The influence of individual trait and
service quality on the actual usage of
mobile payment requests
Name Fleur de Waal
Student number 11421460
Supervisor dr. A. Alexiou
Date of submission 25-1-2018
Master Thesis Business Administration
Specialization Strategy
2
Statement of originality
This document is written by Fleur de Waal who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
Purpose: The relatively new terminology Fintech asks for more research into financial services combined with technology. The exponential increase of users interacting with social media raised attention towards the expansion of mobile payment services. The time of only paying attention to measurements of financial performance is over, as it is the experience of customers that will contribute to successful Fintech companies. Therefore, in order to be successful, it is important to understand consumers’ behaviour.
Design/methodology/approach: This thesis reports the key predictors for actual mobile payment request usage from the individual perspective. The 32- item instrument measures four dimensions of actual usage: accessibility, personal innovativeness in technology, security and perceived usefulness. Survey data were collected from 282 participants in the Netherlands using Facebook and WhatsApp. Results were analysed with SPSS.
Findings: The findings suggest that actual usage of mobile payment requests in the Netherlands are determined by personal innovativeness in technology and perceived usefulness. Therefore, this study can conclude that individual traits influence the actual usage of mobile payment users. Moreover, the direct relation of perceived usefulness and usage corresponds with prior literature.
Research limitations and implications: The theoretical and managerial implications are included in this thesis. They may provide further insights and suggestions for mobile banking companies.
Originality/value: The originality of this thesis lies within its investigation of a technological mobile banking application which has affect behind traditional boundaries of banking and extends the traditional core model upon usage behaviour.
Keywords: Accessibility, Actual usage, Fintech, Mobile banking, Mobile payment requests, Personal innovativeness in technology, Perceived Usefulness, Security, User experience.
4 “There is only one boss, the customer. And he can fire everybody in the company from the chairman on down, simply by spending his money somewhere else” – Tomas Edison.
5
Table of content
1. Introduction ...6
1.1 The upswing of mobile banking ...6
1.2 Two examples of successful disruption in mobile banking ...9
1.3 Purpose and structure of this thesis ... 10
2. Literature review ... 12
2.1 Mobile banking definition ... 12
2.2 Customer behaviour ... 13 2.3 Traditional theories ... 15 2.4 Individual traits ... 17 2.5 Relevance ... 19 3. Theoretical framework ... 23 3.1 Actual usage... 23 3.2 Accessibility ... 24 3.3 Personal innovativeness ... 26 3.4 Moderators ... 28 3.5 Security ... 28 3.6 Perceived usefulness ... 30 4. Methodology... 33
4.1 Study context and sample ... 33
4.2 Survey instrument ... 34 4.3 Operationalization of variables ... 36 5. Results ... 38 5.1 Correlation ... 38 5.2 Pre-test ... 39 5.3 Regression analysis ... 40 6. Discussion ... 45
7. Theoretical and managerial implications... 50
7.1 Theoretical implications ... 50
7.2 Managerial implications ... 51
8. Limitation and future research ... 53
9. Conclusion ... 55
Appendix ... 56
6
1. Introduction
1.1 The upswing of mobile banking
The world is rapidly changing due to constant technological advances and digitalization, resulting in new applications and transformations in various key business operations. The increased interest in digitalization affects multiple aspects of the process, product and organization.(Matt, Hess, & Benlian, 2015) This also affects the strategy of organizations. Traditional companies have implemented different digital technologies over the years and have transferred into digital organizations. As a result of the continuing
technological developments and digitalization the strategy of organizations needs to be regularly evaluated so that necessary lessons need be learned.
The globalization of products and ideas combined with new technologies and
digitalization creates opportunities for start-ups and non-sector related technical providers to enter the market. (Dapp, Slomka, AG, & Hoffmann, 2014). The consequence of this
development is affecting the financial sector and requires necessary transformations in the strategy of financial organizations and especially banks. These challenging transformations include different aspects such as payment solutions, online and mobile banking, allocation of venture capitals to start/ups and customer credits (Dapp et al., 2014). This thesis will focus on the transformation of mobile banking as a relevant example of strategic technological
innovation by banking organizations.
The history of mobile banking started with online banking in 1995. Online banking appeared with the upcoming technological innovations which made it possible to use internet for banking (Hernan E. Riquelme & Rosa E. Rios, 2010). Since then, new technological tools were created to increase the efficiency of online banking and usefulness for both the banks
7 and their customers (Hernan E. Riquelme & Rosa E. Rios, 2010). Online banking is defined as “an Internet portal, through which customers can use different kinds of banking services ranging from bill payment to making investments” (Pikkarainen, Pikkarainen, Karjaluoto, & Pahnila, 2004, p. 224). However, internet banking started to be outdated with the introduction of mobile banking. Since the early 2000s, mobile banking services became popular and were introduced all over the world (Dahlberg, Mallat, Ondrus, & Zmijewska, 2008). Currently, the use of mobile banking is more common than the use of internet banking (Changchit, Lonkani, & Sampet, 2017).
(Non) banking companies started delivering alternative mobile services to customers in the financial market. They used new technologies, such as block chain or wireless
payments by card and by phone, to combine and communicate with their customers
(Betaalvereniging Nederland, 2017). This new trend also gained attention in the Netherlands. Statistics show that people in the Netherlands used their bank card in 2016, which was 10.6% more than in 2015 (Betaalvereniging Nederland, 2017, p. 23). Moreover, wireless card payments were multiplied by factor five in 2016 compared to 2015. Even mobile payments were extended from 36.000 users in the beginning of 2016 to 150.000 users at the end of 2016. The annual report of Betaalvereniging Nederland (2017) shows the relevance of mobile banking in the Netherlands as 99,7% of the public has access to mobile banking during the whole year. Combined with the result that the use of mobile phones increased from 78,8 % in 2015 to 89% in 2017 in the Netherlands (CBS, 2017), the Dutch use of mobile banking is of importance.
Due to the increased importance of mobile banking, companies shifted their focus to more innovative ideas to attract the mobile banking customers, causing various economic benefits associated with mobile banking (Deventer, Klerk, & Bevan-Dye, 2017). Two main economic advantages of mobile and online banking are first the reduction of personnel costs
8 and second, location costs (Pikkarainen et al., 2004), because less people will visit frequent in-person service location (Althobaiti & Mayhew, 2014). Therefore, less physical locations are needed and additionally, less employees are hired. Moreover, the self-service channel of online and mobile banking reduces time and efforts of consumers (Pikkarainen et al., 2004). Both customers and corporations are favoured with the reduction of costs, time and efforts because of mobile banking innovations (Gupta, Yun, Xu, & Kim, 2017). Consequently, customers appreciate these innovations and banks are challenged to invest in new mobile banking innovations in order to align with and retain their clients.
In order to satisfy customers, it is necessary to understand two phenomena of mobile banking and the changed financial world. The first phenomenon is that companies can better understand customer behaviour and meet the customer’s needs. Customer behaviour is
relevant, because the technology used, makes it easier to screen data and influence this stream of information (van Rijn, 2017). Different studies have explored various factors related to customer behaviour such as intention and use of technological systems (Davis, 1989; Delone & McLean, 2003; Rogers, 1995). These researches contribute to the understanding of
customer behaviour in information systems. For example, the behaviour and perceptions of customers differ between internet and mobile banking. Efficiency, convenience and safety determine the differences in customer value perception of internet and mobile banking (Laukkanen, 2007). The second phenomenon is that customers of mobile banking need to experience privacy, safety and transparency of mobile payment applications (van Rijn, 2017). This requires interaction with banking firms that allows customers access at any time and all places.
Customer behaviour, accessibility, transparency, speed and time appear to be big influences on the transformation strategy of banking companies. Factors such as power and equipment are necessary in these strategies and strengthen the position of innovative mobile
9 banking technologies (Fitzgerald, Kruschwitz, Bonnet, & Welch, 2014). Consequently, the combination of the technology strategy and business strategy is named digital transformation strategy (Bharadwaj, El Sawy, Pavlou, & Venkatraman, 2013). The speed of change in digital transformations requires a stream of constant research that investigates new banking
technologies and customer behaviour. A clear understanding of the processes, accessibility, transparency and speed of the digital transformation contributes to an effective digital strategy. Moreover, digital transformation strategy affects corporations beyond their
traditional boundaries. The exploration of disruptors in the banking world enables academics to get a better picture of the scope, scale, sources and speed across these traditional
boundaries (Bharadwaj et al., 2013).
1.2 Two examples of successful disruption in mobile banking
BUNQ is one of the disruptors who entered the banking world as a non-banking company (van Rijn, 2017). This bank makes it easier to open a bank account and transfer money from the mobile phone. Additionally, BUNQ innovated by introducing the possibility to send pictures with the transaction and was the first to implement the inclusion of pictures to contribute to the banking experience of customers. Due to the possibility to interact through multiple communication sources and share experiences with others, customer experiences are relevant (Lemon & Verhoef, 2016). Another interesting example of a disruptive initiative is Tikkie, which is a service for mobile requests for payment provided by a start-up of ABN AMRO. Tikkie is the first Dutch service to make mobile payment requests easier. This early adopter gives people the opportunity to use an application for reminding others to pay a certain amount by using WhatsApp (Tikkie, 2018). The advantage of this technology is the ease of use in sending your bank account number and the receiver just pays with one click.
10 Before the development of this feature, people had to send their information to others before remit the amount of money. The Tikkie application showed an increase of 500.000 users in one year (de Swart, 2017). As a result, Tikkie became one of the most popular application tools in current mobile payment services. Similar examples that followed the success of Tikkie, include ‘betaalverzoeken’ of ING, ‘slice’ from BUNQ, GRPPY from RABOBANK, which appeared to favour the customer conveniently. The increased attention for mobile banking applications shows the interest of this study. Understanding the behaviour of users and picturing their customer journey contributes to the future and innovative technologies (Lemon & Verhoef, 2016).
1.3 Purpose and structure of this thesis
The purpose of this study is to understand the attractiveness of mobile payment request using a quantitative data analysis. Combined with technological innovations affecting traditional boundaries of banking, the academic world can get a clearer picture of the process, accessibility and speed involved in the digital transformation. Moreover, this insight is
scientifically interesting because the results extend the traditional core models used for intention and actual usage behaviour.
First, this thesis provides an explanation of the relevance of mobile banking services. Subsequently the literature review continues with the determination of mobile payment services success measurements and its focus on customers. It elaborates on the already existing terminology about intentions to use specific innovations and the acceptance of them. Thereafter the problem is discussed and the reader is introduced to the research questions. The theoretical framework presents the variables and explains the stated relationships made by this research. In the methods the study sample is enhanced followed by the demographic
11 information and survey instruments. Additionally, the reliability of all variables are tested to check validity and consistency. The next paragraph, results, will present the findings from the data. As a result, the discussion elaborates on the outcomes. The thesis ends with a conclusion in which the theoretical and managerial implications are clarified. Limitations and future research should give the reader an idea of suggestions for follow up studies.
12
2. Literature review
This part of the study explores existing literature upon mobile banking. It overviews the most important theories mentioned in the literature and introduces the reader into the research topic. Moreover, the literature review introduces the reader to the research questions of this study.
2.1 Mobile banking definition
A service delivery evolution has occurred in the financial world as innovations emerged so quickly (Wessels & Drennan, 2010). The newest technologies challenge firms to adapt quickly to changing circumstances (Schuchmann & Seufert, 2015). The urgency for traditional banks to develop new innovations appeared because they did not fulfil the requirements of today’s fast changing society. The understanding of successes and failures with mobile banking usage gained attention due to the increased competition. If traditional banks will not adapt to innovations, non-banking firms, such as BUNQ, conquer and may overtake the mobile payment market. Mobile banking is widely accepted because mobile phones are integrated into people’s lives (Laukkanen, 2007). Almost 100% of the Dutch population has access to mobile phones and 89% Dutch customers use mobile phones in 2017 (CBS, 2017). Also, financial services are more and more delivered by electronic channels such as mobile phones (Loonam & O’Loughlin, 2008). For example, the appearance of applications and tools for mobile banking led to the disappearance of psychical banks. Consequently, the use of mobile banking applications enables consumers to request their account balance easier. Therefore, the use of mobile banking applications contributes to an increased value for customers (Laukkanen, 2007). The services delivered by mobile banks can
13 be defined as “different electronic handling of payments” (Schierz, Schilke, & Wirtz, 2010). In particular, mobile banking refers to “the interaction in which a customer is connected to a bank via a mobile device such as cell phone, smartphone or person digital assistant (PDA)” (Laukkanen & Kiviniemi, 2010, p. 373). Another aspect of mobile banking includes the advantage of wireless interaction and communication technologies (Dahlberg et al., 2008). All definitions argue the mobile phone as a main characteristic of communication in mobile payment use (Laukkanen, 2007). The main purpose of mobile payment services is the transfer of monetary value using mobile phones and communication technologies. The side effect of mobile payment services is that the applications generate lots of data about the use of mobile banking by customers (van Rijn, 2017). Due to this data, banking corporations may better understand and respond to the attitudes and behaviour of their customers.
A new era with new terminology has started within mobile payment services. Fintech is the name for combining technology with financial services (Kim, Park, & Choi, 2016). It refers to “the industrial changes forged from the convergence of financial services and IT” (Kim et al., 2016, p. 1058). Different tools such as banking companies, social media, mobile phones and the internet of things are used and amalgamated in Fintech. The combination of all these aspects lead to mobile banking services participating in Fintech.
2.2 Customer behaviour
Since there are so many forms of mobile banking companies, they all include different aspects of strategy such as digital information systems, marketing activities and e-commerce (Yoon, 2010). It is difficult to decide what the main focus of banking companies should be because of the multiple strategy aspects. Moreover, the developments in mobile banking have changed relationships of banking companies with their extensive range of diverse
14 stakeholders which makes it more difficult to satisfy every stakeholder (Atkinson,
Waterhouse, & Wells, 1997). The most well-known stakeholders of a company are customers, suppliers, employees, owners and the community presented in figure 1 (Atkinson et al., 1997). Firms may focus on all stakeholders, but providing attention to one specific group helps determine the success of banking companies (Pikkarainen et al., 2004). Different
measurements of success could be used to understand the different perspectives and values of stakeholders. For example, traditional financial measurements are a return on investment (Petter, DeLone, & McLean, 2008). However, only paying attention to the financial
performance measurements is not sufficient in today’s society. Focusing on financial factors such as effectiveness and cost reduction may lead to ignorance of customers’ perception, values and satisfaction (Laukkanen, 2007). Due to this potential lack of ignorance customers’ perceptions, companies need to reinforce their relationships with customers. The
strengthening of the relation with customers reflects and contributes to the achievement of the primary objectives of the firm; financial performance (Atkinson et al., 1997). The first part of the strategical process is of importance and starts with the identification of needs,
expectations and requirements of customers. Understanding the user behaviour of banking customers contribute to the development of banking services (Laukkanen, 2007). For
instance, if companies understand how users act and behave, they can adapt to their needs and wishes to achieve the company’s targets. Therefore, the customer experience plays a big role in the successful performance of organizations.
15
Figure 1
2.3 Traditional theories
Previous research has focussed on the intention of mobile use, online payments and the acceptance of these technologies from the perspective of customers (Laukkanen, 2016; Lee, Park, Chung, & Blakeney, 2012; Sheng, Wang, & Yu, 2011). Acceptance is the required precondition to success (Petter et al., 2008). Acceptance is important for the success and experience of mobile banking services, but the dynamic environment may also significantly influence the experience of customers (Lemon & Verhoef, 2016). In order to increase the performance successes of organizations, various theories and models are designed to interpret user behaviour. The success of information systems is explained by the model of Delone & McLean (Delone & McLean, 2003). The technology acceptance model (TAM) explains the individual acceptance behaviour of technological innovations (Venkatesh & Davis, 2000). The theory of reasoned action (TRA) (Fishbein & Ajzen, 1977) and the theory of planned behaviour (TPB) (Ajzen, 1991) form the basis for the TAM-theory.
The framework of Delone & McLean is one of the models which measures the success of information systems by distinguishing system quality, information quality, service quality,
16 user satisfaction and system use (Delone & McLean, 2003). One of the most important aspect in the success model of Delone and McLean (2003) is system use. This is defined as “the degree and manner in which customers use the capabilities of an information system” (Petter, DeLone, & McLean, 2013, p. 11). All components together measure the technical, semantic and efficiency success of information systems. The main contribution of this model is the demonstration of the importance of context for the success of companies (Delone & McLean, 2003). However, the first version of the success model of Delone & McLean suffered a lot of criticism. One of the objections to the first version of the model was concerned with the absence of service quality as an important implication of customer satisfaction. Service quality exists of what an organizations offers and provides combined with the experience and feelings of customers about how they present these offers (Parasuraman, Zeithaml, &
Malhotra, 2005). Moreover, service quality contributes to the context of experiences and measures aspects of the customer’s journey (Lemon & Verhoef, 2016). Therefore, Delone & McLean added service quality as a construct to their revised success model (Pitt, Watson, & Kavan, 1995). The importance of the success model is that Delone & McLean recognize the importance of service qualities in user behaviour. Similar to e-service quality, mobile banking services need to offer efficient and effective services (Arcand, PromTep, Brun, &
Rajaobelina, 2017). Critical for an efficient and effective service is accessibility (Jun & Cai, 2001). Unlimited access to mobile payment services are gained due to technological
innovations. The customers’ decision is motivated by the accessibility to use mobile banking applications frequently (Ramseook-Munhurrun & Naidoo, 2011) . Therefore, access to mobile banking application is important in the experience of customers.
Another study concerning information systems is the technology acceptance model (TAM) (Venkatesh & Davis, 2000). Central to this theory are the reasons of consumers’ acceptance of information technology (Sheng et al., 2011). TAM explains why people reject
17 or accept a technology (Venkatesh & Davis, 2000) using two core variables to explain;
perceived usefulness and perceived ease of use (Davis, 1989). The reasoning of TAM is based on the TRA (Fishbein & Ajzen, 1977) and TPB (Ajzen, 1991). Both theories include attitudes, behaviour and subjective norms of customers in order to better understand the behaviour of customers (Deventer et al., 2017; Luarn & Lin, 2005). The TAM shows a positive relationship between the intention to use and attitude towards using the technology (Schierz et al., 2010). Various research used TAM as the basic framework for research in mobile banking (Lu, Liu, Yu, & Wang, 2008; Mohammadi, 2015; Wessels & Drennan, 2010).
Figure 2
2.4 Individual traits
The ignorance of individual characteristics in these theories is perceived as problematic (S. Yang, Lu, Gupta, Cao, & Zhang, 2012). For example, the TAM model implies that individuals do not experience any barriers for using information systems (Luarn & Lin, 2005). However, persons differ in their preferences and how they experience
information systems. Therefore, different researchers include numerous components from outside the core assumption of the traditional core theories (Laukkanen, 2016; Lee et al., 2012; Montazemi & Qahri-Saremi, 2015; Pikkarainen et al., 2004; Sheng et al., 2011).
Theories
Technological acceptance model Consumer acceptance of information technology
Measurements: intention to use, perceived usefulness, ease of use and usage behaviour Venkatesh & Davis, 2000
Delone & McLean model Success of information systems
Measurements: system quality, information quality, service quality, user satisfaction and system use
18 Different individual characteristic variables may influence how people behave (Agarwal & Prasad, 1998). As there are so many individual behaviour characteristics, the five factor model (FFM) theory (openness, conscientiousness, neuroticism, extraversion and
agreeableness) is created to have an integrative personality classification (Barnett, W Pearson, Pearson, & Kellermanns, 2015). This group of characteristics are used in order to understand individual behaviour. Results show that only conscientiousness and neuroticism were found significantly affect the actual usage of IT systems (Barnett et al., 2015). However, there are more ideas about categorizing individual characteristics which are difficult to measure because they might be latent and invisible.
Cognitive absorption is another theory whereby personal traits are considered to perform as important intrinsic antecedents to the individual’s beliefs about technology. Cognitive absorption affects the use behaviour (Agarwal & Karahanna, 2000). This holistic approach can be seen as the engagement of individuals in technology environment. Three basic thoughts reason the stream of cognitive absorption (Saadé & Bahli, 2005). First is the trait of absorption which explains the state of attention to the experienced event. Some people may experience attention or the state of absorption of subjects more than others. Second inter-related item is concerning the degree to which a person is involved in the activity (Jun & Cai, 2001). This element is named the theory of flow and measures the total experience, control of loss, loss in time and the intension of concentration. Thirdly, the concept of engagement contains different dimensions of intrinsic interest. For instance, personal innovativeness is mentioned as essential in the explanation of user behaviour in mobile payments (S. Yang et al., 2012). This individual characteristic in technological systems is interpreted as an intrinsic antecedent of cognitive absorption (Agarwal & Karahanna, 2000). The experience of
19 who are involved in technological innovations and who are intrinsically motivated by their inner-drive.
Personal innovativeness in technology is also explained as part of the diffusion of innovation theory (IDT) (Agarwal & Prasad, 1998). This theory implies different adopter categories which divides the customers’ level of adoption. IDT consists of different categories: innovators, early adopters, early majority, late majority and laggards (Rogers, 1995). All categories differ in personality characteristics. It implies that innovators and early adopters behave differently than late adopters. Innovators and early adopters have more certain behaviour such as mass media exposure, subjective evaluation of others in their social environment is less important and they are informative seekers for new innovations (Rogers, 1995). In addition, specific for the early adopter groups are their ability to abstract things, to have a greater empathy and intelligence and they show favourable behaviour towards change (Hoffmann, Probst, & Christinck, 2007). Whereas late adopters behave as expectants. Thus, the different innovation adoption categories show that they might affect the digital
transformation strategy. In particular, early adopters and innovators will enhance technological innovations.
2.5 Relevance
Given the fact that mobile banking is such a broad and well accepted terminology (Yoon, 2010), the study focuses on the mobile requests for payments to narrow down the field of study. The relevance of examples such as Tikkie are popular and interesting to investigate, because of the growth of users in one-year time. According to the IDT the most important elements are innovation, communication channels, time and social influence (Malaquias & Hwang, 2016). Mobile payment request services include all aspects of this notion since
20 innovation is the ability of people to send requests by mobile phone to others, using social media in the fast-changing technological environment. Therefore, innovation and customer behaviour aspects are interesting to research.
Existing research shows a connection between personality traits and actual usage in IT systems (Barnett et al., 2015). However, the link between individual characteristics and actual usage can be further extended in the mobile banking technology literature as new applications appear more rapidly as well as their importance for diffusion innovation literature
(Montazemi & Qahri-Saremi, 2015). Additionally, few research report aspects of the service quality at the individual level (Jun & Cai, 2001; Petter et al., 2008). As accessibility is part of the service quality (Jun & Cai, 2001), it facilitates easy access to applications for mobile requests for payments at the individual level. Therefore, this thesis investigates service quality measurements at the individual level of experience. The study is limited to mobile payment requests, otherwise no statement about individuals’ experience can be done. Moreover, this thesis chooses for using the concept ‘system use’ from Delone and McLean (2003). ‘System use’ can be approached in different ways (Petter et al., 2013). One of the approaches is actual usage. Use is relatively important because Petter et al. (2013) show that 80% of the value and success of an information system is realized because of its system’s use. This proves the importance of the measuring actual use.
Consequently, these theoretical insights and the facts of exponential growth of Dutch users lead to a combination of the following aspects: service quality, individual traits and actual application use in mobile payment requests. These factors combined direct to the following research question:
Q1: To what extend are service quality and personal traits predictors for a successful actual use of mobile payment request service?
21 Security threats affect mobile banking because of its vulnerability (Gupta et al., 2017). Hackers around the world try to break in to mobile banking systems to obtain sensitive
information and in worst case create financial losses (Lin, Wang, Wang, & Lu, 2014). Different access control systems such as password systems and authentication, are
implemented to protect from mobile banking frauds (Gupta et al., 2017). These threats imply that technological control, such as accessibility relates to security. Moreover, security also relates to the perceived risk of which people are willing to take. People with a high personal innovativeness are characterized as open minded and risk-taking users (Zhou, 2012). Theory about security states that risk and security conflict with each other (Lim, 2003). It is
reasonable that security will affect the relationship of personal innovativeness in technology to the actual usage of mobile payment requests.
As perceived usefulness is part of the TAM, Davis (1989) argues that perceived usefulness influence the intention to use, and indirectly the usage behaviour of people. IDT named perceived usefulness as relative advantage which is part of the five fundamental characteristics: relative advantage, complexity, compatibility, trialability and observability (Rogers, 1995, 2002). Perceived usefulness and relative advantage are both about the gains and acceptance of technological innovation (Shih & Huang, 2009). Since personal
innovativeness is an individual characteristic and perceived usefulness is the experience of an individual, it seems logic that they both will affect each other. Besides, service quality is also related to the experience in utility of the service. Technological innovations will not be successful if users do not see the utility of the application (Pikkarainen et al., 2004). This implies that perceived utility will contribute to a better service quality experience. People will experience any effort from the organization based on the perception of usefulness. Therefore, this study argues the following question:
22 Q2: Do security and perceived usefulness moderate the relationships between service quality, personal traits and the actual usage of mobile requests for payment?
23
3. Theoretical framework
The questions addressed in the literature review lead to the following model; presented in figure 3.
Figure 3
The following section frames the literature into theoretical statements. Each variable is introduced by existing literature and gives a definition of their meaning in the context of mobile payment requests. Figure 4 shows an overview of the stated hypotheses in the theoretical framework.
3.1 Actual usage
Different authors studied use in mobile and online payment services (Deventer et al., 2017; Montazemi & Qahri-Saremi, 2015; Zandhessami & Geranmayeh, 2014; Zhou, 2012). For example, Deventer et al. (2017) showed that perceived integrity has a significant positive relationship to mobile banking use. Furthermore, trust is significantly related to the
24 dependent variable actual usage of mobile payment requests, the services can be seen as a necessary predictor of the success of payment systems (Petter et al., 2013; Pikkarainen et al., 2004). However, some authors disagree with usage as a predictor for success (Delone & McLean, 2003). Seddon (1997) claims that usage is behaviour and cannot be included in the model of Delone & McLean and therefore is less relevant. Delone & McLean (2003) disagree with this statement and according to them, usage is a good measurement of success. They argue use as relevant and essential because variation exists in the use of mandatory success systems. Additionally, Devaraj & Kohli (2003) argue that users and technology need to have a strong relationship to understand usage behaviour. Without customers using the technology, companies experience losses in success. Therefore, actual use of technological innovations is critical for the financial stability of companies. Use is explained as “the degree and manner in which customers utilize the capabilities of an information system” (Petter et al., 2008, p. 239). Differences in use can be detected; system use differs from user’s satisfaction because
satisfaction measures the contentment of customers with the technological system (Petter et al., 2008). System use measures aspects such as frequency of use, the amount of use and appropriateness of actual use. For this research, the idea of actual usage as a measurement of success is utilised. Moreover, actual usage in mobile requests for payments is possible because customers use mobile banking request applications voluntary and are not forced (Urbach & Müller, 2012). Thus, the actual usage of mobile payment request is a subject of this thesis.
3.2 Accessibility
Accessibility is defined as the construct with physical access to technology and the
ability to use an information system successfully (Culnan, 1985; Karahanna & Straub, 1999). Also mentioned as ‘the approachability and ease of contact of services’ (Jun & Cai, 2001).
25 Davis(1989) shows that perceived ease of use is also part of service quality, just like
accessibility. Service quality is the difference between the performance of firms and the customer’s expectations and their feedback on service (Parasuraman, Zeithaml, & Berry, 1985). Both concepts contain aspects of differences between how the firm performs and what the expectations of customers are. According to Jun & Cai (2001) accessibility also highlights aspects of easiness. Perceived ease of use is the perception of individuals that the technology is free of any effort (Davis, 1989). Especially, both ideas are related to the effort of using technological innovation systems (Venkatesh, 2000). This implies that people are more likely to use new banking innovations if the application is free of effort and customers benefit from user-friendly image (Veríssimo, 2016). Jun and Cai (2001) explain accessibility as part of the ease of use in the online quality system . The similarity between both variables contains aspects of the perception in effortless using a particular system (Veríssimo, 2016). Hence, perceived ease of use has interrelated ties and similar effects as accessibility used in this study because accessibility also refers to the delivering of effort in the company’s performance (Jun & Cai, 2001).Consequently, accessibility is an important factor for the success of
corporations.
Furthermore, individuals also use the perception of access in their decision to use a specific service (Culnan, 1985). According to Jun and Cai (2001), accessibility is essential for both traditional and technological banking. Moreover, it is a critical condition to use
technological information systems (Karahanna & Straub, 1999) because people expect technology to be available and to be responsive (Z. Yang, Cai, Zhou, & Zhou, 2005). This is in line with the purpose of mobile banking, which is to offer easier, more quickly, more efficiently and more convenient access to financial services for customers (Deventer et al., 2017). Prior research demonstrates the strong relationship of perceived accessibility with specific technology information sources over other sources (Culnan, 1985). It means that
26 individuals’ choice for one technology system depends on the entry of this technology.
Companies should keep their technology services open and easy to access. Therefore, one of the key focus of mobile banking is accessibility, because it influences the satisfaction or dissatisfaction of users (Mols, 2000). Culnan (1985) proves that perceived accessibility is a predictor for satisfaction which indirectly implies the success of a system. For instance, if the service of a banking firm has a high accessibility, it is more likely that potential users will adopt this service more than alternative options. Moreover, when people already use the banking service, they will stay because of the easy access to the application. Deventer et al. (2017) show that perceived ease of use is significantly related to mobile banking use behaviour. Thus, there may be assumed that a higher perceived accessibility will positively influence the actual usage of a mobile request for payment. The hypotheses are presented in figure 4.
Hypothesis 1: Perceived accessibility has a positive relationship with the actual usage of mobile request for payments.
3.3 Personal innovativeness
Personal innovativeness in technology (PIIT) is important for the understanding of
individual traits in technological adoption and usage behaviour (Agarwal & Prasad, 1998). This construct helps to understand the perceptions and roles customers play in the usage of services. Personal innovativeness is defined as “the willingness of an individual to try out any new information technology” (Agarwal & Prasad, 1998, p. 206). This implies that users with a high personal innovativeness are more willing to try new technology and take more risks in using innovations (Lu et al., 2008; Zhou, 2012). The risk-taking effect of high personal innovativeness is categorized as the “innovators” and “early majority”. These labels come from the adopter types in the IDT (Malaquias & Hwang, 2016; Rogers, 1995, 2002). The
27 innovation diffusion theory also adopted other types such as “the late majority” and
“laggards”. Innovators and early adopters are also defined as trendsetters in other theories (Chitungo & Munongo, 2013). The innovator or trendsetter is the user with high personal innovativeness who will influence others in their decision process to adopt technological innovation. This theory is in line with the social influence effect of personal innovativeness (Malaquias & Hwang, 2016; Rogers, 2002). Customers with high personal innovativeness are more likely to have a social impact onto others. Therefore, early adopters will impact the actual use of mobile payment requests. This personal characteristic proves the relevance of understanding the customer’s behaviour.
Different research demonstrates the positive effect of PIIT on user behaviour. For example, PIIT has a positive relation with intention to adopt wireless services in China (Lu et al., 2008). Another research shows that people with a higher personal innovativeness develop more positive perceptions about the innovation (Agarwal & Prasad, 1998). Personal
innovativeness also affects the attitude of customers towards the use of mobile banking (Mohammadi, 2015). Moreover, PIIT has a direct influence on the intentional behaviour of people using mobile services (Chitungo & Munongo, 2013; S. Yang et al., 2012). Thus, people with high personal innovativeness develop more positive perception about mobile requests for payment that will positively influence the actual usage. Positive perceptions help users to stay satisfied about the service delivered and continue the application use.
Consequently, higher personal innovativeness is strong related to the higher actual usage of mobile requests for payment and not only related to the intention to use.
Hypothesis 2: Personal innovativeness has a positive relationship with actual usage of mobile requests for payment.
28
3.4 Moderators
In the second part of the model two variables are implemented as moderators (figure 3). The moderator is used to show variables that enhance or diminish the effect of the independent variable to the dependent variable. Various research has reported results regarding security and perceived usefulness as fundamental constructs in user behaviour (Agarwal & Karahanna, 2000; Linck, Pousttchi, & Wiedemann, 2006; Luarn & Lin, 2005; Pikkarainen et al., 2004; Schierz et al., 2010; Venkatesh & Davis, 2000; S. Yang et al., 2012; Zhou, 2012).
3.5 Security
Security issues are experienced as the greatest concerns of mobile banking services
(Gupta et al., 2017; Luarn & Lin, 2005). Mobile transactions are exposed to great uncertainty and risks, because of virtuosity, temporality and anonymity (Schierz et al., 2010; Zhou, 2012). Mobile devices are vulnerable to risks such as hack attacks and lacks of control (Zhou, 2012). Therefore, security is related to risk-taking. Another author defines perceived credibility as no threats from security and privacy (Luarn & Lin, 2005). Moreover, security risks concern the willingness of sharing privacy information of consumers (Lwin, Williams, & Wirtz, 2007). Linck et al. (2006) connect security with the objective and subjective security. Objective security is the technical characteristic in different solutions responding towards different security objectives such as risk-taking activities (confidentially, authentication, integrity, authorization, non-repudiation). Subjective security is more about the feeling of users about the safety of mobile payments. It can be defined as ‘the degree of the perceived sensation of procedures” (Linck et al., 2006, p. 6). From the perspective of consumers, security is
29 (Gupta et al., 2017; Linck et al., 2006; Schierz et al., 2010). Bank accounts are made up of multiple sensitive information of customers, which means that they need to be processed securely. If companies fulfil the requirements for safety and security, people will accept and use mobile payment services more. Linck et al. (2006) show the importance of security as essential for the usage of mobile payments.
Security can be seen as the opposite of perceived risks which is seen as the expectation of some kind of loss (Lim, 2003). In general, more risk will lead to less use of services.
Therefore, companies enhance strategies to give safe and secure mobile services.
Nevertheless, people with a high personal innovativeness have a higher tolerance of risk-taking (S. Yang et al., 2012; Zhou, 2012). The downside of people with a high risk-risk-taking characteristic is that they pay less attention to security issues. High security could conflict the choice of people with personal innovativeness to use mobile banking because of the low taking use involved with high security. According to Yang et al. (2012), the search for risk-taking adventures lead to interest in adopting newer innovations and distract from the actual usage of present innovative mobile banking technologies. Security is interpreted by
individual’s innovativeness as an obstacle. Therefore, security will moderate the relation between personal innovativeness and actual usage.
Hypothesis 3: The relationship between personal innovativeness and actual usage of mobile requests for payment will be moderated by security.
Perceived accessibility influences and is influenced by different contextual factors (Culnan, 1985). Security is contributed to the part of only system qualities (Jun & Cai, 2001), People with a high perceived security will experience a higher perceived accessibility,
because the accessibility of technological systems depends on the security of these IT systems (Gupta et al., 2017). Therefore, security will moderate the positive relationship between accessibility and actual usage of mobile request payments.
30 Hypothesis 4: The relationship between accessibility and actual usage of mobile requests for payment will be moderated by security.
3.6 Perceived usefulness
Perceived usefulness find its roots in the TAM model (Venkatesh & Davis, 2000). It is
part of the core idea and is defined as ‘a person believes that using the system will enhance his or her job performance’ (Venkatesh & Davis, 2000, p. 187) which is important for the use of innovative technology. Moreover, research (Deventer et al., 2017; Tornatzky & Klein, 1982; S. Yang et al., 2012) shows that perceived usefulness is similar to the concept relative advantage. For example, online banking tools which are not perceived as useful do weaken the position and success of the bank instead of reinforce them (Pikkarainen et al., 2004). One of the findings shows that perceived usefulness has a significant effect to the system
utilization (Ha & Stoel, 2009). This effect is explained as the existence of user’s belief in an use and performance relationship (Davis, 1989). The statement suggests that perceived usefulness and use are strongly related. However, Shih & Huang (2009) demonstrate no significant relationship between perceived usefulness and actual usage. Literature shows both the importance as insignificant results of perceived usefulness. Therefore, this study
investigates the moderating effect of perceived usefulness on actual usage.
Literature suggests that personal innovativeness is related to perceived usefulness (Agarwal & Prasad, 1998). Yang et al. (2012) show the positive significant relationship of personal innovativeness and relative advantage. Since perceived usefulness is about the individual’s assessment of the utility of technologies (Lu et al., 2008), the relationship with personal innovativeness seems logic. Because both variables contain aspects of individual behaviour, the effect of perceived usefulness will enhance the relationship between personal
31 innovativeness and actual usage. Therefore, the presence of perceived usefulness will
moderate the relation of individual innovativeness to mobile banking usage positively.
Hypothesis 5: The relationship between personal innovativeness and actual usage of mobile requests for payment will be moderated by perceived usefulness.
Perceived usefulness has been understand as the function of the information quality system of Delone & McLean (Chuan-Chuan Lin & Lu, 2000). Given this fact, there is a relationship between accessibility and perceived usefulness. Accessibility includes aspects of perceived ease of use and both variables belong to online system quality (Jun & Cai, 2001). It seems that accessibility may be influenced by perceived usefulness. Different research show the direct and indirect relationship of ease of use to intentional behaviour affecting perceived usefulness (Agarwal & Prasad, 1998; Davis, 1989; Venkatesh & Davis, 2000). Agarwal & Karahanna (2000) argue that perceived ease of use has an indirect effect through perceived usefulness. For example, someone can have a great access to mobile payment requests but if they do not see the purpose of use in this service, perceived usefulness may downsize the effect of accessibility to actual usage. Therefore, this thesis states that perceived usefulness moderates the relationship between accessibility and actual mobile payment request usage. The moderating effect of accessibility and perceived usefulness will affect actual usage of mobile payment requests.
Hypothesis 6: The relationship between accessibility and actual usage of mobile requests for payment will be moderated by perceived usefulness moderates.
32
Figure 4
Hypotheses
H1: Accessibility has a positive direct effect on the actual usage of mobile request for payments.
H2: Personal innovativeness in technology has a positive direct effect on the usage of mobile requests for payments.
H3: The effect of personal innovativeness on the actual usage of mobile requests for payments will be moderated by security.
H4: The effect of accessibility on the actual usage of mobile requests for payments will be moderated by security.
H5: The effect of personal innovativeness on the actual usage of mobile requests for payments will be moderated by perceived usefulness.
H6: The effect of accessibility on the actual usage of mobile requests for payments will be moderated by perceived usefulness.
33
4. Methodology
This section of the research gives an overview of the used methods. The data is conducted by collecting surveys and elaborates on the main instruments used for the survey responses. Furthermore, this chapter goes further into the operationalization and validation of the variables.
4.1 Study context and sample
This thesis uses an online questionnaire surveys in a quantitative approach to investigate the main two questions. The sample was drawn from the population of the Netherlands. The subject are users participating in Facebook groups and using WhatsApp. This method chosen for a population with affinity to the use of mobile tools and applications. The distribution of mobile payment requests is extended by using the social media channel WhatsApp. Only respondents who owned a mobile device were relevant for this research. The choice for conducting data from Facebook groups is because the reach of WhatsApp is
limited. Another explanation for the choice of Facebook and WhatsApp is that this research particularly focuses on the overall impression of mobile requests for payment. Consequently, the choice to use Facebook as the distributor gives opportunities for mixed answers about different companies offering mobile payment request services. Nevertheless, careless
responding can disturb the research and affect the data (Meade & Craig, 2012). Therefore, the Facebook groups were picked with attention to the usage of payment requests. Five Facebook groups with high potential use of mobile payment requests were targeted. Each group existed of +/- 200 persons and were selected for their differing age and profession. For example, one group existed of employees of KLM and another group contained students. The response rate was 13% of the total population. In addition, various WhatsApp groups were added to the
34 study for more responses to the survey. Together, these groups were responsible for 155 responses.
The survey includes 32 items and a total response of N=285. Only 253 responses are valid and useful for the research mobile payment requests. 32 responses are not included in the study, because they do not use mobile payment request at all or they skipped some questions. In that case the incomplete surveys were removed from the data.
The demographic information about the sample exists of 164 women and 89 men in which 2/3rd was female. 15,8% is below 21 years, the category of 22-29 years exists of 41,9%, 30-39 years is 9,1%, in the category 40-49 years 17,4% participate, 14,2% filled in for the group of 50-59 years and above 60 years is 1,6% of the total response. The age groups show that most respondents are between 22-29 years with 41.9%. The data also points out that most people are student or working which makes the distribution of profession almost 50/50 except from some retirements and unemployment’s. Education proved that most people finished a WO education. The demographic information is listed in table 1.
4.2 Survey instrument
The items of the survey measure the actual usage of mobile payment applications for request in order to understand the indirect predictors of success and failure in the applications. Survey items from prior research are included for measuring the variables of this study.
Actual usage is the dependent variable of this thesis and measures the frequencies of use,
amount of use and the appropriateness of the use payment request applications (Delone & McLean, 2003; Petter et al., 2008). The items of actual usage were adapted, combined and validated from van der Heijden (2003), Moon & Kim (2001) and Zhou (2012). One dummy variable was added to distinct non-users from users. The question aims to get a better
35 use mobile payment requests? “. This study only used the responses from people that said “yes” to the use of mobile payment requests. The variable actual usage was measured with a different seven-type Likert system to interpret the frequency of usage (1 = several times each day, 7 = not at all). Other items were also measured by a seven-type Likert scale, but the answer categories differed from 1 = “extremely frequently” to 7 = “not extremely frequently” and 1= “strongly agree” to 7 = “strongly disagree”.
The independent variables of this research are accessibility and personal
innovativeness in technology. The construct of accessibility was produced by Aladwani &
Palvia (2002) and was measured by a seven-item questions. It enhanced both the physical access as the ability to use an information system successfully (Culnan, 1985). This thesis used accessibility as the approachability of the application perceived by the user, but is produced by its owner. Furthermore, personal innovativeness is influenced by the
characteristics of the user. Personal innovativeness in technology has its origin in the theory of Agarwal & Prasad (1998). They produced a four-item scale which is widely accepted and used by other researchers (Lu, Yao, & Yu, 2005; Mohammadi, 2015; S. Yang et al., 2012; Zhou, 2012).
The two moderators, security and perceived usefulness, were used to investigate their effect on the relationship between accessibility, personal innovativeness and actual usage. Security contains the idea of users that they feel save with their transactions using mobile payment requests (Linck et al., 2006). The items of security were adapted from Schierz et al. (2010) who combined different items from different authors (Luarn & Lin, 2005;
Parasuraman et al., 2005). Taylor & Todd (1995) developed and validated two of the six-item scale of perceived usefulness. The other four items were adapted from Schierz et al. (2010) who composed them from previous research. This construct contributes to the idea that the specific application is advantageous for users (Venkatesh & Davis, 2000). All independent
36 variables and moderators were also measured by the seven-type Likert system similar to the dependent variable. Responses were scaled from 1 for “strongly agree” to 7 for “strongly disagree”. The items used in this study are listed in Appendix.
The two control variables used in this research are age and gender. According to the theory, gender can be a predictor for the adoption of mobile banking, because men are more likely to adopt (Laukkanen, 2007). Females are more reserved and have less positive attitudes towards internet activities (Malaquias & Hwang, 2016). Due to the different thinking and reasoning of men and women, it may affect how they experience the online environments (Rodgers & Ann Harris, 2003). The other control variable is age, because research proves that age matters to mobile innovations (Luo, Li, Zhang, & Shim, 2010). Therefore, these two control variables were used in order to test the hypotheses of this study.
4.3 Operationalization of variables
This thesis used statistical procedure SPSS from the company IBM to guarantee the validity of statistical tests of the hypothesis in different ways. Both, the reliability check and factor analysis were done to assess the reliability and validity of the scales. The reliability measures the consistency of items with the particular variable. The Cronbach Alpha needs to be > 0.7 (Nunnally & Bernstein, 1978) and the corrected item-total correlation above 0.3. All variables were tested in reliability to understand the cohesion in questions. The construct reliability of USE (α = 0.76) was already sufficient, but > 0.8 is better, thus use_4 was removed to obtain a higher Cronbach Alpha of 0.837. ACC had a reliability of α = 0.772 and all the corrected item-total correlation were > 0.3. Deleting one of the questions would not improve the Cronbach Alpha. In addition, the reliability of variable PIIT showed a Cronbach Alpha of 0.741 > 0.7. The reliability of SEC showed a cohesion of α = 0.701 and is used as construct participating in this study. Without deleting anything, PU has a high Cronbach Alpha of 0.83.
37 The factor analysis is conducted to perform the Kaiser – Meyer – Olkin (KMO) measurement. There is a KMO sampling adequacy of 0.768 which is > 0.6. This means that the sampling adequacy is sufficient to continue. The Bartlett’s Test of Sphericity showed an x² (406) of 2800.836 and p = 0.000. P < 0.05 which means that there can be conclude that the correlation between the items is significant.
Table 1
Frequencies demographic information
Frequency Percent Valid Percent
Cumulative Percent
Gender Valid Female 164 64,8 64,8 64,8
Male 89 35,2 35,2 100,0 Total 253 100,0 100,0 Age Valid <21 40 15,8 15,8 15,8 22-29 106 41,9 41,9 57,7 30-39 23 9,1 9,1 66,8 40-49 44 17,4 17,4 84,2 50-59 36 14,2 14,2 98,4 60> 4 1,6 1,6 100,0 Total 253 100,0 100,0
Profession Valid Student 107 42,3 42,3 42,3
Working 137 54,2 54,2 96,4 Retired 5 2,0 2,0 98,4 Unemployed 3 1,2 1,2 99,6 Other 1 ,4 ,4 100,0 Total 253 100,0 100,0
Education Valid High school 26 10,3 10,3 10,3
MBO 25 9,9 9,9 20,2
HBO 82 32,4 32,4 52,6
WO 120 47,4 47,4 100,0
38
5. Results
The intent of this thesis was to see if there exists an effect of different dependent variables on the actual usage behaviour of mobile payment requests. The study’s framework tried to explain the actual usage behaviour and the influence of accessibility and personal innovativeness in technology in mobile requests for payment settings. The correlation matrix is conducted to understand the relations between two variables. Table 2 presents the results of the correlation between variables. To keep consistent with previous research, the study uses a regression analysis. The multiple regression analysis technique is used to test the hypotheses and are produced in table 3. Figure 5 gives an overview of the relationships between all independent, moderating and dependent variables.
5.1 Correlation
The correlation analysis was conducted to show the relations between variables. Significances were explained at the level of p < 0.05 and p < 0.01. Table 2 shows a
correlation of 4.3% variance explained between personal innovativeness and actual usage of mobile payment requests (r = 0.207, p = 0.001). Both variables are significant correlated p < 0.01. The relation between perceived usefulness and personal innovativeness was also
considered as significantly correlated; p < 0.01 (r = 0.223, p = 0.000). Personal innovativeness and accessibility (r = 0.195, p = 0.002) and personal innovativeness and security (r = 0.165, p = 0.009) showed a significant effect p < 0.01. The variance of correlations in determination was explained for 3.8% by personal innovativeness, and accessibility explained 49.4% of the variation (r = 0.195). The relationship between personal innovativeness and security was explained for 2.7% (r = 0.165). Accessibility and actual use showed a low negative correlation (r= 0.018, p = 0.776) and a coefficient of determination of 0.2%. There is no
39 significant relation between accessibility and use. However, accessibility has a significant relationship with perceived usefulness < p = 0.01 (r = 0.322, p = 0.000). Accessibility also correlated significantly with security (r=0.180, p=0.004). This table shows the presence of different noteworthy correlations between the different variables. Additionally, it shows the significant correlation between use and security and use and personal innovativeness. The range of USE is from r = .018 till r = 0.207 and shows the likelihood that different factors influence the actual usage of mobile requests for payment.
Table 2
5.2 Pre-test
Certain assumptions need to succeed before executing the regression analysis, otherwise these assumptions may violate the data interpretation. First, the most important assumption is that the sample size need to be large enough to consider a multiple regression analysis. The formula is N > 50 + 8m (m is the number of independent variables) invented by Tabachnick & Fidell (2013). Therefore, this study satisfies the sample size assumption. Second
assumption is that multicollinearity, which refers to the relationship among different independent variables, may not exist. Multicollinearity is examined by controlling the variation inflation factor (VIF) in the linear regression. VIF < 5, thus there may be assumed
Correlations
ACC USE SEC PU PIIT
ACC 1 ,018 ,180** ,322** ,195**
USE ,018 1 ,125* -,121 ,207**
SEC ,180** ,125* 1 ,101 ,165**
PU ,322** -,121 ,101 1 ,223**
PIIT ,195** ,207** ,165** ,223** 1
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
40 that no multicollinearity exists. Thirdly, the outliers are checked and removed from the data. Another important assumption is that the data should have linearity; relationship with the dependent variable (Tabachnick & Fidell, 2013). Homoscedasticity is also required to execute the regression analysis; it enhances the prediction of actual usage scores which need to be similar for all other predicted scores. Also, the homoscedasticity meets the assumptions for multiple regression analysis. Additionally, the normality and linearity are compared to detect any asymmetry and understand the distribution. The independent, dependent and moderating variables are measured for Skewness and Kurtosis. Generally, values between -2 and 2 are accepted as reference for normality (Groeneveld & Meeden, 1984; Tabachnick & Fidell, 2013). Only USE is negative higher than 1 in the Skewness table and PU is above 1 for Kurtosis. However, they are still acceptable in the range of -2 till 2. All results show that they are within the boundaries of normal distributed data. Thus we may assume that the data normally is distributed. Therefore, the study may continue with the regression analysis.
5.3 Regression analysis
A hierarchical regression analysis was conducted to state the above hypotheses (figure 4) with USE as dependent variable and ACC and PIIT as independent variables. Moderating variables consisted of SEC and PU. Control variables gender and age were used in this thesis. Several models were tested to see the effect of these different variables on the actual usage of mobile payment requests. The first model only includes the control variables gender and age. This model showed an increased R² of 0.034 which states that gender and age explain 2.7% of the variance is related to the actual usage of mobile payment requests. F (2, 250) = 4.382; p = 0.013; the p < 0.05, which suggests the significant effect of gender and age on actual usage (USE) in the first model. The beta is 0.108 for age (p > 0.05) and -.147 for gender (p < 0.05).
41 The second model added personal innovativeness in technology (PIIT) and
accessibility (ACC) to the actual usage of mobile payment requests. The addition of PIIT and ACC to the prediction of USE led to statistically significance increase in R² of 0.044, F(2, 248) = 5.961 and p = 0.003 After controlling for gender and age, these two independent variables explained 4.4% of the total variance of actual usage. Accessibility was found to have a positive direct effect on usage (β = 0.01, p = 0.992). Personal innovativeness was found to have a positive direct effect of β = 0.190 (p = 0.001). If personal innovativeness increases with one, actual usage will also increase with a standardized deviation of 0.215. PIIT has a statistical significant effect on the actual usage of mobile payment request, whether accessibility does not affect actual usage in model 2.
The two standardized Z-score moderating variables entered the regression analysis in model three. Model three shows results that the R² led to an increase of 0.032, F (2, 246) = 4.471; p = 0.012. Thus, 3.2% of the total variance in actual usage is explained by model three compared to model two. Also the addition of security (SEC) and perceived usefulness (PU) in model three showed a significant effect in the model summary. However, the coefficient table shows a difference between perceived usefulness and security. PU had a statistical significant effect on actual usage (β = -.167, p = 0.012), but security did not show a statistical significant effect (β = 0.103, p = 0.097). In conclusion, this result implies that there is no linear
relationship between security and actual usage. There can be concluded that only perceived usefulness is directly significantly related to actual usage.
Model four includes two interaction terms consisting of ACC * PU and ACC * SEC. The overall effect of the R² is increased with 0.01; F (2, 244) = 1.416; p = 0.245, which explains 1% of the total variance in actual usage compared to model three and did not have an overall significant relation; p > 0.05. The coefficient table shows a β for the interaction term is -.05 (β = -.05, p = 0.460) which can be explained as perceived usefulness combined with
42 accessibility had a negative effect on actual usage. However, the effect of the other interaction term security and accessibility increased actual usage with β = 0.103 (p = 0.101). Thus, security * accessibility was found to be a positive effect on actual usage. Both interaction terms revealed statistical significant effect as they are both p > 0.05. Consequently, security did not have any moderating effect to the actual usage of mobile payment requests.
To test the effect of the two moderators SEC and PU on PIIT, a new hierarchical regression was run by SPSS. After model three, this step changed the entry of other standardized z-scores for the interaction terms PIIT * PU and PIIT * SEC. In table 3, the fourth model of the second regression analysis is named model five for convenience of the reader. Since model one till three have the same outcomes, model five explains 9% of the total actual usage variance (F (2, 244) = 0.975; p = 0.379) compared to model three. The effect of the standardized coefficient PIIT * PU showed β = 0.086 and p = 0.170. The interaction term for personal innovativeness * perceived usefulness was found to have a positive effect on the actual usage. The interaction term of PIIT * SEC was found to have a negative effect of β = -0.018 and the p value = 0.761 which is > 0.05. In conclusion, both interaction terms and moderators do not significantly influence the relation PIIT and USE.
Hence, the results present a significance of model one, two and three. It means that age and gender significantly explain a part of the total variance of actual use. After the addition of accessibility and personal innovativeness, the model still represents significance. The addition of security and perceived usefulness shows a significance of the total variance in use. A closer look at the coefficient table shows the direct significance of personal innovativeness and perceived usefulness to the actual usage of persons (figure 6). Therefore, hypothesis 2 is supported by the data presented in this thesis.
43
44
Table 3
Hierarchical Regression Model of Actual Usage
Model t Sig. R R² R² Change B Std. Error Beta 1 ,184 ,034 ,026 Age ,080 ,046 ,108 1,737 ,084 Gender -,314 ,133 -,147 -2,370 ,019 2 ,280 ,078 ,063 Age ,105 ,046 ,142 2,289 ,023 Gender -,260 ,132 -,122 -1,964 ,051 ACC ,001 ,095 ,001 ,010 ,992 PIIT ,190 ,056 ,215 3,381 ,001 3 ,332 ,111 ,089 Age ,106 ,045 ,143 2,350 ,020 Gender -,178 ,133 -,084 -1,339 ,182 ACC ,043 ,098 ,028 ,434 ,665 PIIT ,207 ,057 ,234 3,650 ,000 PU -,228 ,090 -,167 -2,544 ,012 SEC ,115 ,069 ,103 1,668 ,097 4 ,347 ,121 ,092 Age ,106 ,046 ,143 2,329 ,021 Gender -,163 ,133 -,076 -1,221 ,223 ACC ,067 ,101 ,045 ,662 ,508 PIIT ,200 ,057 ,226 3,522 ,001 PU -,225 ,091 -,165 -2,465 ,014 SEC ,127 ,069 ,114 1,833 ,068 int1 -,042 ,057 -,050 -,741 ,460 int2 ,103 ,063 ,103 1,644 ,101 5 ,343 ,118 ,089 Age ,102 ,045 ,137 2,241 ,026 Gender -,182 ,133 -,086 -1,369 ,172 ACC ,030 ,099 ,020 ,308 ,759 PIIT ,214 ,057 ,242 3,756 ,000 PU -,199 ,092 -,146 -2,169 ,031 SEC ,111 ,069 ,100 1,603 ,110 Int3 ,091 ,066 ,086 1,378 ,170 Int4 -,019 ,064 -,018 -,304 ,761
a. Dependent Variable: use
1. Predictors: (constant), Age, Gender.
2. Predictors: (constant), Age, Gender, ACC, PIIT. 3. Predictors: (constant), Age, Gender, ACC, PIIT, SEC, PU
4. Predictors: (constant), Age, Gender, ACCC, PIIT, SEC, PU, int1: PU*ACC, int2: SEC*ACC. 5. Predictors: (constant), Age, Gender, ACC, PIIT, SEC, PU, int3: PU*PIIT, int4: PU*PIIT...
