activate different traits. For exam-
ple, a business meeting might trigger sociability in an extraverted
person and cooperativeness in an agreeable person. The multiple
aspects of a complex environment such as Second Life will trigger
differences in personality expression across users. We now turn to a
discussion of how SL affords the opportunities to express predomi-
nant personality traits which then shape users’ virtual world experi-
ences – that is, the strength of their relationship to the environment,
identification with, and resemblance to their avatars – and in turn
how those experiences affect real life changes.
3.1. Agreeableness
Second Life is essentially a social environment, and it has been
found that social motivations top the lists of reasons that people
join (Hassouneh & Brengman, 2014; Verhagen, Feldberg, van den
Hooff, Meents, & Merikivi, 2012). The social nature of SL should
thus provide a cue that triggers the expression of both Agreeable-
ness and Extraversion. A further cue that should differentiate
Agreeableness is the ease of creating groups and communities
and strong norms toward reciprocity, volunteerism and helpful-
ness. SL should therefore activate the interest in other people
and pro-community orientation characteristic of people high in
Agreeableness (cf., John & Srivastava, 1999). Agreeableness should
therefore lead to strong connection to the social virtual environ-
ment and identification with avatars.
Avatars especially should activate Agreeableness because the
avatar is a person’s means of connecting with other social actors
in the environment (Nagy & Koles, 2014). In support of this idea,
Yee, Ducheneaut, Nelson, and Likarish (2011) reported that World
of Warcraft players high in Agreeableness used their avatars to
express positive emotive behaviors such as hugging and waving.
We do not expect Agreeableness to predict avatar–user similarity,
however. Though agreeable people are likely to create avatars that
would be pleasing and attractive to other people, this does not
imply the avatars would be similar to themselves (Dunn &
Guadagno, 2012). We test the following hypotheses:
H4a. Agreeableness will be positively associated with relationship to
the environment.
H4b. Agreeableness will be positively associated with identification
with the avatar.
3.2. Extraversion
Although it might be expected that extraverts would show a
strong relationship to the SL environment and their avatars
because the many opportunities to meet and interact easily with
other people offers the chance to express the sociable facets of
Extraversion (e.g., Laarni, Ravaja, Saari, & Hartmann, 2004), there
is equally compelling reasoning and supporting empirical evidence
that introverts would also feel a strong connection to the environ-
ment and to their avatars. Alsina-Jurnet and Gutiérrez-Maldonado
(2010) showed a positive relationship between presence and
Introversion, and argued that it was because in virtual
environments introverted people could easily control the pace of
interactions. Thus, extraverts and introverts may be expected to
be equal in strength of association with the environment and
identification with their avatars, albeit for different reasons. We
therefore do not hypothesize that Extraversion would predict the
strength of relationship to the environment or identification with
the avatars.
We do expect, however, that Extraversion will be positively related
to similarity between users and avatars due to the straightforward
observation that Extraversion is more socially desirable than Introver-
sion (e.g., Pavot, Diener, & Fujita, 1990). Therefore, on the average
people’s avatars are likely to be more extraverted than introverted,
and similarity will be highest for people who actually are extraverts.
Consistent with this reasoning, Dunn and Guadagno (2012) found that
user–avatar similarity increased with Extraversion. Further, Messinger
et al. (2008) found that avatars that were more attractive than the real-
life users were also more extraverted. We will test the following
hypothesis.
H5. Extraversion will be positively related to user–avatar similarity.
3.3. Intellect/Openness to Experience
Second Life provides numerous opportunities for intellectual
pursuits (e.g., classes, lectures and discussion), for creative expres-
sion (e.g., musical performance, building structures, designing
clothing), and for exploration and experimentation (e.g., avatar
customization, having ‘‘impossible’’ experiences such as being
inside of an atom). These affordances should provide a cue to acti-
vate the trait of Intellect (Openness to Experience). The limited
empirical evidence is consistent with an expectation that people
high in Intellect would have a strong connection to a virtual envi-
ronment. Hou, Nam, Peng, and Lee (2012) found that users’ immer-
sion tendency – a trait that has been described as closely related to
Openness to Experience (Church, 1994; Goldberg, 1993; McCrae,
1994) – predicted the experience of presence. Weibel et al.
(2010) similarly found that Openness was related to users’ feelings
of immersion. Users high in Intellect should also identify with their
avatars because customizing an avatar provides opportunities for
creativity and to explore issues such as identity. Yee et al. (2011),
for example, found that World of Warcraft users who scored high
in Openness exhibited behaviors associated with exploration and
curiosity, such as having multiple avatars. We do not expect, how-
ever, that Intellect will predict similarity between users and ava-
tars. Although the customization possibilities of avatars may
result in wide variance in avatar characteristics across people high
in Intellect, this does not imply wide differences between individ-
ual users and their avatars. Instead, we would not expect any
systematic relationship across users based on Intellect. We test
the following hypotheses:
H6a. Intellect will be positively associated with relationship to the
environment.
H6b. Intellect will be positively associated with identification with the
avatar.
3.4. Conscientiousness
The virtuality itself of Second Life is the principal reason we
argue that Conscientiousness will have direct as well as indirect
effects on real life. Because the connection with this environment
is mediated through a computer, real life is impacted at the very
least through the explicit choice to log on and to plan for the real
life adjustments needed to accommodate the time spent in virtual
world interaction. Turel and Zhang (2010) for example argued that
Conscientiousness is especially important in virtual work settings,
where the absence of direct interaction with colleagues may
reduce work motivation. Virtuality may provide a cue that triggers
expressions of the inherent achievement orientation associated
with Conscientiousness. Further, no matter how absorbed they
may be when in SL, people high in Conscientiousness will maintain
a higher degree of awareness of how that time affects real life, and
62 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
will be motivated to organize their lives – physical and virtual –
accordingly. Due to this planned and organized approach, we
expect conscientious people to report fewer real life changes
stemming from their participation in a virtual environment. An
additional reason we expect Conscientiousness to be negatively
related to the extent of real life changes relates to the facets of Con-
scientiousness associated with morality, impulse control, openness
and honesty (cf., John & Srivastava, 1999; McCrae & Costa, 1987). A
certain degree of notoriety is associated with Second Life due to
highly publicized accounts of events such as real life marital break-
ups caused by relationships developed in SL (e.g., Adams, 2008).
Awareness of this reputation would trigger the conscientious per-
son to make deliberate choices about whether to engage in such
behaviors and to plan for guarding against negative effects to real
life (cf., McLeod & Leshed, 2011). To our knowledge, this direct
effect of Conscientiousness on real life changes of virtual world
involvement has not been empirically tested.
Conscientiousness will also influence real life indirectly through
the three virtual world experience factors. We argue that part of
the strategy for controlling the effects of virtual world engagement
on real life would be to maintain a certain psychological distance
from the virtual world. This implies that Conscientiousness would
be negatively related to strength of relationship to the environ-
ment and identification with the avatar. Meanwhile, similar to
findings reported by Dunn and Guadagno (2012), we expect
Conscientiousness to be positively related to user–avatar
similarity. In maintaining social ties within the virtual environ-
ment a person high in Conscientiousness would be focused on
ways of developing interpersonal trust and confidence. This can
partly be achieved by having high similarity between the user
and the avatar. Though one may not reveal all details of real life
identity, making it known that one’s avatar is similar to the real
self has been shown to be a factor in developing trust in
relationships within Second Life (McLeod & Leshed, 2011). We will
test the following hypotheses:
H7a. Conscientiousness will be positively associated with similarity to
the avatar.
H7b. Conscientiousness will be negatively associated with identifica-
tion with the avatar.
H7c. Conscientiousness will be negatively associated with relation-
ship to the environment.
H8. Conscientiousness will be negatively associated with experiencing
real life changes of virtual world experiences.
3.5. Emotional Stability/Neuroticism
The unstructured and open nature of Second Life combined
with its ‘‘anything goes’’ reputation should provide situational cues
that activate facets of Emotional Stability. Though terms of service
policies exist
2
these rules are not designed to prevent extremes of
behaviors, but rather serve to prevent the extremes from impinging
on those not wanting to be exposed to them (Boellstorff, 2008;
Malaby, 2009). There is nevertheless a relatively high probability
of casually encountering potentially emotionally disturbing content
or behaviors (e.g., seeing someone with slaves on a leash while in
a shop buying some clothes). We argue that the unpredictability of
the environment with respect to such encounters would activate
the need for calm and even-temperedness, principal facets of
Emotional Stability (cf., John & Srivastava, 1999). One way to main-
tain emotional calmness may be to limit the extent of emotional
investment in the virtual environment, which may imply lower
feelings of immersion and presence and weaker identification with
one’s avatar. With respect to user–avatar similarity, there is
empirical evidence suggesting that overall life satisfaction, a facet
of Emotional stability, is positively associated with similarity
(Trepte & Reinecke, 2010), and that Neuroticism predicts unrealistic
online self-presentation (Michikyan, Subrahmanyam, & Dennis,
2014). Therefore it seems reasonable to expect that Emotional
Stability would predict user–avatar similarity.
Finally, we expect that because of the general emotional
volatility and reactance associated with the Emotional Stability–
Neuroticism dimension, no matter the strength of relationship to
the environment or identification with and similarity to the avatar,
Emotional Stability will directly predict low likelihood of experi-
encing changes to real life. For example, Amichai-Hamburger
et al. (2002) found that Internet experiences were associated with
general well-being more strongly for people high in Neuroticism
than for people high in Emotional Stability. We will test the follow-
ing hypotheses:
H9a. Emotional Stability will be negatively associated with identifi-
cation with the avatar.
H9b. Emotional Stability will be positively associated with similarity
to the avatar.
H9c. Emotional Stability will be negatively associated with relation-
ship to the environment.
H10. Emotional Stability will be negatively associated with real life
changes.
4. Method
4.1. Overview
Data were collected through two online surveys of users of the
virtual world application, Second Life. The first survey contained
items about users’ virtual world experiences. Participants who
completed this first survey were invited two months later to take
the second survey containing the personality measures.
4.2. Participants and procedure
Participants were recruited through a market research firm spe-
cializing in virtual worlds.
3
This firm maintained a panel of research
volunteers who agreed to furnish basic information about their real
life identity, which includes sex, age, occupation and country of res-
idence, but not their real names. Invitations were sent to 1093 U.S.
based
4
volunteers to participate in an online survey on ‘‘Second Life
and You,’’ in exchange for compensation of 500 Linden dollars, the
currency used in Second Life (approximately $2.50 U.S.). Three-hun-
dred surveys were returned for an initial response rate of 27%.
In order to reduce the effects of common method variance
resulting from having self-report data from the same respondents
about both their virtual world experiences and their personality,
we developed a separate survey for the personality measures and
2
http://lindenlab.com/tos.
3
Market Truths Ltd.; www.markettruths.com.
4
Our decision to limit the sample to U.S. residents was designed to control for the
interaction between national culture and personality (Church, 2000). We do not,
however, have data on how many of our respondents may have been expatriates.
P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70 63
administered it two months later (Podsakoff, MacKenzie, Lee, &
Podsakoff, 2003). Invitations were sent to the individuals who
returned the first survey to participate in a ‘‘Personality Survey,’’
in exchange for compensation of 500 Linden dollars. These partic-
ipants were not told that the two surveys were linked. Two hun-
dred twenty-five of these surveys were returned, for a response
rate of 75% on the second survey. Two surveys were removed from
the final data set because the respondents reported their real life
age as being under 18. The final sample contained 223 participants
for an overall study response rate of 20%.
4.3. Survey instruments
The first survey contained items we developed for this study to
measure relationship to the environment, identification with the
avatar, similarity to the avatar and real life changes. For the rela-
tionship to the environment factor, we adopted some items from
pre-existing measures related to presence, immersion, and absorp-
tion (e.g., Witmer & Singer, 1998), and added items to reflect our
arguments that the relationship to the environment transcends
the time that users are connected. For the avatar identification
and similarity items, after consulting with previous studies related
to identification in online gaming (e.g., Van Looy et al., 2012), we
developed items related specifically to identification with avatars
in Second Life, and user–avatar similarity. Previously published
items related to real life changes have tended to focus on a specific
behavior such as violence (e.g., Bartholow et al., 2005), or healthy
eating (e.g., Fox et al., 2009). Because we were interested in a broad
array of real life changes, we developed items for this study drawn
from focus group and interview research done specifically in SL on
users’ general experiences (Market Truths, 2009; Market Truths,
Ltd., 2008
5
;McLeod & Leshed, 2011). Participants in those studies
gave multiple examples of changes to their real lives as a result of
their experiences in SL ranging from finding it easier to socialize in
real life to applying in real life a technical skill learned in SL, and
the items we developed represent the kinds of real life changes
reported. All the items, presented in Table 1, were in Likert format
with responses ranging from 1 = Strongly Disagree to 5 = Strongly
Agree.
A principle components factor analysis with varimax rotation
confirmed that the instrument measured these four constructs
(please see Table 1 for the factor loadings), but two additional fac-
tors nevertheless emerged. Close inspection of the items suggested
that the relationship to the environment scale could be divided
into two sub-scales. The first, which seems to capture emotional
reactions to the environment, we label emotional involvement;
the second, which includes items that seem to capture attention
to the virtual environment to the exclusion of other stimuli, we
label absorption (cf., Weibel et al., 2010). The similarity to avatar
scale also seemed to comprise two sub-scales – physical and behav-
ioral similarity. These subscales were included in the main analyses
testing the study hypotheses. The reliabilities for each of the scales
were
a
= .70 for relationship to the environment overall (7 items),
a
= .67 for emotional involvement, and
a
= .59 for absorption;
a
= .76 for identification with the avatar (5 items);
a
= .77 for sim-
ilarity to the avatar overall (8 items),
a
= .70 for physical similarity,
and
a
= .77 for behavioral similarity;
a
= .85 for real life changes (6
items). Also included in this survey were items related to real life
occupation, motivations for joining Second Life, the amount of time
per week spent in SL (SL Time), the amount of time since opening
the SL account (SL Age), and users’ activities within SL (SL Activi-
ties). In the event that participants used more than one avatar, they
were asked to answer with respect to the one they considered their
principal avatar.
The second survey consisted of Goldberg’s personality
inventory (Goldberg, 1999). We used the version with 10-items
per factor with response scales ranging from 1 = Strongly Disagree
to 5 = Strongly Agree. A factor analysis using varimax rotation of
the personality items confirmed the five factors as conceptualized
by Goldberg, and we obtained comparable reliabilities:
a
= .87 for
Extraversion;
a
= .80 for Agreeableness;
a
= .78 for Conscientious-
ness;
a
= .89 for Emotional Stability; and
a
= .80 for Intellect.
5. Results
5.1. Sample description
The real life age of the participants ranged from 18 to 71 with a
mean of 35.71, and 54% were female. Similar to other studies (e.g.,
Koles, 2012), gender of the avatars closely matched real life
sex – 55% reported having female avatars. Respondents had been
involved in Second Life between 127 and 2813 days (M= 1,189,
SD = 498.99), thus most respondents had had their accounts for a
little over 3 years. They reported spending a mean of 33 h per week
in SL (range = 0–168, SD = 32.32), and of the time spent
there, participants reported that 68% of it was spent in social or lei-
sure activities (e.g., making friends, going to parties, listening to
music), 22% was spent on task-oriented activities (e.g., taking or
teaching a class, running a retail business), and the remaining time
was spent on miscellaneous or specialized activities such as learn-
ing to use the SL scripting language. Consistent with how they
divided their time there, 79% of the participants reported that their
number one motivation for joining SL was for social purposes,
whereas only 12% reported that their primary motivation was
task-related.
5.2. Trait activation validation
To support the arguments about the relevance of various
aspects of Second Life to the different personality traits, we fol-
lowed procedures reported in Tett and Guterman (2000) and
obtained independent ratings of situation-trait relevance.
Twenty-one respondents who rated themselves as expert users
of SL were recruited through the Second Life Educators and Second
Life Researchers listservs, and one thousand respondents from the
general public were recruited online from Amazon Mechanical
Turk.
6
These respondents rated descriptions that could apply to five
different aspects of SL and were asked to rate the likelihood that
each aspect would provoke behavioral expressions of the five per-
sonality traits. For example, the description related to Extraversion
was: Imagine a situation that involves a lot of social interaction where
it is possible to easily meet many new people,and where people try hard
to show off their best side; the one related to Intellect was: Imagine a
situation that offers boundless opportunities for personal growth,crea-
tivity and exploration,with many outlets for artistic and intellectual
pursuits. Each description was rated for all five of the traits using
adjectives drawn from Goldberg (1993). For example, relative to
Extraversion, respondents rated the likelihood that each description
would ‘‘provoke someone to behave in an outgoing, active, and
assertive manner,’’ and for Conscientiousness, in a ‘‘responsible, reli-
able, careful and honest manner.’’
The ratings were made on five-point scales where Very
Unlikely =2; Unlikely =1; Equally Likely and Unlikely =0;
5
These two reports were produced by the market research firm that supplied the
sample used in this study. The cited reports are no longer publicly available because
the company has ceased operations, but copies are available from the corresponding
author.
6
www.MTurk.com.
64 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
Likely = 1; and, Very Likely = 2. The results of ANOVAs showed that
each set of behavioral cues, with the exception of Agreeableness,
was rated as being significantly most likely to provoke the behav-
ioral response corresponding to the expected personality trait (see
Table 2 for the means). The description related to Agreeableness
(described as Imagine a community where friendliness,interest in
other people and helpfulness are highly valued,and where there are
strong sanctions against selfishness,rudeness and unfriendliness)
was rated as most likely to provoke behaviors related to Conscien-
tiousness. It should also be noted that the description related to
Extraversion was rated as nearly as likely to provoke Agreeable-
ness. This is not entirely surprising given that Agreeableness com-
prises facets relevant to the sociability aspects of Extraversion and
the cooperation aspects of Conscientiousness. We will return to
this result in the discussion of the study findings. The validation
results nevertheless largely support the reasoning for the hypoth-
esized effects of Second Life to provide cues for the activation of
the five personality factors.
5.3. Tests of hypotheses
Table 3 presents the intercorrelations and descriptive statistics
of all the study variables. We combined the hypotheses into a sin-
gle structural model, presented in Fig. 1, and tested the hypotheses
through structural equation modeling with PASW-AMOS v. 21,
using the maximum likelihood method to estimate the parameters.
The data approximated a reasonably good fit to the model, fol-
lowing recommendations by Kline (2011). The chi-square was not
significant,
v
2
(10) = 8.08, p= .62, and the CFI (the comparative fit
index) = 1.00, the GFI (goodness-of-fit index) = 0.99 were above
the recommended cutoff of 0.95; the RMSEA (root mean square
error of approximation) = 0.00 was below the cutoff of .05; finally,
the SRMR (standardized root mean square residual) = 0.03 was
below the recommended cutoff of 0.08. Several of the hypothesized
paths nevertheless were not significant: the data showed no sup-
port for the hypothesized effects of Agreeableness on identification
with the avatar (H4b), or on absorption with the environment
Table 1
Items and reliabilities of Second Life experience and real life changes scales.
Item Factor loading
Scale 1: Emotional Involvement (RTE1) (
a
= .67)
I have become angry or upset by things that have happened in SL .72
Things that happen in SL can put me in a good mood .72
I often get emotionally involved with things happening in SL .72
I sometimes get so involved with what I am doing in SL that I lose track of time .63
Scale 2: Absorption (RTE2) (
a
= .59)
I can quickly switch from doing something in SL to doing something in RL. (R) .79
Even when I am involved in something in SL I am generally aware of what is happening around me in RL. (R) .75
I sometimes am so involved with something happening in SL that people in my RL environment have difficulty getting my attention .63
Scale 3: Identification with Avatar (
a
= .76)
When things happen to my avatar, it feels as though they happen to my RL self .71
I like my avatar .58
I feel a close bond with my avatar .86
It’s hard to say where my avatar stops and where my RL self starts .61
I strongly identify with my avatar .83
Scale 4: Behavioral Similarity (
a
= .77)
My avatar behaves the way that I would in similar situations in RL .71
People would probably not notice much difference in how I behave in RL and how my avatar behaves .80
My avatar shares most of the strengths and weaknesses of my RL personality .67
My avatar spends time with avatars whose interests and behavior are similar to those of my RL friends and acquaintances .61
My avatar uses language (vocabulary, grammar, etc.) similarly to the way I do in RL (except for typos in text) .58
Scale 5: Physical Similarity (
a
= .70)
People would be unlikely to see much physical similarity between the RL me and my avatar. (R) .84
My avatar is similar to me in body size, shape, and attractiveness .75
My avatar has similar coloring (skin, hair, and eyes) as I have in RL .73
Scale 6: Real Life Changes (
a
= .85)
Over the time I have been involved with SL, my ways of behaving in RL has come to more closely resemble that of my avatar .54
Over the time I have been involved with SL, the way I communicate in RL has come to more closely resemble the way my avatar communicates .48
Things I have experienced in SL have changed the way I perceive things in RL .86
Things I have experienced in SL have changed the way I behave in RL .78
Participating in SL has exposed me to new possibilities for my real life .77
Participating in SL has helped me discover my ‘‘true’’ RL self .67
Note: RTE = Relationship to the Environment.
Table 2
Mean trait relevance ratings for Second Life aspect descriptions.
Traits
Extraversion Agreeableness Emotional Stability Conscientiousness Intellect
SL aspects
a
Related to E 1.3 1.1 0.85 1.05 1.0 p< .001
Related to A 0.05 0.96 0.80 1.30 0.7 p= .004
Related to ES 0.20 0.10 0.85 0.0 0.1 p= .01
Related to C 0.85 1.2 0.90 1.65 1.10 p< .001
Related to I 1.15 1.25 1.05 1.15 1.80 p< .001
Note: E = Extraversion; A = Agreeableness; ES = Emotional Stability; C = Conscientiousness; I = Intellect.
a
Description of SL environment aspect relevant to specific trait.
P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70 65
(H4a); for the effects of Intellect on absorption with the environ-
ment (H6a) or identification with the avatar (H6b); for the effects
of Conscientiousness on identification with the avatar or either
sub-scale of relationship to the environment (H7b and H7c); or
for the effects of Emotional Stability on similarity to the avatar or
directly on real life changes (H9b and H10). The fit statistics that
resulted after deleting these paths did not significantly differ from
the initial indices.
Finally, the effects of the three control variables were evalu-
ated. SL Age (length of time since SL account was opened) and
SL Activities (work, social or other) had no effect on any of the
virtual world experience variables or on real life changes. Amount
of time per week spent in SL (SL Time) significantly predicted
relationship to the environment, behavioral similarity, and
identification, but none of the structural paths were significantly
changed with the addition of this control variable to the model,
suggesting that real life changes were not due merely to the
amount of time spent in SL. The fit indices of the final model
are summarized in Fig. 2.
In support of H1–H3, the three virtual world experience factors
(relationship to the environment, identification with the avatar,
and behavioral, but not physical similarity to the avatar) were pos-
itively related to reports of real life changes. Agreeableness was
positively related to emotional involvement with the environment,
but not to absorption, thus H4a was partially supported. In support
of H5, Extraversion showed positive relationships to physical and
behavioral similarity to the avatar. In partial support of H6a, Intel-
lect showed a positive relationship to emotional involvement with
the environment, but no relationship to absorption (H6a), nor was
the expected positive effect on identification with the avatar
observed (H6b). As predicted Conscientiousness was positively
related to both physical and behavioral similarity to the avatar
(H7a), and the predicted negative direct effect on real life changes
(H8) was also found, but no effect on identification with the avatar
(H7b) or relationship to the environment (H7c) was observed.
Finally, support was found for the hypothesized negative associa-
tion between Emotional Stability and identification with the avatar
(H9a) and the negative association with relationship to the envi-
ronment (H9c), but no relationship to user–avatar similarity
(H9b) or direct effects on changes to real life (H10).
5.4. Mediation analysis
Following recommendations in the psychology and communi-
cation literatures that mediation effects be tested directly (e.g.,
Hayes, 2009; Preacher & Hayes, 2008; Rucker, Preacher, Tormala,
& Petty, 2011), we used the bootstrapping function of Amos to con-
firm that the virtual world experience factors mediated the effects
of personality traits on the likelihood to report real life changes.
The indirect effects of Extroversion (b= .03, p= .01), Intellect
(b= .04, p= .01), Agreeableness (b= .03, p= .04), Emotional Stabil-
ity (b=.16, p< .001), and Conscientiousness (b= .03, p= .01) on
real life changes were all significant.
6. Discussion
The results of this study support the assertion that individual
differences in personality traits predict the extent to which people
report that virtual world experiences change their real lives. The
strength of their relationship with the environment, identification
with, and similarity to their avatars mediated effects of Agreeable-
ness, Extraversion, Intellect, Conscientiousness, and Emotional Sta-
bility, and Conscientiousness also directly predicted fewer real life
changes of virtual world experiences. These relationships held
when motivations for joining Second Life, how much time people
spent there, and how they spent that time were taken into account.
Close examination of the final model presented in Fig. 2 sug-
gests that the degree of emotional involvement with the environ-
ment and user–avatar similarity were the key mediators among
the SL experience factors; all of the paths from personality went
through one or both of these factors. We argued earlier that the
overall relationship to the environment factor assumes that the
relationship extends beyond the specific period of being physically
logged into SL. Our observation of the two sub-factors of emotional
involvement and absorption supports this argument. Absorption
ends once people log off, but they carry their emotional involve-
ment away from the computer screen. Our findings suggested that
emotional attachment was a more important determinant of the
likelihood of experiencing changes to real life than was the
extent of absorption while physically online. In fact, emotional
Table 3
Bivariate correlations, sample means and standard deviations.
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1. Extraversion .32
***
.06 .24
***
.19
**
.06 .08 .03 .21
**
.18
**
.02 .07 .24
**
.05 .01 .08
2. Agreeableness .15
*
.18
*
.26
***
.16
*
.04 .12 .18
**
.10 .06 .05 .10 .01 .03 .06
3. Conscientiousness .42
***
.09 .06 .11 .03 .22
**
.21
**
.08 .03 .02 .01 .01 .00
4. Emotional Stability .02 .25
***
.15
*
.23
**
.12 .04 .16
*
.11 .30 .02 .04 .02
5. Intellect .24
**
.06 .17
*
.16
*
.09 .13 .04 .08 .11 .21
**
.17
*
6. Emotional Involvement
(RTE1)
.29
***
.55
***
.29
***
.02 .47
***
.03 .15
*
.08 .09 .03
7. Absorption (RTE2) .25
***
.20
**
.05 .26
***
.03 .13 .06 .11 .09
8. Identification with Avatar .35
**
.15
*
.55
***
.04 .16
*
.03 .03 .01
9. Avatar Behavioral Similarity .36
***
.36
***
.04 .21
**
.12 .08 .04
10. Avatar Physical Similarity .24
***
.11 .10 .05 .05 .01
11. Real Life Consequences .05 .13 .08 .11 .07
12. Control: SL Age .02 .07 .06 .01
13. Control: SL Hours .07 .03 .13
14. Control: Work Activities .77
***
.18
**
15. Control: Social Activities .49
***
16. Control: Other Activities
M29.93 39.67 51.98 31.93 39.72 15.98 6.46 17.83 19.05 8.58 30.20 1178.82 33.50 21.20 69.65 9.15
SD 7.35 5.14 5.68 7.78 5.21 2.78 2.29 3.67 4.04 3.24 6.63 502.53 31.44 24.97 28.14 18.34
Note: RTE = Relationship to Environment.
*
p< .05.
**
p< .01.
***
p< .001.
66 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
involvement was the single strongest predictor in the entire model
of real life changes. To be sure, emotional involvement and absorp-
tion are closely related, but absorption may have a less central role
because of its shorter term effects. Past research has found that
presence affects the relationship between online and offline expe-
riences (e.g., Behm-Morawitz, 2013; Fox et al., 2009), and our
results support the importance of recent examinations of the dif-
ferential effects of the various dimensions of presence (e.g., Jin,
2011; Wirth, Hofer, & Schramm, 2012).
The other key mediator was resemblance to the avatar. Person-
ality was more predictive of similarity to the avatar than of identi-
fication with the avatar. Meanwhile, identification with the avatar
showed a strong relationship to real life changes. An interpretation
of this pattern is that with the exception of Emotional Stability (to
be discussed shortly), we did not find that personality predicted
the likelihood of identification with the avatar or that identifica-
tion would lead to changes in real life. Rather, personality exerted
effects on similarity to the avatar. More work would be needed in
order to better understand this difference. From a trait activation
perspective (Tett & Burnett, 2003) we might speculate that a wide
array of cues, applicable across multiple personality factors, trigger
avatar identification, but that user–avatar similarity may be more
specific to certain traits. But such speculation would need to be
subject to systematic investigation in future studies. We also note
that although personality predicted both behavioral and physical
similarity to avatars, only behavioral similarity predicted reports
of real life changes. We speculate further that the ease of changing
physical appearance is the reason it is less predictive of real life
changes; although the SL environment may trigger frequent
appearance changes to their avatars among people with certain
personality dispositions, only the more lasting changes associated
with behaviors would tend to spill over into real life. Another fruit-
ful direction for future research would be to examine further the
impact of various dimensions of user–avatar similarity on the
boundary between people’s virtual and real lives (e.g., Koles, 2012).
Among the personality factors, Conscientiousness and Emo-
tional Stability were the strongest predictors of people’s likelihood
to report changes to their real lives as a result of their experiences
in Second Life. Most of the paths observed in the final model orig-
inated from these two personality factors, and the coefficients of
the paths from these two factors were higher than those from
the other personality factors. Conscientiousness was the strongest
predictor of user–avatar similarity, and Emotional Stability was the
strongest predictor of emotional aspects of people’s SL experiences
(i.e., low emotional involvement with the environment and low
identification with the avatar). These emotional aspects of SL expe-
riences in turn were the strongest direct predictors of real life
changes, and they fully mediated the effects of Emotional Stability.
The results from this sample thus suggest that high stability may
be associated with relatively low emotional investment in SL, thus
creating a buffer from real life.
Meanwhile, Conscientiousness appears to exert its effects pri-
marily through the degree of resemblance between people and
their avatars. High similarity to the avatar may be an expedient
way to express the conscientious motivation for transparency
and honesty. At the same time, we also found a modest but signif-
icant direct path indicating that Conscientiousness tends to pre-
vent changes to real life. The role of Conscientiousness in
regulating people’s boundaries between virtual and real worlds
would be a useful avenue to pursue in future research, particularly
related to people’s ability to separate online taboos (e.g., violence,
immoral acts) from their real lives (e.g., Whitty, Young, &
Goodings, 2011; Young & Whitty, 2010).
The effects for Extraversion, Intellect and Agreeableness were
not as strong, but the observed paths for these personality factors
were nevertheless consistent with the hypotheses. As expected,
people high in Extraversion tended to have high similarity to their
avatars. For the extraverted person, the avatar is the means of
expressing characteristics related to sociability that are triggered
by the essentially social nature of Second Life. At the same time,
-.11*** (H9a)
Identification
with Avatar
.12** (H7a)
-.09* (H8)
.05* (H4a)
.08** (H6a)
.07*(H5)
.29* (H1)
.39*** (H2)
Extraversion
Conscientiousness
Emotional
Stability
Intellect
Real Life
Changes
-.10*** (H9c)
Agreeableness
Behavioral
Similarity
Physical
Similarity
Emotional
Involvement
Absorption
.50*** (H1)
.11* (H7a)
-.04* (H9c)
.08*(H5)
.21** (H3)
Fit Statistics
χ2 (23) = 20.24, p = .62
CFI = 1.00
GFI = .98
RMSEA = 0.00
SRMR = 0.04
Fig. 2. Final model. Note:
*
p< .05;
**
p< .01;
***
p< .001.
P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70 67
introverted people will be motivated to create avatars that are
somewhat more extraverted than themselves, largely due to gener-
alized social pressures that favor extraversion (Morgeson, Reider, &
Campion, 2005). The similarity will be highest for people who actu-
ally are extraverts, which explains the positive effects we observed.
Intellect positively impacted real life changes through strong
emotional involvement with virtual world experiences. This pat-
tern is also consistent with expectations based on trait activation
theory that the unstructured nature of Second Life combined with
the availability of multiple tools for creating new content activates
the propensity of people high in Intellect toward creativity and
exploration. Emotional engagement can be reasonably expected
to result from the ability to express these personality facets. Mean-
while, however, we did not see evidence for the more cognitive
effects related to being absorbed in the virtual world experiences.
Although the observed results for Agreeableness were consis-
tent with the hypotheses, the results for this factor were generally
weak. Agreeableness related only to emotional involvement with
the environment but not to absorption, and the predicted effect
on identification with the avatar was not found. Keeping in mind
that the results of the trait activation validation analysis were also
weak for this personality factor, this suggests that the Second Life
environment may not activate Agreeableness in the way we
argued. Trait activation theory offers a possible explanation. Tett
and Burnett (2003) argued that situation strength affects the predic-
tiveness of personality. Situation strength refers to environmental
characteristics that exert strong demands – through high rewards
or punishments – for manifesting certain behaviors. In SL, the
rewards for friendliness are high and reasonably predictable,
therefore creating a strong environmental demand for Agreeable-
ness. Low variation across individuals in the expression of this trait
would be expected as a result, and Agreeableness would therefore
not be strongly predictive of behavior differences.
Further explanations for the failure to find support for some of
the hypothesized relationships suggests directions for possible
future research on how different virtual environment features
may activate different facets of personality. We speculate that
one reason we found the predicted effects of Conscientiousness
on avatar similarity but not on identification with the avatar or
on relationship to the environment may be due to a difference
among our three virtual world experience factors along a dimen-
sion of external signaling. That is, avatar appearance and behavior
are easily visible external signs related to people’s virtual world
experiences whereas their degree of attachment to the environ-
ment and identification with the avatar are internal states and
therefore not directly observable. We argue further that Conscien-
tiousness (and Extraversion, which we will discuss presently) is
manifested by a set of relatively specific observable behaviors,
compared to Agreeableness, Intellect and Emotional Stability. For
example, Conscientiousness is typically associated with the behav-
iors of planning, organizing, and completing tasks (Jackson et al.,
2010), whereas Emotional Stability is defined in terms of the inter-
nal state of calmness (cf., John & Srivastava, 1999). Although it is
possible to observe that someone is calm, in contrast to Conscien-
tiousness there are not specific behaviors associated with that
state. Like Conscientiousness, the trait of Extraversion tends to be
defined in terms of specific observable behaviors, such as talkative-
ness and activeness (Funder & Sneed, 1993).
Our finding that Extraversion predicted the observable environ-
mental aspect of avatar similarity but not the internal aspects
could be explained with similar reasoning. We thus speculate that
environmental experiences associated with internal states (e.g.,
identification with avatars) may more strongly activate personality
traits also associated with internal states (e.g., Emotional Stability),
and that environmental experiences associated with specific
observable behaviors (e.g., avatar appearance) may more strongly
activate traits associated with observable behaviors (i.e., Conscien-
tiousness and Extraversion). This speculation is consistent with
arguments derived from trait activation theory that predictiveness
of personality may be improved by focusing on narrow facets of
the Big Five traits (e.g., Tett & Burnett, 2003), and points to the
need for further research on the interaction of specific virtual envi-
ronmental features and the expressions of personality traits.
6.1. Limitations
First, the direction of causality and the assumption of linearity
in the model we tested are major questions to consider. Although
we are on firm ground in modeling the personality factors as
exogenous variables (cf., Landers & Lounsbury, 2006), and we feel
confident that the virtual world experience variables must precede
real life changes of those experiences, we also recognize that over
time the changes that users experience in their lives will eventu-
ally affect their perceptions and experiences in the virtual environ-
ment. For example, there is growing interest in factors leading to
users’ practices of reproducing real life experiences and events in
virtual environments (e.g., Lomanowska & Guitton, 2014). It would
be important in future research to examine more closely the direc-
tionality of influence between people’s material and virtual lives,
and the effects of individual difference factors. Second, our data
do not allow us to differentiate between positive and negative
changes, and thus another fruitful direction for research would
be to examine that difference.
Finally, there was limited variance in our sample on the motiva-
tions for joining Second Life and the way people spent their time.
The sample consisted of voluntary users with pleasure-oriented
more than task-oriented motivations. This is the most likely
explanation for our finding that these two factors had very little
effects on individuals’ in-world experiences or impact on real life.
Examining differences in motivations may reveal variations in
the patterns of personality trait activation, given that people’s
motivational orientation may affect their sensitivity to and
interpretation of environmental cues (cf., Tett & Burnett, 2003).
Despite these limitations, the results of this study show mecha-
nisms by which personality can explain differences in people’s
experiences in virtual environments, and add to the growing
evidence that people’s Internet experiences have implications for
life offline.
Acknowledgements
The authors gratefully acknowledge Mary Ellen Gordon for
extensive help in the design and conduct of the study, analysis of
the data, and helpful comments on the manuscript. We thank Mike
Shapiro for helpful comments on the manuscript.
References
Adams, W. L. (2008). UK couple to divorce over affair in Second Life. Time,
Friday, November 14, 2008. <http://content.time.com/time/world/article/
0,8599, 1859231,00.html>.
Alsina-Jurnet, I., & Gutiérrez-Maldonado, J. (2010). Influence of personality and
individual abilities on the sense of presence experienced in anxiety triggering
virtual environments. International Journal of Human-Computer Studies, 68,
788–801. http://dx.doi.org/10.1016/j.ijhcs.2010.07.001.
Amichai-Hamburger, Y., Wainapel, G., & Fox, S. (2002). ‘‘On the Internet no one
knows I’m an introvert’’: Extroversion, neuroticism, and internet interaction.
CyberPsychology & Behavior, 5, 125–128. http://dx.doi.org/10.1089/
109493102753770507.
Bagozzi, R. B., Dholakia, U. M., & Pearo, L. R. K. (2007). Antecedents and
consequences of online social interactions. Media Psychology, 9, 77–114.
http://dx.doi.org/10.1080/15213260709336804.
Bargh, J. A., & McKenna, K. Y. A. (2004). The internet and social life. Annual Review
of Psychology, 55, 573–590. http://dx.doi.org/10.1146/annurev.psych.55.
090902.141922.
68 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
Bartholow, B. D., Sestir, M. A., & Davis, E. B. (2005). Correlates and consequences of
exposure to video game violence: Hostile personality, empathy, and aggressive
behavior. Personality and Social Psychology Bulletin, 31, 1573–1586. http://
dx.doi.org/10.1177/0146167205277205.
Behm-Morawitz, E. (2013). Mirrored selves: The influence of self-presence in a
virtual world on health, appearance, and well-being. Computers in Human
Behavior, 29, 119–128. http://dx.doi.org/10.1016/j.chb.2012.07.023.
Bell, M. W. (2008). Toward a definition of ‘‘virtual worlds’’. Journal of Virtual Worlds
Research, 1(1). http://dx.doi.org/10.4101/jvwr.v1i1.283.
Biocca, F., Harms, C., & Burgoon, J. K. (2003). Towards a more robust theory and
measure of social presence: Review and suggested criteria. Presence:
Teleoperators and Virtual Environments, 12, 456–480. http://dx.doi.org/
10.1162/105474603322761270.
Boellstorff, T. (2008). Coming of age in Second Life: An anthropologist explores the
virtually human. Princeton, NJ: Princeton University Press.
Church, A. T. (1994). Relating the Tellegen and five-factor models of personality
structure. Journal of Personality and Social Psychology, 67, 898–909. http://
dx.doi.org/10.1037/0022-3514.67.5.898.
Church, A. T. (2000). Culture and personality: Toward an integrated cultural trait
psychology. Journal of Personality, 68, 651–703. http://dx.doi.org/10.1111/1467-
6494.00112.
Correa, T., Hinsley, A. W., & de Zúñiga, H. G. (2010). Who interacts on the Web? The
intersection of users’ personality and social media use. Computers in Human
Behavior, 26, 247–253. http://dx.doi.org/10.1016/j.chb.2009.09.003.
Dunn, R. A., & Guadagno, R. E. (2012). My avatar and me: Gender and personality
predictors of avatar-self discrepancy. Computers in Human Behavior, 28, 97–106.
http://dx.doi.org/10.1016/j.chb.2011.08.015.
Evans, S. (2012). Virtual selves, real relationships: An exploration of the context and
role for social interactions in the emergence of self in virtual environments.
Integrative Psychological and Behavioral Science, 46, 512–528. http://dx.doi.org/
10.1007/s12124-012-9215-x.
Fox, J., Bailenson, J., & Binney, J. (2009). Virtual experiences, physical behaviors:
The effect of presence on imitation of an eating avatar. Presence: Teleoperators
and Virtual Environments, 18, 294–303. http://dx.doi.org/10.1162/pres.
18.4.294.
Funder, D. C., & Sneed, C. D. (1993). Behavioral manifestations of personality: An
ecological approach to judgmental accuracy. Journal of Personality and Social
Psychology, 64, 479–490. http://dx.doi.org/10.1037/0022-3514.64.3.479.
Gilbert, R. L., Murphy, N. A., & Ávalos, B. A. (2011). Communication patterns and
satisfaction levels in three-dimensional versus real-life intimate relationships.
Cyberpsychology, Behavior, and Social Networking, 14, 585–589. http://dx.doi.org/
10.1089/cyber.2010.0468.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. American
Psychologist, 48, 26–34. http://dx.doi.org/10.1037/0003-066X.48.1.26.
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory
measuring the lower-level facets of several five-factor models. Personality
Psychology in Europe, 7, 7–28. <http://projects.ori.org/lrg/PDFs_papers/
A%20broad-bandwidth%20inventory.pdf>.
Graham, L. T., & Gosling, S. D. (2013). Personality profiles associated with different
motivations for playing World of Warcraft. Cyberpsychology, Behavior, and Social
Networking, 16, 189–193. http://dx.doi.org/10.1089/cyber.2012.0090.
Guitton, M. J. (2012). The immersive impact of meta-media in a virtual world.
Computers in Human Behavior, 28, 450–455. http://dx.doi.org/10.1016/
j.chb.2011.10.016.
Hamburger, Y. A., & Ben-Artzi, E. (2000). The relationship between extraversion and
neuroticism and the different uses of the Internet. Computers in Human
Behavior, 16, 441–449. http://dx.doi.org/10.1016/S0747-5632(00)00017-0.
Hartmann, T., Toz, E., & Brandon, M. (2010). Just a game? Unjustified virtual
violence produces guilt in empathetic players. Media Psychology, 13, 339–363.
http://dx.doi.org/10.1080/15213269.2010.524912.
Hassouneh, D., & Brengman, M. (2014). A motivation-based typology of social
virtual world users. Computers in Human Behavior, 33, 330–338. http://
dx.doi.org/10.1016/j.chb.2013.08.012.
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the
new millennium. Communication Monographs, 76, 408–420. http://dx.doi.org/
10.1080/03637750903310360.
Hou, J., Nam, Y., Peng, W., & Lee, K. M. (2012). Effects of screen size, viewing angle,
and players’ immersion tendencies on game experience. Computers in Human
Behavior, 28, 617–623. http://dx.doi.org/10.1016/j.chb.2011.11.007.
Jackson, J. J., Wood, D., Bogg, T., Walton, K. E., Harms, P. D., & Roberts, B. W. (2010).
What do conscientious people do? Development and validation of the
Behavioral Indicators of Conscientiousness (BIC). Journal of Research in
Personality, 44, 501–511. http://dx.doi.org/10.1016/j.jrp.2010.06.005.
Jin, S. A. (2011). ‘‘It feels right. Therefore, I feel present and enjoy’’: The effects of
regulatory fit and the mediating roles of social presence and self-presence in
avatar-based 3D virtual environments. Presence: Teleoperators and Virtual
Environments, 20, 105–116. http://dx.doi.org/10.1162/pres_a_00038.
John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History,
measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.),
Handbook of personality: Theory and research (2nd ed., pp. 102–138). New York:
Guilford Press.
Kline, R. B. (2011). Principles and practice of structural equation modeling. New York:
Guilford Press.
Koles, B., & Nagy, P. (2012). Who is portrayed in Second Life: Dr. Jekyll or Mr. Hyde?
The extent of congruence between real life and virtual identity. Journal of Virtual
Worlds Research, 5(1). http://dx.doi.org/10.4101/jvwr.v5i1.2150.
Kolotkin, R. A., Williams, M. M., Lloyd, C., & Hallford, E. W. (2012). Does loving an
avatar threaten real life marriage? Journal of Virtual Worlds Research, 5.<http://
journals.tdl.org/jvwr/index.php/jvwr/article/view/6296>.
Laarni, J., Ravaja, N., Saari, T., & Hartmann, T. (2004). Personality-related differences
in subjective presence. In M. Alcañiz & B. Rey (Eds.), Proceedings of the seventh
annual international workshop presence 2004 (pp. 88–95). Valencia: Ed. UPV.
Landers, R. N., & Lounsbury, J. W. (2006). An investigation of Big Five and narrow
personality traits in relation to Internet usage. Computers in Human Behavior, 22,
28–293. http://dx.doi.org/10.1016/j.chb.2004.06.001.
Lee, K. M. (2004). Presence, explicated. Communication Theory, 14, 27–50. http://
dx.doi.org/10.1111/j.1468-2885.2004.tb00302.x.
Linares, K., Subrahmanyam, K., Cheng, R., & Guan, S. A. (2013). A second life within
Second Life: Are virtual world users creating new selves and new lives?
International Journal of Cyber Behavior, Psychology and Learning, 1, 50–71. http://
dx.doi.org/10.4018/ijcbpl.2011070104.
Liu, X., & LaRose, R. (2008). Does using the Internet make people more satisfied with
their lives? The effects of the Internet on college students’ school life
satisfaction. CyberPsychology & Behavior, 11, 310–320. http://dx.doi.org/
10.1089/cpb.2007.0040.
Lomanowska, A. M., & Guitton, M. J. (2014). My avatar is pregnant! Representation
of pregnancy, birth, and maternity in a virtual world. Computers in Human
Behavior, 31, 322–331. http://dx.doi.org/10.1016/j.chb.2013.10.058.
Lombard, M., & Ditton, T. (1997). At the heart of it all: The concept of presence.
Journal of Computer-Mediated Communication, 3.http://dx.doi.org/10.1111/
j.1083-6101.1997.tb00072.x.
Malaby, T. M. (2009). Making virtual worlds: Linden Lab and Second Life. New York,
NY: Cornell University Press.
Market Truths, Ltd. (2008). Psychographic segments and media consumption.
Unpublished Technical Report.
Market Truths, Ltd. (2009). Psychographic segments and media consumption.
Unpublished Technical Report.
McCrae, R. R. (1994). Openness to experience: Expanding the boundaries of Factor
V. European Journal of Personality, 8, 251–272. http://dx.doi.org/10.1002/
per.2410080404.
McCrae, R. R., & Costa, P. T. Jr., (1987). Validation of the five-factor model of
personality across instruments and observers. Journal of Personality and Social
Psychology, 52, 81–90. http://dx.doi.org/10.1037/0022-3514.52.1.81.
McCrae, R. R., Jang, K. L., Ando, J., Ono, Y., Yamagata, S., Riemann, R., et al. (2008).
Substance and artifact in the higher-order factors of the big five. Journal of
Personality and Social Psychology, 95, 442–455. http://dx.doi.org/10.1037/0022-
3514.95.2.442.
McCreery, M. P., Krach, S. K., Schrader, P. G., & Boone, R. (2012). Defining the virtual
self: Personality, behavior and the psychology of embodiment. Computers in
Human Behavior, 28, 976–983. http://dx.doi.org/10.1016/j.chb.2011.12.019.
McLeod, P. L., & Leshed, G. (2011). As long as they don’t know where I live:
Information disclosure strategies for managing identity in Second Life. In A.
Peachy & M. Childs (Eds.), Reinventing ourselves: Contemporary concepts of
identity in virtual worlds (pp. 191–211). London, UK: Routledge.
Messinger, P. R., Ge, X., Stroulia, E., Lyons, K., Smirnov, K., & Bone, M. (2008). On the
relationship between my avatar and myself. Journal of Virtual Worlds Research,
1(1). http://dx.doi.org/10.4101/jvwr.v1i1.291.
Messinger, P. R., Stroulia, E., Lyons, K., Bone, M., Niu, R. H., Smirnov, K., et al. (2009).
Virtual worlds – past, present, and future: New directions in social computing.
Decision Support Systems, 47, 204–228. http://dx.doi.org/10.1016/
j.dss.2009.02.014.
Michikyan, M., Subrahmanyam, K., & Dennis, J. (2014). Can you tell who I am?
Neuroticism, extraversion, and online self-presentation among young adults.
Computers in Human Behavior, 33, 179–183. http://dx.doi.org/10.1016/
j.chb.2014.01.010.
Mischel, W. (1977). The interaction of person and situation. In D. Magnusson & N.
Endler (Eds.), Personality at the crossroads: Current issues in interactional
psychology (pp. 333–352). Hillsdale, NJ: Erlbaum.
Morgeson, F. P., Reider, M. H., & Campion, M. A. (2005). Selecting individuals in team
settings: The importance of social skills, personality characteristics, and
teamwork knowledge. Personnel Psychology, 58, 583–611. http://dx.doi.org/
10.1111/j.1744-502005.655.x.
Nagy, P., & Koles, B. (2014). The digital transformation of human identity: Towards a
conceptual model of virtual identity in virtual worlds. Convergence: The
International Journal of Research into New Media Technologies.http://dx.doi.org/
10.1177/1354856514531532 (Advance online publication).
Pavot, W., Diener, E., & Fujita, F. (1990). Extraversion and happiness. Personality and
Individual Differences, 11, 1299–1306. http://dx.doi.org/10.1016/0191-
8869(90)90157-M.
Persky, S., & Blascovich, J. (2007). Immersive virtual environments versus
traditional platforms: Effects of violent and nonviolent video game play.
Media Psychology, 10, 135–156. http://dx.doi.org/10.108/15213260701301236.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: A critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88, 879–903. http://
dx.doi.org/10.1037/0021-9010.88.5.879.
Preacher, K. J., & Hayes, A. F. (2008). Contemporary approaches to assessing
mediation in communication research. In A. F. Hayes, M. D. Slater, & L. B. Snyder
(Eds.), The Sage sourcebook of advanced data analysis methods for communication
research (pp. 13–54). Thousand Oaks, CA: Sage Publications.
Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011). Mediation analysis
in social psychology: Current practices and new recommendations. Social and
P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70 69
Personality Psychology Compass, 5, 359–371. http://dx.doi.org/10.1111/j.1751-
9004.2011.00355.x.
Slater, M. (1999). Measuring presence: A response to the Witmer and Singer
presence questionnaire. Presence: Teleoperators and Virtual Environments, 8,
560–565. http://dx.doi.org/10.1162/105474699566477.
Steen, F., Davies, M., Tynes, B., & Greenfield, P. (2006). Digital dystopia: Player
control and strategic innovation in the sims online. In R. Schroeder & A. S.
Axelsson (Eds.), Avatars at work and play (pp. 247–273). Netherlands: Springer.
Sung, Y., & Moon, J. H. (2011). Actual self vs. avatar self: The effect of online social
situation on self-expression. Journal of Virtual Worlds Research, 4(1). http://
dx.doi.org/10.4101/jvwr.v4i1.1927.
Tamborini, R., & Skalski, P. (2006). The role of presence in the experience of
electronic games. In P. Vorderer & J. Bryant (Eds.), Playing video games: Motives,
responses, and consequences (pp. 225–240). Mahwah, N.J: Lawrence Erlbaum
Associates.
Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of
job performance. Journal of Applied Psychology, 88, 500–517. http://dx.doi.org/
10.1037/0021-9010.88.3.500.
Tett, R. P., & Guterman, H. A. (2000). Situation trait relevance, trait expression, and
cross-situational consistency: Testing a principle of trait activation. Journal of
Research in Personality, 34, 397–423. http://dx.doi.org/10.1006/jrpe.2000.2292.
Trepte, S., & Reinecke, L. (2010). Avatar creation and video game enjoyment. Journal
of Media Psychology: Theories, Methods, and Applications, 22, 171–184. http://
dx.doi.org/10.1027/1864-1105/a000022.
Turel,O., & Zhang, Y. (2010). Does virtualteam compositionmatter? Trait and problem-
solving configuration effects on team performance. Behaviour & Information
Technology, 29, 363–375. http://dx.doi.org/10.1080/01449291003752922.
Van Looy, J., Courtois, C., De Vocht, M., & De Marez, L. (2012). Player identification in
online games: Validation of a scale for measuring identificationin MMOGs. Media
Psychology, 15, 197–221. http://dx.doi.org/10.1080/15213269.2012.674917.
Verhagen, T., Feldberg, F., van den Hooff, B., Meents, S., & Merikivi, J. (2012).
Understanding users’ motivations to engage in virtual worlds: A multipurpose
model and empirical testing. Computers in Human Behavior, 28, 484–495. http://
dx.doi.org/10.1016/j.chb.2011.10.020.
Weibel, D., Wissmath, B., & Mast, F. W. (2010). Immersion in mediated
environments: The role of personality traits. Cyberpsychology, Behavior, and
Social Networking, 13, 251–256. http://dx.doi.org/10.1089/cyber.2009.0171.
Whitty, M. T., Young, G., & Goodings, L. (2011). What I won’t do in pixels: Examining
the limits of taboo violation in MMORPGs. Computers in Human Behavior, 27,
268–275. http://dx.doi.org/10.1016/j.chb.2010.08.004.
Wiederhold, B. K. (2013). Avatars: Changing behavior for better or for worse?
Cyberpsychology, Behavior, and Social Networking, 16, 319–320. http://dx.doi.org/
10.1089/cyber.2013.1517.
Wirth, W., Hartmann, T., Böcking, S., Vorderer, P., Klimmt, C., Schramm, H., et al.
(2007). A process model of the formation of spatial presence experiences. Media
Psychology, 9, 493–525. http://dx.doi.org/10.1080/15213260701283079.
Wirth, W., Hofer, M., & Schramm, H. (2012). The role of emotional involvement and
trait absorption in the formation of spatial presence. Media Psychology, 15,
19–43. http://dx.doi.org/10.1080/15213269.2011.648536.
Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: A
presence questionnaire. Presence: Teleoperators and Virtual Environments, 7,
225–240. http://dx.doi.org/10.1162/105474698565686.
Wolfendale, J. (2007). My avatar, myself: Virtual harm and attachment. Ethics and
Information Technology, 9, 111–119. http://dx.doi.org/10.1007/s10676-006-
9125-z.
Yee, N., Ducheneaut, N., Nelson, L., & Likarish, P. (2011). Introverted elves &
conscientious gnomes: The expression of personality in World of Warcraft. In
Proceedings of the SIGCHI conference on human factors in computing systems
(CHI’11) (pp. 753–762). doi:10.1145/1978942.1979052.
Young, G., & Whitty, M. T. (2010). Games without frontiers: On the moral and
psychological implications of violating taboos within multi-player virtual
spaces. Computers in Human Behavior, 26, 1228–1236. http://dx.doi.org/
10.1016/j.chb.2010.03.023.
70 P.L. McLeod et al. / Computers in Human Behavior 39 (2014) 59–70
