Investigating the influence of mobile game addiction on in-app purchase intention in PUBG mobile: the mediating roles of loyalty, negative e-WOM and perceived risk

Abstract

This study focuses on in-app purchase intention in Players Unknown Battleground (PUBG) Mobile to examine the relationship between mobile game addiction and in-app purchase intention in PUBG Mobile. Moreover, this study explores the moderating role of loyalty, negative electronic word-of-mouth (e-WOM) and perceived risk. A survey of 347 PUBG mobile game players was conducted to collect the data through a convenience sampling technique. Partial least square-structural equation modeling (PLS-SEM) was used to test the proposed conceptual framework. Our findings indicated that online game addiction significantly influences loyalty, negative e-WOM, perceived risk and inapp purchase intention. Moreover, online game loyalty, negative e-WOM and perceived risk significantly affect in-app purchase intentions in online games. The sample was exclusively comprised of PUBG players from Indonesia. Including participants from developing and developed countries could offer valuable cross-country comparisons, enhancing the diversity and generalizability of the study’s findings. This study emphasizes that loyalty, negative electronic word of mouth and perceived risk mediate the influence of online game addiction on in-app intention. Research provides information on factors influencing purchase intentions in online game applications. This study is among the few to explore the mediating role of perceived risk from three dimensions, namely, online payment, financial and performance risk. This work enriches the existing literature, particularly regarding the insight of the Indonesian context.

IMPACT STATEMENT

Online mobile games have become popular among youth, especially in the digital era. Although it is often free for players, it occurs that there are new mobile game apps come to the market every single day. Among the most played mobile games, PUBG mobile is one of the leading global mobile game markets; it attracts players due to its several in-app features. This study aims to understand the influence of mobile game addiction on loyalty, perceived risk, negative e-WOM and in-app intention to purchase. The survey was conducted in Indonesia, and the data collected were analyzed using structural equation modeling. We found that the in-app purchase intention is not only determined by the players’ game addiction level but also influenced by their loyalty, perception of risk and negative e-WOM. Besides, these factors intervened in the link between addiction and in-app purchase intention. This study offers theoretical insight into existing literature and provides a practical guide for mobile game marketers.

Keywords:

• Addiction
• loyalty
• negative e-WOM
• perceived risk
• in-app purchase intention

Reviewing Editor:

• Len Tiu Wright, De Montfort University Faculty of Business and Law, United Kingdom

SUBJECTS:

• general psychology
• consumer psychology
• cognitive psychology

1. Introduction

Technological developments have increased rapidly in recent years, drastically changing our behavior, routine and daily habits. Technological innovation is inevitable, transforming into a preferred entertainment avenue for many individuals. Technological advancements allow individuals to socialize with friends on social media, stream content on Maxstream and Netflix and engage in online mobile games. Electronic commerce managers have recognized the importance of online gamers as they have consistently grown over the years (Liao & Teng, 2017). The digital era presents significant potential for online game products specifically in Indonesia (Wiguna, 2024). Despite the prevalence of free-to-play models, most profits in online gaming are generated through in-app purchases (Ravoniarison & Benito, 2019). Besides, sustaining the duration of engagement among gamers is crucial for game managers. Correspondingly, designing online games to be more addictive is a strategy implemented and considered effective over decades (Ravoniarison & Benito, 2019). Nevertheless, the impact of game addiction has raised an interesting debate among scholars, especially regarding its influence on gamers’ behavior and decision-making.

Mobile games are among the many technological advancements that have experienced phenomenal growth (Liao et al., 2020). Wijman (2021) reported that the mobile games market has reached US$77.2 billion, with many players reaching 3.5 billion people. Further, Wijman (2021) also documented that the Asian continent attracted the intention of the expanding mobile gaming industry, representing nearly 54% of players worldwide. In particular, Yasir and Agus (2021) emphasize the significant growth of the gaming industry in Indonesia. Indonesia is one of the largest mobile game markets in the Asian region, and 43% of the Asian mobile games market encompasses 212.25 million users (Wijman, 2021). Correspondingly, Trihendrawan (2021) illustrates that in Indonesia, mobile games contributed the most significant revenue in Asia, reaching US$2.08 billion (Trihendrawan, 2021). The remarkable expansion of the online gaming market is attributed not only to external factors, such as internet accessibility but also to the elevated level of interest in online gaming products and technological addiction among the youth (Purwaningsih & Adison, 2017).

The Tencent-developed Players Unknown Battleground (PUBG) mobile game is currently one of the most-played mobile e-sports titles. It is classified as the biggest earning and most-played game worldwide in 2020–2021 (Hendrawan, 2021). PUBG Mobile is a first-person shooter (FPS) game that simulates player movement and depicts authentic in-game experiences. Most FPS games follow a similar pattern, replicating player actions and experiences (Adjorlu, 2024). According to the Esports Scharts reports that PUBG mobile is popular among Indonesian youth due to inter-campus competitions in Indonesia called the PUBG Mobile Campus Championship Indonesia 2020 and 2021, which attended numerous campuses in Indonesia (Esports Scharts, 2020). On 14 December 2021, PUBG Mobile recorded a revenue of approximately IDR 39.8 trillion. Recognized for its substantial earnings, PUBG Mobile is the most profitable and popular mobile game on both the Google Play Store and App Store (SensorTower, 2021).

PUBG Mobile dominates the global gaming market; as a free apps with sophisticated features and an in-app purchase model (Hendrawan, 2021). Aligned with the case study conducted by Gillani (2021). Evaluation of Games Monetization Approaches: A case study on Players Uknown’s Battlegrounds (PUBG). Gillani (2021) noted that PUBG mobile games dominate the market in free game applications accompanied with in-app purchases. In-app purchases are mainly in-game and free download services, known as freemium applications. Nevertheless, the game apps can be downloaded for free; several PUBG mobile features demand in-app purchases from players to unlock particular features or levels by purchasing unknown cash, including clothes, skins, coins and guns. In this sense, in-app purchases are the primary metric for evaluating a game’s success. In marketing research, in-game purchases are considered consumers’ willingness to pay for features in future events and intentions to pay are indicators of gamers’ attitudes toward features (M. L. Cheung et al., 2021). This study adopts the concept of in-app purchase in the context of games, defining in-app purchase as the player’s readiness to buy in-game products such as unknown cash or skins and outfits in a future event (Hamari et al., 2017). Hence, it is essential to comprehend the factors influencing in-app purchasing intention among players, especially in online gaming, which is known for its addictive nature. Exploring the impact of addiction on purchasing behavior becomes crucial in this phenomenon.

Previous studies have identified the determining factors of in-app purchases, including value (Hsiao & Chen, 2016; Hsu & Lin, 2016), the experience of consumer (Liao et al., 2020), consumer innovation (Su et al., 2016), motivation (Hamari et al., 2017), lifestyle (Akkaya, 2021), pricing (Weisstein et al., 2016) and subjective norms and behavioral control (Renny et al., 2013). Moreover, the existing studies have employed a psychological point of view; prior studies have delved into the association between mobile game buying behavior in terms of impulsive and compulsive actions (Maraz et al., 2016). Undeniably, in-app purchases are unlike other online purchases; in-game purchases require long-term momentum for gamers to engage in games with high-quality relationships and immersion bonds (Kuss & Griffiths, 2012). Consequently, it can cause negative impacts on behavior, which is called addiction (Griffiths, 2010). However, numerous studies demonstrate that Game addiction drives in-app purchases (Andreassen et al., 2015; Balakrishnan & Griffiths, 2018; Griffiths, 2010; Hsiao & Chen, 2016; Liao et al., 2020). To the best of our knowledge, research on online game addictions and their implications for human behavior and psychology is limited. This study aims to investigate the connections between mobile game addiction, loyalty, negative electronic word-of-mouth (e-WOM) and perceived risk in relation to in-app purchase intention.

Online purchasing behavior studies highlighted that loyalty is one determinant affecting in-app purchases (Hsu & Lin, 2015). User loyalty is crucial for game developers because loyal users will make in-app purchases and ensure the sustainability of the game (Hsu et al., 2016). The emergence of online game addiction has attracted the attention of scholars, prompting an interest in understanding the relationship between loyalty and in-app purchase intention (Hsu & Lin, 2015). Putra and Wahyudi (2022) explored the link between satisfaction and loyalty in studying mobile game addiction. Similarly, Balakrishnan and Griffiths (2018) look at the function of addiction and loyalty toward in-app purchases. The findings highlight that loyalty mediated the relationship between addiction and in-app purchases.

Moreover, Baker et al. (2016) found that negative electronic word of mouth can reduce purchase intentions and that user-generated criticism is more credible than game developers’ positive e-WOM. Likewise, Fu et al. (2015) further stressed the importance of Negative online comments made by players/gamers. In this regard, E-WOM differs from face-to-face communication but plays a vital role in online gaming, influencing purchasing behavior and preventing gamers’ online addiction (Puspitasari et al., 2018; Silaban et al., 2022; Teng, 2018). Also, Haque and Tarofder (2020), followed by Silaban et al. (2023), found that negative e-WOM significantly influences consumer behavior toward purchase intentions more than positive e-WOM. Furthermore, perceived risk constitutes a significant factor in online purchases (Chen, 2010). Nevertheless, engaging in online games has different adverse consequences (Chen et al., 2012). In addition, existing empirical evidence highlighted that addiction plays a crucial role in purchasing decisions and is among the detrimental effects associated with playing online games (Balakrishnan & Griffiths, 2018; Hsiao & Chen, 2016). Based on the above empirical evidence and previous findings, this study proposes a conceptual framework to investigate the impact of game addiction on in-app purchase intention in the context of PUBG mobile in Indonesia (Figure 1).

This study aims to explore and examine the determinants of gamers’ in-app purchasing behavior in PUBG mobile. It also investigates the influence of game addiction on loyalty, perceived risk and negative e-WOM in-app purchasing behavior. Moreover, this work tests the mediating role of loyalty, perceived risk and loyalty. In doing so, we conducted a survey of PUBG mobile players in Indonesia. The present study provides two-fold contributions. First, it aims to fill the gap in the existing literature by investigating the mediating role of loyalty, perceived risk and negative e-WOM. The second contribution of this study is for the practitioners; marketing professionals can use this result to gain additional knowledge on improving in-app purchasing behavior.

2. Literature review and hypothesis developments

2.1. Mobile gaming

Previous literature discussed online game addiction (Balakrishnan & Griffiths, 2018; Haberlin & Atkin, 2022). Fournier et al. (2023) assert that addiction in online games consists of seven core components: sense-importance, tolerance, mood modification, withdrawal, relapse, conflict and problems. Online game addiction means uncontrollable behavior and adverse effects (Kuss & Griffiths, 2012). Addiction to online gaming stresses the conflict between personal aims and conduct, which is one of the two manifestations of the cause. Online game addicts attempt to reduce gaming time but fail since they cannot control their behavior. A study by Balakrishnan and Griffiths (2018) explores the psychological aspects of mobile games, focusing on gaming addiction in terms of loyalty and in-app purchases. Their study explains the significant relationship between game addiction and loyalty. More explicitly, once game addiction increases because of the time spent on the mobile game, players become more attached and spend money on mobile game features. In addition, H. Wei et al. (2017) support that online game addiction can increase user loyalty. A similar finding has also been discovered (Putra & Wahyudi, 2022). In the same context, some studies refer to loyalty as an addictive behavior (Fournier et al., 2023; Lee et al., 2016; Liao et al., 2020; Weinstein et al., 2016). In comparison, another study discovered the role of addiction and how it relates to peer attachment and social seeking (Blinka & Mikuška, 2014). Thus, we assume that online game addiction and loyalty are intertwined variables, as addictive behavior can drive loyalty in online mobile games.

The present study focuses on the role of loyalty in online games toward in-app purchasing intention. However, the study by P. S. Wei and Lu (2014) focuses on the user and social aspects, and the other focuses on the determinants affecting purchase intention in gaming applications through understanding users’ perceived value and loyalty toward mobile games (Hsiao & Chen, 2016). It emphasizes the essence of loyalty in influencing player-purchasing intention in online games. In addition, in-game purchases are the primary outcome of measuring game performance and soundness. Marketing research defines customer purchase as a consumer’s will to pay for an item in the future and intention to pay refers to an indication of a customer’s attitude toward that product (Al Mamun et al., 2023). The present study adopts the concept of purchase intention in game applications, conceptualizing in-game purchase intention as an individual’s readiness to purchase game-related products including virtual currency or tickets in the future (Hamari et al., 2017). Loyalty enhances players’ engagement, fidelity and support in the mobile game (Albatati et al., 2023). Thus, there is a strong correlation between loyalty and purchasing intention. As a result, game addiction and loyalty are genuine reasons to have intentions for and behaviors related to in-app purchases when playing a particular game.

Based on prior literature, this current work developed a theoretical framework using variables of online game addiction, loyalty and in-app purchase intention using the seven dimensions of game addiction: salience, tolerance, mood modification, withdrawal, relapse, conflict and problem. We understand the influence of addiction and loyalty in increasing purchase intention in online game applications and the interconnectedness between addiction and loyalty. We suggest the following hypothesis below:

Hypothesis 1

(H1): Online game addiction positively increases online game loyalty.

Hypothesis 2

(H2): Loyalty to online games will increase purchase intention in online game applications.

In online gaming context, online gaming players can establish a new community because the platform allows them to interact with other players (Kim et al., 2022). Huang et al. (2017) explain that mobile games play a significant role in forming a social community because they allow gamers to interact and build links. The electronic word of mouth (e-WOM) enables mobile game players to discuss their thoughts and experiences with other users and contact new players to join, ultimately driving the creation of communities (Huang et al., 2017). Huang et al. (2017) further describes that e-WOM is the engine power behind the development of online communities. It shapes participants in online game forums so they can open up about their emotions and experiences with other users and invite new ones. According to an Massively Multiplayer Online Role-Playing Games (MMORPG) study, game developers must generate positive opinions to obtain positive e-WOM because game user experience can spread negative e-WOM in response to a game system failure (Albatati et al., 2023). It emphasizes how addictive behavior might become aggressive. In addition, Israeli et al. (2019) pointed out that addiction will channel harmful behavior through negative e-WOM. Thus, either negative or positive e-WOM can be explicitly necessary for continuing or stopping online game activation (Teng et al., 2024). Meanwhile, the unpleasant experience of the user during the online game may augment unfriendly comments and vice-versa.

Therefore, the studies investigating the consequences of online gaming addiction are scant. Since research on this relevant topic is limited, the correlation between the negative e-WOM and in-app game purchase intention has yet to be identified. However, Baker et al. (2016) stated that negative e-WOM significantly reduces purchase intentions. Similarly, negative e-WOM significantly influences consumers’ evaluation of purchase intention more than positive e-WOM (Almohaimmeed, 2019). Meanwhile, Ally et al. (2020) argued that game addiction significantly impacts negative e-WOM since the users have a wide range of online communities, which allows the quick spread of negative e-WOM. Bad experiences, negative comments and system failure greatly influence purchase intention in online gaming applications. Based on these prior studies, this study extends the consequences of addiction to online games on negative e-WOM and purchase intention in-game applications. We construct our hypothesis as follows:

Hypothesis 3

(H3): Addiction to online games positively affects negative e-WOM.

Hypothesis 4

(H4): Negative e-WOM positively influences purchase intention in online gaming applications.

Preventing game addiction is an essential goal and determining the perceived risks that can be involved in game addiction is the basis for setting preventive measures (Kaya et al., 2022). Concerning the scope of the research, there is a lack of empirical studies examining the relationship between game addiction and perceived risk. This study explores the association between game addiction and perceived risk. Some studies refer to online gaming as a type of internet usage that does not positively impact our society (Chopdar et al., 2018). Indeed, online game addiction has long been the subject of concern due to potential negative impacts that could jeopardize society’s values and interests. Previous studies by Syvertsen et al. (2022) asserted that the risk of online gaming addiction had become part of every life, such as the risk of overspending, the risk of being socially excluded and the risk of mental health. Online game addiction incorporates risks, although it may carry some benefits. Align with that, the study of Toozandehjani et al. (2021) stated the significant effect of addiction on perceived risk in online games.

In the literature on information systems, perceived risk has been identified as a significant element in lowering user risk in behavioral intentions across a variety of scenarios, including virtual banking (Tham et al., 2017), online applications (Gumasing et al., 2023) and social media (Nikolinakou et al., 2024). In online gaming, when a person perceives a high degree of risk related to behavioral intention, ie playing games, users tend to be reluctant to play excessive games (Dahabiyeh et al., 2020). Perceived risk is a potential obstacle to the overuse of online games (Alrawad et al., 2022). The high-perceived risk could minimize the excitement of the game, reducing the impact on play intention and in-game purchases (Albatati et al., 2023). Marakanon and Panjakajornsak (2017) support this argument that perceived risk influences online buying behavior. Therefore, online purchase intentions could influence users to make decisions in situations of discomfort or concern about online payment, financial and game performance risks.

According to the above literature, we conclude the hypothesis of this study as follows:

Hypothesis 5

(H5): Addiction to online games affects the perceived risk of online games.

Hypothesis 6

(H6): Perceived risk from online games affects purchase intentions in online game applications.

Several studies have suggested expanding the correlation between addiction and purchasing intention (Andreassen et al., 2015; Carvalho et al., 2023; Fournier et al., 2023; Haberlin & Atkin, 2022; Lee et al., 2016; Weinstein et al., 2016). However, empirical evidence is finite in the online game addiction setting. In-app purchase intention is an activity that demands ability and motivation (Hsiao & Chen, 2016). In-app purchases are crucial in mobile gaming, especially in games that use the freemium business model and the caching system. In-app purchases are among the game providers’ most established revenue samples (Hsu & Lin, 2015). In the gaming industry, the profitability and sustainability of the game depend heavily on the in-app purchases. Indeed, the number of players is important as well.

Tarafdar et al. (2020) defined addiction as an affective, cognitive, behavioral and psychological reaction bias. Technological advancement stimuli influence the user’s intention to operate it frequently. For instance, mobile game application builders provide products on a seven-day trial basis (Hsu et al., 2016). As a result, the more often the user uses the game, the more attached and emotional it becomes. Interestingly, users who engage with game apps start playing more often and become addicted. Based on the decision-making process theory (Schwartz, 2016), extending use and sticking to objects may be influenced, especially when players are cognitively and affectively engaged in products and services. As a result, players will be dependent on the application. A previous study explained the connection between game addictions and in-app purchasing intentions (Balakrishnan & Griffiths, 2018). They support this hypothesis that addictions affect in-app purchase intentions in online mobile games. Unquestionably, a user’s addiction to mobile games may cause them to overspend and overuse regardless of the consequences.

Based on this study, this research investigates how online game addiction and in-app purchase intentions are by taking seven dimensions of game addiction: salience, tolerance, mood modification, withdrawal, relapse, conflict and problem. Based on the literature, we presented the following hypothesis:

Hypothesis 7

(H7): Addiction to online games positively increases purchase intention in online gaming applications.

3. Methodology

This quantitative research use primary data. The population for this study is the PUBG mobile game players. We chose PUBG Mobile due to its popularity and as most downloaded games on the Play Store and App Store, with an excessive profit among other mobile games (Gillani, 2021). The data were collected in early 2022; the self-administered questionnaire was distributed online through LimeSurvey. The final sample used for this research were 347 after removing the 33 invalid data (91.3% response rate). The samples satisfy the minimum requirements for the SEM test (Balakrishnan & Griffiths, 2018). A non-probability sampling, the convenience sampling technique was used in this study. This research implemented a survey with a questionnaire distributed using LimeSurvey through the gaming community on WhatsApp and Telegram groups, where game top-up vouchers were awarded to 10 participants who participated.

The survey questionnaire was initially designed using English and then translated to Indonesian and back to English to ensure the accuracy and rightness of each question (Podsakoff et al., 2012). The surveys were divided into three sections. The first section includes introduction of the objective of the survey, instruction and respondent’s consent. Second section consists of demographic data, and the third section consists of the 44 self-report questions about attitudes.

All items were adopted from previous studies to guarantee the reliability and accuracy of the measurement of the variables (Tsang et al., 2017). The measures of variables used to measure Addiction, Loyalty, Negative E-WOM, Perceived Risk and Purchase Intention variables are extracted from previous literature such as (Balakrishnan & Griffiths, 2018; Chopdar et al., 2018; Fournier et al., 2023; Hegner et al., 2017; Hong & Cha, 2013; Hsiao & Chen, 2016). Our study employs the seven dimensions adopted from Fournier et al. (2023) to measure game addiction Sense-importance refers to expressing the players’ feelings about their importance or the importance of something. Tolerance is defined as the excitement of players to play more and more. Mood modification relates to people’s subjective sensations from playing video games. Withdrawal relates to the uncomfortable emotional states or physical repercussions that arise when one abruptly stops playing games. Relapse describes the tendency for reversions to earlier patterns of gaming. Conflict refers to conflict between players and the people around them because of the time spent on online gaming. Problems incorporate social, physical and physiological issues generated by online gaming. The model evaluation follows the recommended two-stage process (Hair et al., 2019) assessing the outer and inner model. We measured the items using five Likert scales ranging from 1 (strongly disagree) to 5 (strongly agree).

The data were analyzed using Smart PLS software. The Confirmatory factor analysis was conducted to assess the reliability and validity of the measures used in this study. The discriminant validity test is carried out to evaluate the relationship between two variables. Next, hypothesis testing is applied in partial least squares-structural equation modeling (PLS-SEM).

4. Result

4.1. Sample description

Table 1 describes the sample’s demographic characteristics of 347 PUBG mobile players. The sample demonstrate that 53.3% of the participants were males and 46.7% were females. The majority of the participants were aged between 21 and 25 years old (56.5%), followed by participants aged between 16 and 20 years old (36.0%), participants above the age of 30 represented 4.3%, and only 3.2% of the participants were aged between 26 and 26 years old. Besides, in term of education background, 201 (57.9%) respondents have a Bachelor’s degree, 131 respondents (37.8%) have a Senior High School diploma, 12 (3.5%) respondents have a master’s degree and 3 (0.9%) respondents have a Junior High School level. Regarding income, 44.1% of the participants have an income lower than Rp 1.000.000, 36% of them earned Rp between Rp 1.000.000 and Rp 3.000.00, 14.4% of them earned between Rp 3.000.000 to 6.000.000 and the remaining 14% have an income above 6.000.000. Regarding the time spent on playing online gaming, 32.0% of the participants spend less than 30 min on mobile gaming, but the remaining 68.0% spend more than 30 min, and some are even more than 120 min. In-app purchasing, 44.1% of respondents never purchase game features, 29.4% have made purchases 1 to 5 times, 12.1% have purchased 5 to 10 times and 14.4% have actively purchased more than 10 times.

Table 1. Demographic characteristics of the sample.

Profile	Category	n	%
Gender	Male	185	53.3
	Female	162	46.7
Age (years old)	16–20	125	36.0
	21–25	196	56.5
	26–30	11	3.2
	31 or above	15	4.3
Education	Junior high school	3	0.9
	Senior high school	131	37.8
	Bachelor’ degree	201	57.9
	Master’ degree	12	3.5
Monthly income (IDR)	Below 1 million	153	44.1
	1−3 million	125	36.0
	3–6 million	50	14.4
	Above 6 million	19	5.5
Time spent playing mobile games (minutes a day)	29 or below	111	32.0
	30–60	63	18.2
	60–90	35	10.1
	90–120	48	13.8
	120 or above	90	25.9
Number of in-app purchases of mobile games (in a month)	Never	153	44.1
	1–5	102	29.4
	5–10	42	12.1
	11 or above	50	14.4
Total		347	100

Note: Percentage (%), frequency (n), (IDR) Indonesian Rupiah.

4.2. Measurement evaluation

The convergent validity was assessed based on Hair et al. (2019, 2020) indicating the loading factor value of each item must be above 0.70, average variance extracted (AVE) values must exceed 0.5. Table 2 indicates that all items used in this study met the criteria. The statistics described that each indicator in each construct tested is greater than the error value in that construct (Haryanto et al., 2022). Therefore, all the items were validated measures of the construct.

Table 2. Convergent validity test.

Variable	Dimension	Indicator	Loadings	AVE dimension	AVE variable
Mobile game addiction	Salience	SLC01	0.883	0.818	0.541
		SLC02	0.939
		SLC03	0.890
	Tolerance	TLR01	0.895	0.806
		TLR02	0.931
		TLR03	0.867
	Mood modification	MM01	0.770	0.724
		MM02	0.880
		MM03	0.897
	Relapse	RLS01	0.876	0.745
		RLS02	0.839
		RLS03	0.873
	Withdrawal	WDR01	0.895	0.846
		WDR02	0.923
		WDR03	0.941
	Conflict	CFL01	0.891	0.806
		CFL02	0.913
		CFL03	0.888
	Problems	PRL01	0.858	0.659
		PRL02	0.835
		PRL03	0.738
Mobile game loyalty		LYT01	0.864	–	0.805
		LYT02	0.903
		LYT03	0.909
		LYT04	0.914
Perceived risk	Online payment risk	OPR01	0.941	0.904	0.708
		OPR02	0.962
		OPR03	0.949
	Financial risk	FR01	0.932	0.843
		FR02	0.928
		FR03	0.895
	Performance risk	PFR01	0.926	0.853
		PFR02	0.924
		PFR03	0.920
Negative E-WOM		NEW01	0.819	–	0.699
		NEW02	0.885
		NEW03	0.846
		NEW04	0.762
		NEW05	0.864
In-app purchase intentions in mobile game		NBA01	0.876	–	0.751
		NBA02	0.856
		NBA03	0.815
		NBA04	0.879
		NBA05	0.905

Note: SLC: salience; TLR: tolerance; MM: mood modification; RLS: relapse; WDR: withdrawal; CFL: conflict; PRL: problems; LYT: loyalty; OPR: online payment risk; FR: financial risk; PFR: performance risk; NEW: negative e-WOM; IPI: in-app purchase intentions; AVE: average variance extracted.

Reliability test was conducted to assess the accuracy and consistent of the instrument to measure the variable (G. W. Cheung et al., 2023). Cronbach Alpha and composite reliability were used to evaluate the reliability of all constructs. Table 3 shows that all variables met the required threshold criteria recommended by Hair et al. (2019), suggesting that the Cronbach alpha and composite reliability value should be greater than 0.70. Thus, the reliability test results conclude that the measurements of the study were reliable.

Table 3. Reliability test.

Variable	Cronbach’s alpha	Composite reliability
Game mobile addiction	0.957	0.961
Game mobile loyalty	0.919	0.943
Perceived risk	0.948	0.956
Negative E-WOM	0.892	0.921
In-app purchase intentions game mobile	0.917	0.938

Note: The independent variable is addiction game mobile, the dependent variable is in-app purchase intentions game online and the mediating variables are loyalty, perceived risk and negative e-WOM.

Table 4 shows the discriminate validity test result demonstrating the correlation between variables. This research adopted follow the recommendation of Fornell and Larcker (1981), suggesting that the correlation coefficient between each variable should not exceed the value of the square root of the average variance extracted (AVE). The result indicated that all construct have a correlation value smaller that the square root AVE in which explain that there is no discriminate validity issue detected (Hair et al., 2019).

Table 4. Discriminant validity.

	1	2	3	4	5
Mobile game addiction	0.736
In-app purchase intention	0.649	0.867
Mobile game loyalty	0.659	0.716	0.897
Negative e-WOM	0.584	0.459	0.349	0.836
Perceived risk	0.448	0.382	0.288	0.478	0.841

Note: The italics number denote the correlation between constructs and its indicators.

4.3. Structural model testing

The R-square (R²) was used to estimate the structural model for the construct (Henseler & Sarstedt, 2013). The R-square value determines the variance power of dependent variable on independent variable. It explains that a greater R² value shows a better prediction. The result suggest that 58.8% of the variance power of in app purchase intention explain by mobile game addiction, loyalty, negative e-WOM and perceived risk moderately explain the variant of in-app purchase intention. Moreover, Mobil game addiction explain 43.5% of the variance power of loyalty. Besides, the negative variable value of e-WOM is 0.341 or equal to 34.1%, which means that addiction to mobile games weakly explains the negative variant of e-WOM. Therefore, the remaining 65.9% is explained by external factors. Finally, 20% of the variance power of perceived risk explained by mobile game addiction. Table 5 demonstrates the result of the seven proposed hypothesis and show the coefficient, t-value and p value. Besides, Figure 2 presents the results of the structural equation model (SEM).

Figure 1. Conceptual framework.

Figure 2. Structural equation model results.

Table 5. Hypothesis testing.

Hypothesis	Original sample	Sample mean	Standard deviation	T statistics	p Values
AGM → LYT	0.659	0.657	0.033	20.076	.000**
LYT → IPI	0.518	0.517	0.057	9.052	.000**
AGM → NEW	0.584	0.585	0.043	13.635	.000**
NEW → IPI	0.121	0.120	0.054	2.232	.026**
AGM → PR	0.448	0.447	0.061	7.347	.000**
PR → IPI	0.087	0.088	0.042	2.060	.040**
AGM → IPI	0.649	0.648	0.037	17.371	.000**

Note: AGM: addiction game mobile; IPI: in-app purchase intention; LYT: loyalty game mobile (LYT); NEW: negative e-WOM; PR: perceived risk.

*Represents test statistical significance at 10% level.

**Represents test statistical significance at 5% level.

***Represents test statistical significance at 1% level.

The hypothesis was tested using structural equation modeling, the path significant was evaluated based on the t-statistic above 1.96, and the p value below .001, .01 and .05 (Sekaran & Bougie, 2016). Table 5 demonstrates the hypothesis results confirm that all the hypotheses H1, H2, H3, H4, H5, H6 and H7 proposed in were supported.

5. Discussion

This study examines the impact of game addiction, loyalty, negative e-WOM and perceived risk on in-app purchase intention in online games. Our empirical results revealed a significant positive impact of mobile game addiction on purchase intentions in game applications, consistent with previous findings (Balakrishnan & Griffiths, 2018). This result aims to enrich the existing literature in this context particularly and provide and empirical insight and empirical evidence from Indonesia. This study stresses the importance of addiction in enforcing in app purchase intention in mobile game context particularly PUBG. In addition, it proves the essence of mobile loyalty, perceived risk and negative e-WOM in improving in app purchase intention.

The results indicate that when users are addicted to games, they are willing to spend money to purchase in-game items to optimize and beautify the appearance of their characters with the costumes available in the game. Moreover, our study found a significant positive impact of mobile game addiction on game loyalty. The research results show that the more addicted someone is to games; the more loyal they are to games (Liao & Teng, 2017). This finding is supported by Hsiao and Chen (2016), indicating that online game loyalty significantly affects purchase intention in online game applications. When an individual has a strong addiction to a mobile game, the more they stay loyal to the game. It is vital to recognize that mobile game addiction does not always leads to adverse consequences but also it assists in increasing in loyalty of player toward the mobile game.

Besides, in the online game context, communities are formed because online games provide a platform for players to engage and interact with one another (Huang & Hsieh, 2011). Huang et al. (2017) also argued that negative e-WOM underlies the formation of online game communities and influences game players to share their feelings and experiences with other players. The present empirical study revealed that there is a significant positive impact of mobile game addiction on negative e-WOM. Consistent with prior study, this work confirmed the positive influence of game addiction on negative e-WOM (Ally et al., 2020). Correspondingly, game addiction evoke detrimental behavior in the form of negative e-WOM (Israeli et al., 2019). Moreover, this study demonstrates that when online game players are addicted to the games, they often exhibit similar behaviors, frequently resorting to the use of rude, toxic language and profanity, particularly among those who are addicted to gaming (Bae et al., 2016). Accordingly, Fu et al. (2015) stated that negative e-WOM in online games can be explicitly important for someone to continue or stop playing online games.

Besides, the results of this study found a significant positive impact of mobile game addiction on perceived risk. In the context of online games, when users perceive a level of risk associated with online games, they are reluctant to play excessive games (Dahabiyeh et al., 2020). In this regard, Kaya et al. (2022) stated that preventing online gaming addiction is an important goal in identifying gamers’ perceived risks and contributing to preventative measures for excessive purchasing in gaming addiction. Similarly to prior evidence from Syvertsen et al. (2022) and Toozandehjani et al. (2021). This study shows that perceived risks, including payment risk, inappropriate costs incurred, risks from slow connections and operating systems, can be considered when designing useful strategies to reduce risks and increase purchasing intention in-game applications. Managing risk is important for user well-being and ethical game design. Therefore, this study implies that that the level of addiction of the mobile game player determine their perception of risk.

Furthermore, supporting the prior studies (Albatati et al., 2023; Ally et al., 2020; Almohaimmeed, 2019; Hamari et al., 2017; Marakanon and Panjakajornsak, 2017). This study reveals that online game addiction has direct and indirect consequences in influencing loyalty, negative e-WOM and perceived risk on purchase intentions in-game applications. This study underscores the importance of game developers understanding the ramifications of game addiction to pay attention to factors influencing purchase intentions in game applications. Game developers should design game systems with sufficient device capacity so that users loyal to the game can cease playing if their device cannot support proper gameplay. While game developers aim for financial benefits and user comfort, the phenomenon observed in mobile game players is concerning, as they may abandon the game due to compatibility issues with specific devices. Therefore, it is crucial for game developers to focus on users’ time spent playing and encouraging in-game purchases and consider the device’s playing capacity, user reviews and feedback to retain loyal players.

In addition, this study emphasizes the direct relationship between online game addiction and in-app purchase intention. Additionally, the association between mobile game addiction and in-app purchase intention is intervened by online game loyalty, negative e-WOM and perceived risk. The findings suggest that individuals addicted to online gaming exert indirect effects on mobile game loyalty, negative e-WOM and perceived risk. This study is consistent with prior research concluding that online game loyalty, negative e-WOM and perceived risk can mediate the relationship between online game addiction and in-app purchase intention (Balakrishnan & Griffiths, 2018; M. L. Cheung et al., 2021). Therefore, the implications of this research have enriched the relevance of management and psychological literature concerning the relationship between game loyalty, negative e-commerce and perceived risk in mediating the connection between game addiction and application purchase intentions. This study further emphasizes the increasingly important role of game developers in addressing addictive behavior associated with spending time on games, which should be acknowledged and addressed. Games can offer tips to limit playing hours, promote moderation and remind players of the primary purpose of gaming – namely, to have fun and seek entertainment. The gacha rate of purchased items and characters can be increased, influencing purchase intentions in-game applications. In sum, the significance of these key findings underscores current research trends in this field. It aligns with specific insights into online gaming addiction and its repercussions on purchase intentions within online gaming applications.

6. Conclusion

In recent years, the mobile games industry has expanded exponentially and preventing gaming addiction has become a major challenge for many societies. Our study focuses on gaming addiction, taking the case of PUBG mobile game players in Indonesia. Our study finds that (1) online game addiction positively and significantly affects online game loyalty. It indicates that addiction augments player loyalty in mobile games. (2) Online game loyalty positively and significantly affects in-app purchase intentions. The willingness to buy some mobile game features depends greatly on loyalty. Players are likely to purchase when their degree of loyalty increases. (3) Online game addiction has a positive and significant negative effect on e-WOM. (4) Negative e-WOM positively and significantly affects in-app purchase intentions. The spread of negative online reviews or information about the game may block players from making in-app purchases. Players are more sensitive to negative e-WOM compared to positive e-WOM. (5) Online game addiction positively and significantly affects perceived risk. Addiction can lead to a lack of rationality and objectivity, which may result in uncontrolled behavior. (6) Perceived risk positively and significantly affects in-app purchase intentions. Players avoid risk and evaluate the possible consequences before any transaction occurs. (7) Online game addiction positively and significantly affects in-app purchase intentions. Addictive behavior can lead to overspending on in-app purchases. Furthermore, this study emphasizes the connection between mobile game addiction and in-app purchase intention, mediating with the perceived risk, mobile game loyalty and negative e-WOM. The results underline that these constructs partially mediated the influence of in-app purchase intention.

This study provides insights into how the psychological and behavioral of gamers are interconnected. It gives us an understanding of the importance of mobile game addiction, loyalty, negative e-WOM, and perceived risk significantly affecting purchase intention. The increased number of players and the number of transactions done by users on the game apps is crucial for the sustainability and efficiency of the game developers. Indeed, game developers need to understand these aspects to construct new tactical strategies to reduce the related risks in mobile gaming by understanding the perceived risks, ie (payment risks, unfeasibility of costs incurred, risks from slow connection and operating systems). For this reason, game developers need to ensure that the game is safe and secure and will not cause any harm to the players. In addition, marketers can develop related marketing strategies to reduce risks and increase the likelihood of in-game purchases. For instance, in their marketing campaigns, they should raise awareness of the mobile game’s potential positive and negative impacts. If addiction to mobile games is used as a strategy to increase in-app purchases, it tends to engender some serious ethical issues. According to Kaya et al. (2022), preventing online game addiction is a crucial objective in identifying the risks that players/gamers experience and can contribute to game addiction as a preventative measure for over-purchasing. Individual players could understand the limit, control themselves and be aware of the negative impact of overplaying mobile games.

Therefore, some limitations arise from this study. Future studies will follow the observed phenomena of mobile gaming players and employ exact measurements. This study focuses on mobile games played by more than two players simultaneously (four people). Future research can be carried out on mobile games played individually or single-player. Furthermore, it is advisable to maintain the research model and develop one that includes game design, social experience, type of devices used (Android and iPhone) and gender difference.

Disclosure statement

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

Data availability statement

The data that support the findings of this study are available from the corresponding author, Rakotoarisoa Maminiaina Heritiana Sedera, upon reasonable request.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Ronaldo Yolanda Putra

Ronaldo Yolanda Putra is a lecturer in marketing; he earned a master’s degree in management marketing at the Universitas Sebelas Maret. His research areas are in marketing, particularly digital marketing and technological adoption.

Rakotoarisoa Maminiaina Heritiana Sedera

Rakotoarisoa Maminiaina Heritiana Sedera is a Junior Lecturer and Researcher at the Institut Supérieur de la Communication, des Affaires et du Management (ISCAM), Madagascar. Having pursued his academic journey at Sebelas Maret University in Indonesia, he has been driven by a relentless pursuit of intellectual enrichment and professional advancement. Currently, he plays a pivotal role as an active member of the ‘Centre de Recherche ISCAM’ while spearheading the Entrepreneurship and Innovation Pole, which is dedicated to fostering the development of startups within the campus.

Rakotoarisoa Maminirina Fenitra

Rakotoarisoa Maminirina Fenitra, Lecturer and Postdoc Researcher at the Universitas Indonesia. His research interests are in consumer psychology and behavior, including Halal, digital and green Marketing.