Gamification in the Metaverse: How Design Attributes Shape User Preferences Across Age Groups

Abstract

We examine how gamification attributes shape user preferences for metaverse platforms and how these relationships vary across age groups. Using rank-ordered logit on 304 metaverse users from the Korean Media Panel Survey, we code platform features into four domains—character customization, experience/skill systems, social networking, and economic systems—and link them to stated preference rankings of leading services. Results show that realistic avatars and expressive behaviors are positively associated with preference, whereas complex body/environment customization is not. Within experience/skill systems, quest presence, content creation, and real-world–mirroring quests relate positively to preference, while excessive freedom/option breadth does not. In social networking, close interactions and group conversation capacity are valued, but rigid chat-window styles are not. Users also prefer low device dependency and real-world task utility. Age heterogeneity emerges: teens show stronger interest in appearance customization, whereas users in their twenties and thirties value mirroring quests, conversational freedom, and monetization. We provide design guidelines for segment-sensitive gamification and discuss implications for inclusive metaverse retail and service strategy.

1. Introduction

Immersive platforms have rapidly transformed the metaverse into a proving ground for retail and service innovation. Yet managers still face three implementation challenges when deploying gamification at scale: first, feature uncertainty—avatar customization, quest systems, social functions, and monetization are often adopted without clear evidence of their behavioral impact; second, segment misfit—preferences differ across demographic groups, but actionable, demographic-sensitive design rules remain scarce; and third, operational constraints—for example, device dependency can undermine consistent experiences across contexts. To address these gaps, we focus explicitly on user preferences (rather than satisfaction or decision stages) and link platform-level gamification attributes to ranked service preferences reported by metaverse users.

While the metaverse has been heralded for its commercial and societal promise, its conceptualization and empirical validation remain uneven. Prior work frequently treats metaverse environments as sophisticated online games, overlooking their user-driven, sociotechnical nature and the diversity of interaction and commerce they enable [1,2,3]. This limits our ability to identify which specific gamification attributes actually matter for users—and for whom. Moreover, little is known about whether the influence of these attributes varies by age and gender, two segmentation variables that are central to marketing practice.

This study directly examines how design attributes shape metaverse preferences and how these relationships differ across demographic segments. Building on recent discussions of avatar-mediated self-expression and social shopping—where pre-purchase exploration and post-purchase engagement are intertwined [4,5]—we classify observable platform features into four domains that recur across leading services: (1) character customization, (2) experience/skill systems, (3) social networking systems, and (4) economic systems. We then test how attributes within these domains relate to user preferences and whether effects are moderated by age and gender.

We study the Korean metaverse context, characterized by advanced digital infrastructure, rapid adoption of emerging technologies, and a vibrant gaming culture. Prominent domestic platforms such as ‘Zepeto’ and ‘ifland’ provide rich, real-world–inspired virtual environments in which gamification is salient and comparable across services. Using data drawn from the 2022 Korean Media Panel Survey, we analyze a subsample of 304 users who provided ranked preferences over 10 major metaverse services (listed in Appendix A Table A1). Platform attributes were coded into the four domains described above, and their associations with preference ranks were estimated using a rank-rank-ordered logit framework.

Our research is guided by the following questions:

• RQ1. Which gamification attributes characterize leading metaverse services?
• RQ2. How do attributes in character customization, experience/skill systems, social networking, and economic systems relate to user preferences?
• RQ3. Do these attribute–preference relationships differ by age?
• RQ4. Do these attribute–preference relationships differ by gender?

This paper makes three contributions. Conceptually, it aligns theorization and measurement by treating preference—not satisfaction or choice outcomes—as the focal dependent variable, and by organizing design features into a parsimonious, literature-grounded taxonomy. Empirically, it links coded platform attributes to ranked preferences using nationally collected panel data, quantifying which elements matter and for whom. Managerially, it offers segment-sensitive design implications for metaverse services—clarifying when to emphasize appearance customization versus progression/quest mechanics, how to structure social interaction affordances, and where economic features or device dependencies are likely to help or hinder adoption.

2. Literature Review

2.1. Metaverse and Digital Society

A metaverse—combining “meta” (transcendence) and “universe”—is a socio-technical service space in which virtual and real worlds interact and coevolve, enabling social, economic, and cultural activities that generate value [6]. Prior work reflects three core characteristics: (1) convergence of virtual and physical realities [2], (2) interaction among users, artifacts, and environments [7], and (3) value creation through production and consumption in digitally mediated contexts [8,9,10]. Convergence blurs the boundaries between online and offline life; interaction encompasses communication and shared experiences; and value creation reflects economic, social, and cultural exchanges at scale.

Unlike digital games that primarily target entertainment, metaverse services embed entertainment within broader social, cultural, and commercial exchanges. Brands and retailers, therefore, experiment with immersive marketplaces and staged events to differentiate experiences. Illustrative cases include the Travis Scott concert in Fortnite, drawing more than 12 million attendees to a virtual venue [11], and virtual idol groups achieving chart success in mainstream platforms [12]. Fashion and beauty brands have been particularly active: Nike’s Roblox activation paired avatars and NFTs with celebrity collaborations to market personalized virtual items [13]. Industry reports further anticipate growth in luxury-oriented virtual consumption—especially among younger users—via digital ownership and NFTs, with the metaverse economy projected to expand substantially by 2030.

As Generation Z spends more time in immersive environments [14], new payment and ownership logics reshape brand experiences and purchase pathways. At the same time, the metaverse reconfigures how consumers interact, learn, and enjoy digital experiences, foregrounding the need to identify which design features align with user needs and motivations. Building on this, the present study examines how gamification attributes of metaverse services relate to user preferences, and how these relationships differ by age and gender [4,5].

2.2. Understanding Gamification in the Metaverse

Metaverse platforms are often associated with gaming due to shared technology stacks (e.g., Unreal, Unity) and the prominence of services such as Roblox and Minecraft [15]. Yet they diverge from traditional games that emphasize mission completion for entertainment. Metaverse environments prioritize interactive, participatory experiences, enabling users to create and monetize digital assets, collaborate socially, and engage in diverse cultural practices [16,17].

In this context, gamification—the use of game-based mechanics and thinking to enhance engagement—has been shown to affect user involvement, loyalty, and preference across domains such as education, sustainability, retailing, and services [18,19,20,21]. Elements including points, badges, leaderboards, avatar customization, quests, and competition can structure challenges, provide feedback, and scaffold competence and relatedness, thereby sustaining participation [18]. Importantly, the goal is to craft a gameful experience—a psychologically resonant state—rather than to bolt on isolated mechanics [20,21].

Emerging evidence specific to metaverse contexts suggests that quests/challenges, social competition/cooperation, and user-generated creation can deepen engagement and clarify user motivations [22,23]. However, we still know relatively little about which concrete attributes of metaverse services are associated with user preferences, and for whom these effects are strongest. Addressing this gap, our study extends offline/online gamification findings into immersive, avatar-mediated settings, with explicit attention to age and gender as segmentation variables.

2.3. Gamification Components of the Metaverse: A Preference-Focused View

Guided by prior syntheses, we organize metaverse gamification into four domains—character customization, experience/skill systems, social networking systems, and economic systems—as a parsimonious taxonomy for linking observable design features to user preferences [24]. Below, we summarize each domain and its theorized relevance.

2.3.1. Character Customization

Avatars function as digital proxies that carry identity cues and enable self-expression, shaping how users interact with others and with commercial content. Alignment between an avatar’s representation and a user’s desired self can heighten immersion and support self-actualization [25]. Avatar imagery can also inform brand attitudes and product evaluations when visual relevance and identity fit are high [26]. Prior work indicates that greater customization—of appearance, body type, realism, and expressive behaviors—enhances identity salience and presence, fostering emotional attachment to virtual goods and brands [27,28]. Accordingly, we assess how specific customization attributes are associated with metaverse preferences in both individual and shared settings.

2.3.2. Experience and Skill System

Gamification frequently deploys quests and challenges that offer progression and competence signals, sustaining participation over time. Users’ self-efficacy—confidence in overcoming challenges—supports immersion and persistence [29]. In metaverse contexts, perceived realism and interactivity further encourage repeat visits, aided by items or mechanics that enhance experience/skills. Setting clear goals can increase enjoyment and, by extension, preferences for services that facilitate progress [30]. We therefore examine how quest presence and complexity, autonomy in goal achievement, content-creation potential, and task repetition relate to user preferences.

2.3.3. Social Network System

A distinctive affordance of the metaverse is its high-bandwidth sociality, enabling relationship formation and community building unconstrained by time and place [31]. Interaction mechanisms influence performance and positive affect in immersive settings [7]. In retail, customer interaction and community participation can strengthen loyalty and continued use, with downstream effects on preference and purchase [32]. We focus on Rigid chat-window style and freedom, frequency/scale of group interaction, and closeness of ties, positing that these features will be differentially valued across user segments and associated with preferences for particular platforms.

2.3.4. Economic System

Metaverse economies blur producer–consumer boundaries, enabling a participatory market in which users create, trade, and own digital assets. Core mechanisms include digital creation, digital assets, and digital trading markets [33]. Scholars describe this as a novel phase of the digital economy, linked to virtual currencies, digital legal tender, and asset-centered business models [34]. The rise of NFTs further integrates virtual and real economies [35]. In this study, we assess how monetization opportunities, device dependency, and the applicability of virtual activities to real-world tasks are associated with user preferences for metaverse services.

3. Methodology

3.1. Data and Analysis Sample

For this analysis, we used the Korean Media Panel Survey. We focused on the Korean metaverse market because it is supported by advanced digital infrastructure, rapid adoption of emerging technologies, and a strong gaming culture, making it an ideal context for examining gamification features and user preferences. This survey was initiated to measure cross-media usage behavior across various devices, platforms, and content packaging (combinations). It tracks changes in the media environment and usage behavior by accumulating long-term data from the same sample. Since 2010, the survey has collected extensive data on media device ownership and connectivity, broadcast communication service subscriptions and expenditures, and media usage behavior at both individual and household levels. Consequently, it provides a robust framework for designing a measurement methodology that reflects new broadcasting and communication environments and analyzing its impact on the media usage behavior of households and individuals.

In this study, we utilized responses regarding the use of metaverse services from the media usage survey conducted in 2022. Table 1 shows the items in the media usage questionnaire and their descriptions. First, the survey asked users whether they had used metaverse services. Those who had used metaverse services were then asked to rank their favorite services as 1, 2, or 3. The survey covered 11 metaverse services, encompassing almost all the metaverse platforms available at the time of the 2022 survey.

Table 1. Survey items related to the use of metaverse services in media usage.

Questionnaires	Item	Description
Have used a metaverse service	1	Yes
	2	None
1. Most used metaverse service (Main service)   2. Second Most used metaverse Service   3. Third Most used metaverse Service	1	Zepeto
	2	Animal Crossing
	3	Roblox
	4	Minecraft
	5	ifland
	6	Play Together
	7	Gather.town
	8	Second Life
	9	Metapolis
	10	Fortnite
	11	Zwift
	12	Etc.

When asked if they had used metaverse services, 304 people (approximately 3%) out of 9941 surveyed in 2022 responded affirmatively. This study aimed to investigate which attributes of metaverse services influence user preference-based choices. In this study, choices refer to the ranking of metaverse service options based on respondents’ stated preferences, as measured through a rank-ordered survey design, rather than observed behavioral selections. The distributions of the first, second, and third-most preferred metaverse services are listed in Table 2. The top three choices were ZEPETO (29%), Animal Crossing (28%), and Minecraft (19%). The second-tier preferences were Animal Crossing (25%), Minecraft (21%), Roblox (17%), and ZEPETO (11%). The third-tier preferences were Animal Crossing (28%), Play Together (20%), and ifland (11%). We excluded Zwift from the analysis because it was not selected in any of the top three positions, and Other (Etc.) did not receive many selections. Table 3 summarizes the sample characteristics by age and gender (n = 304).

Table 2. Distribution of choices by metaverse service rank (Unit: %).

Metaverse Service	1st	2nd	3rd
ZEPETO	29	11	7
Animal Crossing	28	25	28
Roblox	13	17	5
Minecraft	19	21	8
ifland	4	7	11
Play Together	2	9	20
Gather.town	2	1	2
Second Life	0	1	4
Metapolis	2	7	3
Fortnite	1	3	11
Zwift	0	0	0
Etc.	1	1	0

Table 3. Sample characteristics (n = 304).

(a) Age
Age group	n	(%)
Teens (13–19)	52	17.1
Twenties (20–29)	98	32.2
Thirties (30–39)	82	27.0
Forties (40–49)	48	15.8
50+	24	7.9
(b) Gender
Gender	N	(%)
Female	168	55.3
Male	136	44.7

To analyze the selection factors of metaverse services, we identified and extracted the characteristics of services from Zepeto to Fortnite. For this purpose, we conducted a qualitative study of all metaverse services and organized the attribute items that differentiate these services.

Table 4 presents the factors related to various forms of metaverse gamification. First, in character customization, we distinguished services based on whether users can individually adjust elements, including the character’s face, body, and background design. We also evaluated the characters’ realism and freedom of expression. Second, in the experience and skill system, we measured the difficulty of achieving in-game goals and the freedom to do so. This includes whether users can create their own content and how closely the quests in the service are aligned with real-world experiences. Third, the social network system examines how users communicate with each other within the metaverse, including aspects such as the format and frequency of conversations, the number of participants, and the freedom within the conversations. Finally, the economic system includes the monetization function, including whether items can be converted into virtual currency, the costs for devices or membership, and the extent to which the metaverse service can be used for real-world tasks.

Table 4. Attributes and factors of metaverse service gamification.

Main Category	Subcategories	Description	Coding	Details
Character customization	Creating a Character	Character Face	0: None 1: Yes	Availability of character face and hair design
		Character Body	0: None 1: Yes	Availability of character body type and skin color design
	Creating Character Environments	Character backgrounds Design	0: None 1: Yes	Whether users can design your character’s background, profile, and personal space
	Character Realism	Character Realism Reflectivity	0: low 1: Medium 2: High	Step-by-step assessment of the degree of anthropomorphism of character users 1: An animal-shaped or created object character that does not exist in reality 2: A character that looks like a person but has a body type, gender, skin color, etc. that does not reflect me in reality 3: A character that mirrors its own appearance
	Character Freedom	Freedom of character expression	0: low 1: Medium 2: High	How much character movement there is and Whether a variety of expressions can be expressed through the character’s movements
Experience and skill system	Difficulty of achieving the goal	Quest presence/absence	0: None 1: Yes	Whether or not the character has a defined purpose and story in the metaverse - Fortnite aims to level up and improve skills through repeated battle games - Bondee aims to maintain a free daily life that reflects reality without irritating elements
	Freedom to achieve purpose	Game freedom	0: Very low 1: Low 2: Normal 3: High 4: Very high	Whether the user can play a variety of games depending on their purpose and degree of freedom when playing the game - Minecraft allows you to play various characters and achieve goals without any special purpose or story. In particular, even watching the ending is unnecessary depending on the goal
	Content Creation	Whether to create content	0: None 1: Yes	A system that allows you to create your own content in the game without boundaries between producers and users
	Mirroring quest	Whether to mirror quest	0: Very low 1: Low 2: Normal 3: High 4: Very high	The degree to which the user’s actions or quest performance is far from reality - Zepeto: Dance challenge with your character and friends - Gather.town: Video conference, virtual space construction - Minecraft: gathering, exploration, design, etc.
Social network system	Rigid chat-window style	Chat window function	0: None 1: Yes	Ability to have a conversation using a chatbot or chat window - Animal Crossing: Chatbot, chatting with conversational keyboard
	Closeness of Conversation	Personal space	0: None 1: Yes	A way to invite friends to your space or chat alone with other users
	Frequency of Conversation Participation	Number of people Allowed to participate in conversation	0: 1 person 1: 10 people or more 2:30 people or more 3: 50 people or more 4: More than 100 people	The number of people who can participate in or check the actual conversation, not just one-on-one conversation with the character
	Conversation Freedom	Whether there is freedom of conversation	0: None 1: Yes	Whether it is possible to talk to people you meet for the first time virtually, excluding people you know or belong to the same group in real life
Economic system	Availability of monetization	Item Switch to virtual currency	0: None 1: Yes	Whether it can be converted to virtual currency through quest completion or items (currency, gems, gold bars, etc.)
	Device Dependency	Whether devices and paid services	0: None 1: Yes	Whether there are purchase costs for devices (game consoles, PCs) and membership registration to use the platform. - Animal Crossing operates as a paid membership system with device purchase - Zepeto and ifland can sign up and participate for free
	Real-world task utilization	Business function	0: None 1: Yes	- Metapolis and Kethetown are virtual offices, allowing for seminars and video conferences necessary for actual work. - Gather.town, Roblox has fun and functions through a pentagram-making game

3.2. Qualitative Coding of Platform Attributes

We conducted a structured content analysis of ten leading metaverse services (Appendix A Table A1). Building on gamification taxonomies (e.g., Schöbel, Janson, & Söllner, 2020 [24]), we developed a literature-grounded codebook across four domains—character customization, experience/skill systems, social networking systems, and economic systems—with operational definitions and binary/ordinal codings. Two trained coders independently coded all platforms following a calibration round on two services. Inter-rater reliability (Cohen’s κ/Krippendorff’s α) was assessed for each sub-attribute and met conventional thresholds (κ = 0.82 overall; 0.76–0.90 by domain; α = 0.85 overall; 0.78–0.92 by domain). Discrepancies were resolved via adjudication to produce a consensus coding used in the main analysis. The expanded codebook with example screenshots is provided in Appendix A. The attribute values assigned by the researchers based on metaverse service attributes are listed in Table 5.

Table 5. Attribute Table for Metaverse Service Gamification Factors.

Metaverse Service	Character Customizing					Gamification				Social Networking				Economic System
	Face	Body	Environment	Reality	Freedom	Quest	Freedom	Content creation	Mirrored quest	Text balloon	Close conversation	Group talk	Conversation freedom	Virtual currency	Device dependency	Productivity tool
Zepeto	1	1	1	2	2	1	2	1	3	1	1	2	1	1	0	1
ifland	1	0	1	2	1	1	1	1	3	1	1	3	1	1	0	2
Gather.town	1	0	0	1	0	0	1	1	1	1	1	2	0	0	0	3
Metapolis	0	0	0	1	0	0	0	1	2	0	0	3	0	0	0	2
Bondee	1	1	1	2	2	0	3	0	3	1	1	1	1	1	0	0
Roblox	1	1	0	0	2	2	3	1	0	1	0	3	1	1	1	0
Animal Crossing	1	1	1	0	2	2	3	0	0	1	1	0	1	1	1	0
Play Together	1	1	1	0	2	1	3	1	1	1	1	1	1	1	0	1
Minecraft	1	0	0	0	1	3	1	0	0	1	0	3	1	1	1	0
Fortnite	0	0	0	0	0	3	0	0	0	1	0	0	1	1	1	0

3.3. Rank-Ordered Logit Model

The rank-ordered logit model is a statistical model used to analyze ordered choices. Ordered categorical data are common in scenarios like preference surveys or preference ratings. A rank-ordered logit model leverages order information when a categorical dependent variable has an order.

The dependent variable is the preference-based choice of metaverse service, derived from ranked preference ratings in survey responses. We defined the utility function of a metaverse service user as shown in Equation (1). It represents how much the respondent (i) prefers each choice (j).

$$U_{ij}={\alpha }_j+X_i\beta +{\epsilon }_{ij}$$

(1)

where $U_{ij}$ is the utility of the i-th observation belonging to the j-th category, ${\alpha }_j$ is the threshold of the j-th category, $X_i$ is the vector of attribute variables of the metaverse service j, $\beta$ is the parameter vector of the attribute variables of metaverse j, and finally ${\epsilon }_{ij}$ is the error term. The probability of each metaverse service being selected can now be expressed as in Equation (2).

$$P\left(Y_i=j|X_i\right)=P({\alpha }_{j-1}+X_i\beta <U_{ij}\le {\alpha }_j+X_i\beta )$$

(2)

This can be expressed using a logit link function, as shown in Equation (3).

$$P\left(Y_i=j|X_i\right)={\displaystyle \frac{e^{({\alpha }_j+X_i\beta )}}{1+e^{({\alpha }_j+X_i\beta )}}}-{\displaystyle \frac{e^{({\alpha }_{j-1}+X_i\beta )}}{1+e^{({\alpha }_{j-1}+X_i\beta )}}}$$

(3)

The above equation describes the probability that a metaverse service is selected using the threshold and parameters that lead to the selection of an attribute alternative j. The parameters are estimated to interpret the results. The model estimation uses the maximum-likelihood method, with the likelihood function provided in Equation (3). The estimation process employs the Newton-Raphson method, which is used to find the maximum value of the log-likelihood function.

$$L\left(\alpha ,\beta |Y,X\right)={\prod }_{i=1}^n(P\left(Y_i=j|X_i\right)={\displaystyle \frac{e^{({\alpha }_j+X_i\beta )}}{1+e^{({\alpha }_j+X_i\beta )}}}-{\displaystyle \frac{e^{({\alpha }_{j-1}+X_i\beta )}}{1+e^{({\alpha }_{j-1}+X_i\beta )}}})$$

(4)

In addition to analyzing the entire sample for these estimated threshold and parameter values, we analyzed age- and gender-specific samples to determine how attribute preferences varied by age and gender.

4. Results

Table 6 presents the estimations derived from the rank-ordered logit model regarding the gamification attributes of metaverse services. These estimations reveal the preference coefficients associated with character customization attributes in metaverse service gamification. Notably, character realism (1.62, p < 0.001) and expression (dimensions of behavior and expression; 0.77, p < 0.05) exhibited positive preference coefficients. Conversely, body customization (−1.45, p < 0.001) and environmental customization (−1.55, p < 0.001) manifested negative preference coefficients. The preference for character realism and expressive behavior likely stems from the enhanced immersion facilitated by avatars or characters with realistic and nuanced expressions. In contrast, the aversion toward extensive body and environmental customization suggests that overt complexity or intricacy may detract from overall user experience.

Table 6. Estimated results for rank-ordered logits.

					10s			20s			30s			40s			50s or Above
VARS		Coef.	S.E.		Coef.	S.E.		Coef.	S.E.		Coef.	S.E.		Coef.	S.E.		Coef.	S.E.
Character customization	Face	−0.30	0.17		−0.02	0.22		−1.34	0.48	**	−0.84	0.74		0.09	0.53		−1.48	1.15
	Body	−1.45	0.28	***	0.04	0.49		−3.75	0.59	***	−3.43	0.93	***	−0.25	0.88		−3.74	1.41	**
	Environments	−1.55	0.30	***	−1.11	0.56	*	−1.60	0.53	**	−2.61	0.99	**	−1.74	0.84	*	−0.42	1.00
	Realism	1.62	0.14	***	1.78	0.23	***	1.65	0.24	***	1.96	0.47	***	1.32	0.38	**	0.85	0.42	*
	Expression	0.77	0.23	**	−0.82	0.42		2.89	0.53	***	3.05	0.83	***	0.19	0.68		2.33	1.23
Experience and skill system	Quest presence	0.16	0.07	*	−0.25	0.11	*	0.78	0.13	***	0.93	0.23	***	0.04	0.22		0.42	0.31
	Game freedom	−0.37	0.05	***	−0.63	0.08	***	−0.14	0.09		0.06	0.14		−0.33	0.14	*	−0.27	0.20
	Content Creation	1.22	0.14	***	1.52	0.27	***	1.10	0.24	***	1.14	0.36	**	1.17	0.47	*	0.65	0.60
	Mirroring quest	0.60	0.05	***	0.36	0.08	***	1.08	0.10	***	1.07	0.16	***	0.27	0.16		0.63	0.24	*
Social network system	Rigid chat−window style	−1.35	0.22	***	−1.54	0.34	***	−2.43	0.55	***	−1.02	0.87		−0.47	0.69		−2.29	1.30
	Conversational closeness	1.92	0.15	***	2.24	0.27	***	2.27	0.26	***	1.71	0.38	***	1.04	0.45	*	1.42	0.64	*
	Frequency of conversation participation	0.99	0.07	***	0.91	0.10	***	1.23	0.12	***	1.06	0.18	***	0.65	0.20	**	0.90	0.30	**
	Conversation freedom	0.05	0.15		−1.00	0.19	***	2.04	0.44	***	1.78	0.63	**	−0.61	0.46		1.62	1.09
Economic system	Monetization function	0.04	0.11		−0.04	0.23		2.37	0.34	***	2.78	0.58	***	0.05	0.54		1.51	0.85
	Device dependencies	−0.43	0.21	*	−0.93	0.28	**	−0.75	0.28	**	−0.52	0.43		−0.18	0.52		0.31	0.74
	Real−world task utilization	1.32	0.19	***	0.78	0.13	***	0.84	0.16	***	0.83	0.25	**	0.55	0.29		0.92	0.45	*
gender	Male	0.03	0.09
Age	Twenties	0.13	0.19
(ref.: Teenager)	Thirties	0.24	0.21
	Forties	0.05	0.21
	Over Fifty	0.12	0.25
Education	Middle school graduate or higher	0.04	0.17
(ref.:	High school graduate or higher	−0.07	0.16
Elementary school or higher)	College graduate or higher	−0.14	0.21
	0|1	1.44	0.12	***	2.15	0.35	***	3.98	0.47	***	4.23	0.77	***	2.07	0.79	**	3.58	1.25	**
	1|2	1.65	0.12	***	2.37	0.35	***	4.24	0.47	***	4.59	0.78	***	2.25	0.79	**	3.85	1.26	**
	2|3	2.22	0.13	***	3.04	0.35	***	4.86	0.49	***	5.26	0.80	***	2.81	0.80	***	4.44	1.29	**

Note: Stars represent: * p < 0.05, ** p < 0.01, *** p < 0.001.

Regarding experience and skill system attributes, the presence of quests (0.16, p < 0.01), content creation opportunities (1.22, p < 0.001), and mirroring quests (0.60, p < 0.001) were positively correlated. Meanwhile, freedom in quest selection exhibited a negative preference coefficient (−0.37, p < 0.001). The positive coefficients for quest presence, content creation, and mirroring quests possibly reflect users’ preferences for engaging in goal−oriented activities and the availability of creative outlets. The negative coefficient for quest freedom may indicate that users prefer clarity and direction provided by constrained quest paths, whereas excessive options could engender confusion among users.

In the social network system, attributes such as conversation closeness (1.92, p < 0.001) and the facility for group conversations (0.99, p < 0.001) were positively valued, while the style of conversation received a negative preference (−1.35, p < 0.001). The favorable view of close and group conversations underscores the significance of social interactions and communal involvement in the metaverse. This suggests that users value meaningful interaction and active community participation. Conversely, the negative preference for specific interaction styles might be related to the limitations imposed by certain communication methods, such as chatbots or chat windows, which could restrict users’ sense of comfort or freedom.

In the economic system category, the monetization feature did not exhibit statistical significance, presenting a negative correlation with device dependency (−0.43, p < 0.05) and a positive preference for the application of real-world tasks (1.32, p < 0.001). Reluctance toward device dependency highlights a user’s preference for a consistent experience across different platforms, without being confined to particular devices. The positive preference for real-world task integration suggests an appreciation of the metaverse’s linkage to tangible outcomes, its practical utility, and the notion that achievements within the virtual domain can translate into rewards in the real world.

In the domain of character customization within metaverse services, the analysis revealed a predominantly negative disposition toward body customization across most age groups, with the exception of teenagers. Although not statistically significant, teenagers exhibited a positive inclination toward face customization, a trend not observed in other age groups. This age-specific preference suggests a heightened engagement with avatars, emphasizing the desire for extensive personalization of appearance. The favorable disposition toward both face and body customization among teenagers may be attributed to the developmental importance of self-exploration and expression during adolescence—a period characterized by identity formation and a pronounced focus on appearance and individualism.

Concerning the experience and skill system attributes, the data indicate that individuals in their twenties and thirties prefer mirrored quests—activities that resonate with real-world experiences—over fantastical or abstract quests, which are preferred by teenagers and older age groups. This trend suggests that individuals more engaged with real-world social dynamics may seek similar experiences within the metaverse. The distinct preferences of those in their twenties and thirties could also reflect a mature sense of social responsibility and a stronger connection to real-life contexts. Conversely, teenagers exhibited a lower preference for the presence and complexity of quests, suggesting an inherent desire for autonomy within the metaverse. When quests are introduced, teenagers favor those with clear objectives, reflecting their exploratory spirit and preference for unbounded expression.

In social networking systems, a notable variance in preference for conversational freedom across different age groups was observed. Conversational freedom—defined as the ability to engage virtually with previously unknown individuals—is valued by individuals in their twenties, thirties, and fifties, but not by teenagers or those in their forties. This divergence may reflect each group’s unique perspective on social interactions and the importance of anonymity within the metaverse. The reluctance of teenagers and individuals in their forties to form virtual relationships contrasts with the value placed on anonymity by those in their twenties, thirties, and fifties. Although teenagers and individuals in their forties exhibited similar preference patterns, further research is warranted to ascertain the underlying causes, considering the potential influence of sample size limitations.

Within the economic system domain, all age groups demonstrated a preference for utilizing the metaverse as a productivity tool, with younger participants expressing a stronger desire to avoid device dependency. This trend suggests a generational shift toward valuing technological flexibility and the freedom to access digital environments across multiple devices. Moreover, individuals in their twenties and thirties exhibited a pronounced preference for converting achievements within the metaverse into virtual currency, reflecting an intersection between virtual experiences and real-world economic engagement among these age cohorts.

Gender differences in attribute preferences for metaverse services are further elucidated in the Appendix A (see Figure A1). Significant disparities were identified in several areas: content creation within gamification attributes (with a higher preference among females), conversational closeness within social networking systems (preferred more by females), the ability to invite users into personal spaces (favored by females), and the conversion of achievements into virtual currency within the economic system (also preferred more by females). These findings highlight the nuanced ways in which gender influences engagement and preferences in metaverse services.

5. Discussion and Conclusions

5.1. Summary of Findings

This study examined how gamification attributes of metaverse services relate to user preferences across age segments (teens to fifties) and derived design strategies for retail and service contexts.

Character customization. Preferences increased with avatar realism and expressive behaviors (e.g., mood/gesture expression), but decreased with complex body/environment customization. Users in their twenties and thirties particularly valued realism and expressive affordances, while showing a lower preference for time-consuming face/body/environment configuration. We interpret this pattern as a desire for recognizable, expressive selves with low setup burden, and we relate the aversion to heavy configuration to digital fatigue from device operation and prolonged online tasks.

Experience and skill systems. Preferences were positively associated with quest presence, content creation, and real-world–mirroring quests, while excessive option breadth or unrestricted quest freedom lowered preference. Teens, twenties, and thirties favored the presence of quests, content creation, and mirroring elements; attitudes toward broad “do-anything” freedom diverged by age, suggesting heterogeneous tolerance for unstructured progression.

Social networking systems. Conversational closeness and group discussion capacity increased preference. In contrast, rigid chat styles (overly constrained formats) were less favored. All age groups valued close interaction and group exchange; users in their twenties and thirties additionally appreciated greater conversational freedom, while teens (and some users in their twenties) reacted negatively to certain stylized or restrictive interfaces—indicating sensitivity to interaction design fit.

Economic systems. Preferences rose with monetization opportunities and real-world task utility, and fell with device dependency. Real-world applicability was valued across all ages; monetization was especially preferred by users in their twenties and thirties; device dependency was disliked by teens and twenties.

Across domains, findings emphasize expressive identity with minimal friction, structured yet flexible progression, social immersion, and tangible utility. The 16 sub-attributes operationalized in this study provide a concrete blueprint for preference formation in metaverse services.

5.2. Theoretical Implications

First, we align theorization and measurement by focusing on preferences—rather than satisfaction or downstream choice—thus clarifying how design attributes map to stated rankings in avatar-mediated settings. This preference-centric stance helps reconcile mixed evidence in prior metaverse work that often treated services as game analogs while overlooking user-driven, sociotechnical dynamics.

Second, we contribute a parsimonious taxonomy—character customization, experience/skill systems, social networking, and economic systems—linking observable attributes to preference signals. The differentiated effects (e.g., realism/expression ↑ vs. heavy body/environment customization ↓; quest presence ↑ vs. excessive freedom ↓) nuance the common assumption that “more features” uniformly increase engagement.

Third, we foreground age heterogeneity in gamified environments. Segment-specific patterns (e.g., twenties–thirties valuing expressive realism and monetization; teens’ aversion to rigid conversational styles and device dependency) suggest that demographic moderators are not peripheral but constitutive of metaverse preference formation.

5.3. Managerial Implications

Translating results into design heuristics, we propose the following:

• Character customization. Prioritize recognizable realism and expressive behaviors; streamline or pre-bundle body/environment configuration to minimize digital fatigue. Provide quick-start defaults plus optional depth for power users.
• Experience and skill. Ensure clear quest presence with adaptive difficulty, immediate feedback, and visible rewards (e.g., badges, virtual goods, social recognition). Combine open-ended exploration (to sustain curiosity, especially for younger users) with milestone-based progression (appealing to goal-oriented users).
• Social networking. Design for closeness (small-group affordances, proximity cues) and scalable group interaction. Avoid rigid, one-size-fits-all chat formats; offer conversational freedom with safety/etiquette scaffolds and transparent controls.
• Economic system. Elevate real-world utility (bridge virtual tasks to tangible outcomes) and fair, transparent monetization pathways. Reduce device lock-in by enabling cross-device continuity and graceful degradation on lower-end hardware.

These guidelines support segment-sensitive product roadmaps and marketing that align feature bundles to the needs of teens vs. twenties–thirties vs. older users.

5.4. Limitations and Future Research

This study is research-question-driven and does not advance formal hypotheses. While the literature review was strengthened accordingly, future work should derive and test hypotheses based on the attribute–preference relationships identified here to improve explanatory power and generalizability.

Second, our context is Korea. Cross-cultural research should examine whether gamification preferences—and tolerance for configuration burden, conversational styles, or monetization—vary across markets with different platform ecologies and norms. Incorporating finer demographic and psychographic measures would also sharpen segmentation.

Third, longitudinal designs are needed to track how preferences evolve as users gain familiarity and as platforms iterate. Finally, integrating emerging technologies (e.g., AI-driven personalization, immersive haptics, blockchain-based economies) into the attribute framework could clarify their effects on adoption, retention, and preference across industries.

5.5. Conclusions

By linking a literature-grounded, four-domain taxonomy of gamification to ranked user preferences for leading metaverse services, this study clarifies which design attributes matter, for whom, and why. The evidence points to metaverse experiences that combine expressive identity, structured progression with optional freedom, social immersion, and real-world utility, while avoiding unnecessary configuration burden and device dependence. The resulting 16 sub-attributes offer actionable levers for developers and decision-makers seeking to optimize user adoption and sustained engagement in a rapidly evolving metaverse ecosystem.