Examining self congruence between user and avatar in purchasing behavior from the metaverse to the real world

Journal of Global Fashion Marketing

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/rgfm20

Examining self-congruence between user and
avatar in purchasing behavior from the metaverse
to the real world

Jinsu Park & Naeun (Lauren) Kim

To cite this article: Jinsu Park & Naeun (Lauren) Kim (2024) Examining self-congruence
between user and avatar in purchasing behavior from the metaverse to the real world, Journal
of Global Fashion Marketing, 15:1, 23-38, DOI: 10.1080/20932685.2023.2180768
To link to this article: />
Published online: 03 Apr 2023.
Submit your article to this journal
Article views: 1953
View related articles
View Crossmark data
Citing articles: 9 View citing articles

Full Terms & Conditions of access and use can be found at
/>
JOURNAL OF GLOBAL FASHION MARKETING
2024, VOL. 15, NO. 1, 23–38
/>
Examining self-congruence between user and avatar in
purchasing behavior from the metaverse to the real world

Jinsu Park and Naeun (Lauren) Kim

Department of Design, Housing, and Apparel, College of Design, University of Minnesota, Minneapolis,
MN, USA

ABSTRACT ARTICLE HISTORY
Received 6 October 2022
The aim of this study was to examine the role of “avatar” in the Revised 3 January 2023
metaverse, specifically, by exploring how appearance similarity with Accepted 5 February 2023
an avatar affects virtual product purchase intention, which is
mediated by self-congruence and flow. Further, this study also KEYWORDS
attempted to find how human behavior in the virtual world could Metaverse; avatar;
affect behavior in the real world. An online survey was administered self-congruency; flow; virtual
to a total of 164 users of a popular South Korean metaverse plat­ product purchase intention
form ZEPETO. The findings indicated that the appearance similarity
with the avatar influenced virtual product purchase intention, and
this relationship was fully mediated by avatar-self congruence and
flow. This study also found that virtual product purchase intention
could influence real product purchase intention. Based on these
findings, it is suggested high-quality avatar customization is critical
for users, and retail brands can utilize metaverse as a new market­
ing tool and as a testbed for new products.

1. Introduction

In the disastrous pandemic of COVID-19, many activities that had traditionally been
conducted face-to-face made a major shift to online. As a result, the metaverse has gained
considerable attention due, in part, to the increased popularity of non-face-to-face
activities. The “metaverse” is a virtual world extending the physical universe where
various activities occur (Papagiannidis et al., 2008). The concept of the metaverse is
not new as many video games have taken the form of the metaverse. In addition,
advances in 3D virtual programs and hardware technology have introduced a more

realistic metaverse world. For example, the “Animal Crossing” game has recently been
described as a representative metaverse platform. After Nintendo of Japan developed and
released Animal Crossing in March 2020, 32.63 million units of Animal Crossing were
sold worldwide as of 31 March 2021 (Bankhurst, 2021). In this game, players can build
and decorate their towns through various hunting-gathering activities. Animal Crossing
is considered a metaverse platform particularly because of its interactive features. People
can visit other players’ towns, and engage and interact with other users, such as sharing,
trading, and communicating (Zhu, 2021). According to the PricewaterhouseCoopers

CORRESPONDENCE TO Naeun (Lauren) Kim Department of Design, Housing, and Apparel,
College of Design, University of Minnesota, 240 McNeal Hall, 1985 Buford Ave. St. Paul, Minneapolis, MN 55108-61367, USA

© 2023 Korean Scholars of Marketing Science

24 J. PARK AND N. (L.) KIM

(PwC) global consulting firm, the size of the metaverse market was valued at
$148.5 billion dollars in 2021 and is estimated to grow to $15,423 billion dollars by
2030 (Hwang, 2021). Therefore, retail industries have paid considerable attention to this
new media and are trying to utilize the metaverse as a new retail channel (Bourlakis et al.,
2009).

Virtual channel such as metaverse is also being utilized by traditional high-end
companies to promote their products and gain capital, and this trend has accelerated
during the COVID-19 pandemic (Loranger & Roeraas, 2022). A popular metaverse
platform, “ZEPETO”, was created by Naver in South Korea in 2018. Two-hundred-
million people worldwide have used this platform, and 80% of the users are from
generation Z (Lancaster, 2021). The global luxury brand Gucci made a “Gucci Villa” in
ZEPETO, allowing users to visit their virtual space. Users can try on virtual Gucci
products and can buy a virtual Gucci product for their avatar, which is much cheaper

than a real (i.e. tangible) product (Tamola, 2021). This trend illustrates that entering the
metaverse is beneficial for product advertisement and can be quite lucrative.

What makes the metaverse special compared to other media outlets is the virtual
interaction between users. All of the interactions are accomplished through a digitalized,
psychological being called an avatar, which represents the digital self of the user. These
avatars are interconnected with each other as they could be in the real world but with
fewer physical limitations (e.g. time and space). Digital-self avatars have been studied in
video game contexts, and video game players are known to feel self-congruence with their
avatars, which subsequently affects the players’ perception and behavior in the digital
world (Ko & Park, 2020). The role of the avatars in the metaverse is similar to those of
video games such as portraying users in the virtual world and helping users to immerse in
the virtual world. Thus, self-congruence with avatar and flow should be discussed in the
metaverse, but these concepts have not been studied extensively.

The biggest difference between video games and the metaverse comes from the
proximity to the real world. Unlike video games, the metaverse is highly attached to
the real world. It is not hard to find a virtual version of the product in the metaverse,
which exist in the real world. For example, global luxury brands such as Gucci, Polo
Ralph Lauren, and Givenchy have launched their digital form of products in the
metaverse, and most of them are less expensive than the real world counterparts (Bein,
2022). The affordability of virtual products makes people easy to purchase the products,
bringing advantages to both consumers and retailers. Virtual products provide consu­
mers greater access to tangible products by virtually trying on a product on their avatar
and, in succession, create the chance of increasing brand recognition and loyalty (Miles,
2022). However, few studies have focused on how consumer behavior in the metaverse
impacts the real world (Belk, 2016; Papagiannidis et al., 2014). In addition, while several
studies on virtual products have identified consumer behavior in the “virtual reality”
realm (Barhorst et al., 2021), few studies have been conducted in the metaverse where the
focus is interaction features.

The purpose of this quantitative study is twofold. First, based on the extended-self
theory, this study aims to analyze the effect of the appearance similarity between users
and their avatars on self-congruence and understand how such congruence between self
and avatar increases the flow state of the user while playing in the metaverse. The second
aim is how these effects, derived from the relationship between the avatar and the user,

JOURNAL OF GLOBAL FASHION MARKETING 25

promote purchase behavior intention for their avatar in the metaverse. We also examine
how virtual interaction and virtual purchase behavior influence purchase behavior
intention for the user in the real world.

The findings of this study provide a deeper understanding of the nature of the
metaverse which remains as a nascent field of research for its stakeholders (i.e. users,
retailers, and scholars). Since the development and commercialization of the metaverse
are still in the early stage, studies on the metaverse are largely lacking. Our study will fill
this gap in the literature and, specifically, will illuminate how the user’s perception of
their avatar develops under the extended-self theory, comparing it with the previous
studies on video games. This study will also reveal how self-congruence with the avatar
can affect purchase behavior in both the virtual world and the real world. In addition, to
the best of the authors’ knowledge, this is the first study to examine the underlying
mechanism behind the effect of appearance similarity with an avatar on user behavior.
Furthermore, the findings will provide implications to the fashion retail brands who are
questioning the marketability of the metaverse and suggest ways in which the metaverse
can be effectively utilized as an additional retail channel.

2. Literature review and hypotheses development

2.1. Metaverse

The word “metaverse” was first coined by Neal Stephenson in 1992 and appeared in his
novel, “Snow Crash”, stating that the metaverse is a new version of future virtual reality
(Stephenson, 1993). Malaby (2006) later described the metaverse as a synthetic world
where users can create content by themselves and own the intellectual property rights of
the virtual products they make. Papagiannidis et al. (2008) suggested a broader concept,
defining the metaverse as a virtual world extending the physical universe by adding new
dimensions and domains for economic, social, and leisure activities. Dionisio et al. (2013)
conceptualized the metaverse as a more technological concept, describing it as a fully
immersive 3D digital environment in contrast to cyberspace, which is more inclusive and
reflects the totality of shared online space across all dimensions. Most recently, Sparkes
(2021) took a broader view, defining the metaverse as a shared online space that
integrates 3-dimensional graphics on a screen or in virtual reality. These changes in the
definition indicate that people initially considered the metaverse as simply a “synthetic
world”. It was then expanded to a broader concept referring to a “shared online space”.
The definition has become broader rather than concrete because it was believed that
unique technologies would be used to implement the metaverse.

Although the definition of the metaverse has continued to change, several prominent
features of the metaverse have been maintained. Ubiquity is a central component of the
metaverse (Dionisio et al., 2013), which comes from the major criterion that the
metaverse must provide an environment for culture and interaction like in the physical
world (Dionisio et al., 2013). Gilbert (2011) also maintained that the metaverse should
support massive multi-user remote interactivity simultaneously and be immersive, giving
a sense of psychological presence. Tools and functions should also be provided to
generate activities, which share the same goals as an open-ended setting rather than
a specific quest (Gilbert, 2011). Based on these perspectives, we infer that an essential

26 J. PARK AND N. (L.) KIM

factor in the metaverse is the interaction among users with their own personas (i.e.
psychological presence), and this interactivity is accomplished through their extended
self, the avatar.

2.2. Extended self and avatar-self congruence

Traditionally, the extended self can be comprised of a person’s body, mind, physical
possession, and close affiliation group, including family and friends (Belk, 1988).
However, with the advent of information technology (IT), there have been more ways
of representing “self” than ever and, especially, the virtual world has been stimulating the
way of expressing oneself with the digital version of myself, an avatar (Belk, 2016). Belk
(2016) asserted that the concept of extended self needs to be modified because of
digitalization, and one of the changes is re-embodiment, which concretely expresses
the concept of an avatar in the digital world. In a digital age, people can be emancipated
from their physical bodies and be whatever they want to be. In addition, the relative
freedom of creating an avatar has led people to think that the avatar they made represents
themselves, and when people perceive a gap between the self in the real world between
that in the virtual world, they tend to modify their self-representation which is called
“proteus effects” (Fox et al., 2013; Yee & Bailenson, 2007). Interestingly, this self-
discrepancy between virtual and real can have effects on offline behavior as well. For
example, a person who has an attractive avatar behaves warmly and is more social in real
life (Yee et al., 2009).

An avatar is a digital representation of the user through which the user interacts with
and relates to others in the virtual environment (Hooi & Cho, 2014). In most virtual
worlds, users create an avatar or customize their character as they want to express
themselves (Sung et al., 2011), and almost every interaction has to do with the digital
vicarious self, the avatar (Hollebeek et al., 2020). According to Belk (2016), regardless of
the level of immersion, users are attached to their avatars, and users become their avatars
while playing games, where avatar-self congruence occurs. Self-congruence was devel­

oped to explain the way in which consumers tend to choose a brand that is congruent
with their self-concept (Malär et al., 2011). Because of the malleability in describing how
people behave in a way to keep the consistence of their self-image, this concept has been
used in a wide range of studies, including video game fields. According to Davis and Lang
(2013), who studied the effect of self-congruence with video game itself on game usage
and purchase behavior, they found that people who perceive high self-congruence with
the game they play are more likely to play and purchase games. Similarly, Ko and Park
(2020) who examined the effect of self-congruence with the avatar in video game context,
suggested that people choose an avatar that is congruent with their self, as consumers
tend to choose a brand or product that accords with their self-image.

Self-congruence with an avatar begins to develop when the user first creates their
avatar. Avatar creation is often one of the very first tasks users are given in the metaverse.
During this process, a user’s main consideration is the physical appearance of the avatar,
and one of the most available references would be their own physical traits in shaping the
avatar’s appearance. This phenomenon has been clearly observed in several studies.
Nowak and Rauh (2005) noted that people tend to prefer avatars that align with their
own gender and choose avatars featuring characteristics that are similar to their own.

JOURNAL OF GLOBAL FASHION MARKETING 27

Williams (2010) also suggested that users are more likely to identify with avatars that are
physically similar to the users.

Once the avatar creation is done, the next step is to meet and interact with other users’
avatars in the metaverse. As users increasingly interact with other avatars and spend time
in the metaverse, the importance and attachment to their own avatars grow. Avatars help
users carry out all activities in the virtual world, including communication, game playing,
and transactions. When performing these activities, avatars will exhibit the user’s persona
such as personality, communication style, and behavioral traits in addition to physical

qualities (Procter, 2021). This, in turn, makes the users further reduce the gap between
the self and avatar. Such a phenomenon can also be explained by self-verification theory.
Self-verification theory refers to a tendency to maintain a consistent self-concept, avoid­
ing psychological anarchy (Swann, 1987). Therefore, users may perceive their avatar in
a way to lessen the inconsistency between the avatar and the self, and this unconscious
effort is inextricably related to interaction with other users. Given that the self-
congruence effect is often attributed to the underlying motivation of socialness (Davis
& Lang, 2013), the more people interact with other users in the metaverse, the greater
self-congruence with the avatar could occur. Thus, the following hypothesis is developed.

H1: Appearance similarity with one’s avatar is positively related to avatar-self
congruence.

2.3. Flow

Given the nature of the metaverse that provides users with an immersed experience, it is
crucial to understand the flow experience in the metaverse. Technology-mediated flow
refers to a situation in which people are fully immersed, absorbed, and engaged by
mechanical interactions, in the meantime, losing self-consciousness and experiencing
enjoyment (Barhorst et al., 2021). Flow provides an experience of holistic immersion for
users (Ha et al., 2007). The metaverse can induce a flow state in two ways. The first way
stems from the equipment. Sherry (2004) asserted that the flow occurs when players
ignore the real world and increase concentration through visual and auditory stimuli.
The perfect form of the metaverse requires various equipment including headgear that
blocks the environment around the users. Users cannot see the real world or hear sounds
nearby, but they can see the virtual world and hear the sounds that are digitally created.
However, few metaverse platforms now need this 3D virtual equipment. Many metaverse
platforms can be played with a small digital device with a screen such as a mobile phone,
tablet, or computer that only allows limited immersion (Wiederhold, 2022). Therefore,
this study did not consider the flow induced by 3D equipment. Flow can also come from

the content that users participate in and enjoy. Activities in the metaverse mainly focus
on leisure activities such as decorating their avatar, playing with other users in the virtual
world, and visiting famous places that are hard to visit in the real world. These activities
deeply immerse users in the metaverse. In addition, according to Barhorst et al. (2021)
study, the novelty of the metaverse also increases the flow, and users step into a flow state
when playing in the metaverse.

To understand the metaverse, one must examine studies related to games, since the
metaverse shows similar characteristics to games in that users are involved in the

28 J. PARK AND N. (L.) KIM

metaverse or the game with their digital representatives: their avatars. The similarity with
the avatar is highly related to attention and concentration (Liao et al., 2019). In addition,
as users identify with a character, their consciousness is transported into the avatar they
manipulate, making users lose self-reflection, which is an aspect of flow (Jennett et al.,
2008). Furthermore, since the extent of self-congruence affects people’s level of attention
(Ding & Hung, 2021), the extent of flow may be determined by the level of congruence
with the avatar. Therefore, the following hypothesis is developed.

H2: Self-congruence with the avatar is positively related to flow.

When people fall into a flow state, it results in specific human behavior or
stimulates a certain intention. For example, shopping in the virtual environment
could induce a state of flow, making people stay in the virtual store and facilitating
purchase behavior (Bleize & Antheunis, 2019). Domina et al. (2012) found that
enjoyment and control, which are major constructs of flow are positively related to
purchase intention in the virtual world. This effect of a flow state on purchase
intention can also be found in the game field. Jin et al. (2017) asserted that people
buy virtual products (i.e. items) due to social presence and user engagement,

which are some of the main attributes of the flow. Specifically, people who are
in a flow state are more likely to purchase virtual game items (Liu & Shiue, 2014).
Goli and Vemuri (2021) also found that a flow experience in a mobile game has
a positive effect on virtual game items. Thus, it is plausible to infer that the flow
experience can influence the virtual product purchase intention in the metaverse.
Therefore, the following hypothesis was developed.

H3: Flow is positively related to virtual product purchase intention.

Given the fact that the metaverse is a three-dimensional virtual world, avatar
and flow must be crucial factors in explaining the purchasing behavior in the
metaverse. As described earlier, an avatar is considered a digital version of self in
a digital world, and the extent of homogeneity with an avatar influences human
perception and intention. Thus, it can be inferred that the more users perceive
their avatars as similar to themselves, the more they are likely to dive into the
virtual world and buy virtual products to decorate their avatars, which, in turn,
makes users identify their avatars more synchronized with themselves.
Furthermore, considering the self-verification theory, people attempt to verify
and enhance their self-concept, and one of the easiest to do this is to purchase
items that could help achieve these goals (Stuppy et al., 2020). In video games,
people buy virtual products which make their virtual characters more powerful
(Domina et al., 2012). However, compared to video games, buying virtual products
does not directly enhance the ability of their avatar in the metaverse. In the
metaverse, where virtual characters’ ability to compete with other users and
villains is not necessary, the role of virtual products is to decorate the avatar to
express their identity, which is closely associated with social motivation. As Davis
and Lang (2013) asserted, underlying self-motives of socialness is one of the

JOURNAL OF GLOBAL FASHION MARKETING 29

functions of self-congruence. In other words, in the metaverse, where the trait of
“interactivity” is stressed, the virtual item is a tool for maintaining self-
congruence, which, in turn, makes users buy virtual products. As Belk (1988)
asserted, the concept of “self” can be defined by what they possess, and this also
can be applied to the virtual world. The behavior of possessing virtual products is
a way of representing and defining their avatar, implying the user’s self-concept.
Therefore, this study hypothesizes that appearance similarity with the avatar has
a positive effect on virtual product purchase intention through self-congruence
with avatar and flow successively.

H4: Self-congruence with avatar and flow serially mediate the relationship between
appearance similarity and virtual product purchase intention.

2.4. Purchase intention (virtual product and real product)

One of the distinct features of the metaverse from video games is that the
metaverse is highly related to the real world. The metaverse is a virtual world
that is extended from the real world (Papagiannidis et al., 2008). In other words,
users’ behavior in the metaverse can affect their real life and vice versa. For
instance, Papagiannidis et al. (2014), who identified the impact of virtual driving
experience on automobiles in the real world, revealed that people who experienced
a simulated product were willing to buy the real-world product. Nevertheless, to
the best of our knowledge, studies that have found a direct connection between the
virtual and the real worlds are scarce. However, several industry cases have
indicated a strong relevance between the two worlds. Louis Vuitton’s limited
edition (physical) products worn by virtual characters in the League of Legends
game were all sold out in less than an hour after launch (Phelps, 2021). After
Balenciaga released a brand showcase in the virtual game Afterworld, the number
of searches for the brand rose by 41% within 48 hours (Tashjian, 2021). Given this,
we can infer that consumer behavior in the virtual world is somewhat related to

the real world. Therefore, the following hypothesis is developed. Figure 1 shows
the conceptual framework of the study.

H5: Virtual product purchase intention is positively related to real product pur­
chase intention.

Self-congruence H2

with Avatar Flow

H1 H3
H4

Appearance Virtual product H5 Real product
Similarity with
Purchase Intention Purchase Intention
Avatar

Figure 1. Conceptual framework for the study.

30 J. PARK AND N. (L.) KIM

3. Methodology

3.1. Samples

To capture the experience of actual metaverse users, this study recruited people who had
prior experience in the popular metaverse platform, ZEPETO. ZEPETO is one of the
most famous metaverse platforms with 200 million users worldwide (Lancaster, 2021).
Since ZEPETO was developed and is operated by a Korean company, Naver, we recruited

participants from South Korea. A total of 164 participants were recruited through
a survey panel of a market research company, and participants were given a voucher
for USD $2.00 as compensation for their participation. All of the participants were in
their 20s and 30s, representing generation Z and millennials, who are the major users of
the metaverse (Locke, 2021). Among the 162 participants, 132 were female, and 32 were
male, and the age range was 15 to 39 years (M = 27.97, SD = 6.36) (see Table 1).

3.2. Procedures and measurements

An online survey questionnaire was used, and all survey questions were translated and
checked by two bilingual researchers. In the initial stage of the survey, participants
were asked whether they had used ZEPETO and had created their own avatars. Then,
several survey items related to ZEPETO were presented to stimulate their memories of
using ZEPETO, such as the date of their first use, the number of their avatars, the
amount of time they spent creating their avatars, and their experience of participating
in the virtual event in ZEPETO. Participants then proceeded to the main survey
items.

The main survey measurement items were borrowed from previously validated scales.
The survey items were modified from existing measures to fit the context of this study

Table 1. Demographic profiles. n %

Gender 131 19.1
Female
Male 31 80.9

Age 19 11.7
10s
20s 71 44.1

30s
72 44.2
Annual income
Less than $10,000 49 30.2
$10,000–$30,000
$30,000 - $50,000 39 24.1
$50,000 - $70,000
$70,000 - $90,000 53 32.7
Over $90,000
15 9.3
ZEPETO usage
Once a week 3 1.9
Twice a week
Three times a week 3 1.9
Frequently used but not using it anymore
37 22.8
Experience in buying a virtual item in ZEPETO
Never 15 9.3
Have experience
12 7.4
Total
98 60.5

89 54.9

73 45.1

162 100.0

JOURNAL OF GLOBAL FASHION MARKETING 31

Table 2. Measurement items and exploratory factor analysis results.

Items FL CA CR AVE

Appearance similarity .870 .912 .723

My avatar in ZEPETO is as close to my real self as can be made. .915

My avatar’s body is similar to that of my real self. .896

My avatar in ZEPETO is generally recognizable as my real self. .815

My avatar’s style of clothes is similar to that of my real self. .767

Self-congruence with avatar .823 .894 .739

The avatar in ZEPETO is consistent with how I see myself. .888

The avatar in ZEPETO is a mirror image of me. .865

The avatar in ZEPETO is close to my own personality. .824

Flow .828 .897 .744

I was deeply engrossed when I played ZEPETO. .910

I was absorbed in the experience when I played ZEPETO. .907

My attention was not focused on the experience. (Reversed) .763

Virtual product purchase intention .907 .907 .764

I am likely to consider purchasing virtual items to decorate my avatar in ZEPETO. .902

I am likely to consider purchasing virtual items (e.g. clothes, accessories) for my avatar in .889

ZEPETO.

I would like to recommend the virtual items I bought for my avatar to my friends who know .830

ZEPETO.

Real product purchase intention .857 .874 .698

If the virtual products I have purchased for my avatar are available in the real world . . .

The experience in ZEPETO would increase my intention to buy the real items I used in the .864

virtual world.

I would consider buying real products for myself. .825

The experience in ZEPETO would help me make a purchase decision about real products. .817

FL = Factor loadings, CA = Cronbach’s alpha, CR = Composite reliability, and AVE = Average variance extracted.

(see Table 2). Four items related to appearance similarity with their avatar, including
body, style of clothing, and overall look (Hooi & Cho, 2013). The scales for self-
congruence with the avatar included three items (Ko & Park, 2020; V. Kumar &

Kaushik, 2022). Three items were used to measure virtual product purchase intention
by modifying items from Huang (2012) and Cha (2011). Three items were used to gauge
their flow state when using ZEPETO (Barhorst et al., 2021). For real product purchase
intention, three items were used from N. Kumar et al. (2022) and Papagiannidis et al.
(2014). All items were rated on a 7-point Likert scale ranging from 1 (strongly disagree)
to 7 (strongly agree).

4. Results

4.1. Reliability and validity

Before the main analysis, an exploratory factor analysis (EFA) was conducted using SPSS
28. All values of factor loadings exceeded the recommended threshold of 0.7 (Hair et al.,
2010), confirming indicator validity. Internal consistency was checked with Cronbach’s
alphas of the constructs, and the values exceeded the recommended threshold of 0.7. In
addition, convergent validity was confirmed with average variance extracted (AVE), and
the values exceeded the threshold of 0.5 (see Table 2). Discriminant validity was also
established by comparing the square roots of the AVE values with the correlation values
(Fornell & Larcker, 1981) (see Table 3).

32 J. PARK AND N. (L.) KIM

Table 3. Correlation matrix.

Variables Mean SD 1 2 3 4 5

Appearance similarity 3.81 1.33 .850

Self-congruence with avatar 3.64 1.35 .747** .860

Virtual product purchase intention 3.93 1.54 .404** .508** .863

Real product purchase intention 4.16 1.38 .371** .534** .592** .874

Flow 3.72 1.38 .389** .521** .698** .578** .835

The values of the lower triangles of the matrix are the correlation coefficients between the constructs. The diagonal
values (italics values) represent the square root of the average variance extracted of each construct.

4.2. Measurement model

Partial least square (PLS) modeling through the SmartPLS 3.0 software was used for the
data analysis as this study is exploratory in nature (Hair et al., 2011). A total of 164
responses were collected and two insincere responses were discarded from the data. The
minimum number of participants is 20 for PLS-SEM with ten times the largest number of
structural paths directed from a particular construct (Hair et al., 2011). Therefore, 162
samples were enough for PLS-SEM analysis. The significance of the path coefficients was
tested using a nonparametric bootstrapping procedure, and the entire model fit was
tested with the indices of SRMR and R2. The recommended SRMR should be lower than
0.08 (Hu & Bentler, 1998), and this model showed a value of 0.067, being considered
a good model fit. In addition, all the R2 values of each endogenous construct are greater
than the recommended threshold of 0.1 (Falk & Miller, 1992).

The analysis revealed that appearance similarity with the avatar significantly influ­
enced self-congruence with the avatar (β = .760, p < .001), supporting H1, and self-
congruence with the avatar significantly influenced the flow (β = .521, p < .001), support­
ing H2. Sequentially, flow significantly influenced virtual product purchase intention (β
= .645, p < .001), hence, H3 was supported. Virtual product purchase intention signifi­
cantly influenced real product purchase intention (β = .595, p < .001), supporting H5 (see
Figure 2). The analysis results also revealed that self-congruence with the avatar and flow

fully mediated the relationship between appearance similarity with the avatar and virtual
product purchase intention. Besides self-congruence with the avatar and flow, appear­
ance similarity with the avatar also significantly influenced virtual product pur­
chase intention (β = .255, p < .001); however, the indirect effect of appearance
similarity on virtual product purchase intention was not significant (β = .149, p
= .087), when including self-congruence with the avatar and flow in the model,
supporting H4 (see Figure 2).

H1 Self-congruence H2 Flow H3
.760*** with Avatar .521*** R2 = .266 .645***

Appearance R2 = .575 H4 Virtual product
Similarity with Purchase Intention
H5 Real product
Avatar .595*** Purchase Intention

.149n.s

R2 = .350 R2 = .510

Figure 2. PLS results of the conceptual model. n.s. means not significant.

JOURNAL OF GLOBAL FASHION MARKETING 33

5. Discussions

This study aimed to understand the role of avatars in the metaverse, and how self-
congruence with the avatar and flow can be a bridge to virtual product purchase
intention. This study also examined the possibility of an extension of purchase intention
from the metaverse to the real world. To do so, this study conducted a survey to gauge the

perceptions of a famous metaverse platform, ZEPETO users.

First, this study found a positive effect of avatar similarity on self-congruence with the
avatar. The more people perceive that their avatar is identical to their own appearance, the
more they perceive self-congruence with their avatar. As Jin (2010) suggested that people
feel more connected when their avatar is their ideal instead of their real image, our findings
imply that appearance similarity can also enhance such connection. Second, users are more
likely to be immersed in the metaverse when they perceive that their avatar’s self-concept
accords with that of the real self. This study supports previous studies showing significant
causality between self-congruence with the avatar and flow (Ding & Hung, 2021). Third,
this study also found that the flow experience in the metaverse could prompt the user to
buy virtual products for their avatar. This finding is consistent with Syahrizal et al. (2020)
who identified the drivers of virtual item purchases in video games. Importantly, we
discovered that while the appearance similarity with the avatar does not directly affect
virtual product purchase intention, it does so indirectly by allowing users to perceive self-
congruence with their avatars, thereby increasing their immersion (i.e. flow) into the
metaverse world. Lastly, to the best of our knowledge, this study is the first attempt to
show that consumer behavior in the virtual world can extend to behavior in the real world.
Specifically, people who intend to buy virtual products for their avatars are more likely to
buy the real versions of the products for themselves.

5.1. Theoretical implications

This study suggests how the researchers understand the metaverse itself and the
dynamics in it. Several studies to date have examined consumer behavior in the virtual
world, especially through video games, and many psychological concepts have been
adopted including self-congruence with the avatar (Ko & Park, 2020), flow (Li & Lwin,
2016), and behavior related to virtual products (Wu & Hsu, 2018). Yet, the literature
surrounding the metaverse is still scarce. In filling this literature gap, this study attempted
to provide a deeper understanding of the metaverse by establishing the definition and

applying the concepts from the previous studies to this nascent field of research. Our
findings verified that most concepts from video game-related studies work similarly in
the metaverse. First, both video games and metaverse are constructed in the virtual world
in which users can be easily immersed. Second, all activities are accomplished by their
digital form of avatar and users perceive their avatar as their representations, feeling
homogeneity with it. Third, the users purchase virtual products (items) which give users
specific benefits. Nonetheless, discrepancies between video games and the metaverse also
exist. The findings of our study imply that the extent of appearance similarity may play
a greater role in the metaverse compared to video games. It may be because, in the
metaverse, virtual items are much more realistic in their appearances and consist of items
from the real world such as t-shirts, hoodies, hats, etc. Therefore, users in the metaverse

34 J. PARK AND N. (L.) KIM

are able to perceive similarity with their avatars more so than in video games, which, in
turn, boosts self-congruence with their avatar and flow. Another distinguishing facet of
the metaverse, as discovered in our study, is the connection of purchasing behavior
between the virtual and real world. Studies have yet to show how the consumption of
virtual items in video games influences the real world; yet, according to our findings, the
purchasing behavior in the metaverse can be extended easily to that of the real world.

Moreover, this study used the “extended self” theory by Belk (2016) to understand the role
of an avatar as a digital version of the self in the metaverse. This study revealed the self-
concept transformation, that is, appearance similarity with the user would be developed into
self-congruence with the avatar. This finding is meaningful because whether people make
their avatars identical to their actual appearance or to their ideal image has been debated.
Hence, this study confirms Belk’s (2016) notion that the identification with one’s digital self
can be increased with an avatar who is similar to the owner. In addition, as Belk (2016)
suggested that the experience of using an avatar in the virtual world has an effect in real life,
this study attests to the “proteus effects” by finding the link between the real world and the

virtual world. Specifically, to the best of the authors’ knowledge, this study is the first attempt
to demonstrate the possibility that purchase intention in the metaverse can influence that of
the real world. This link should be examined further in aspects other than consumption.

5.2. Managerial implications

The findings of this study also have several implications for metaverse developers and retail
brands. For developers, it is crucial to allow users to engage in high-quality avatar customiza­
tion. The more people feel that their avatars are similar to their real appearance, the more
likely they are to buy virtual products. One of the reasons ZEPETO has been popular with
users is that there are numerous options (e.g. body shape and various facial features) they can
choose for their avatar, which is not common in other metaverse platforms. Thus, developers
should focus on creating sophisticated customization features.

Retail brands that want to utilize the metaverse as a new marketing tool can also find
marketing cues in this study. Many retail brands, especially apparel brands, are already
using the metaverse as a retail channel to roll out virtual and real-world products. They
have created virtual versions of products that have the exact same appearance as real
products in the metaverse. This study shows that users’ purchase intention of virtual
products can lead to purchase intention of the real versions of the products in the real
world. Since the virtual versions of the products are far more affordable than the real
versions, users can easily access new luxury products in the virtual world. Brands could also
use the metaverse as a testbed for new products. If users show high interest in the newly
launched virtual products, brands can interpret this success as a sign to manufacture real
versions of the products. Moreover, brands can combat the industry-wide problem of
product returns by allowing users to virtually try on items in their digital boutiques. Given
the user-avatar appearance similarity, sizing and fit issues can be alleviated, and brands can
encourage customers to buy items with greater confidence. Lastly, the metaverse can be
used as a virtual brand community where users interact with the brand and with other
consumers. Through the highly interactive and immersive nature of the metaverse, users

can experience the brand in an exciting way (e.g. watch a virtual fashion show, play
branded games), in turn, increasing brand familiarity and loyalty.

JOURNAL OF GLOBAL FASHION MARKETING 35

5.3. Limitations and future studies

This study has several limitations. The first limitation stems from the metaverse platform
we used. We recruited participants who had experience using ZEPETO. While ZEPETO
is a popular metaverse platform, there are other major metaverse platforms in the market,
and each offer slightly different features (e.g. the degree of avatar customization, con­
tents, etc.) Although metaverse platforms have common features such as the use of
avatars and interaction between users, the quality of avatar customization, and the goal
of each metaverse platform are different. To strengthen the generalizability, we recom­
mend recruiting users from various metaverse platforms and even the non-users to
understand their perceptions and experiences. The second limitation concerns the
demographics of the sample. The number of female participants was much larger than
male participants, and the sample was limited to users in South Korea. Since metaverse
platforms are enjoyed by males and females alike and from around the world, we
encourage future studies to increase the diversity of the sample demographics. Third,
this study examined the path from virtual to real product purchase intention. However,
future research should consider the opposite direction. For instance, people who have
purchased real products may also consider buying virtual versions of the products for
their digital self (i.e. an avatar) in the metaverse. The bi-directional nature of the purchase
intentions warrants further examination. Finally, even though this study only considered
the “appearance” (i.e. physical) similarity between the user and the avatar, additional
dimensions of similarity, such as personal characteristics should be considered in a future
study (Hooi & Cho, 2014).

Disclosure statement

No potential conflict of interest was reported by the author(s).

ORCID

Jinsu Park />Naeun (Lauren) Kim />
References

Bankhurst, A. (2021, October 19). Animal crossing: New horizons is another case of Nintendo
stopping support too soon. IGN. /> -still-huge-and-nintendo-shouldnt-stop-updating-it-so-soon

Barhorst, J. B., McLean, G., Shah, E., & Mack, R. (2021). Blending the real world and the virtual
world: Exploring the role of flow in augmented reality experiences. Journal of Business Research,
122, 423–436. />
Bein, K. (2022, July 11). Five luxury brands with explorable space in Roblox. Gotham. https://
gothammag.com/ralph-lauren-gucci-nike-tommy-hilfiger-givenchy-roblox

Belk, R. (1988). Possessions and the extended self. The Journal of Consumer Research, 15(2),
139–168. />
Belk, R. (2016). Extended self and the digital world. Current Opinion in Psychology, 10, 50–54.
/>
36 J. PARK AND N. (L.) KIM

Bleize, D. N., & Antheunis, M. L. (2019). Factors influencing purchase intent in virtual worlds:
A review of the literature. Journal of Marketing Communications, 25(4), 403–420. https://doi.
org/10.1080/13527266.2016.1278028

Bourlakis, M., Papagiannidis, S., & Li, F. (2009). Retail spatial evolution: Paving the way from
traditional to metaverse retailing. Electronic Commerce Research, 9(1), 135–148. /> 10.1007/s10660-009-9030-8

Cha, J. (2011). Exploring the internet as a unique shopping channel to sell both real and virtual
items: A comparison of factors affecting purchase intention and consumer characteristics.
Journal of Electronic Commerce Research, 12(2), 115. /> 12_2_p01.pdf

Davis, R., & Lang, B. (2013). Does game self‐congruity increase usage and purchase? Young
Consumers, 14(1), 52–66. />
Ding, H. M., & Hung, K. P. (2021). The antecedents of visitors’ flow experience and its influence
on memory and behavioral intentions in the music festival context. Journal of Destination
Marketing & Management, 19, 100551. />
Dionisio, J. D. N., Iii, W. G. B., & Gilbert, R. (2013). 3D virtual worlds and the metaverse: Current
status and future possibilities. ACM Computing Surveys (CSUR), 45(3), 1–38. /> 1145/2480741.2480751

Domina, T., Lee, S. E., & MacGillivray, M. (2012). Understanding factors affecting consumer
intention to shop in a virtual world. Journal of Retailing and Consumer Services, 19(6), 613–620.
/>
Falk, R. F., & Miller, N. B. (1992). A primer for soft modeling. University of Akron Press.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable

variables and measurement error. Journal of Marketing Research, 18(1), 39–50. /> 10.1177/002224378101800104
Fox, J., Bailenson, J. N., & Tricase, L. (2013). The embodiment of sexualized virtual selves: The
proteus effect and experiences of self-objectification via avatars. Computers in Human Behavior,
29(3), 930–938. />Gilbert, R. L. (2011). The PROSE project: A program of in-world behavioral research on the
metaverse. Journal of Virtual Worlds Research, 4(1), 3–18. /> 2108
Goli, M., & Vemuri, V. V. (2021). Users’ in-game purchase intention: The effects of flow
experience and satisfaction. Journal of Electronic Commerce in Organizations, 19(4), 1–19.
/>Ha, I., Yoon, Y., & Choi, M. (2007). Determinants of adoption of mobile games under mobile
broadband wireless access environment. Information & Management, 44(3), 276–286. https://
doi.org/10.1016/j.im.2007.01.001
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of
Marketing Theory and Practice, 19(2), 139–151. />Hair, J. F., William, C. B., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th

ed.). Prentice Hall.
Hollebeek, L. D., Clark, M. K., Andreassen, T. W., Sigurdsson, V., & Smith, D. (2020). Virtual
reality through the customer journey: Framework and propositions. Journal of Retailing and
Consumer Services, 55, 102056. />Hooi, R., & Cho, H. (2013). Deception in avatar-mediated virtual environment. Computers in
Human Behavior, 29(1), 276–284. />Hooi, R., & Cho, H. (2014). Avatar-driven self-disclosure: The virtual me is the actual me.
Computers in Human Behavior, 39, 20–28. />Hu, L. T., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to
underparameterized model misspecification. Psychological Methods, 3(4), 424. /> 10.1037/1082-989X.3.4.424
Huang, E. (2012). Online experiences and virtual goods purchase intention. Internet Research, 22
(3), 252–274. />
JOURNAL OF GLOBAL FASHION MARKETING 37

Hwang, E. Y. (2021, July 1). Are you over meme stocks? How about a few metaverse plays? Korea
JoongAng Daily. />
Jennett, C., Cox, A. L., Cairns, P., Dhoparee, S., Epps, A., Tijs, T., & Walton, A. (2008). Measuring
and defining the experience of immersion in games. International Journal of Human-Computer
Studies, 66(9), 641–661. />
Jin, S. A. A. (2010). I feel more connected to the physically ideal mini me than the mirror-image
mini me: Theoretical implications of the “malleable self” for speculations on the effects of avatar
creation on avatar–self connection in Wii. Cyberpsychology, Behavior and Social Networking, 13
(5), 567–570. />
Jin, W., Sun, Y., Wang, N., & Zhang, X. (2017). Why users purchase virtual products in
MMORPG? An integrative perspective of social presence and user engagement. Internet
Research, 27(2), 408–427. />
Ko, D. W., & Park, J. (2020). I am you, you are me: Game character congruence with the ideal self.
Internet Research, 31(2), 631–634. />
Kumar, N., Garg, P., & Singh, S. (2022). Pro-environmental purchase intention towards
eco-friendly apparel: Augmenting the theory of planned behavior with perceived consumer
effectiveness and environmental concern. Journal of Global Fashion Marketing, 13(2), 134–150.
/>
Kumar, V., & Kaushik, A. K. (2022). Engaging customers through brand authenticity perceptions:

The moderating role of self-congruence. Journal of Business Research, 138, 26–37. https://doi.
org/10.1016/j.jbusres.2021.08.065

Lancaster, A. (2021, September 15). Four new metaverse players: Are you in? Banyan Hill. https://
banyanhill.com/4-new-metaverse-players/

Li, B. J., & Lwin, M. O. (2016). Player see, player do: Testing an exergame motivation model based
on the influence of the self avatar. Computers in Human Behavior, 59, 350–357. /> 10.1016/j.chb.2016.02.034

Liao, G. Y., Cheng, T. C. E., & Teng, C. I. (2019). How do avatar attractiveness and customization
impact online gamers’ flow and loyalty? Internet Research, 29(2), 349–366. /> 1108/IntR-11-2017-0463

Liu, H. J., & Shiue, Y. C. (2014). Influence of Facebook game players’ behavior on flow and
purchase intention. Social Behavior and Personality: An International Journal, 42(1), 125–133.
/>
Locke, T., (2021, December 15). Millennials and Gen Z are planning to spend thousands on
cryptocurrency, NFTs and metaverse land as holiday gifts. CNBC. /> 12/15/millennials-gen-z-plan-to-spend-thousands-crypto-nfts-metaverse-land-as-holiday-
gifts.html

Loranger, D., & Roeraas, E. (2022). Transforming luxury: Global luxury brand executives’ percep­
tions during COVID. Journal of Global Fashion Marketing, 14(1), 1–15. /> 20932685.2022.2097938

Malaby, T. (2006). Parlaying value: Capital in and beyond virtual worlds. Games and Culture, 1(2),
141–162. />
Malär, L., Krohmer, H., Hoyer, W. D., & Nyffenegger, B. (2011). Emotional brand attachment and
brand personality: The relative importance of the actual and the ideal self. Journal of Marketing,
75(4), 35–52. />
Miles, S. (2022, July 11). 6 real-world metaverse benefits for multi-location brands. Street Fight.
/> brands/

Nowak, K. L., & Rauh, C. (2005). The influence of the avatar on online perceptions of anthro­
pomorphism, androgyny, credibility, homophily, and attraction. Journal of Computer-Mediated
Communication, 11(1), 153–178. />
Papagiannidis, S., Bourlakis, M. A., & Li, F. (2008). Making real money in virtual worlds:
MMORPGs and emerging business opportunities, challenges and ethical implications in
metaverses. Technological Forecasting and Social Change, 75(5), 610–622. /> 1016/j.techfore.2007.04.007

38 J. PARK AND N. (L.) KIM

Papagiannidis, S., See-To, E., & Bourlakis, M. (2014). Virtual test-driving: The impact of simulated
products on purchase intention. Journal of Retailing and Consumer Services, 21(5), 877–887.
/>
Phelps, N. (2021, December 9). Louis Vuitton’s new capsule with League of Legends brings French
high fashion to online gaming—and vice versa. VOGUE. /> vuittons-new-capsule-with-league-of-legends

Procter, L. (2021). I am/we are: Exploring the online self-avatar relationship. The Journal of
Communication Inquiry, 45(1), 45–64. />
Sherry, J. L. (2004). Flow and media enjoyment. Communication Theory, 14(4), 328–347. https://
doi.org/10.1111/j.1468-2885.2004.tb00318.x

Sparkes, M. (2021). What is a metaverse. New Scientist, 251(3348), 18. /> S0262-4079(21)01450-0

Stephenson, N. (1993). Snow crash. Bantam Books.
Stuppy, A., Mead, N. L., & Van Osselaer, S. M. (2020). I am, therefore I buy: Low self-esteem and

the pursuit of self-verifying consumption. The Journal of Consumer Research, 46(5), 956–973.
/>Sung, Y., Moon, J., & Lin, J. S. (2011). Actual self vs. avatar self: The effect of online social situation
on self-expression. Journal of Virtual Worlds Research, 4(1). /> 1927
Swann, W. B. (1987). Identity negotiation: Where two roads meet. Journal of Personality and Social
Psychology, 53(6), 1038–1051. />Syahrizal, A., Purwandari, B., Mishbah, M., & Dzulfikar, M. F. (2020, January). Buying the unreal:

Drivers of virtual item purchase in video games. Proceedings of the 3rd International Conference
on Software Engineering and Information Management, 203–209. /> 3378936.3378948
Tamola, K. (2021, February 8). Gucci offers consumers stylish avatars in gaming collaboration.
Luxury Daily. />Tashjian, R. (2021, December 2), Why is fashion so obsessed with the metaverse? GQ. https://www.
gq.com/story/metaverse-fashion-explainer
Wiederhold, B. K. (2022). Ready (or not) player one: Initial musings on the metaverse.
Cyberpsychology, Behavior and Social Networking, 25(1), 1–2. /> 2021.29234.editorial
Williams, K. D. (2010). The effects of homophily, identification, and violent video games on
players. Mass Communication and Society, 14(1), 3–24. /> 15205430903359701
Wu, S. L., & Hsu, C. P. (2018). Role of authenticity in massively multiplayer online role playing
games (MMORPGs): Determinants of virtual item purchase intention. Journal of Business
Research, 92, 242–249. />Yee, N., & Bailenson, J. (2007). The proteus effect: The effect of transformed self-representation on
behavior. Human Communication Research, 33(3), 271–290. /> 2958.2007.00299.x
Yee, N., Bailenson, J. N., & Ducheneaut, N. (2009). The Proteus effect: Implications of transformed
digital self-representation on online and offline behavior. Communication Research, 36(2), 285–
312. />Zhu, L.(2021). The psychology behind video games during COVID‐19 pandemic: A case study of
animal crossing: New horizons. Human Behavior and Emerging Technologies, 3(1), 157–159.
/>