Perceptions and Acceptance of Cashless Gambling Technology: An Empirical Study of U.S. and Australian Consumers

UTAUT and UTAUT2: Factors Influencing Behavioral Intention

Venkatesh et al. (2003) proposed UTAUT and examined the utility of this theory in two empirical studies of employee technology acceptance, using data collected from six organizations. Their findings support that an individual’s behavioral intention to use a system/technology is influenced by performance expectancy, effort expectancy, social influence, and facilitating conditions. Later, Venkatesh et al. (2012) reviewed consumer behavior studies and altered the prior perspective, by adjusting the UTAUT model to establish a new framework for consumers (i.e., UTAUT2). More specifically, they introduced three new predictors for consumer technology adoption as follows: hedonic motivation, price-value, and habit. To test the UTAUT2 model, Venkatesh et al. (2012) collected data from Hong Kong residents within the mobile internet technology context. They found that hedonic motivation was a stronger determinant than performance expectancy while also revealing that the other two predictors (i.e., price-value and habit) were significant determinants of behavioral intention to use mobile internet technology. UTAUT2 has been adopted for examining consumer acceptance of various types of technology, ranging from smart mobile devices (C. Huang & Kao, 2015) to automatic shuttles (Madigan et al., 2017).

Venkatesh et al. (2012) further explained that UTAUT took an approach that highlighted the importance of extrinsic motivation (i.e., performance expectancy). To complement this perspective from motivation theory, they added hedonic motivation to the UTAUT2 model. Employees in an organization evaluate their time and effort in creating opinions about the overall effort related to adopting a specific technology. Unlike employees, consumers are often required to bear the costs related to the use of a system/technology. Thus, UTAUT2 proposes that if an individual perceives that the benefits of utilizing a particular technology are greater than the costs, s/he would positively rate the price-value relationship of using it, which in turn increases his or her intention to adopt the technology (Venkatesh et al., 2012). Consistent with this argument, previous research has examined the relationship between price and perceived value, or the influence of such a relationship on the consumer’s behavioral intentions (Dodds et al., 1991; Zeithaml, 1988). For example, Zeithaml (1988, p. 11) stated, “monetary price is not the only sacrifice consumers make to obtain products,” and other costs (e.g., time, effort, search, and psychic costs) generally enter into their perception of sacrifice. In the cashless gambling context, where the price-value supersedes the costs (i.e., a net benefit to the gaming experience), there will be a higher propensity for consumers to adopt this technology.

Based on the aforementioned theories and previous studies, we propose that the six constructs in UTAUT2 (i.e., performance expectancy, effort expectancy, social influence, facilitating conditions, hedonic motivation, and price-value) will influence intention to use cashless gambling technology (see Figure 1). The definitions of these six constructs are provided in Table 1.

OPEN IN VIEWER

Figure 1.

Research Model.

OPEN IN VIEWER

Table 1.

Definitions of the Constructs.

Construct	Definition	Source
Performance expectancy	the extent to which using cashless gambling technology will “provide benefits to consumers in performing” gambling-related activities	Venkatesh et al. (2012, p. 159)
Effort expectancy	“the degree of ease associated with consumers’ use of” cashless gambling technology	Venkatesh et al. (2012, p. 159)
Social influence	“the extent to which consumers perceive that important others (e.g., family and friends) believe they should use” cashless gambling technology	Venkatesh et al. (2012, p. 159)
Facilitating conditions	“consumers’ perceptions of the resources and support available to” use cashless gambling technology	Venkatesh et al. (2012, p. 159)
Hedonic motivation	“the fun or pleasure derived from using” cashless gambling technology	Venkatesh et al. (2012, p. 161)
Price-value	“consumers’ cognitive tradeoff between the perceived benefits of” cashless gambling technology and the “cost for using” this technology	Venkatesh et al. (2012, p. 161)

TAM3: Determinants of Performance Expectancy and Effort Expectancy

As listed in Table 1, the UTAUT2 model included a wide range of predictors of behavioral intention. However, researchers have acknowledged the relevance of investigating the interrelationships between these predictors (e.g., Bervell & Umar, 2017; Nordhoff et al., 2020). More specifically, TAM3 (Venkatesh & Bala, 2008) suggested perceived ease of use (equivalent to effort expectancy in UTAUT2) and subjective norm (equivalent to social influence in UTAUT2) as key determinants of perceived usefulness (equivalent to performance expectancy in UTAUT2). TAM3 was developed by integrating TAM2 (Venkatesh & Davis, 2000) and the model of the antecedents of perceived ease of use (Venkatesh, 2000). Venkatesh and Bala (2008) explained that perceived ease of use captures the effect of “cognitive instrumental process” on perceived usefulness, with the core theoretical argument holding that individuals “form perceived usefulness judgment in part by cognitively comparing what a system is capable of doing with what they need to get done in their job” (Venkatesh & Davis, 2000, p. 190). In other words, an individual’s mental evaluation of the fit between work objectives and the outcomes of performing tasks utilizing a system/technology provides a basis for creating his or her perceptions of the usefulness of the system/technology (Venkatesh & Davis, 2000). In the context of this study, if a customer perceives that using cashless gambling technology requires too much effort, it could unfavorably influence his or her perceived benefits of this technology.

Venkatesh and Bala (2008) further elucidated that internalization is related to social influence. Internalization is equivalent to informational social influence in Deutsch and Gerard (1955, p. 629), defined as “an influence to accept information obtained from another as evidence about reality.” In the context of the current study, if important others (e.g., friends and family) suggest that cashless gambling technology could be useful, the individual might come to believe that it truly is useful and, in turn, positively influence his or her intention to use this technology. In addition, Venkatesh and Bala (2008) found that perceived enjoyment (equivalent to hedonic motivation in UTAUT2) did not have a significant direct effect on perceived usefulness. Instead, they found that perceived enjoyment had a significant direct influence on perceived ease of use and that perceived ease of use had a significant direct influence on perceived usefulness. Such relationships indicate the possibility of an indirect effect of perceived enjoyment on perceived usefulness via perceived ease of use of cashless gambling technology.

Venkatesh (2000, p. 344) proposed a model identifying potential determinants of perceived ease of use, based on the behavioral decision theory, which suggests that “anchoring and adjusting” is a critical decision-making process generally used by individuals. More precisely, when users do not have enough information about a new system, they tend to depend on general information that serves as an “anchor.” When they obtain additional information about the system (e.g., having experience with the system), they tend to adjust their perceptions to reflect the system, although still depending on the anchoring criteria (Venkatesh, 2000). TAM3 included both perceptions of external control (equivalent to facilitating conditions in UTAUT2) and perceived enjoyment as the determinants of perceived ease of use (Venkatesh & Bala, 2008). For instance, consumers will have control beliefs concerning the availability of knowledge, resources, and support, which can facilitate the usage of cashless gambling technology. In addition, consumers who perceive using cashless gambling technology to be enjoyable will likely have favorable perceptions of this technology regarding user-friendliness. In sum, UTAUT, UTAUT2, TAM3, and the aforementioned prior studies offer a useful theoretical foundation for the research model advanced in this study (see Figure 1).

Relationships Among Performance Expectancy, Effort Expectancy, and Intention to Use

In developing UTAUT (Venkatesh et al., 2003), the performance expectancy and effort expectancy constructs were adapted from the perceived usefulness and perceived ease of use constructs in Davis’s (1989) TAM. Several studies have proposed that these constructs are influential factors in technology adoption behavior while also testing the relationships with different technologies such as mobile payment (Khalilzadeh et al., 2017) and mobile hotel booking (Ozturk et al., 2016). Several previous studies have also revealed that effort expectancy was an antecedent of performance expectancy in using diverse technologies, such as internet devices (Bruner & Kumar, 2005), online teaching technology (R. Huang et al., 2011), and mobile data services (Yang, 2010). Furthermore, prior research infers that performance expectancy directly affects intention to use and that effort expectancy directly influences behavioral intention as well as indirectly via performance expectancy (Davis et al., 1989). Within the hospitality context, Ozturk et al. (2016) demonstrated that perceived ease of use positively influenced performance expectancy, which in turn had a positive impact on intention to use mobile hotel booking.

Conversely, other studies reported that effort expectancy did not have a statistically significant effect on behavioral intentions (e.g., Madigan et al., 2017; Subramanian, 1994). Subramanian (1994) explained that it is possible that the data failed to support this relationship because the two communication technologies examined were much less complicated to utilize than the software packages studied in previous research. In other words, if consumers perceive using technology as relatively effortless, the ease-of-use factor would have little or no influence on their behavioral intentions. Such inconsistent findings suggest the importance of investigating the role of effort expectancy in technology adoption with different types of technologies, which was also recommended by Venkatesh et al. (2012), the developers of UTAUT2. Based on the aforementioned theories and findings from the empirical studies, the current study expects a consumer’s performance expectancy and effort expectancy to have direct effects on his or her behavioral intention to use cashless gambling technology. In addition, we propose effort expectancy to be a salient predictor of performance expectancy. Therefore, we advanced the following hypotheses:

Effects of Social Influence on Performance Expectancy and Intention to Use

Social influence is apparent in hospitality/service settings due to the intangibility and variability characteristics of this sector (Xu et al., 2017). For instance, when an individual plans to travel, s/he usually seeks the opinions of others to decide which hotel to book, or reviews opinions posted on social media to ascertain the hotel’s quality (Flanagin et al., 2014). Similarly, customers could feel uncomfortable with adopting a new technology implemented by a casino (e.g., cashless gambling) because the efficiency and accuracy of the technology is uncertain to them. In this case, they will consult with others concerning their technology adoption decisions (Abrahamson & Rosenkopf, 1993). Prior research supported this notion by proposing that individuals tend to depend on social norms when adopting new technology (Lu et al., 2005; Venkatesh & Morris, 2000).

Furthermore, a customer will need to use cashless gambling technology via kiosks, gaming tables, or slot machines located on the casino floor where his or her social others can share their opinions about cashless gambling and recommend the use of this technology. These social others could also affect the customer’s perceptions of the specific technology and his or her intention to use it before, during, and after visiting the casino (Venkatesh & Davis, 2000; Venkatesh et al., 2012). For instance, if customers read their family members’ positive comments pertaining to cashless gambling experiences, it will positively affect their perceptions of cashless gambling technology and, in turn, increase their intention to use it. Therefore, we expect that positive opinions held by reference groups will have a positive effect on a customer’s usefulness perceptions of cashless gambling technology as well as his or her intention to adopt this technology. In step with these expectations anchored in the aforementioned theories and prior studies, we advanced the following hypotheses:

Effects of Facilitating Conditions on Intention to Use and Effort Expectancy

In UTAUT (Venkatesh et al., 2003), the facilitating conditions construct was hypothesized to affect actual usage of technology within an organization but not the intention to use. They explained that the absence of the link between the facilitating conditions and the intention to use constructs was proposed because the core concept of facilitating conditions would be mainly captured by the effort expectancy construct. In contrast, in UTAUT2 (Venkatesh & Bala, 2008, p. 162), the facilitating conditions construct was hypothesized to influence intention to use per the following reason: “The facilitation in the environment that is available to each consumer can vary significantly across application vendors, technology generations, mobile devices, and so on.” Thus, facilitating conditions will behave more like perceived behavioral control in the consumer context and influence not only the actual use of but also the intention to use the technology. Specifically, consumers with a favorable set of facilitating conditions are more likely to demonstrate a higher level of intention to use a technology.

In the context of cashless gambling, consumers have different levels of access to information/resources that can facilitate the usage of cashless technology. For example, consumers with different phones experience somewhat different features, internet connections, and data usage, which could influence their intention to use this technology. The impact of facilitating conditions (or perceived behavioral control) on the acceptance of cashless gambling technology has rarely been explored; however, this relationship has been examined in the self-service technology context (Bobbitt & Dabholkar, 2001). For instance, perceived control in using a technology-based self-service option was found to be an important predictor of service quality and intention to use (Dabholkar, 1996).

As noted in the theoretical background section, the facilitating conditions factor was proposed as a vital determinant of effort expectancy in TAM3 (Venkatesh & Bala, 2008), and their finding supported this relationship. Recent empirical studies further provided evidence for this relationship by examining learning management systems (Bervell & Umar, 2017) and automated shuttles (Nordhoff et al., 2020). The current study builds on these findings and expects that consumers who believe that they have the necessary information and resources to use cashless gambling technology will be more likely to provide higher ratings on effort expectancy. As discussed in the previous section, the current study expects effort expectancy to influence performance expectancy (i.e., Hypothesis 3). Accordingly, this study expects facilitating conditions to have an indirect effect on performance expectancy through effort expectancy, hence the following hypotheses:

Hedonic Motivation, Intention to Use, Effort Expectancy, and Performance Expectancy

Recent technology adoption studies, such as UTAUT2, insisted that consumers adopt new technology to enhance not only their performance but also enjoyment (Venkatesh et al., 2012). Several prior technology adoption studies have also demonstrated that hedonic motivation (or perceived enjoyment) acts as an intrinsic motivator in accepting new technology (e.g., Bruner & Kumar, 2005; Rosenbaum & Wong, 2015). For example, a hedonic aspect was found to be a stronger predictor of handheld internet device adoption than a utilitarian aspect (Bruner & Kumar, 2005). Collier and Barnes (2015) criticized that previous self-service research often focused on the utilitarian elements of a service experience (e.g., faster transactions) while overlooking the hedonic elements. Accordingly, they included both utilitarian and hedonic elements in their model, producing findings that revealed hedonic motivation as a key factor underlying customer delight in a self-service setting. Similarly, we expect that when customers perceive using cashless gambling technology as a hedonic experience, they will be more likely to utilize this technology.

In the context of cashless payment, Koenig-Lewis et al. (2015) claimed that perceived enjoyment could stem from the user’s novelty-seeking and instant gratification. Based on the data collected from residents in France, they provided empirical evidence for the positive impacts of hedonic motivation on both performance expectancy and effort expectancy. Their finding is in line with a previous study that found the cognitive absorption (i.e., intrinsic motivation) of using an information technology acts as a predictor of both perceived ease of use and perceived usefulness of the technology (Agarwal & Karahanna, 2000). Consistent with these previous studies, we expect that the hedonic motivation associated with using cashless gambling technology will act as a predecessor of both effort expectancy and performance expectancy.

As discussed earlier, we hypothesized that effort expectancy influences performance expectancy (i.e., Hypothesis 3). In this section, we discussed the effect of hedonic motivation on effort expectancy and performance expectancy. In accordance, the present study postulates that consumers’ hedonic motivation related to cashless gambling technology will positively affect their effort expectancy and, in turn, their performance expectancy associated with this technology. Based on the abovementioned theoretical background and extant studies, we advanced the following hypotheses:

Effects of Price-Value on Intention to Use

The price-value construct compares the cost/price of obtaining a product/service with its quality/value (Venkatesh et al., 2012; Zeithaml, 1988). Recent research in the information technology field has stressed the importance of this concept and applied it to examine the adoption of smart mobile devices (C. Huang & Kao, 2015). Their findings indicated that the price-value factor is critical for the consumer adoption of new mobile technology. In UTAUT2 (Venkatesh et al., 2012), the price-value condition for using mobile internet technology involved a tradeoff between the consumer’s monetary costs and the economic benefits of using the technology. Prior research conducted in hospitality settings also applied the concept of price-value to technology with no monetary costs for the consumer. For example, Escobar-Rodríguez and Carvajal-Trujillo (2014) adapted UTAUT2 to analyze the acceptance of low-cost carrier websites, revealing that the price-value construct significantly influenced consumer intention to use these websites.

For the consumer, the monetary costs associated with using cashless gambling technology can vary across vendors and operators. For instance, The D Las Vegas implemented ACS’s cashless wagering system (i.e., PlayOn) which charges a transaction fee of US$4, plus 2.5% of the withdrawal amount (Katz, 2020). Resorts World Las Vegas uses a cashless gambling system (i.e., Play+) that requires a 2.95% bank card/eCheck/PayPal load fee, every time customers fund their cashless accounts (Resorts World Las Vegas, 2021). One reason for using cashless gambling technology could be the high fees to withdraw cash from ATMs in casinos. For example, in 2020, the ATM fees in Las Vegas casinos were reported to range from US$7.99 to US$8.99 (Gilbertson, 2020). If a customer loads US$100 utilizing Play+, instead of withdrawing US$100 from these ATMs, the fee will be US$2.95. This would save the customer between US$5.04 and US$6.04. If customers believe that lowering such fees outweighs the effort and risks of using a new cashless gambling technology, they may be more likely to adopt it.

As using cashless gambling technology can offer potential benefits to customers (e.g., faster, safer, less expensive, and more convenient transactions), they may perceive it to add value to the overall gaming experience. That is, consumers will compare these benefits against the monetary and non-monetary costs (e.g., time, learning, and emotional costs) that are required to use cashless gambling technology (Zeithaml, 1988). Thus, we expect that customers who perceive that the benefits supersede the costs (i.e., a net benefit to the gaming experience) will be more likely to use this technology. Accordingly, we advanced the following hypothesis:

Survey Instrument

We developed the survey questions based on prior research related to new technology adoption. More specifically, we adapted the established survey instrument used in UTAUT2 (Venkatesh et al., 2012) to make it appropriate for the context of this study. To measure the performance expectancy, effort expectancy, and facilitating conditions factors, 12 items were used (four items for each factor). The social influence, hedonic motivation, and price-value factors were measured by nine items (i.e., three items for each factor). In addition, three items were used to measure the intention to use factor. All items were measured on a five-point Likert-type scale, where 1 = “strongly disagree” and 5 = “strongly agree.” In addition, we reviewed different types of cashless gambling technology and provided descriptions of each type at the beginning of the survey, to ensure that the respondents had a global understanding of the technology (see Online Appendix A). The following demographic questions were included in the survey: age, gender, and educational attainment. The face and content validity of the survey was confirmed by an executive from a gaming manufacturing company that provides cashless gambling technology, and a casino operator intimately familiar with this technology.

Samples and Procedure

At the time of this study, the early adopters of cashless gambling technology were casino operators in the United States and Australia; thus, we targeted U.S. and Australian patrons who had gambled within the previous 12 months of the survey date. More precisely, U.S. consumers who had gambled at a North American land-based casino in the last 12 months were qualified to participate in this study. Similarly, the qualified Australian consumers were those who had gambled at a club/casino in Australia over the last 12 months. We employed a quota sampling method, to align the basic demographic characteristics (i.e., gender and age) of our samples with those of the U.S. and Australian populations, based on the national demographic statistics (Australian Bureau of Statistics, 2020, U.S. Census Bureau, 2020).

An online survey software company, Qualtrics, facilitated the data collection through its panel members. The surveys were sent to 4,439 consumers residing in the United States and 6,186 consumers in Australia. The survey period began on December 14, 2020, and closed on January 30, 2021. Of the 1,069 U.S. consumers who started the survey, 486 respondents were qualified and completed the survey. Out of the 967 Australian consumers who began the survey, 405 were qualified and completed the survey. Accordingly, a total of 891 responses were used for data analysis. The sample characteristics are shown in Online Appendix B.

Measurement Model

The result of testing the CFA yielded a good fit to the data: S-Bχ² (231) = 475.53, p < .05, χ²/df = 2.06, CFI = .985, IFI = .985, TLI = .982, RMSEA = .034. To test convergent validity, all loadings should be statistically significant, and the “standardized loading estimates should be at least .5 or higher, and ideally .7 or higher” (Hair et al., 2006, p. 777). Accordingly, we removed one item from the performance expectancy factor with a loading score of <.5 (i.e., “I believe the information collected by cashless gambling technology will lead to more personalized and targeted offers”). After this respecification, the model fit improved and demonstrated a good fit to the data: S-Bχ² (209) = 387.97, p < .05, χ²/df = 1.86, CFI = .989, IFI = .989, TLI = .986, RMSEA = .031. All factor loading estimates were also statistically significant, and all standardized loading estimates exceeded .66 (see Online Appendix C). As demonstrated in Online Appendix D, we calculated the average variance extracted (AVE) estimates to evaluate convergent validity and found that all estimates were above the threshold value of .5. All factors also showed acceptable scale reliability, as the composite reliability values were higher than the recommended value of .7 (Hair et al., 2006). Finally, the results showed strong discriminant validity, as the squared correlations between a pair of latent factors were less than the AVE of each factor (Fornell & Larcker, 1981), with the exception of the following five pairs: PE-EE, PE-FC, PE-HM, PE-PV, and EE-FC.

As recommended by Bagozzi and Yi (1988) and Hair et al. (2006), the discriminant validity was further addressed by combining these two factors into one, conducting CFA with the combined model, and performing the χ² difference test on the values of the combined and uncombined models. The results showed that the χ² differences were significant at p < .001 for all five pairs of constructs, and the other fit indices in the combined models were inferior to those from the uncombined model as shown by the following measures: (a) PE-EE, Δχ² (6) = 280.70; CFI = .978, IFI = .978, TLI = .974, RMSEA = .043; (b) PE-FC, Δχ² (6) = 316.02; CFI = .976, IFI = .976, TLI = .972, RMSEA = .045; (c) PE-HM, Δχ² (6) = 412.30; CFI = .972, IFI = .972, TLI = .967, RMSEA = .048; (d) PE-PV, Δχ² (6) = 346.04; CFI = .974, IFI = .974, TLI = .970, RMSEA = .046; and (e) EE-FC, Δχ² (6) = 163.25; CFI = .982, IFI = .982, TLI = .979, RMSEA = .038. These results indicated that the uncombined model was more appropriate, suggesting that each pair of factors represented two separate constructs. Overall, the reliability, convergent validity, and discriminant validity of the measurement scales were adequate.

To identify common method bias, we used an analytical technique with a common latent factor in EQS. The EQS output demonstrated that the unstandardized loading values for all variables were .53. The common method variance is the square of this loading (.53² = .28), which was below the threshold (i.e., 50%). The result suggested that common method bias did not pose a serious threat to the findings of the current study (Eichhorn, 2014).

Structural Equation Modeling

The structural model with the data yielded a good model fit: S-Bχ² (213) = 401.69, p < .05, χ²/df = 1.89, CFI = .988, IFI = .988, TLI = .986, RMSEA = .032. As illustrated in Figure 2, an examination of path estimates (β) revealed that performance expectancy and social influence had significant direct effects on intention to use cashless gambling technology (β = .19, β = .26, respectively; p < .05), supporting H1 and H4. Price-value also had a significant direct effect on the respondents’ intention to use (β = .46; p < .05), supporting H13. Effort expectancy, facilitating conditions, and hedonic motivation did not have significant direct effects on intention to use (β = −.11, β = .05, β = .12, respectively; p > .05), failing to support H2, H6, and H9.

OPEN IN VIEWER

Figure 2.

Results of Testing the Hypotheses.

In terms of the determinants of effort expectancy, both facilitating conditions and hedonic motivation directly influenced effort expectancy (β = .76, β = .20, respectively; p < .05), supporting H7 and H10. With regard to the determinants of performance expectancy, effort expectancy and hedonic motivation were found to be significant determinants, but not social influence, supporting H3 and H11 (β = .55, β = .39, respectively; p < .05), but not supporting H5 (β = .07, p > .05). In addition, facilitating conditions and hedonic motivation indirectly affected performance expectancy, via effort expectancy, supporting H8 and H12 (β = .42, β = .11, respectively; p < .05).

Theoretical Implications

The present study addresses the lacunae of existing technology research within the hospitality domain, by developing a model to better understand the influential factors associated with the adoption of cashless gambling technology. In addition, we offer an empirical test of this model with samples obtained from U.S. and Australian casino patrons. To date, no study has examined the effects of the six constructs in the UTAUT2 (Venkatesh et al., 2012) model on the intention to use cashless gambling technology. Furthermore, the current study extended the UTAUT2 model by adding the potential determinants of performance expectancy and effort expectancy, based on TAM3 (Venkatesh & Bala, 2008), as well as adding effort expectancy as a mediator. By addressing this knowledge gap, our findings contribute to the literature in both the hospitality and the technology fields.

In line with the findings of previous studies (e.g., Bruner & Kumar, 2005; R. Huang et al., 2011; Yang, 2010), effort expectancy was found to be a key determinant of performance expectancy. More specifically, effort expectancy had a stronger effect on performance expectancy than hedonic motivation. This finding denotes that the complexity of cashless gambling technology should be reduced to increase the perceived benefits of using it. Similar to the findings in Venkatesh et al. (2012), performance expectancy had a significant effect on the intention to utilize cashless gambling technology. This finding emphasizes the importance of implementing an accurate and reliable cashless gambling system that can assist customers in completing transactions faster and increase the convenience of their gambling experience.

Prior research indicated that social influence could play an important role in technology acceptance (e.g., Lu et al., 2005; Venkatesh & Morris, 2000); however, a theory-driven study that explores the influence of others (e.g., family, friends, and colleagues) on cashless technology adoption in a casino setting is comparatively rare. By addressing this gap, our research yields new insights into the hospitality and technology adoption literature by increasing the understanding of the relationship between social influence and consumer adoption of cashless gambling technology. Our findings revealed that social influence significantly and positively affected the intention to use cashless gambling technology. In line with previous studies (e.g., Lu et al., 2005; Venkatesh & Morris, 2000), this finding implies that the positive opinions of others are essential to the adoption of cashless technology.

In addition, we found that social influence failed to significantly impact performance expectancy. Although this finding is not consistent with the findings in TAM3 (Venkatesh & Bala, 2008), it corroborates the finding of Bervell and Umar (2017) who examined technology acceptance among Ghanaian distance education tutors. Our outcome could be related to the novelty of cashless gambling technology, as it was not widely available at the time of this study. As people tend to observe the behavior of others and attempt to adjust their own attitudes and behavior (Nysveen et al., 2005), future research may need to re-examine this relationship after more casinos implement cashless gambling technology.

The positive impact of facilitating conditions on effort expectancy implies that patrons who think they have the necessary resources are more likely to consider cashless gambling technology easy to use. This finding highlights the need to provide proper resources and support for consumers to utilize this technology. It also aligns with previous studies that revealed a positive effect of facilitating conditions (or perceived behavioral control) on effort expectancy (or perceived ease of use; Bervell & Umar, 2017; Nordhoff et al., 2020; Venkatesh & Bala, 2008). In addition, our study revealed that facilitating conditions were a stronger determinant of effort expectancy than hedonic motivation, which corroborates the finding in TAM3 (Venkatesh & Bala, 2008), as well as a recent study of automated vehicles (Nordhoff et al., 2020).

Our study further demonstrated that facilitating conditions had a substantial indirect effect on performance expectancy, through effort expectancy. These findings support Venkatesh and Bala (2008) who claimed that organizational support (e.g., providing necessary infrastructure, help desks, hiring experts, and user training) can play a vital role in determining the perceived ease of use and usefulness of new systems. However, they did not test the mediating role of effort expectancy between facilitating conditions and performance expectancy. Also, the types of technologies they tested were employee-facing (e.g., a proprietary system to manage daily operations and a customer account management system), which were quite different from cashless gambling technology for consumers. Thus, this study contributes new insights into the mediating role of effort expectancy in the technology acceptance literature.

Our study also contributes to the literature by examining an intrinsic factor, hedonic motivation, within the hospitality/gaming context. The related findings suggest that hedonic motivation is a key determinant of both effort expectancy and performance expectancy. Furthermore, the results showed that hedonic motivation had a significant direct impact on performance expectancy, via effort expectancy. In sum, our outcomes echoed the findings of prior studies demonstrating that the perceived enjoyment derived from utilizing specific technology can lead to positive perceptions of the technology (Agarwal & Karahanna, 2000; Koenig-Lewis et al., 2015). Venkatesh and Bala (2008) found a significant direct effect of hedonic motivation on effort expectancy, and a significant direct effect of effort expectancy on performance expectancy; however, they did not examine the indirect effect of hedonic motivation (i.e., the mediating role of effort expectancy between hedonic motivation and performance expectancy). Thus, our findings offer insight outlining the mechanism in which these factors interact to create an individual’s perceptions of new technology. Specifically, it suggests that customers tend to perceive cashless gambling technology to be easier to use and more useful when they believe that using it will provide an entertaining and pleasant experience.

We found that price-value produced a significant impact on the intention to use cashless gambling technology. This infers that the greater the cost savings or net benefits that a customer thinks s/he may earn by using cashless gambling technology, the greater her or his intention will be to adopt it. This is congruent with the findings of previous research (e.g., Escobar-Rodríguez & Carvajal-Trujillo, 2014; C. Huang & Kao, 2015) suggesting that the price-value factor plays an important role in driving consumer intention to use a new technology. It is important to note that these prior studies adapted the price-value construct in UTAUT2 to particular technologies that are notably different from cashless gambling (i.e., low-cost carrier websites and phablets such as Galaxy Note). Our study confirmed the specific application of this construct to the hospitality and gaming context, following the recommendation of Venkatesh et al. (2012) concerning the need to examine the UTAUT 2 constructs by applying them to various technologies.

Finally, effort expectancy failed to produce a significant effect on behavioral intentions toward cashless gambling technology, which suggests that this relationship originally proposed/tested in UTAUT (Venkatesh et al., 2003) and UTAUT2 (Venkatesh et al., 2012) may not be applicable in all domains. This finding is in line with previous studies that also failed to support a significant effect of perceived ease of use on behavioral intentions (e.g., Madigan et al., 2017; Subramanian, 1994). Our study offers further evidence of the difficulties in comprehending how effort expectancy may influence behavioral intention when technology is relatively simple. As the previous studies explained, it is possible that cashless gambling technology worked in a similar fashion to other types of cashless technology (e.g., mobile payment at grocery stores); hence, the respondents may have thought that new skills or continuous input/effort would not be necessary to use this technology effectively.

Practical Implications

First, performance expectancy was positively influenced by effort expectancy. If customers think that learning how to use cashless gambling technology is easy, clear, and understandable, they will be more likely to expect higher utilitarian performance from this technology. Thus, cashless gambling developers should ensure that the system is easy to use, and casino operators should emphasize the ease of use when they are marketing this new system to consumers. Second, this study revealed that performance expectancy is a significant predictor of intention to use cashless gambling. This finding suggests that it is vital for casino operators to acknowledge the need for a user-friendly cashless system to improve ease of use and to lower user resistance. Furthermore, providing effortless cashless gambling technology will not be sufficient to improve the intention to use it if patrons do not consider the technology useful. That is, to increase patrons’ intent to use cashless gambling technology, operators should concentrate on its usefulness (e.g., benefits associated with convenience and efficiency). Marketers can also attempt to put more emphasis on these benefits when sending emails to customers. For instance, to highlight the time-saving benefit, marketers may track the average time customers spend to cash out at the casino cage in the absence of cashless technology. Benefits such as this one could then be emphasized in marketing materials. They could communicate other beneficial features of cashless gambling technology (e.g., receiving offers via the company’s mobile wallet app). When social media reveals problems with the technology and/or negative reviews related to its functionality, we recommend promptly addressing any such issues and sending emails apologizing for the inconvenience. Casino operators can also utilize the survey questions offered in this study to obtain customer feedback on the perceived usefulness of the technology (see Online Appendix C).

Third, this study revealed that important others can influence the intention to use cashless gambling technology. This finding indicates that gaming companies should realize that the adoption of this technology could be enhanced by increased positive word of mouth. During the system development and deployment phases, we recommend that gaming operators invite potential users or influencers to their properties to allow them to experience the new cashless system and share their positive experiences with other consumers. This approach will also provide an opportunity for operators to receive feedback regarding their negative experiences (e.g., incompatible mobile devices, inconvenient features, or unclear instructions). After the deployment of the cashless gambling system, we suggest operators and vendors continue to collect sentiments from the users of this system. For example, when management sends confirmation or appreciation emails to customers, links to the company’s social media site can be embedded to facilitate the sharing of customer experiences with the system. Then management can use the positive reviews and testimonials in marketing materials. As most gaming companies have loyalty programs, management could award social media users with reward points when they share useful videos, photos, or comments related to their cashless gambling systems.

Fourth, although facilitating conditions did not have a significant direct impact on intention to use cashless gambling, the data revealed that this factor significantly affected effort expectancy directly as well as performance expectancy indirectly. These findings imply that a supportive arrangement for consumers to use cashless gambling technology is likely to give rise to positive opinions about its perceived ease of use and usefulness. A supportive arrangement could include the provision of accessibility enhancements, such as the following: (a) offering different ways to place wagers (e.g., loyalty cards, mobile wallets, and tickets); (b) ensuring an uncomplicated registration and payment process; (c) setting-up help desks on the casino floor where consumers can obtain information or ask questions about cashless gambling technology; and (d) training slot attendants and dealers to assist customers in their use of this new technology.

Fifth, our results revealed that hedonic motivation can have positive impacts on effort expectancy as well as performance expectancy. Therefore, gaming companies should attempt to understand which functions/features of cashless gambling technology can offer a pleasant and entertaining experience for customers. When gaming management conducts a customer satisfaction survey, cashless gambling-related questions can be included, to identify enjoyable functions/features and suggest new ones. Based on this type of customer feedback, the gaming company will be able to provide more personalized and engaging features, such as incorporating social casino games into its mobile wallet app. If such adjustments or modifications are made, it would be important to measure the effect of the changes on the hedonic motivation, effort expectancy, performance expectancy, and behavioral intention constructs. It may be helpful for casino executives to utilize the questions provided in the current study to evaluate the impacts and effectiveness of any such changes.

Sixth, we found that the price-value construct played a significant role as a direct driver of intent to use cashless gambling. This finding indicated that patrons who believe they will gain greater cost savings and net benefits from using this technology will demonstrate a stronger intention to use it. Thus, we suggest that casino management develop strategies to increase consumer awareness of cost savings and other benefits. For example, a casino could interface its loyalty programs with a cashless gambling system. Such an interface would allow the system to utilize information from the customer database, which includes customer preferences, spending history, demographics, and much more. For instance, when a customer uses a mobile wallet app to play a slot machine, the app can be programmed to automatically deliver promotional and informational messages, based on his or her profile and location. Marketers should try to convey the price-value benefits through a variety of channels (e.g., social media, emails, in-room TVs/magazines, and digital billboards). Messages should emphasize cost-saving traits likely to inspire interest in the technology, such as no-fee or low-fee access to money. This would present a clear and recognizable comparison in the minds of gamblers who have paid the higher fees charged by traditional modes of access (e.g., ATMs in casinos). Subsequent to this type of messaging, it may be helpful for marketers to use the price-value scale provided in this study to examine the traction of the messaging as well as the general importance of this construct. If they discover that the scores of the price-value items are low, they should carefully assess the related functions/features of the technology and consider conducting further targeted inquiries to better understand customer expectations.

Limitations and Future Research

There were some limitations that warrant future research. Although we used quota sampling to ensure variability in the data, we only gathered samples from two different countries. It is recommended that future researchers replicate this study with consumers in other countries. Another limitation of this study could be related to our cross-sectional design. Although the literature review revealed that this design has been frequently utilized in technology acceptance studies, other designs may provide additional insights or yield different results. For example, we recommend that future researchers conduct a longitudinal study to explore whether and how the relationships between the factors change over time. Researchers could find a casino that recently introduced cashless gambling technology and survey its regular users via the questions provided in this study (see Online Appendix C). After a certain period of time (e.g., 6 months), the casino could survey the same users, asking the same questions, to explore any changes in their perceptions and acceptance of this technology. In this case, user experience could be added as a moderator to the research model. Alternatively, researchers could select a casino that plans to alter features/functions of its existing cashless gambling system. Such an experimental condition would allow them to measure the relationships in our model, both before and after making the changes.