Consumers
Acceptance towards
NFC Mobile
Payments
A research conducted by a Business Information System 2013-‐2014 student of the University of Amsterdam for his Master Thesis
Student: Shailin Mohan 100078754 Supervisor: Dick Heinhuis Second Assessor: Tom van Engers
………... ………... ………...
1. Introduction
Mobile payments is a combination of payment systems with mobile devices and services to provide users with the ability to initiate, authorize and complete a financial transaction over mobile network or wireless communication technology (Chandra et al., 2010). The boundless invasion of mobile devices and its
closeness to the user, with the far-reaching specification of the devices, makes them applicable for many payment scenarios and even for carrying everything that would fit into a wallet. This can provide the mobile network operators the opportunity to develop a new business model and increase their revenues (Chen, 2008). Mobile payments consist of many types of payments, like payment via a mobile app (called MyOrder), bank transfer via a mobile app, payment by sms etc. However this research will focus on near field communication (NFC). NFC is a two-way, short-range communication method. This method facilitates the transaction between two devices, the mobile devices and the payment terminal, when in close range of each other. The transactions can provide service
providers information on the consumers’ preferences, which can be used to offer personalized discounts, coupons etcetera.
Nowadays large international companies also make use of NFC mobile payments (Slade et al., 2014). For example, Google has the GoogleWallet and MasterCard with Samsung’s Smart Ticket app. Also on a national level in the Netherlands companies are more focusing on NFC mobile payments: Vodafone recently introduced the SmartPass and Rabobank has the MyOrder Cashless payment. Despite the investment of the providers made, worldwide adoption of NFC has been very low (Gartner, 2013). This proposethat NFC mobile payments
providers need to better understand the stimulators of consumers acceptance of NFC mobile payments to adjust their strategies according to consumer needs (Schierz et al., 2010). In addition, foreign business models of mobile payment cannot directly be applicable to different cultural contexts due to the different market constraints in terms of economic, technology and social aspects. NFC mobile payment adoption in the context of the Netherlands, where to date no similar research has been undertaken, is also important.
NFC mobile payments have a number of advantages, because of the ubiquity of the device, there is no fuss about bankcards, receipts and tickets; everything is available on the mobile device. Also the payment itself will be quicker and the extra information, such as account balance, will be easier to see. However, there
are uncertainties and risks involved due to the vulnerability of both the devices and the network to hacker attacks (Zhou, 2014). For example, the public transportation card (in Dutch: OV-chipkaart) in the Netherlands also works with NFC technology. This card was hacked shortly after it’s release (the card could be upgraded for free). The above-mentioned advantages and risks can influence the acceptance of NFC mobile payments.
In search for the factors, which affects the acceptance of NFC payments, several acceptance models and theories are available for research. For example,
Diffusion of Innovation (Rogers, 1962 & 1995), Theory of Planned behavior (Ajzen, 1991), Technology Acceptance Model (Davis, 1989) and the Unified Theory of Acceptance and Use of Technology (Venkatesh et al., 2003 & 2012). Most of these acceptance models are focused on the employee acceptance towards technology in a workplace context. However the focus of this research is the consumer’s acceptance of the use of technology for mobile payments. For the consumers focus, the Unified Theory of Acceptance and Use of Technology will be well suited for this research, because the model focuseson the consumer acceptance of technology. This model consists of constructs from different models on acceptance of technology. Moreover the construct selection is based on the consumers perspective and needs of the user of today.
Taking the above in consideration, the following research question will be central in this research:
“What factors influence the intention to use NFC mobile payments by consumers in the Netherlands?”
The research question will be supported by the following sub questions:
- Which models/theories are available for technology acceptance? - Which model/theory is well suited to determine the factors, which
influence the intention to use NFC mobile payments?
- Which factor is the strongest for acceptance of NFC mobile payments?
The remainder of the research is organized as followed:
Chapter 2 will describe the technology acceptance models. First, the models that are related to the UTAUT. Secondly, the UTAUT model will be described. Finally, the UTAUT2 model, which is an extension of the UTAUT, will be described.
Chapter 3 will describe the methodology of this research. First, the hypothesis will be described. Secondly, the conducted survey will be described.
Chapter 4 will present the results of the survey and research. First, demographic information about the respondents will be described. Secondly, a factor analysis will be conducted. Thirdly, the reliability of the results will be tested. Fourthly, a regression analysis will be conducted. Finally, the multicollinearity will be checked amongst the variables to test whether they measure redundant information. Chapter 5 will present the discussion section where explanation will be discussed on topics.
Chapter 6 will present the conclusion with the future research.
2. Theory
2.1 Literature review
Studies on the user’s acceptance of technology mobile payments are quite extensively. There have been developed several models and theories for technology acceptance: Diffusion of Innovation (DOI), Theory of Planned Behavior (TPB), Technology Acceptance Model (TAM) and Unified Theory of Acceptance and Use of Technology (UTAUT). These models are selected from literature of studies on mobile payments. The following section will describe the models and one will be suitable for investigate the factors that influence the intention to use NFC mobile payments. Table 1 gives an overview of the relevant literature on mobile payments with their location and subject.
Table 1 - Overview of acceptance research of mobile payments.
Source Theory Application Location
Brown et al., 2003 DOI Mobile banking South Africa Chen, 2008 TAM Mobile payment United-States of America Cheong et al., 2004 TAM Mobile payment Korea
Dahlberg & Oorni, 2007 UTAUT Mobile payment Finland Goeke & Pousttchi, 2010 TAM Mobile payment Germany Hongxia et al., 2011 UTAUT Mobile payment China Kim et al., 2010 UTAUT Mobile payment Korea Leong et al., 2013 UTAUT2 Mobile payment Malaysia Puschel et al., 2010 D-TPB Mobile banking Brazil Schierz et al. 2010 TAM Mobile payment Germany Shin, 2010 TAM Mobile payment United-States of America Slade et al., 2014 UTAUT2 Mobile payment United Kingdom Suoranta, 2003 DOI Mobile banking Finland Wang & Yi, 2012 UTAUT Mobile payment China Wu & Wang, 2003 TAM Mobile payment Taiwan Zmijewska et al., 2004 UTAUT Mobile payment Australia
Diffusion of Innovation (DOI)
First, Rogers (1962,1995) developed the DOI, although this model was not
designed for Information System (IS) research it has been used for explaining the acceptance of IS. The DOI suggests that when consumers perceive the
innovation to have a greater relative advantage, observability, trialability and compatibility, the rate of technology adoption will increase. Brown et al. (2003) and Suoranta (2003) are the only two quantitative studies in mobile banking adoption, which have used the DOI as their core theory. According to Slade et al. (2013) DOI was relative, in comparison to other studies, which applied different research models, unsuccessful due to its low percentage of variance in behavior intention. The DOI will be insufficient for the acceptance of NFC mobile
payments, due to the purpose of the model which is explaining the acceptance of IS and not NFC mobile payments.
Theory of Planned Behavior (TPB)
The Theory of Planned Behavior suggests that behavior is a direct function of behavioral intention which itself is driven by an individual’s attitude, subjective norms and perceived behavioral control (Ajzen, 1991). The TPB is extended, also known as D-TPB, by decomposing the antecedents of attitudinal beliefs. Puschel et al. (2010) is the only study which has used the D-TPB as the core theory. In this study there is a high percentage of variance in behavioral intention to adopt mobile banking. According to Slade et al. (2013) components such as subjective norm have been included by other research (Schierz et al, 2010; Sripalawat et al, 2011). The component subjective norm has also been used in the UTAUT2 model of Venkatesh et al. (2012). This model will be insufficient for the
acceptance of NFC mobile payments due to the context of the application of TPB, which is in the health care and not NFC mobile payments.
Technology Acceptance Model (TAM)
The Technology Acceptance Model (TAM) is developed for IS by Davis (1989). According to TAM, usage is a direct function of behavioral intention, which itself is influenced by attitudes towards the IS formulated from the innovation’s perceived usefulness and perceived ease of use (Davis, 1989). The model is originally intended to predict employee the acceptance and the usage of technology in organizational context. Schepers & Wetzels (2007) are the first who applied the
TAM to examine individual acceptance of technology in a consumer context, but have not empirically validated their research. According to Slade et al. (2013), TAM is the most used amongst all theories in research for mobile payments. Since the development of the TAM (Davis, 1989) studies have used the model with mobile payments.For example, Dahlberg, Mallat, & Öörni, (2003) is one of the first mobile payments adoption study.They added in their study the factor ‘trust’ to the TAM to better describe consumer acceptance of mobile payment solutions. The studies (e.g. Dahlberg et al., 2003; Lee & Warkentin, 2004) on the adoption of mobile payments before 2003 were mostly qualitative or descriptive of nature. In 2004 quantitative research on mobile payment adoption began to emerge. Cheong & Park, (2004) was one of the first who did a quantitative research on mobile payment adoption using the TAM, after Cheong & Park (2004) several studies (Chen, 2008; Goeke & Pousttchi, 2010; Shin, 2010) examining followed.
Wu & Wang (2003) used the TAM2 (Venkatesh & Davis, 2000) to model users acceptance of using mobile payments. The TAM2 is an extension of the TAM, as you can see in figure 1. In this extension the social influences and cognitive instrumental processes are added. However, this model is also intended to predict employee acceptance of technology and the usage of technology in organizational context.
Figure 1 - TAM2 (Venkatesh & Davis, 2000)
Unified Theory of Acceptance and Use of Technology (UTAUT)
After the TAM2 model Venkatesh et al. (2003) developed the Unified Theory of Acceptance and Usage of Technology (UTAUT). The UTAUT model is derived from several theories and models. From these models several key constructs were derived:
• Performance expectancy was derived from the TAM’s perceived
usefulness and from the DOI’s relative advantage;
• Effort expectancy was derived from TAM’s perceived ease of use and
from DOI’s complexity;
• Social influence was derived from the TPB’s subjective norm and DOI’s
image and
• Facilitating conditions was derived from the DOI’s compatibility and TPB’s
perceived behavioral control (Venkatesh et al., 2003).
These constructs on behavioral intention of use behavior are moderated by different combinations of gender, age, experience and voluntariness of use. Hongxia et al. (2011) and Wang & Yi (2012) have empirically validated UTAUT in the mobile payment context, but excluded the UTAUT moderators. Several studies (Dahlberg & Oorni, 2007; Hongxia, Xianhao, & Weidan, 2011; Kim,
Mirusmonov, & Lee, 2010; Zmijewska, Lawrence, & Steele, 2004) have used the UTAUT in mobile payment adoption.
The TAM, TAM2 and UTAUT models were originally developed to explain employee technology acceptance within an organizational context, for the
consumer context Venkatesh et al (2012) developed the UTAUT2 model, which is an extension of the UTAUT model. This model is tailored to the consumer
technology acceptance context, which is also the context of this research and therefore will be used in this research. In the study of Leong, Hew, Tan, & Ooi (2013) the UTAUT2 is used to determine the factors influencing the adoption of Near Field Communication (NFC)-enabled mobile credit card in Malaysia.
Another study by Slade et al. (2014) used the UTAUT2 to show that performance expectancy is the strongest predictor in their research on mobile payments in the UK.
According to Slade et al. (2013) mobile payment adoption research is still in its infancy with regard to the TAM, UTAUT and UTAUT2 models. The relevant studies, mentioned earlier, have taken place across different countries. The table gives an overview of the studies with the theories they have used for their
research. Moreover, it shows that there haven’t been conducted a research on mobile payment adoption in the Netherlands context.
2.2. Unified Theory of Acceptance and Use of Technology (UTAUT)
Venkatesh et al. (2003) developed the Unified Theory of Acceptance and Use of Technology (UTAUT) model. This model is based on eight acceptance models, each with a different set of acceptance determinants. These eight models are the theory of reasoned action (TRA), the technology acceptance model (TAM), the motivational model (MM), the theory of planned behavior (TPB), a model
combining the technology acceptance model, the theory of planned behavior, the model of PC utilization (MPCU), the innovation diffusion theory (IDT) and the social cognitive theory (SCT). These models are described in appendix A, with their core constructs and it’s definition.
Of all the constructs from the different models there are four constructs that will play a significant behavioral role as direct determinants of intention and usage: performance expectancy, effort expectancy, social influence and facilitating conditions (see figure 2).
Figure 2 - The UTAUT model (Venkatesh et al., 2003)
Performance expectancy
According to Venkatesh et al. (2003), the determinant performance expectancy can be defined as the degree to which an individual believes that using the system will help him or her to attain gains in job performance. From the different models, five constructs relate to performance expectancy: perceived usefulness from the TAM, extrinsic motivation from the MM, job-fit from the MPCU, relative
advantage from the IDT and outcome expectation from the SCT. These
constructs are all related to enhancing the job performance. As visible in the figure2, the relationship link between performance expectancy and behavioral intention is moderated by gender and age. Based on research, men tend to be more task-oriented (Minton & Schneider, 1980) and therefore performance expectancy is strongly noticeable to men because the focus of performance expectancy is on task accomplishment (Venkatesh et al., 2003). Comparable to gender, age is to play a moderating role according to research (Hall & Mansfield, 1975; Porter, 1963).
Effort expectancy
Effort expectancy is the degree of ease associated with the use of the system (Venkatesh et al., 2003). This determinant captures three constructs from the different models of acceptance: perceived ease of use from the TAM, complexity from the MPCU and ease of use of the IDT. These constructs (see appendix A) have similarities in the definitions; all of them are related to the usability of a system. Also effort expectancy has moderators, these are gender, age and experience. Venkatesh & Morris (2000) suggests that effort expectancy is more notable for women than for men. According to earlier research (Plude & Hoyer, 1985), when men get older their ability to process complex stimuli and allocating attention to information on the job tends to be difficult. These are necessary when using software systems.
Social influence
The degree to which an individual perceives that important others believes he or she should use the new system, is called social influence (Venkatesh et al., 2003). This determinant captures three constructs from the different models of acceptance: subjective norm from the TRA, social factors from the MPCU and
image from the IDT. Similarity can be seen in the definitions of these constructs,
all are associated with influence of the status of a person or group. This
determinant, social influence, is the only one that has 4 moderators: gender, age, experience and voluntariness of use. Venkatesh et al. (2000) suggests that women look to be more sensitive to others opinion and find social influence to be more notable when forming an intention to use new technology. According to Rhodes (1983) that affiliation needs is to be increased with age. Older workers are more likely to place increased importance on social influence with the effect of decrease with experience.
Facilitating conditions
The last determinant is facilitating conditions, this is the degree to which an individual believes that an organizational and technical infrastructure exists to support the use of the system. Facilitating conditions is a determinant of use behavior. This determinant consists of three different constructs: perceived
behavioral control from the TPB, facilitating conditions from the MPCU and compatibility from the IDT. The similarity among these constructs is that,
according to Venkatesh et al. (2003), they are used to include aspects of the technological and/or organizational environment that are designed to remove barriers to use the system. The influence of facilitating conditions of usage will be moderated by age and experience (Venkatesh et al., 2003).
According to Venkatesh et al. (2003), behavioral intention will have a positive influence on technology usage. This is consistent with the underlying theory of all the intention models mentioned before. This is also visible in the model (see figure 2) by the relation between behavioral intention and use behavior. As mentioned before the UTAUT model is originally developed to explain employee technology acceptance and use. The focus of this research is on the consumer use context, therefore the extension of the UTAUT by Venkatesh et al. (2012), also called UTAUT2, will be appropriate.
UTAUT2
Venkatesh et al. (2012) developed an extension of the UTAUT, called the
UTAUT2. This model focuses on the consumer use and context. Venkatesh et al. (2012) added three new constructs: hedonic motivation, price value and habit. Various constructs related to hedonic behavior (for example: enjoyment) are important in consumer product and/or technology use (Brown & Venkatesh 2005; Holbrook & Hirschman 1982; Nysveen et al. 2005; van der Heijden 2004).
According to Venkatesh et al. (2012), hedonic motivation will complement UTAUT’s strongest predictor that emphasizes utility. Furthermore, contradictory to workers in an organization consumers have to pay the costs and these can be of an influence to the consumer’s decisions (Brown & Venkatesh 2005; Chan et al. 2008; Coulter & Coulter 2007). By adding a construct that is related to price and costs will complement UTAUT’s existing resource considerations that focus only on time and effort. Finally, habit as another critical predictor of technology use will complement the focus on intentionality as the overarching mechanism and key driver of behavior (Venkatesh et al., 2012).
As regards to the moderators, the moderator voluntariness from the UTAUT model is dropped, because the most consumer behaviors are completely voluntary (Venkatesh et al., 2012).
Figure 3 - UTAUT2 model (Venkatesh et al., 2012)
Hedonic motivation
According to Venkatesh et al. (2012), hedonic motivation is defined as the fun or pleasure derived from using a technology. Hedonic motivation plays an important role in determining technology acceptance and use (Brown & Venkatesh, 2005). Research on IS (van der Heijden 2004; Thong et al. 2006) showed that hedonic motivation can be of an influence on the technology acceptance and use. Also in consumer context is hedonic motivation an important determinant of technology acceptance and use (Brown & Venkatesh 2005).
Price value
In an organizational use setting the employees do not have the costs of the technology, in contrary to consumers who do have costs of use. According to Chan et al. (2008), cost and price may have a convincing impact on consumers’ technology use. Venkatesh et al. (2012) define price value as consumer’s cognitive tradeoff between the perceived benefits of the applications and the
monetary cost for using them. When the perks of using a technology are perceived to be greater than the monetary costs and such price value has a positive impact on intention, then the price value can be marked as positive.
Habit
Habit is defined as the extent to which people tend to perform behaviors automatically because of learning (Limayem et al., 2007). According to
Venkatesh et al (2012), habit is operationalized in two ways: habit is viewed as prior behavior and is measured as the extent to which an individual believes that behavior has to be automatic.
The models, DOI, TPB, TAM, UTAUT and UTAUT2 are models to investigate the technology acceptance. To determine which model is well suited to investigate what factors influence the intention to use NFC mobile payments is it important to look at the focus of this research and the models. The focus of this research is on the consumers, the factors of the intention of the consumers to use NFC mobile payments are going to be investigated. Therefor, The UTAUT2 model of
Venkatesh et al. (2012) is the best model to use to investigate the factors of the intention to use NFC mobile payments for consumers in the Netherlands. The focus of this model is on the consumers whereas the focus of the other models is on the employee in a workplace, which implies that the context is also different. Also the voluntariness is different between the employee context and consumers context. Whereas, the employee is somewhat obligated to use and accept the technology and the consumer is free to use and accept the technology.
Furthermore, the UTAUT2 model is developed in 2012 and therefor is it more up-to-date then the other models, this will provide the research more representative results. This is also answers sub question one and two.
3. Methodology
In this chapter the research design (see paragraph 3.1) and the survey design (see paragraph 3.2) will be described. In paragraph 3.1 hypotheses will be described. The hypotheses are each based on every construct of the UTAUT2 model. Paragraph 3.2 will describe the construction of the survey.
3.1 Research Design
Figure 4 presents the model for this research. It is derived from the UTAUT2 model of Venkatesh et al. (2012). Every hypothesis corresponds to a construct or
moderator. The hypotheses are divided into two groups: hypotheses on the constructs that are directly connected to the intention to use NFC mobile
payments and hypotheses on the moderators. These hypotheses are tested with the help of a survey (see appendix B). The participants of the survey were recruited by using online networks such as Facebook, Twitter and email. The questions in the survey were based on the hypotheses that are described in the next paragraph. For example, in appendix B is the survey presented and the questions are coded like PE1, PE2, PE3, EE1, EE2, EE3, EE4, etcetera. PE1, PE2 and PE3 correspond to the construct performance expectancy and the same principle with the other. The survey will give quantitative results for the UTAUT2 model. According to Slade et al. (2014), a survey is a good way to gather
information for the model.
Figure 4 - Research model
3.1.1 Hypotheses
The hypotheses were implemented from the constructs and moderators of the UTAUT2 model (Venkatesh et al., 2012). In the paper of Venkatesh et al. (2012) the constructs were in the context of mobile internet. In this research the context
will be NFC mobile payments. Therefore, the hypotheses will be formulated in the context of mobile payments. Paragraph 3.1.1.1 will describe the hypotheses that are directly related to the constructs and paragraph 3.1.1.2 will describe the hypotheses that are related to the moderators.
3.1.1.1 Hypotheses -‐ Direct links
In the consumer’s context, performance expectancy is the degree to which using a technology will provide benefits to consumers in performing certain activities (Venkatesh et al., 2012). In UTAUT, the original model, Venkatesh et al. (2003) found performance expectancy to be the strongest predictor of intention, this applies in the employee context. However, in the consumer context, Venkatesh et al. (2012) found in the UTAUT2 model hedonic motivation and habit the strongest predictors of behavior intention. Performance expectancy has been supported in the mobile payment context by studies of Hongxia et al. (2011) and Wang & Yi (2012). As NFC mobile payment could lead to the end of carrying cash and cards and offer a quicker payment method, then it will offer useful benefits that are likely to be important drivers of adoption (Slade et al., 2014). Taking the above mentioned in consideration, the first hypothesis is formulated as follows:
H1: Performance expectancy (PE) has a positive influence on the intention to use NFC mobile payments
The degree of ease associated with consumers use of technology is defined as effort expectancy by Venkatesh et al. (2012). Wang & Yi (2012) found that effort expectancy is the most significant predictor of intention to use mobile payments. Hongxia et al. (2011) did not find support for the significant effect of effort
expectancy on behavioral intention. Nevertheless, as NFC mobile payments use different technology to existing payment systems, it is likely that the perceived degree of ease associated with using NFC mobile payment will affect behavioral intention (Slade et al., 2014). Therefore, the second hypothesis is:
H2: Effort expectancy (EE) has a positive influence on the intention to use NFC mobile payments
According to Venkatesh et al. (2012), social influence is the extent to which consumers perceive that important others believe they should use a particular
technology. The belief is that people tend to turn to their social network to reduce any doubt, which starts due to uncertainty of a new technology. Social influence is, of the four original UTAUT constructs, the most tested construct in the context of mobile payments, and its effect on behavioral intention has acquired more support (Hongxia et al., 2011; Tan et al., 2014; Yang et al., 2012) than rejection (Shin, 2010; Wang & Yi, 2012). According to Slade et al. (2014), non-users of NFC mobile payments are more concerned about financial risks associated with a new payment system then they are likely to seek reassurance from important others. Thus, the third hypothesis is formulated as follows:
H3: Social influence (SI) has a positive influence on the intention to use NFC mobile payments
Facilitating conditions is defined as consumers perceptions of the resources and support available to perform a behavior (Venkatesh et al., 2012). According to Slade et al. (2014), the effect of facilitating conditions on behavioral intention has gained support in the mobile payment context, although the connection has not been widely examined. As NFC mobile payments use new technologies and offerings are currently fragmented, then logically facilitating conditions are likely to affect behavioral intentions. Therefore, the fourth hypothesis is:
H4: Facilitating conditions (FC) has a positive influence on the intention to use NFC mobile payments
Venkatesh et al. (2012) added price value to the UTAUT2, which is defined as consumer’s cognitive tradeoff between perceived benefits of the applications and the monetary costs for using them. According to Hongxia et al. (2011), financial costs have been found to negatively affect behavioral intention. Yang et al. (2012) found that financial costs negatively affect behavioral intention for non-users, but was not significant for actual users. Tan et al. (2014) found the effect of financial costs to be insignificant. The financial costs of acquiring an NFC enabled device and subscribing to network charges can be weighed against the perceived benefits of having a convenient payment system (Slade et al., 2014). Hence, the fifth hypothesis is formulated as follows:
mobile payments
Habit is the tendency to automatically use a technology as a result of learned behavior (Venkatesh et al., 2012). Habit is found to have a more significant effect on behavioral intention than the other constructs (Venkatesh et al., 2012).
However, the opportunity to form habit can only occur when consumers use a technology. It is impossible for non-users of NFC mobile payments to have formed an use habit, therefore it is impossible to measure habit in the concept of Venkatesh et al. (2012). Nonetheless, as a type of mobile service, NFC mobile payments do use mobile internet, which consumers have already adopted on a much wider scale, therefore habit in the sense of mobile internet use can be examined. Hence, the sixth hypothesis is:
H6: Mobile Internet habit (H) has a positive influence on the intention to use NFC mobile payments
The fun of pleasure derived from using a technology is the definition of hedonic motivation (Venkatesh et al., 2012). Hedonic motivation is found to be the second strongest predictor of behavioral intention in UTAUT2. Despite that hedonic motivation has not been tested in the mobile payment context, the effect of perceived enjoyment on behavioral intention has gained support in the mobile commerce (Zhang et al., 2012). Unlike mobile commerce, where hedonic
motivation may be associated with perceived enjoyment or fun, in the context of NFC mobile payment hedonic motivation may be derived from consumers innovativeness and novelty-seeking (Slade et al., 2014). Therefore, the seventh hypothesis is:
H7: Hedonic motivation (HM) has a positive influence on the intention to use NFC mobile payments
3.1.1.2 Hypotheses -‐ Moderating variables
The constructs of UTAUT2 are moderated by variables such as age, gender and experience.
Age is to moderate the links between performance expectancy (PE), effort expectancy (EE), social influence (SI), facilitating conditions (FC), hedonic motivation (HM), price value (PV), habit (H) and the dependent construct
behavioral intention (Venkatesh et al., 2003). The effect of performance
expectancy on intentions is stronger for younger people, but the effects of effort expectancy and social influence were noticeable for older people (Venkatesh et al., 2003). According to Khechine et al. (2014) younger people are more
confident about their capabilities in mastering technologies. In the case of NFC mobile payments, the positive effect of the construct is stronger for younger people. Taking the above mentioned in consideration the hypothesis related to the moderating effect of age is formulated as follows:
H8: The positive effect of the constructs on the intention to use NFC mobile payments is moderated by age.
According to research (e.g. Venkatesh et al., 2003; Venkatesh et al., 2012) gender has a moderating effect on the relationship between effort expectancy (EE), social influence (SI), facilitating conditions (FC), hedonic motivation (HM), price value (PV), habit (H) and the dependent variable behavioral intention. According to research (e.g. Cheng et al., 2011; Venkatesh et al, 2003; Venkatesh & Morris, 2000), the effect of the construct effort expectancy and social influence were more noticeable for women than men. The hypothesis related to the
moderating effect of age is formulated as follows:
H9: The positive effect of the constructs on the intention to use NFC mobile payments is moderated by gender.
According to Limayem et al. (2007), the connection between experience and habit is formed and strengthened as a result of repeated behavior. Habit can be learned behavior and only after a long period of practice can it be stored in long-term memory and override other behavior patterns (Lustig et al., 2004). Although it is possible for habit to be formed through repetition in a short period of time (Venkatesh et al., 2003). Consumers with more experience of using a technology will develop a cognitive lock-in that creates a barrier to behavioral changes (Murray & Haubl, 2007). The response to possible new trends becomes stronger with increasing experience with a technology. The hypothesis related to the moderating effect of experience is formulated as follows:
mobile payments is moderated by experience.
3.2 Survey Design
Google Form is the tool that has been used for the survey. The survey was online available via Facebook, Twitter and email. This means that respondents were mostly family, friends and colleagues.
The first part of the online survey consists of general questions on mobile
payments. The second part of the survey consists of questions about NFC mobile payments. These questions are based on the UTAUT2 model and are grouped per construct. Also these questions were adapted from Venkatesh et al. (2012) (see appendix B). The final part of the survey consists of questions on the demographic information of the respondent.
The survey makes use of a 7-point Likert-scale, ranging from 1 (strongly agree) to 7 (strongly disagree). This scale was chosen over the 11-point Reysen
Likability Scale (Reysen, 2005), in order to keep the survey simple by not offering too many response alternatives. According to Matell & Jacoby (1971), the
reliability and validity are independent of the number of scale points used for Likert-type items. Therefore, a 7-point scale can illustrate diverse enough results and should be manageable for deciding on answers within a reasonable
timeframe. The Likert-scale ranged from strongly agree to strongly disagree, this is an ordinal scale, but in order to compute an average the variable have to be an interval. But in the research literature it can occur that the scale strongly agree to strongly disagree is between ordinal and interval. This research is conducted with consideration of this issue.
4. Research results
The results of the survey were extracted with the tool Statistical Package for the Social Sciences (SPSS). In this chapter an overview of the statistical information on the respondents will be presented. Furthermore, the factor analysis and the reliability will be described. Finally, the regression analysis will be described for the constructs of the UTAUT2 model.
4.1 Respondents
Table 2 gives an overview of the demographic profile of the respondents. The 166 respondents consisted of 57,8% males and 42,2% females. The majority of the respondents are between the age of 25 and 34 years old and most of them
are single. Most of the respondents have a HBO education and the second most of the respondents have a WO education. Out of the 84,3% of the respondents who have a smartphone, only 55,4% made a payment by using their mobile device. The most used app is the mobile app of the bank, for example Rabo bankieren, ING bankieren etcetera. 44,6% have not used their mobile device to make a payment, because of security reasons or due to the absence of the smartphone. 51,1% of the respondents will use NFC mobile payments in the future.
Table 2 - Demographic profile of respondents
Demographic Frequency % Gender Male Female 96 70 57,8 42,2 Age 12 - 17 18 - 24 25 - 34 35 - 44 45 - 54 55 - 64 3 59 70 18 11 5 1,8 35,5 42,2 10,8 6,6 3,0 Marital status Married/relationship
Single 35 131 21,1 79,9 Education Basic education
VMBO, MAVO HAVO, VWO MBO HBO WO 1 1 7 17 79 61 0,6 0,6 4,2 10,2 47,6 36,7 Use of a smartphone Yes
No 140 26 84,3 15,7 Made a payment by mobile Yes
No
92 74
55,4 44,6 Will use NFC mobile payment in future Yes
No Don’t know 85 44 36 51,5 26,7 21,8
4.2 Factor analysis
According to the literature on mobile payments acceptance and UTAUT2, before a regression analysis is conducted a factor analysis is mandatory. A factor analysis is been conducted, this is to determine whether two or more variables measures the same (Bruin, 2006). To determine whether a factor analysis is appropriate a Kaiser-Meyer-Olkin test is necessary. This test is a way to indicate the suitability of the data for structure detection (Bruin, 2006).
Table 3 shows the Kaiser-Meyer-Olkin (KMO) sampling adequacy and Bartlett’s sphericity test. The value for the KMO test has to be greater than 0.6, which is a commonly used minimum value in the research field (e.g. Hair et al., 2006; Tan et al., 2014). The KMO test in table 3 shows the value 0.825 which is greater than 0.6. A small value of the significance level, in the Bartlett’s Test of Sphericity,
indicates that a factor analysis may be useful with your data. Table 3 shows a significance value of 0.000 which is smaller than 0.05. This suggests a factor analysis should be appropriate.
Table 3 - KMO and Bartlett’s Test
KMO and Bartlett’s Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy
,825 Bartlett’s Test of Sphericity Approx. Chi-Square
df Sig.
3381,089 325 .000
Since the significance value is smaller than 0.05, a factor analysis is conducted. Factor analysis was conducted, with principal component analysis and Varimax method, this is the most common used method in the literature on UTAUT and mobile payments. According to Hair et al (2006) and Nunnally (1978) the factor loading should be greater than 0.5 to confirm the existence of convergent and discriminant validity. In table 3 the components are loaded onto their
corresponding items with factors loading greater than 0.5. Two items have a factor loading which is smaller than 0.5. These are not taken into account with the analysis. These two items are PE3 from performance expectancy and H1 from habit. Table 4 gives an overview of the factor loadings of each item.
Table 4 - Factor loadings - Rotated component matrix
Rotated Component Matrix
1 2 3 4 5 6 7 8 PE1 PE2 PE3 0.928 0.926
EE1 EE2 EE3 EE4
0.852 0.932 0.894 0.916
SI1 SI2 SI3
0.938 0.947 0.927
FC1 FC2 FC3 FC4
0.791 0.839 0.674 0.605
H1 H2 H3
0.921 0.908
PV1 PV2 PV3
0.793 0.874 0.866
HM1 HM2 HM3
0.865 0.871 0.896
BI1 BI2 BI3
0.879 0.948 0.930
PE = performance expectancy, EE = effort expectancy, Si = social influence, FC = facilitating conditions, H = habit, PV = price value, HM = hedonic motivation, BI = behavioral intention.
4.3 Reliability & validity
Following the next step of analysis, according to the literature on mobile payments, is to assess the reliability. The reliability test is conducted with the Cronbach’s Alpha. Cronbach’s Alpha measures the internal consistency, how closely related a set of items are as a group. According to Nunnally (1978), the construct will be satisfactory as the Cronbach’s Alpha value is greater than 0.7. The results of the survey shows that two items must be deleted to get an appropriate Cronbach’s Alpha value. Table 5 shows that the value of
performance expectancy will be satisfactory when PE3 is deleted. This is also shown with table 6, when H1 is deleted from habit the Cronbach’s Alpha value will be satisfactory. Thus the two variables are not taken into account for the analysis.
Table 5 - Cronbach’s Alpha of the Performance expectancy construct Item-‐Total Statistics Scale Mean if Item Deleted Scale Variance if
Item Deleted Corrected Item-Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted PE1 - I would find NFC mobile
payments useful in my daily life 7,85 7,056 ,449 ,598 -,337
PE2 - Using NFC mobile payments would help me
accomplish things more quickly 7,70 7,027 ,434 ,599 -,313
PE3 - Using NFC mobile payments might increase my
productivity 5,37 12,856 -,149 ,022 ,872
Table 6 - Cronbach’s Alpha of the Habit construct
Item-‐Total Statistics
Scale Mean
if Item Deleted
Scale Variance if
Item Deleted Corrected Item-Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted H1 - The use of Internet-based
applications (apps) on a mobile 10,71 13,880 -,109 ,022 ,829
H2 - I am addicted to using Internet-based applications on a
mobile phone 7,63 7,507 ,431 ,510 -,134
H3 - I must use Internet-based
applications on a mobile phone 7,74 6,642 ,496 ,503 -,323
After the removal of the two variables (PE3 & H1) the Cronbach’s Alpha values are increased to satisfactory. Table 7 shows the Cronbach’s Alpha value of each construct of the UTAUT2 model. All constructs have a Cronbach’s Alpha value greater than 0.7, which is satisfactory according to Nunnally (1978).
Table 7 - Reliability of research variable
Construct Cronbach’s Alpha
Performance Expectancy Effort Expectancy Social Influence Facilitating conditions Hedonic Motivation Price Value Habit 0.872 0.918 0.928 0.713 0.850 0.791 0.829
4.4 Regression analysis
Regression analysis is used to test the research hypotheses. Regression analysis checks if there is a connection based on the correlation of the
independent variables and dependent variable and if it can be used to test the hypothesis. In this research there are two types of hypotheses; hypotheses where the independent and dependent variables are directly connected and hypotheses where a moderator is between the independent and dependent variable.
4.4.1 Regression analysis -‐ Direct links
Table 8 shows the regression results of the significant constructs. The three constructs show that they are significant; performance expectancy, price value
and hedonic motivation. The R2 value was 0.605, which means that more than 60% of the variance in the intention to use NFC mobile payments was explained by the three independent constructs.
Table 8 - Regression coefficients and significance without moderating variables
R2 = 60,5%
Construct
B
Corr. Sig.*
VIF
PE PV HM
,266 ,345 ,368
0.484 0.534 0.654
,000 ,000 ,000
2,799 1,159 1,731
*p ≤ 0,05 B = Standardized Coefficients Corr. = Correlations
The construct, performance expectancy (PE), positively affects the intention to use NFC mobile payments. This result supports the first hypothesis and is
consistent with other research results (Pardamean & Susanto, 2012; Slade et al., 2014; Tan, 2013; Venkatesh et al., 2003). As figure 5 presents the findings of the hypotheses, the standardized coefficient of performance expectancy is 0.346, correlation is 0.475 and the Sig. = 0.000 (p ≤ 0.05). Moreover, performance expectancy is not the strongest predictor of behavioral intention, but the weakest of the significant constructs. This is not consistent with the results of other
research (Khechine et al., 2014; Slade et al., 2014; Venkatesh et al., 2003) where performance expectancy construct was the strongest predictor of behavioral intention.
The construct price value (PV) also positively affects the intention to use NFC mobile payments. This result supports the fifth hypothesis and was only
consistent with one research result of Venkatesh et al. (2012). The standardized coefficient is 0.345, correlations is 0.534 and the Sig. = 0.000 (p ≤ 0.05).
However, price value was not the strongest predictor, but the second strongest predictor of behavioral intention. According to Slade et al. (2014), the price value construct was the weakest predictor of behavioral intention.
The construct hedonic motivation (HM) positively affects the intention to use NFC mobile payments. This result supports the seventh hypothesis and is consistent with other research results (Slade et al., 2014; Venkatesh et al., 2012). The standardized coefficient is 0.368, correlation is 0.654 and the Sig. = 0.000 (p ≤ 0.001). Thus, the hedonic motivation construct is the strongest predictor of behavioral intention. This is consistent with other research results (Venkatesh et
al., 2012). Figure 5 gives an overview of the hypotheses and their regression results.
Figure 5 - Research model with findings
4.4.2 Regression analysis -‐ Moderating variables
The results of the moderating effects of age, gender and experience are
presented in figure 5 and table 9. According to research (Khechine et al., 2014; Venkatesh et al. 2012), age did play a moderating role for the construct
performance expectancy, where the effect was stronger for younger people. In this research the moderator age did not have any influence on the constructs. Table 9 shows that the Sig. of the moderator age is greater than 0.001, which implies that there is no significant construct moderated by age. This does not support the eighth hypothesis of this research.
In various research (Al-Gahtani et al., 2007; Lin et al., 2004) gender did not play a moderating role for any of the constructs. However, in the research of the UTAUT2 model the moderator gender did play a moderating role in all of the constructs. Also in this research the moderator gender did not play a moderating role. Table 9 shows the Sig. is greater than 0.05, which implies that there is no
significant construct moderated by gender. This result does not support the ninth hypothesis of this research.
Finally, the moderator experience did not have influence on the constructs.
However, in the research of Venkatesh et al. (2012) experience had a moderating role in most of the constructs. Table 9 shows that the Sig. is greater than 0.05, which implies that there is no significant construct moderated by experience. Also this result does not support the last hypothesis. Table 10 gives an overview of the supported and unsupported hypotheses of this research.
Table 9 - Regression coefficients and significance with moderating variables
R2 = 67,7%
Construct
B
Corr.
Sig.*
VIF
PE x Age EE x Age SI x Age FC x Age H x Age PV x Age HM x Age PE x Gender EE x Gender SI x Gender FC x Gender H x Gender PV x Gender HM x Gender EE x Experience SI x Experience FC x Experience H x Experience HM x Experience
-0.108 0.213 0.126 0.046 -0.008 -0.060 -0.116 0.155 -0.164 0.065 0.111 -0.045 -0.047 -0.074 0.076 0.125 -0.119 -0.038 0.018 0.117 0.190 0.209 0.135 0.084 0.142 0.207 0.010 0.003 0.069 0.033 -0.124 -0.004 0.103 0.003 0.103 0.048 -0.017 0.039 ,236 ,080 ,082 ,619 ,907 ,330 ,248 ,054 ,070 ,377 ,178 ,457 ,412 ,328 ,408 ,077 ,200 ,558 ,817
1,409 1,624 3,510 3,198 6,274 4,460 3,552 2,232 2,277 2,106 3,655 2,913 3,700 1,998 1,579 1,795 4,334 2,458 2,524
*p ≤ 0,05 B = Standardized Coefficients Corr. = Correlations
Table 10 - Summary of findings
Hypotheses Construct Moderator Supported
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 PE EE SI FC PV H HM Age Gender Experience Yes No No No Yes No Yes No No No
4.4.3 Multicollinearity
Multicollinearity is when there is a high correlation identified between two or more variables, suggesting the constructs are not truly independent and thus may be measuring redundant information (Myers, 1990). Variation influence factor (VIF) is used to assess multicollinearity. The maximum recommended VIF value is 10 (Myers, 1990).
Table 8 shows that the VIF values of the constructs are between 1.159 and 3.923, which is lower than the maximum recommended VIF value. The VIF value of the constructs of this research did not suffer from multicollinearity. Moreover, table 9 shows the VIF values of the constructs with their moderators. The VIF values are between 1.409 and 6.274, also these values are lower than the recommended VIF value.
The VIF values are all lower than the recommended VIF value, which is defined by Myers (1990). This means that the constructs and their moderators are independent and did not measure redundant information.
The results show three factors that influence the intention to use NFC mobile payments. These are performance expectancy, price value and hedonic motivation. The strongest factor to influence the intention to use NFC mobile payments is hedonic motivation, this factor showed that the correlation was higher (0.654) than the others (0.345 and 0.266).
5. Discussion
5.1 The respondents
The amount of respondents can be questioned due to the relatively low number of respondents due to the small time frame of conducting the survey. With a larger time frame to conduct the survey, the average amount of respondents in other research (e.g. Slade et al., 2014; Khechine et al., 2014) is between 240 and 380 respondents. It can be argued that the small amount of respondents can be of an influence on the results. For example, the results of the moderators, these were not consistent with other research (Khechine et al., 2014; Venkatesh et al., 2012) where age, gender and experience did moderate the constructs
performance expectancy and facilitating conditions.
Also, the respondents were recruited from social network sites such as
Facebook, Twitter and email. This implies that the respondents consist of family, friends and acquaintances, which can question the representativeness of the sample. These respondents are representing the population of the Netherlands, which are 16.7 million people. Also the data collection was geographically constrained, the respondents were representing the big cities in the province South Holland, Utrecht and North Holland in the Netherlands. The data will be more representable if the collection was cross-national, thus also cities from other provinces.
Furthermore, the respondents are actually mobile phone users, by focusing exclusively on this group may pose potential bias since behavioral differences may be significant between users and non-users (Sarel and Marmorstein, 2003). This suggests that future research should consider non-users as well for
5.2 The UTAUT2 model
The UTAUT2 model of Venkatesh et al. (2012) describes 7 independent constructs, which influence the behavioral intention. The model misses some important constructs such as trust, risks and security. Slade et al. (2014) incorporated trust and risk into their modified model, which were significant constructs for the intention to use NFC mobile payments in the United Kingdom (UK). The construct risk, which was an additional construct, was the second strongest influence on behavioral intention. Some of the respondents responded that they would not use NFC mobile payments due to security reasons, thus suggesting that risk and security should be added to the model for extension for future research.
5.3 The results
The strongest predictor of the intention to use NFC mobile payments is hedonic motivation, this is not consistent with a similar research (Slade et al., 2014) conducted in the UK. Here was performance expectancy the strongest predictor, this was probably due to the focus of the research, which was non-users of NFC mobile payments. The focus of this research was all mobile phone users. The result of this research was consistent with the research of Venkatesh et al. (2012) where hedonic motivation also was the strongest predictor. But the type of
technology being investigated was different, they investigated mobile internet associated with fun applications whereas NFC mobile payments are more utility focused. This implies the motivation to use the technology was different between the two studies.
The construct price value is a predictor of the intention to use NFC mobile payments. This is an unexpected result compared with other similar research (Slade et al., 2014), where price value was not a predictor of the intention to use NFC mobile payments. This is probably because of the example of the price to use the NFC service of Vodafone mentioned in the survey (see appendix B). This example was necessary to mention because NFC mobile payments is a very new service in the Netherlands and for most of the people the price of the use of NFC mobile payment is unknown. With the example the respondents had an idea of the price, which they can refer to.
The constructs of the UTAUT2 model have their moderators; age, gender and experience. The moderators did not play any role in this research, they were
insignificant. However, in other research (Venkatesh et al., 2012) the moderators did play a role. This is probably due to the low amount of respondents in this research. For further research it can be suggested to focus on the moderators that will give a better understanding to the factors, which influence the intention to use NFC mobile payments.
Furthermore, the moderators age, gender and experience are validated by Venkatesh et al. (2012). An additional moderator education could be interesting to see if there is any difference between the levels of education. This would be interesting for marketing purposes in the way marketers can focus on a particular group of consumers.
6. Conclusion
This research presents what factors influence the intention to use NFC mobile payments by consumers in the Netherlands. Several models of technology acceptance were described but one was the best suited to investigate the acceptance of NFC mobile payments.
To answer the research question, what factors influence the intention to use NFC
mobile payments by consumers in the Netherlands, the three sub questions need
to be answered.
- Which models/theories are available for technology acceptance? - Which model/theory is well suited to determine the factors, which
influence the intention to use NFC mobile payments?
- Which factor is the strongest for acceptance of NFC mobile payments?
Paragraph 2.1 described the different theories of technology acceptance that have been used in the literature on mobile payment. These models are DOI, TPB, TAM, UTAUT and UTAUT2.
The UTAUT2 model is well suited to determine the factors, which influence the intention to use NFC mobile payments. The other models focus on the
workplace/employee context whereas the UTAUT2 model focuses on the consumers context. This also implies that the voluntariness of the use of a new technology is very low because the employees are required to use the technology in their work environment. With the use of NFC mobile payments by consumers the voluntariness is high because they choose on their own whether they will use
Furthermore, the UTAUT2 model is used to determine the factors, which
influence the intention to use NFC mobile payments. The results show that there are three factors, which influence the intention to use NFC mobile payments, these are performance expectancy, price value and hedonic motivation. The strongest predictor is hedonic motivation, this is also the strongest predictor according to Slade et al. (2014) and Venkatesh et al. (2013).
The practical application of the results of this research can be used for
practitioners in NFC mobile payments such as mobile communication companies, mobile phone manufacturers, commercial banks and merchants. The
understanding of the factors, which influence the intention to use NFC mobile payments in the Netherlands, can be crucial in the design and development of devices and services.
The factors performance expectancy, price value and hedonic motivation should be promoted. But the focus should be more on price value and hedonic
motivation, because these two were the strongest factors to influence the
intention to use NFC mobile payments. Practitioners should not waste much time on advertising on the ease of use NFC mobile payments. The process of the use of NFC mobile payments is not a problem for the consumers, they care most about the fun and price of NFC mobile payments. The cost of the use of NFC mobile payments should be decreased to enlarge the market share. Information on costs must be presented in details and clearly to lower down the “perceived” costs. The actual cost of NFC mobile payments can be reduced not only from the service provider’s perspective but also from the user’s perspective by multiple categorized rates.
The factor performance expectancy must be focused on the use in daily tasks, for example a payment at the grocery store. NFC payments can be an additional payment method next to cash, pin and chipknip. The chipknip payment method will be removed as a payment method in 2015 in the Netherlands, therefor NFC mobile payments can be a great replacement of the chipknip payment method. Practitioners can focus on this development and introduce NFC mobile payments as a substitute for the chipknip. Furthermore, the fun/pleasure derived from the use of NFC mobile payments should come forward in the use of NFC mobile payments. The results of this research presents that hedonic motivation is the strongest predictor, this means that the consumer find it important whether NFC mobile payments satisfy their pleasurable needs or feelings.
Practitioners can use the factors from this research to design, develop, introduce, improve and promote services of NFC mobile payments.
7. References
• Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179-211.
• Al-Gahtani, S. S., Hubona, G. S., & Wang, J. (2007). Information technology (IT) in
Saudi Arabia: Culture and the acceptance and use of IT. Information and
Management, 44 (8), 681-691. http://dx.doi.org/10.1016/j.im.2007.09.002
• Brown, I., Cajee, Z., Davies, D., & Stroebel, S. (2003). Cell phone banking: predictors
of adoption in South Africa—an exploratory study. International journal of information
management, 23(5), 381-394.
• Brown, S. A., & Venkatesh, V. (2005). Model of adoption of technology in
households: A baseline model test and extension incorporating household life cycle.
MIS quarterly, 399-426.
• Bruin, J. 2006. newtest: command to compute new test. UCLA: Statistical
Consulting Group. http://www.ats.ucla.edu/stat/stata/ado/analysis/. • Chan, K. Y., Gong, M., Xu, Y., & Thong, J. Y. (2008, July). Examining user
acceptance of SMS: An empirical study in China and Hong Kong. In PACIS (p. 294).
• Chandra, S., Srivastava, S., & Theng, Y.-L. (2010). Evaluating the role of trust in
consumer adoption of mobile payment systems: An empirical analysis. Communications of the Association for Information Systems, 27, 561–588.
• Chen, L. D. (2008). A model of consumer acceptance of mobile payment.
International Journal of Mobile Communications, 6(1), 32-52.
• Cheng, Y. S., Yu, T. F., Huang, C. F., Yu, C., & Yu, C. C. (2011). The comparison of
three major occupations for user acceptance of information technology: applying the UTAUT model. iBusiness, 3(02), 147.
• Cheong, J. H., Park, M. C., & Hwang, J. H. (2004, September). Mobile payment
adoption in Korea: Switching from credit card. In ITS 15th Biennial Conference, Berlin, Germany, September (pp. 4-7).
• Compeau, D. R., & Higgins, C. A. (1995). Application of social cognitive theory to
training for computer skills. Information systems research, 6(2), 118-143.
• Coulter, K. S., & Coulter, R. A. (2007). Distortion of price discount perceptions: The
right digit effect. Journal of Consumer Research, 34(2), 162-173.
• Dahlberg, T., & Oorni, A. (2007, January). Understanding changes in consumer
payment habits-do mobile payments and electronic invoices attract consumers?. In
System Sciences, 2007. HICSS 2007. 40th Annual Hawaii International Conference on (pp. 50-50). IEEE.
• Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user
acceptance of information technology. MIS quarterly, 319-340.
• Davis Jr, F. D. (1986). A technology acceptance model for empirically testing new
end-user information systems: Theory and results (Doctoral dissertation, Massachusetts Institute of Technology).
• Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1992). Extrinsic and intrinsic
motivation to use computers in the workplace. Journal of applied social psychology,
22(14), 1111-1132.
• Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An
introduction to theory and research. Addison-Wesley Publishing Company. Reading,
Massachusetts.
• Gartner. (2013). Gartner says worldwide mobile payment transaction value to
surpass $235 billion in 2013 in Gartner Newsroom. Retrieved from http://www.gartner.com/newsroom/id/2504915
• Gefen, D., & Straub, D. W. (1997). Gender Differences in the Perception and Use of
E-Mail: An Extension to the Technology Acceptance Model. MIS quarterly, 21(4). • Hair, J. F., Tatham, R. L., Anderson, R. E., & Black, W. (2006). Multivariate data
• Hall, D. T., & Mansfield, R. (1975). Relationships of age and seniority with career
variables of engineers and scientists. Journal of Applied Psychology,60(2), 201.
• Holbrook, M. B., & Hirschman, E. C. (1982). The experiential aspects of consumption:
consumer fantasies, feelings, and fun. Journal of consumer research, 132-140.
• Hongxia, P., Xianhao, X., & Weidan, L. (2011, May). Drivers and barriers in the
acceptance of mobile payment in China. In E-Business and E-Government (ICEE),
2011 International Conference on (pp. 1-4). IEEE.
• Khechine, H., Pascot, D., & Bytha, A. (2014). UTAUT Model for Blended Learning :
The Role of Gender and Age in the Intention to Use Webinars. Interdisciplinary
Journal of E-Learning and Learning Objects, 10, 33–52.
• Kim, S. S., & Malhotra, N. K. (2005). A longitudinal model of continued IS use: An
integrative view of four mechanisms underlying post adoption phenomena.Management science, 51(5), 741-755.
• Lee, C. P., Warkentin, M., & Choi, H. (2004). The role of technological and social
factors on the adoption of mobile payment technologies. Proceedings of the 10th
Americas Conference on Information Systems (AMCIS), New York, USA, August 6–8
• Limayem, M., Hirt, S. G., & Cheung, C. M. (2007). How habit limits the predictive
power of intention: the case of information systems continuance. Mis Quarterly, 705-737.
• Lin, J., Chan, H. C., & Jin, Y. (2004). Instant messaging acceptance and use among
college students. Proceedings of the Eighth Pacific Asia Conference on Information
Systems, Shanghai, China.
• Lustig, C., Konkel, A., & Jacoby, L. L. (2004). Which route to recovery? Controlled
retrieval and accessibility bias in retroactive interference.Psychological Science,
15(11), 729-735.
• Mathieson, K. (1991). Predicting user intentions: comparing the technology
acceptance model with the theory of planned behavior. Information systems research, 2(3), 173-191.
• Matell, M. S., & Jacoby, J. (1971). Is there an optimal number of alternatives for Likert
scale items? I. Reliability and validity. Educational and psychological measurement.
• Minton, H. L., & Schneider, F. W. (1980). Differential psychology. Brooks/Cole
Publishing Company.
• Murray, K. B., & Häubl, G. (2007). Explaining Cognitive Lock‐In: The Role of Skill‐
Based Habits of Use in Consumer Choice. Journal of Consumer Research, 34(1), 77-88.
• Myers, R. (1990). Classical and modern regression with applications. Boston:
PWS-KENT.
• Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.
• Nysveen, H., Pedersen, P. E., & Thorbjørnsen, H. (2005). Intentions to use mobile
services: antecedents and cross-service comparisons. Journal of the academy of
marketing science, 33(3), 330-346.
• Pardamean, B., & Susanto, M. (2012). User acceptance toward blog technology
using the UTAUT model. International Journal of Mathematics and Computers in
Simulation, 1 (6), 203-212.
• Porter, L. W. (1963). Job attitudes in management: II. Perceived importance of needs
as a function of job level. Journal of Applied Psychology, 47(2), 141.
• Püschel, J., Mazzon, J. A., & Hernandez, J. M. C. (2010). Mobile banking: proposition
of an integrated adoption intention framework. International Journal of Bank
Marketing, 28(5), 389-409.
• Reysen, S. (2005). Construction of a new scale: The Reysen likability scale.Social Behavior and Personality: an international journal, 33(2), 201-208.
• Rhodes, S. R. (1983). Age-related differences in work attitudes and behavior: A
review and conceptual analysis. Psychological bulletin, 93(2), 328.
• Rogers Everett, M. (1962). Diffusion of innovations. Free Press. New York.
• Rogers Everett, M. (1995). Diffusion of innovations, fourth edition. Free Press. New York.
• Schepers, J., & Wetzels, M. (2007). A meta-analysis of the technology acceptance
model: Investigating subjective norm and moderation effects. Information &