• No results found

The effect of trust and personal innovativeness on the intention to use new radical technology

N/A
N/A
Protected

Academic year: 2021

Share "The effect of trust and personal innovativeness on the intention to use new radical technology"

Copied!
47
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Master Thesis

The effect of trust and personal

innovativeness on the intention to use new

radical technology

©Shutterstock Inc.

Author:

Max Valenkamp

Student number: 10684646

Email:

max.valenkamp@student.uva.nl

Supervisor:

Frank Slisser

Faculty:

Faculty of Economics and Business

Study:

MSc Management Studies

Specialization: Marketing Strategy

Date:

January 31st, 2018

(2)

2

Statement of confidentiality

This document is written by student Max Valenkamp 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)

3

Abstract

This study examines the direct and indirect effects of trust and personal innovativeness on the intention to use new radical technology. Due to the rapidly growing popularity of new technologies this field of research is particularly interesting for both practitioners and behavioural scientists. In order to study these relationships a theoretical model was built out of several previous models by combining different constructs in a unique way. To test this model a survey was conducted among 380 respondents. It was then tested to what degree trust and personal innovativeness influenced the intention to use an autonomous car. Also mediated by perceived risk, perceived usefulness and perceived ease of use. The results confirmed all but one of the tested hypotheses. Only the relationship between personal innovativeness and behavioural intention was not found to be mediated by perceived usefulness. All other hypotheses tested in the model were supported by the analyses performed in this study. This leads to the insight that there is sufficient empirical evidence to conclude that trust and personal innovativeness influence the intention to make use of radical new technology, such as autonomous cars. In other words, the results of this study provide evidence on the effect of trust and personal innovativeness on the intention to use radical technology.

Keywords: TAM, radical technology, innovation, autonomous, trust, personal innovativeness, perceived risk, perceived usefulness, perceived ease of use, behavioural intention, self-driving.

(4)

4

Table of contents

Abstract ... 3

1 Introduction ... 5

1.1 New technology and usage barriers ... 5

1.2 Focus of study ... 5 1.3 Research gap ... 6 2 Literature review ... 8 2.1 Introduction ... 8 2.2 Types of technologies ... 8 2.3 Radical technology ... 9

2.4 Innovation Diffusion Theory ... 10

2.5 Technology Acceptance Model ... 11

2.6 Choi & Ji model ... 12

2.7 Hypotheses ... 14 2.8 Personal innovativeness... 15 2.9 Autonomous cars ... 18 3 Methodology... 21 3.1 Research design ... 21 3.2 Data collection ... 21 3.3 Data analysis ... 22 4 Results ... 24 4.1 Descriptive Statistics ... 24 4.2 Correlations ... 24

4.3 Principal Component Analysis ... 25

4.4 Testing the hypotheses with the Process Regression analysis ... 26

5 Discussion... 29

6 Conclusion ... 31

7 Limitations & Future research ... 31

Bibliography ... 32

Appendix A: Research model Choi & Ji ... 39

Appendix B: Reliability Analysis ... 40

(5)

5

The effect of trust and personal innovativeness on the intention to

use new radical technology

1 Introduction

Technology has taken over our lives and we can no longer live without, according to Škrebljin et al., (2013). In all aspects of life technology has intruded rapidly; from healthcare, to trade, agriculture, transport and communication. Think of online banking, mobile payments, mobile health applications, online courses, video conferencing and autonomous vehicles. The list of technological applications seems endless and all segments of society and business appear to be involved (Mckelvey, 2014). Practically almost all industries have an interest in technology and therefore want to explore new technological possibilities further. For example, this is reflected in the amount of money that is invested worldwide in Information Technology (IT), which is more than 2 trillion dollars a year (Carr, 2003).

1.1 New technology and usage barriers

It is clear that technology plays an important role in our society and business. Consumers are frequently confronted with new technologies and innovations with which they are required to perform new behaviours (Roumani et al., 2015). But uncertainties around the performance and usage create barriers for using these technologies (Hoeffler, 2003). For example, with mobile banking there is not only the question how to execute a financial transfer but also what the privacy risk is as a consumer when dealing with this sort of technology (Yu, 2012). These feelings about the user friendliness and perceived risk of course have a great impact on the consumers’ behavioural intention to use new technology (Martins et al., 2014).

For marketers, it is important to have an understanding of such predictors that can influence the usage of new technology products or services. With the right marketing approaches and communication tools companies can make sure that certain products are trusted, so doubts or perceived risks are put aside sooner (Hoeffler, 2003). As with most new technologies, the level of trust that consumers have in technological services and products is understood to be one of the key areas of concern by marketers. This knowledge allows firms to improve their market strategies and financial performances (Choi et al., 2007).

The impression is that uncertainties and barriers related to the usage of technology are more present with certain specific types of technologies, especially new ones. The requirements to perform certain behaviour also seems to differ strongly per type of new technology.

1.2 Focus of study

This study is primarily built on the expanded Technology Acceptance Model (TAM) theory by Choi and Ji (2015). TAM is based on a theory of Davis (1989) that models how users come to accept and use technology. The fundamental knowledge of the theory claims that consumers who perceive using a new technology as useful (perceived usefulness) and easy to use (perceived ease of use) are more likely to have a positive attitude, and end up using a particular technology (Svendsen et al., 2013). Choi & Ji expanded the TAM theory by including other factors to their model.

(6)

6 Apart from the TAM factors, additional key elements which have been integrated in their model are trust, perceived risk and two personality characteristics. The objective of their study is to examine the impact of all these factors influencing the intention to use an autonomous vehicle.

1.3 Research gap

Besides building on the TAM and Choi and Ji’s model, scholars like Oostrom et al. (2013) point in the direction of including personality characteristics, to improve explaining the usage of technology better. Slade et al. (2015) and Devaraj et al. (2008) conclude that the inclusion of personal characteristics in technology acceptance models has largely been ignored. Recent study results indicate positive relationships between personality characteristics and the behavioural intention to make use of new technology such as the autonomous Google car (Nordhoff, 2014). Future research could re-evaluate this phenomenon by first of all selecting specific personality characteristics and by making use of a more neutral survey than the one from Google. Thereby clearing the biased reputation of Google as brand. Choi and Ji (2015) recommend future research to investigate the effect of personality characteristics on the behavioural intention to use autonomous vehicles. This is suggested by specifically including other external variables such as personality characteristics: “Future researches should test the possible inclusion of other external variables (e.g., personality characteristics). Therefore, more research is needed to validate, expand, and generalize these results” (Choi and Ji, 2015, p.700). The earlier mentioned gap of including personality characteristics will be examined in the current study by fitting in personal innovativeness to the research model to investigate the intention to use new technology. According to Prahalad and Rmaswamy (2004) personal innovativeness is a personality characteristic that measures the individual’s ability to adapt and use a new technology. This characteristic is related to gaining insights of individuals’ attitude towards the usage of

innovative technology (Agarwal & Prasad, 1998).

The inclusion of personality characteristics in explaining acceptance of technology seems like a logical additional step. Most models are developed and applied in an organizational research context, whereby company expectancies can largely influence users. Consumers however experience a higher degree of freedom when it comes to technology usage. In other words, consumers tend to have more choices when it comes to deciding to use new technologies. Therefore, focusing on consumer characteristics can create a better understanding of possible patterns in explaining new technology usage.

Besides including personality characteristics to enrichen the knowledge around technology acceptance and usage, it is likely that the type of technology also plays an important role. For instance, it is likely that a new technology will generate more uncertainties and perceived risk because of the unfamiliarity, compared to more familiar older technologies. In line with this assumption the newest type of technology is selected to investigate factors influencing the usage intention, namely radical technology. The motive for selecting this type of innovation is because of its newness, large unfamiliarity and lack of experience by consumers with this technology. The last two motives mentioned (unfamiliarity and lack of experience) can cause higher feelings of uncertainty and risk towards the usage of the technology (Pavlou, 2003).

(7)

7 Objective study

In short, this study is built on the study of Choi and Ji (2015) whom expanded the TAM. The objective of this study is to replicate the model of Choi and Ji and in addition to examine whether personal innovativeness affects the behavioural intention to use new radical technology.

This study will be investigated with the following research question:

To what degree do Trust and Personal Innovativeness affect the behavioural intention to use new radical technology?

This question will be answered during the following chapters:

Chapter 2 Theoretical framework – Research on the factors influencing acceptance and usage

of technology (formulated hypothesis), followed by the definition of the conceptual model to visualize the relationships between the variables. Finally, the literature review of the diffusion theory, radical technology and autonomous cars will be elaborated.

Chapter 3 Methodology – Operationalization of research, how are these factors defined and

how are the variables measured followed by a research and survey design.

Chapter 4 Data and results – Results and analysis of the data followed by testing the

hypotheses.

Chapter 5 Discussion and conclusions – Answers to the research questions and discussion,

(8)

8

2 Literature review

2.1 Introduction

In chapter one, the possible impact of new technology on our society, businesses and consumers has been introduced to get a better understanding of its relevance. The focus for this study is on the theoretical models and aspects that predict the acceptance and usage of new technology. More specifically the focus of interest is to examine the effect of trust and personal innovativeness on the intention to use radical new technology.

In this chapter, the theoretical constructs and relationships of the variables related to technology usage are elaborated more extensively. This is followed with several hypotheses clarifying the relationships which shall be examined. First of all, the TAM and Choi and Ji’s model shall be described and also the application towards this study. This brief illustration shall explain the variables used in the research models and the concluding results. The current paper is derived for a great part from the research of Choi and Ji (2015). Therefore the differences in both studies shall be clarified. To be precise, not all variables in the study of Choi and Ji are examined. The motivation for this decision shall be elaborated upon. At the end of this chapter the conceptual model (Figure 4) is presented in order to investigate the research question for this study.

2.2 Types of technologies

The validity and generalizability of innovation studies using technology acceptance models can be questioned (Yu, 2012). A criticism of innovation research is the assumption that a universal theory applies to all types of innovations. Because of the fundamental differences that exist across types of innovation, a single theory may be inappropriate for generalization and measurement (Wisdom et al., 2014).

There are many different ways to categorize different sorts of technologies (Garcia and Calantone, 2002). In the literature, most categorizations are largely based on eight, five, tetra, triadic and dichotomous categorization types.

Current study focuses on a specific category containing two types of technology. With this dichotomous categorization, technology is either incremental or radical (Martini et al., 2013), as seen in Figure 1. Incremental and radical innovations are opposites when defining the level of newness of a technology. Both innovation types are interconnected with each other as shown in the Figure below. Radical innovation follows the developments of incremental innovation. It then takes a great leap forward in terms of revenue or efficiency, this is opposite of incremental innovation that tends to be more modest (Orcik et al., 2013).

While incremental innovations can be seen as advanced versions of a technology with relatively minor extensions (Dewar and Dutton, 1986), radical innovations are involved with significantly new functions that either don’t exist or require dramatic behavioural changes from consumers. These radical innovations become the foundation upon which future generations of products are manufactured. The usage and adoption of incremental and radical innovations is expected to be very different. As mentioned before, it is expected that radical technologies evoke a more significant response compared to non-radical new technologies (Souto, 2015), which is also a reason radical technology is selected for this study.In the following radical technology will be elaborated on more in detail.

(9)

9

Figure 1: Radical and incremental innovation.

Note. Reprinted from “Customer co-creation throughout the product life cycle”, by Orcik, A., Tekic, Z., & Anisic, Z., 2013, International Journal of Industrial Engineering and Management 4(1), 46.

2.3 Radical technology

A definition for radical technological innovation is the creation of a new line of business in the marketplace. A new product or process refers to unique performance features or with already familiar features that offer greater improvement in performance, or greater reduction in cost (O’Connor and McDermott, 2004).

The usage of radical technology requires new skills and higher levels of understanding. A reliable and valid measure developed by Green et al. (1995) for defining radical new technology incorporates four dimensions: technological uncertainty, technical inexperience, business inexperience, and technology cost. Radical innovations provide the foundation for manufacturing future generations of products (McDermott and O’Connor, 2002). Examples of radical technologies are as indoor plumbing, electric lighting in homes, the automobile, airplane, radio and television. Recent radical innovations are such as Facebook, Twitter and other effective social platforms (Norman & Verganti, 2014). The ultimate goal for firms dealing with radical technology is to create revenue and a competitive advantage (Ritala & Sainio, 2014). In order for firms to succeed the technological innovation needs to provide new benefits for customers. Radical innovations similarly require big changes in the behaviour of consumers. It can create new patterns of behaviour or even substitute established ones. If the radical technology is perceived as difficult to use and understand, adoption will most probably be difficult. The perceived benefits of radical technology seem to be more important than the objective benefits (Oerlemans et al., 2013). Consumers usually evaluate these benefits by its relative advantage towards similar products.

Testing new radical technologies increases the chance of adoption because trials can reduce potential uncertainty of consumers (Maddux & Rogers, 1983). Experience tends to reduce dissonance by reinforcing a favourable attitude towards the innovation or even turn negative attitudes positive (Metin & Camgoz, 2011).

According to Veryzer (1998), several key factors influence consumers’ evaluations of radical new products. First, unfamiliarity may cause resistance and even fear among consumers.

(10)

10 Second, consumers may be encouraged to focus on irrational product attributes because of the newness of these products. This may not correspond to their actual requirements. Third, radical new technologies require customers to invest time and effort in learning to use the product properly. Fourth, uncertainty of the product benefits and the associated risks may encourage resistance. Fifth, visual aspects of the radical new technology can affect customer feelings about product safety and their attraction to it. Sixth, the impact of these products on the customers’ life and consumption patterns can cause resistance.

2.4 Innovation Diffusion Theory

Examining personal innovativeness is in line with the study of Umberger (2016), which predicts how the usage of new technology, such as autonomous cars, will spread through society by analysing specific consumer characteristics.

These consumer characteristics highly correlate with the level of personal innovativeness, which is formulated in the Diffusion of Innovation Theory. The Diffusion of Innovation Theory by Rogers (1962) is one of the most established theories in marketing literature that describes the process of product adoption. This theory explains: “the process by which innovations spread among users” (Johnson, et al., 2011, p. 303). The rate whereby consumers use a new product is defined as the “relative speed with which an innovation is adopted by members of a social system” (Rogers, 2003, p. 22). The theory describes and predicts the adoption of new products in different stages as seen in Figure 2. The adoption process is spread over five stages of different consumer groups. Each innovation group is clustered together on the fact of having similar personality traits within the group.

The consumer groups adopting new products are classified as: innovators, early adopters, early majority, late majority and laggards. These groups are distributed in a normally distribution curve (Rogers, 2010). Each group has specific traits that determine the group characteristics. In the diffusion theory, it has been recognized that people with a high level of personal innovativeness are active information seekers of new ideas (Bin, 2013). It is therefore presumed that this group of people are able to cope with high levels of uncertainty, and develop more positive attitudes towards using and accepting a technology.

Figure 2: Innovation Diffusion Theory

Note. Reprinted from “Diffusion of innovations”, by Rogers, E. M., 2010, Simon and Schuster.

The traits of individuals in each group are explained as following:

1. Innovators: tend to have a higher level of income and education, are young, enjoy a great social participation and have favourable attitudes towards risk (Im et al., 2003).

(11)

11 2. Early adopters: have a higher social status, financial liquidity, advanced education and are more socially forward than late adopters. They are more discreet in adoption choices than innovators (Rogers 1962, p. 283).

3. Early majority: They adopt an innovation after a varying degree of time that is significantly longer than the innovators and early adopters. Early Majority have above average social status, contact with early adopters and seldom hold positions of opinion leadership in a system (Rogers 1962, p. 283)

4. Late majority: They adopt an innovation after the average participant. These individuals approach an innovation with a high degree of scepticism and after the majority of society has adopted the innovation. Late majority are typically sceptical about an innovation, have below average social status, little financial liquidity (Rogers, 2010).

5. Laggards: They are the last to adopt an innovation. Individuals in this category show little to no opinion leadership. Laggards typically tend to be focused on "traditions", lowest social status, lowest financial liquidity, oldest among adopters, and in contact with only family and close friends (Rogers, 2010).

As competitiveness around technology increases, it becomes crucial to examine which consumers tend to be more willing to make use of new products (Venkatesh et al., 2012). One of the important reasons for identifying consumers who adopt new products rapidly, is because they can serve as change agents to stimulate further use of the new technology, so that other consumers will be more willing to use this technology (Kolodinsky et al., 2004).

Because of this, it is vital to find out how to influence the rate of adoption in the best possible way. One seemingly appropriate strategy is to focus on consumer groups that are more likely to adopt new products first. Simply put; with any new product or service, someone has to be first. If the innovation is successful, the rest will follow. Therefore, it can be a competitive benefit to identify the first potential group of users of a new technology.

The speed with which an innovation is used by their consumers often makes the difference between failure and success for firms. Researchers and practitioners in the field of technology therefore deal with the effect of predicting which consumers adopt new technology rapidly.

2.5 Technology Acceptance Model

The most widely applied theoretical model for predicting and explaining the usage of technology is the TAM (Venkatesh & Davis, 2000). Its purpose is to predict the level of consumer readiness concerning the usage of technology. The model is based on a theory whereby factors of influence predict the behavioural intention to use a technology (Davis, 1989). It’s theory claims that consumers who perceive using a new technology as useful (perceived usefulness) and easy to use (perceived ease of use) are eventually more likely to end up using a particular technology (Svendsen et al., 2013). The model consists of different variables that are related to each other and eventually estimate probabilities of an individual performing certain behaviours (Figure 3). These variables are explained below.

(12)

12

Figure 3: Technology Acceptance Model

Note. Reprinted from “User acceptance of computer technology: a comparison of two theoretical models”, by Davis, F. D., Bagozzi, R. P., & Warshaw, P. R., 1989, Management science, 35(8), 985.

External Variables

The external variables include external factors such as system design characteristics, political influences, and organizational structure.

Perceived usefulness

This variable refers to “the degree to which a person believes that using a particular system would enhance his or her performance” (Davis, 1989, p. 320). According to several studies this factor consistently stands out as the main driver for accepting and using technology (King & He, 2006; Ma & Liu, 2004; Schepers & Wetzels, 2007).

Perceived ease of use

Perceived ease of use is described as: “the degree to which a person believes that using a particular system would be free of effort” (Davis, 1989, p. 320). In other words, it is the belief of a user being capable to perform certain behaviour in order to make the technology work.

Many studies have demonstrated that both perceived usefulness and perceived ease of use positively influence behavioural intention (Schepers & Wetzels, 2007). Also, the effect of perceived ease of use on behavioural intention is mediated by perceived usefulness (King & He, 2006).

Behavioural Intention to Use

The dependent variable in this study predicts the user’s likelihood of actual making use of a technology (Davis, 1989; Venkatesh et al., 2003). It is very useful to determine the behavioural intention as the dependent variable instead of actual behaviour from a practical perspective when conducting research. Using this method is less time consuming and at the same time the behavioural intention is a reliable variable because of the high predictability (Wu & Tsang, 2008).

2.6 Choi & Ji model

As previously mentioned in the introduction Choi and Ji expanded the TAM by adding several factors of influence. Besides TAM factors, other key elements that have been added by Choi and Ji are the factors trust, perceived risk and two personality characteristics (locus of control and sensation seeking). The researchers used a radical technology, namely autonomous cars, to investigate the variables in relationship to the usage intention.

Because this research is in line with the study of Choi and Ji, autonomous cars will also be used to investigate the intention to use radical new technology. Paragraph 2.9 will elaborate on the literature review of the autonomous cars and also the usage intention of this technology.

(13)

13 Important factors of Choi and Ji’s model that are added to this study are trust and perceived risk, these are also shown in the conceptual model (Figure 4) at the end of the chapter. The literature review of the variables trust and perceived risk shall be elaborated upon below.

Trust

A major factor of importance in understanding the elements affecting the usage of technology is the role of trust (Li et al., 2008). Trust is an abstract concept that has been widely studied over the decades. This has generated many different meanings and interpretations of trust (Wang & Emurian, 2004). Also, trust is subjective of nature since it deals with personal feelings, thoughts and perceptions which can differ per individual, even in similar contexts (Bagheri & Jamieson, 2004). For example, people can experience trust in many different forms like trust in their own behaviour, other people, relationships, company brands and technology. To put it another way, trust is complex and can be different depending on the context.

Trust is generally seen as an important factor in relationships between consumers and organizations (Martínez & del Bosque, 2013). As with most new technologies, the level of consumer trust is understood to be one of the key areas of concern by marketers. An antecedent of trusting behaviour is the willingness to be dependent of another person or object (Mcknight et al., 2011). This willingness, also known as intention, is influenced by the trustworthiness of the trustee based on ability, benevolence and integrity.

Trust research was initially proposed in the context of interpersonal relationships, where the trustee is a human being (Li et al., 2008). Studies have discussed trust in technology where the trustee is a technological object such as an information system (Corritore et al., 2003). Trust in this context has been defined similarly to interpersonal trust that reflects the willingness of the trustor to depend on software to do a specific task (McKnight, 2005). Models of trust and technology acceptance have been applied to the relationship between humans and technology in similar manner (Komiak and Benbasat, 2006). Benbasat and Wang (2005) state that empirical results support the validity of applying such models to technology, with significant loadings for the three trust dimensions benevolence, competence, and integrity. Other studies have also revealed that trust has the same three primary dimensions (ability, benevolence, and integrity) (Gefen, 2002). An explanation for this is that people tend to humanize technology (Hultman et al., 2015). Studies suggest that trust in technology is a feasible concept and that such models of trust are applicable for technology acceptance research (McKnight, 2005). In behavioural literature, many researchers similar to Jarvenpaa et al. (1999) have adopted these three trust dimensions for determining consumer trust in technology. The three dimensions ability, benevolence, and integrity that measure the level of trust are adapted from the study of Mayer et al. (1995):

1) Ability: the trustor’s perception of trustee’s competencies and knowledge are present or relevant in relation to the expected behaviour. Such perceptions may be based on prior experience.

2) Benevolence: the extent to which a trustee is believed to intend doing good to the trustor, beyond its own profit motive. Benevolence introduces faith and unselfishness in a relationship, which reduces uncertainty.

3) Integrity: the trustor’s perception that the trustee will stick to a set of principles or rules of exchange accepTable to the trustor during and after the exchange. Integrity is similar to honesty, fairness, consistency, predictability, reliability, and dependability dimensions proposed in the literature.

(14)

14 The reason for trust being relevant in regards to using new technology is because it has functional roles that are of use in certain situations. One of its roles is to help make decisions and prevent risky situations that can have negative outcomes. People have limited cognitive resources available when making choices. By applying trust as mental shortcut it helps to reduce uncertainty and complexity (Luhmann, 2017). Trust can be served as a mechanism to reduce complex situations when consumers deal with uncertainty and risk (Luhmann, 2017). Kim (2012) emphasizes the importance of trust in the initial stage, especially in relation to new technologies. This is because perceptions of risk and possible negative outcomes must be faced or overcome, in order to interact with the technology. Several researchers have confirmed the effect of trust both as a direct determinant of behavioural intention and as an indirect influence through perceived usefulness and perceived risk (Carter & Bélanger 2005). Trust is found to negatively influence perceived risk (Lee & Song, 2013). Trust and perceived risk are highly related to each other according to Choi & Ji (2015).

Perceived risk

Risk in general can be seen as the probability of a particular harm in a situation and the consequence of that harm (Lantos & Spertus, 2014). In contrast, perceived risk reflects an individual’s subjective judgment about particular dangers. The judgement may differ considerably from the actual risk

(Smith, 2016). In consumer behaviour research, perceived risk has been formulated as the expectation of

experiencing losses in uncertain situations (Featherman & Pavlou, 2003). Consumers evaluate the potential safety benefits by weighing up the perceived risks and benefits in the relevant contexts. Perceived risk is seen as an uncertainty in a specific situation. Consumers tend to be willing to accept some amount of risk. For instance, when they perceive the possible risk as unlikely to happen. Another situation that can influence consumers to accept risk is when a benefit is expected to outweigh the possible risk. Literature stresses the importance of perceived risk being present in technology acceptance models. In relation to autonomous vehicles this perceived risk plays an important role because it may hinder the usage and development of new technology (Nordhoff, 2014). A study of Adell (2010) shows that it can negatively influence the intention to use driving assistance systems. When dealing with a new radical technology it is expected that consumers experience higher amount of perceived risk compared to less newer technologies.

2.7 Hypotheses

In the conceptual model of this study six variables and twelve hypotheses are shown in Figure 4. The hypotheses shall be tested in order to answer the research question of this study “To what degree do

trust and personal innovativeness affect the behavioural intention to use new radical technology?”.

H1, H2, H3, H4, H5, H6 and H7 are exact similar hypotheses tested in the study of Choi and Joi (2015), the relationships are presented with dotted lines. These hypotheses shall be tested in order to determine if this study will generate similar results.

The other hypotheses H8, H9, H10, H11 and H12 test the possible direct and indirect effect of personal innovativeness on the intention to use new radical technology.

Replication Choi and Ji

Not all variables of Choi and Ji’s model are adopted in this study. The main motivation for excluding the three antecedents of trust is due to the lack of relevance for this study. Exploring the trust backgrounds is not a topic of investigating for this research. Other variables that are also excluded from Choi and Ji’s model are locus of control and sensation seeking.

(15)

15 The reason for not including these two personality characteristics is because they are substituted with an alternative personality characteristic: personal innovativeness. This substitution is in line with the specific research gap mentioned earlier: investigating the effect of another specific personality characteristic on the intention to use an autonomous car.

Results of Choi and Ji’s study indicate that trust and perceived usefulness are very important factors that influence the intention to use an autonomous vehicle. Another finding of their research is that they identified trust affecting perceived risk in a negative direction. In other words; trust decreases the perceived risk. Trust was also found to have a positive effect on perceived usefulness. Furthermore, Choi and Ji confirmed the TAM relations that are known in the literature, namely the positive effect of perceived ease of use on behavioural intention. And the positive effect of ease of use on perceived usefulness. As mentioned before, an objective of this study is to expand the model of Choi and Ji.

Therefore, in line with the hypotheses also tested in Choi and Ji’s model the following is hypothesized:

H1: Trust has a negative effect on perceived risk. H2: Trust has a positive effect on perceived usefulness.

H3: Trust has a positive effect on behavioural intention to use new radical technology.

H4: Perceived ease of use has a positive effect on behavioural intention to use new radical

technology.

H5: Perceived ease of use has a positive effect on perceived usefulness.

H6: Perceived usefulness has a positive effect on behavioural intention to use new radical

technology.

H7: Perceived risk has a negative effect on behavioural intention to use new radical technology.

2.8 Personal innovativeness

Personal innovativeness as direct effect

Research indicates that technology acceptance depends much on individual differences as on the technology itself (Venkatesh et al., 2012). The relationships between differences in personality characteristics and their beliefs are also discussed in social psychology and learning theories (Lu et al., 2005). As mentioned, there is a research gap concerning personal characteristics as a possible determinant on the intention to use new technology. In this study, personal innovativeness is selected to fill this gap.

Innovativeness is derived from the concept of innovation. This is described as an idea or object that is perceived as new by an individual (Rogers, 2010). Individuals are characterized as ‘innovative’ if they are early to adopt an innovation (Agarwal and Prasad 1998). Personal Innovativeness is considered a generalized personality trait and is defined in many different ways (Svendsen et al., 2013). This conceptualization of innovativeness has also been used in the marketing literature and reflects a highly abstract and generalized personality trait (Im et al., 2003). Other definitions across literature include “a willingness to change” (Hurt et al., 1977 p.220) and the openness to new experiences and innovative stimuli (Steel et al., 2012). In the study of Choi & Ji it is defined as the willingness to try out something new. It is also conceptualized as the degree of adopting a new product by an individual (Simons, 2013). Rogers (2010) describe it as how fast an individual adopts an innovation in his or her social system, and as an individual’s ability to be innovative. According to the innovation diffusion theory people react differently to innovation because of their differences in individual innovativeness (Rogers, 2003). Innovation is suggested to be associated with greater risk and

(16)

16 uncertainty (Agarwal and Prasad 1998). Thus, it is reasonable to argue that personal innovativeness can possibly reduce these perceived risk and uncertainties, or have more ability to face these fears better. Individuals with higher levels of personal innovativeness are expected to develop more positive beliefs towards new technologies (Yang et al., 2012).

Kaasinen (2005) concluded that people’s attitude towards innovations will have a positive influence on the intention to use new technologies. The more personally innovative a consumer is, the more able they are to adapt to new technologies and ideas.

Yi et al (2006) conclude that people with a high level of innovativeness are more likely to try out new things and feel comforTable with using a new technology, even with little knowledge of the possible usage outcome. This is similar to other findings of innovators, who are known to have the tendency to use a technology without fully knowing the potential value and the performance (Son & Han, 2011). Research shows that personal innovativeness has a positive effect on both the attitude towards and intention to use a new technology (Amoroso & Lim, 2014).

However, determining the level of innovativeness strongly depends on the type of innovation, according to some researchers.

It is suggested that it can differ per innovation how fast a technology will be adopted by consumers (Chao et al., 2012). For example, a person that is using the latest Microsoft Excel version for personal use can be considered as an innovator of Microsoft Excel.

But the same person in the same situation might use the newest Microsoft Word version in a much later stage, whereby the individual is not considered as innovator for that specific innovation. Thus, although products can fall in a similar product group it strongly depends on the specific technology when classifying consumers as innovators (Rosen, 2005). This makes it challenging to generalize individuals on the level of personal innovativeness because it can differ strongly for different types of innovation. On the other hand, consumer innovators are identified based on their innovativeness across various product classes (Rogers, 2010). Often referred to as general innovativeness (Ha & Stoel, 2004). These contradicting views in literature challenge the generalizability of classifying consumer groups in the stages of the diffusion innovation theory. Numerous of studies have found a significant relationship between personal innovativeness and the behavioural intention to use technology (e.g.

Thompson et al., 2008; Xu and Gupta, 2009).

From the previous literature, it is expected that personal innovativeness is positively associated with the behavioural intention to use new radical technology. Therefore, the following is hypothesized:

H8: Personal Innovativeness positively affects the behavioural intention to use new radical

technology.

Personal innovativeness as indirect effect

Personal Innovativeness & Trust

Users with low personal innovativeness may doubt the credibility of an innovation faster and are therefore more likely to hesitate to use it. This can negatively affect the intention to use a new radical technology. On the other side, users with high personal innovativeness are more willing to experience new technologies. Due to their openness to new technologies and risk-taking attitude, consumers with high personal innovativeness will have more trust in the technology (Zhou, 2011) The following is hypothesized:

(17)

17

H9: The effect of personal Innovativeness on the behavioural intention to use new radical technology

is mediated by trust.

Personal Innovativeness & Perceived usefulness and Perceived Ease of Use

Lewis et al., (2003) found that personal innovativeness has a significant positive relationship with perceived usefulness and perceived ease of use. Lu et al., (2005) propose that personal innovativeness affects the level of perceived usefulness and ease of use, which eventually predicts the intention to use a new technology. In the study of Agarwal and Prasad (1998) personal innovativeness was found significant related with perceived usefulness and the intention to use a technology. Agarwal and Prasad expect personal innovativeness also to have a strong impact on perceived ease of use. People with a high level of innovativeness are expected to develop more positive perceptions about the innovation in terms of ease of use since they have a higher willingness to try out new technology. The following relationships are hypothesized:

H10: The effect of personal Innovativeness on the behavioural intention to use new radical

technology is mediated by perceived ease of use.

H11: The effect of personal Innovativeness on the behavioural intention to use new radical

technology is mediated by perceived usefulness.

Personal Innovativeness & Perceived Risk

Personal innovativeness is suggested to influence the risk-taking propensity of an individual that would eventually lead to a stronger usage intention (Yi et al., 2006). Rogers (2010) argues that innovators are able to cope with high levels of uncertainty better. Similarly, individuals with higher level of personal innovativeness are more willing to take risks. It is reasonable to expect them to develop more positive intentions toward the use of a new technology, compared to people with a lower level of personal innovativeness. This leads to expectation that perceived risk mediates the relationship between personal innovativeness and the intention to use a new radical technology:

H12: The effect of personal Innovativeness on the behavioural intention to use new radical

(18)

18

Figure 4: Conceptual Model

Figure 4 illustrates the proposed research model in this study. Twelve hypotheses are proposed based on literature reviews. As previously mentioned, to examine the behavioural intention to use autonomous cars, this study expanded the model of Choi and Ji with an additional factor: personal innovativeness. In total, there are 6 constructs in this model: perceived risk, perceived usefulness, perceived ease of use, trust, personal innovativeness and behavioural intention as the dependent variable. All hypothesized relationships are, apart from those that involve personal innovativeness, in line with the tested hypotheses of Choi and Ji. The hypotheses in this model are tested to finally generalize knowledge on the behavioural intention to use new radical technology. The autonomous car is the specific technology to represent radical technology in this study. All direct and indirect relations related to personal innovativeness are marked with thick arrows (H8, H9, H10, H11 and H12).

All other hypothesis, derived from Choi and Ji’s model are shown in dotted lines (H1, H2, H3, H4, H5, H6 and H7).

2.9 Autonomous cars

The objective of this paper is to determine the effects of the selected factors on the behavioural intention to use new radical technology. Since radical technology is a broad concept that consists of multiple types of technologies, this research has selected one specific technology, the autonomous car. The results will then be generalized to radical technology as a whole.

In other words, as already mentioned in paragraph 2.6, the case of autonomous cars is selected to investigate the behavioural intention to use new radical technology. Why the autonomous car is a good representative for radical technology will be explained further on in this section.

The autonomous car is a considerably new radical technology which is gradually gaining the public’s attention. These vehicles are not commercialized yet, thus examining the attitudes and usage

(19)

19 intention of consumers towards this new technology is important (Choi and Ji, 2015). In literature, the autonomous car is described as a vehicle that is able to drive autonomously without any intervention from the driver, or anyone else in the vehicle (Payre et al., 2014). The autonomous car is meant to work independently (Rizzi et al, 2014). As time goes by the possibility for consumers to actually use autonomous cars is becoming more realistic, this creates challenges for Marketers how to deal with this specific radical technology.

Another term used for autonomous cars (also known as self-driving- or driverless car) are fully automated vehicles. This is reflected from the highest degree of automation, namely: fully automated. A fully automated vehicle drives by itself without any human intervention, whereby the steering, acceleration and deceleration are completely performed by an automated system built in the car. Scaling autonomous vehicles is based on a range from 0 to 4 (Rödel et al., 2014):

 Level 0: Fully absence of automation, the driver has complete control of the car.  Level 1: Single functions are autonomous.

 Level 2: Automation of at least two primary functions.

 Level 3: The autonomous vehicle may have full control for a period of time.

 Level 4: The vehicle acts on a fully autonomous level, performing all driving functions for an entire trip.

According to the Diffusion of Innovation Theory innovators and early adopters will be the first to make use of autonomous cars. It is expected that these consumers have a high level of personal innovativeness (Amoroso & Lim, 2014). This expectation is thus in line with examining to what degree personal innovativeness predicts the intention to use an autonomous car.

One of the research motives for using this specific technology is because the technology of autonomous car is known as a radical technology (Schreurs & Steuwer, 2016). In other words, this technology is expected to have a significant impact on the automobile industry and on the economic activity of firms in that industry. Another reason for selecting this technology is because it is in line with the suggested recommendation for further research around autonomous cars, as mentioned in paragraph 1.3: “Future researches should test the possible inclusion of other external variables (e.g., personality characteristics). Therefore, more research is needed to validate, expand, and generalize these, results” (Choi and Ji, 2015, p.700).

Since most people have seen or been in a car before, this technology is likely to be easier to understand than other radical technologies.

Also, as prototypes of highly automated vehicles are presented, public and media interest in the usage possibilities of autonomous cars has grown. Speculative benefits of autonomous cars have included increased safety, fewer traffic problems, and the possibility to work or enjoy leisure in the vehicle during the journey (Litman, 2014). To put it another way, the familiarity of this radical technology and its functions seem to exist among a wider audience. This makes the technology of autonomous cars very suiTable for conducting this research.

Besides familiarity of the technology, the selected factors all seem to play a relevant role according to the literature. Trust, risk, usefulness and ease of use all seem to have theoretical associations with this specific technology. It has been shown that trust is a major determinant of automation acceptance (Carter & Bélanger, 2005). According to Ghazizadeh et al. (2012) trust also explains the usage of driving assistance systems, which are important elements of fully autonomous vehicles. Safety and risk concerns have long been considered problematic for the acceptance of new technology (Swinyard & Smith, 2003). Finally, Choi and Ji (2015) strongly suggest that the usefulness

(20)

20

and the difficulty in using an autonomous car can

(21)

21

3 Methodology

The methodology chapter will describe how this research was conducted. More specifically, the methodologies that are applied for this research will be described. The study is made of both primary and secondary data. Furthermore, this chapter shall elaborate the research design and an explanation of the approach for collecting and analysing data. The final step is the operationalization of constructs which explains the process of defining variables into measurable factors.

3.1 Research design

The main purpose of this study is to identify what the influences are of the mentioned independent variables on the dependent variable, behavioural intention to use new radical technology.

Measurements of the independent variables and dependent variable will be measured with a quantitative approach by making use of survey questionnaires similar to the study of Choi and Ji (2015). Questions in this study are focused on specific relationships and hypotheses of existing theories. Edmondson and McManus (2007) also propose this methodological approach to reach a methodological fit. This method also falls within the deductive approach of this study. In the deductive approach hypotheses are developed from existing theory, and then tested using quantitative data (Saunders & Lewis, 2009).

The surveys will be primarily filled online and in addition to this a smaller part is filled in by making use of pen and printed questionnaires on paper in order to create a greater response chance. In both cases the questionnaire will take approximately 5 minutes to complete. Finally, the statistical software program SPSS will use the collected data for analysing the results.

Before the survey will be posted online there will be a pre-test where four fellow master students will complete the survey and give their feedback on the quality and interpretability of the survey. Based on the feedback, the necessary adjustment to the survey questionnaire will be made. Thereafter the survey will be posted online and therefore made available to the public. The survey will be open to the public for a week, after this it will not be possible for respondents to fill in the survey.

There are several motives for choosing an online questionnaire as a method to collect data. These include: reduced costs, rapid deployment, high response likelihood rate, speed to collecting data, and decreased data entry error (McDaniel & Gates, 2005). Another major advantage is that the data can be transferred easily and quickly to SPSS. This approach has also been supported by Wright (2005). Autonomous car is the specific new radical technology that is selected for investigating this study. The motive for selecting this technology, besides the relevance of the mentioned paper (Choi and Ji), is because autonomous cars can be considered as a new radical technology. Also, the familiarity of cars in general is well known to a large audience.

3.2 Data collection

For this study, the online survey tool Qualtrics is selected to collect data of the survey. The data collection started on 30th of December 2017 and the survey was open for respondents up to and

including the 6th of January 2018, in Dutch or English. The online survey, which include a brief

explanation about self-driving cars was distributed in two ways. First, by making use of convenience sampling. With this approach, a link to the survey was posted on Facebook & LinkedIn. Second, the same link was emailed to friends, family, colleagues, teammates, fellow students and other social groups. Using the idea of snowball sampling, the aforementioned respondent group was requested to forward the survey to their personal network.

(22)

22 Therefore, in order to reach a sufficient number of respondents two non-probability sampling techniques have been used, namely: convenience sampling and snowball sampling (Schillewaert et al., 1998). As mentioned, a part of the questionnaire was filled in with pen and paper, these respondents were selected from members of social clubs from the researcher.

Previous studies about trust in self driving car technology received an average of 400 respondents, with a response rate around 15% (Nees, 2016; Choi & Ji, 2015; Armida, 2008; Choi et al., 2007). Considering the given information on the rule of thumb, suggested for getting a minimum response for quantitative data of similar research (Green, 1991). Therefore, the proposed research aims to attain a minimum of 200 respondents.

3.3 Data analysis

Before any data analysis began, all data from Qualtrics has to be exported to SPSS. The next step was to check for errors in the data. This was performed through a frequency and normal distribution test. After this step, a descriptive analysis of the sample distribution was conducted to determine the reliability and the validity of the used measures. Furthermore, these scales were used for testing the hypotheses in SPSS, by making use of multiple regression analysis.

Missing data

Among all the responses there were six questions left unanswered. These missing data were

substituted with the mean value of the concerning variable. It was decided not to remove the concerning respondents that failed to fill in a response on a question. To be sure not to lose power, these missing data were substituted with the mean value of the specific variable.

Operationalisations

The 6 constructs measured in our study were personal innovativeness, trust, perceived risk, perceived usefulness, perceived ease of use and behavioural intention to use new radical technology. Each construct was measured with three items, every single item was adapted from extant literature to improve the content validity. All the items of constructs were measured on a 7-point Likert scale from 1 (strongly disagree) to 7 (strongly agree). Below all scales are shown with an example.

Trust

The items from the trust scale were adapted from Gefen et al. (2003) and Pavlou (2003). An example of a trust item is: “Autonomous cars are dependable”.

Personal Innovativeness

The items from this scale are adapted from Agarwal & Prasad (1998) and Jackson et al. (2013). An example of one of the items is: “If I heard about a new technology, I would look for ways to experiment with it”.

Perceived usefulness, Perceived Ease of Use and Behavioural Intention

The items perceived usefulness, perceived ease of use and behavioural intention were all three adapted from Davis (1989). An example of the item perceived usefulness is: “Using an autonomous car will increase my productivity”, for perceived ease of use: “Learning to operate an autonomous car would be easy for me” and for behavioural intention: “I intend to use autonomous vehicle in the future”.

(23)

23

Demographics

Next to the five variables mentioned above, demographic information such as age, gender, income, education will be presented in the questionnaire. Another important construct which will be shown is the amount of experience with technology in general.

(24)

24

4 Results

In this chapter, the results of the survey will be presented, more specifically the results relevant for the hypothesis testing. There will be an extensive summary of the collected data and an in-depth statistical analysis of these results. This research paper investigates the relation between the variables personal innovativeness (independent variable), trust, perceived usefulness, perceived ease of use (all mediators) and the behavioural intention to use a new radical technology (dependent variable).

4.1 Descriptive Statistics

In the survey, scores could range from 1 (totally disagree) to 7 (totally agree) meaning that 4 was the neutral score. This of course does not count for the control variables age, gender, income and education level. From Table 1 it can be concluded that the sample scored on all of the main constructs, with the exception of perceived usefulness and perceived risk, between neutral and moderately positive. Eventually 380 respondents completed the survey, of which 219 males (57.6%) and 161 females (42.4%), leading to a somewhat even distribution. The mean age was 34,91 years (S=13.58). Furthermore, the modal gross income category of the respondents was between €35.000 - €54.999 which accounts for 23.4% of the sample. With respect to the education level of the respondents, the majority of the respondents enjoyed a university level. In total 63.2% of the respondents completed an Associate (28.2%), Bachelor (14.7%) or Master/MBA (20.3%) degree.

Table 1: Descriptive statistics of the different variables

Variable N Mean Dev. Std.

Behavioural intention 380 4.40 1.620

Personal innovativeness 380 4.21 1.421

Trust 380 4.10 1.231

Perceived usefulness 380 3.95 1.339

Perceived ease of use 380 4.76 1.149

Perceived risk 380 3.83 1.115 Age 380 34.91 13.576 Gender 380 1.42 .495 Income 380 3.00 1.399 Education level 380 5.03 1.453 Driving skills 380 4.96 1.459 Computer skills 380 4.47 1.093 4.2 Correlations

From Table 2 it can be concluded that all construct have significant bivariate associations, which shows potential for confirmation of the different hypotheses. However, testing the hypotheses, especially those about mediation effects, can only be done in a valid way with the more complex, multiple mediation analysis done in the next section. The Table shows no extreme correlations (>0.7), so there is no risk of multicollinearity.

(25)

25

Table 2: Correlations

* Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed)

4.3 Principal Component Analysis

On the total number of eighteen items Principal Component Analysis (PCA) with Varimax Rotation was conducted. With this analysis different scales are identified by means of the rotated component matrix which displays the factor loadings for each variable on each factor calculated after rotation. The next step is to look at the content of the items that load highly on the same factor to try to

identify common themes. Marked with a red line, Figure 5 shows the part of the rotated component matrix that highlights the

3 items used to measure the construct of perceived risk. It can be seen that the first item (-.611) loads in component 5, and the other two components (-.700; -.782) are loaded in component 1. Therefore, it was concluded that with this sample perceived risk was not measured in a valid way, and it was decided to leave perceived risk out of this study. Which means that the hypotheses H1, H7 and H12 were not tested. After the PCA a reliability analysis was conducted which showed that all of

(26)

26

Figure 5: Rotated component Matrix

4.4 Testing the hypotheses with the Process Regression Analysis

To test the different hypotheses a multiple mediation regression analysis was performed with the Process software developed by Hayes (2012). For this analysis, personal innovativeness was the independent variable. Trust, perceived usefulness and perceived ease of use were the mediators and behavioural intention to use a new radical technology the dependent variable. Age, gender, income, education level, driving skills and computer skills were the control variables.

Direct effects

Table 3 shows the results of the PROCESS mediation regression analysis. Behavioural intention was of course the dependent variable, while trust, perceived usefulness and perceived ease of use were the mediators. The format of the PROCESS-output is such that it provides not only the test results of the indirect effects, but also those of the direct effects on the dependent variable and on the mediators. From Table 3 it can be concluded that the effect of trust on perceived usefulness is significantly positive (t(370)=4.68, p<0.001). So, H2 can be accepted. Furthermore, the effect of trust on behavioural intention is also significantly positive (t(369)=4.30, p<0.001). So, H3 can also be accepted. H4 posed the expectation that perceived ease of use has a positive effect on behavioural intention, which is proven by the analysis (t(369)=4.88, p<0.001). Perceived ease of use has a positive effect on perceived usefulness as well (t(370)=3.02, p=0.003), so H5 can also be accepted.

(27)

27 When we consider the effect of perceived usefulness on the behavioural intention, it is shown this effect is, again, significantly positive (t(369)=4.72, p<0.001), which means that H6 is accepted. Finally, personal innovativeness appears to have a significantly positive effect on the behavioural intention (t(369)=2.70, p=0.007), so H8 is accepted.

Table 3: Regression coefficients (with standard errors in parentheses) from the mediation analysis using PROCESS.

* p < 0.05 ** p < 0.01 *** p < 0.001

Indirect effects

The three indirect effects considered in this study are those of personal innovativeness on the behavioural intention mediated by trust (H9), perceived ease of use (H10) and perceived usefulness (H11). The significance of the indirect effects is analysed using bootstrapped confidence intervals of the effects. If such an interval does not contain the value zero, the effect is significant.

Table 4 shows that the indirect effects mediated by Trust and Perceived Ease of Use are significant and the one mediated by Perceived Usefulness is not. H9 and H10 can be accepted and H11 not.

Table 4: Indirect effects

Mediator Effect Boot SE BootLLCI BootULCI Hypothesis

Trust .0862 .0223 .0469 .1367 H9

PEOU .0372 .0193 .0061 .0832 H10

PU .0163 .0174 -.0148 .0560 H11

Dependent

variable Trust PEOU PU BI

Constant 2.92 (0.43)*** 2.89 (0.38)*** 1.41 (0.50)** 0.55 (0.57) Independent variable Personal innovativeness 0.31 (0.51)*** 0.10 (0.05)* 0.06 (0.06) 0.16 ( 0.06) ** Mediators Trust 0.30 (0.06)*** 0.29 (0,06)*** PEOU 0.25 (0.08)** 0.37 (0.07)*** PU 0.28 (0.06)*** Control variables Age 0.01 (0.01) - 0.01 (0.01) - 0.01 (0.01) - 0.01 (0.01) Gender - 0.27 (0.13)* - 0.27 (0.11)* 0.04 (0.14) - 0.37 (0.14)** Income 0.10 (0.05) 0.12 (0.05)** 0.11 (0.06) 0.06 (0.06) Education level - 0.10 (0.05) - 0.06 (0.04) 0.08 (0.05) 0.06 (0.05) Driving skills - 0.07 (0.05) - 0.01 (0.04) - 0.15 (0.05)** - 0.01 (0.05) Computer skills - 0.01 (0.06) 0.08 (0.05) 0.04 (0.06) - 0.06 (0.07) R2 0.17 0.33 0.25 0.46 F 12.10*** 25.11*** 15.44*** 36.25***

(28)

28

Summary hypotheses

To summarize this chapter and thereby the results of this study we created Table 5 to give an overview of all hypotheses and whether they are accepted or rejected. All tested hypotheses were accepted, except for H11.

Table 5: Summary hypotheses

Hypothesis Result

H2: Trust has a positive effect on perceived usefulness. Accepted H3: Trust has a positive effect on behavioural intention to use new radical

technology. Accepted

H4: Perceived ease of use has a positive effect on behavioural intention to

use new radical technology. Accepted

H5: Perceived ease of use has a positive effect on perceived usefulness. Accepted H6: Perceived usefulness has a positive effect on behavioural intention to use

new radical technology. Accepted

H8: Personal Innovativeness positively affects the behavioural intention to

use new radical technology. Accepted

H9: The effect of personal Innovativeness on the behavioural intention to use

new radical technology is mediated by trust. Accepted

H10: The effect of personal Innovativeness on the behavioural intention to

use new radical technology is mediated by perceived ease of use. Accepted

H11: The effect of personal Innovativeness on the behavioural intention to

use new radical technology is mediated by perceived usefulness. Rejected

Figure 6 shows the strength of the hypothesized relationships. It can be seen that all the tested hypotheses, besides H11, are supported by the data. Because perceived risk was not measured in a valid way, it was decided to leave perceived risk out of this study. Therefore, the hypotheses H1, H7 and H12 were not tested.

Figure 6: Strength hypothesized relationships

(29)

29

5 Discussion

This study is conducted to get a better understanding between the relations between factors influencing the intention to use a new radical technology. The case of autonomous cars was selected for testing these relationships. In order to investigate this, a large part of Choi & Ji’s and TAM model was adopted in this study. A major difference with the current model is that the two personality characteristics used in Choi and Ji’s model were substituted with personal innovativeness. Also, this study did not focus on the antecedents of trust. The two personality characteristics used in Choi and Ji’s model were replaced to get a better understanding of the user’s intention to use a new radical technology. A survey was conducted to collect the data, finally the data were analysed. All three hypotheses (H1, H7 and H12) related to the factor perceived risk were not used in this study because they were not measured in a valid way. It can be concluded that all the other tested hypotheses, besides H11, are supported by the data.

Starting with the part of the study that aimed to replicate most of Choi and Ji’s model, it can be concluded that this model is confirmed, with exception of H1, H7 and H12. Just as Choi and Ji, a direct relationship between perceived usefulness and behavioural intention was found. Also, both the study of Choi & Ji and this study, show a direct relation between perceived ease of use and the behavioural intention that is significant, but weaker. According to Choi and Ji, this is in line with research of King & He (2006) and Ma & Liu (2004). Both their studies showed a weaker relationship between perceived ease of use and behavioural intention. One of their assumptions here is that people are more willing to make use of an autonomous car if they find it useful compared to if they find it easy to use. Also, similar to Choi and Ji, trust has significant relations with perceived usefulness and with behavioural intention. Results of other studies investigating the usage of technology also indicate a strong effect of trust on the two mentioned factors (Parasuraman et al., 2008; Carter & Bélanger 2005; Pavlou, 2003). This indicates that trust plays an important role in the usage intention of autonomous cars. Choi and Ji mention the importance of improving the user’s perception on the construct of trust. And finally, with respect to the replication of Choi and Ji, this study confirmed the positive effect of perceived ease of use on perceived usefulness.

Personal innovativeness was selected in this study as a personality characteristic in the model. Its function in the model was to test its effect, direct and indirect, on the behavioural intention to use autonomous cars. This characteristic was found to have a positive direct relation with the behavioural intention, and positive indirect relations with behavioural intention, mediated by trust and perceived ease of use. In other words, personal innovativeness can have an effect on the intention to make use of an autonomous car. However, the mediation of perceived usefulness was not found in this study. A possible explanation of this rejected hypothesis is that in the case of autonomous cars, users have not been able to actually experience related services which can be useful for them. For example, a consumer does not have the possibility to take a test drive in an autonomous car and experience its benefits. Therefore, the lack of experiencing the usefulness of these cars in real life may hinder the mediating role of perceived usefulness. Concluding, the perceived usefulness does not seem relevant in mediating the relation between personal innovativeness and the behavioural intention to use an autonomous car.

According to the Diffusion of Innovations theory, individuals develop beliefs about new technologies by gathering information from different media (Rogers, 2010). Innovators are expected to develop more positive beliefs about a new technology. Therefore, when looking at the proposed relationships between personal innovativeness and the other selected factors it is not surprising to expect significant relationships between them.

Referenties

GERELATEERDE DOCUMENTEN

In episode three, the editor/author utilises bodies and spaces such as the king, the Babylonians, Daniel, the lions’ den, the prophet Habakkuk and food to demonstrate the

Key results include a direct measurement of the magnetoelectric coupling parameter by measuring the magnetic response of the PZT/LSMO system as a function of applied electric field,

Specifically, this paper will study the influence of perceived usefulness, perceived ease of use and perceived enjoyment (adapted from the TAM), innovativeness

Keywords: Sexting, Online experiment, Perceived legitimacy, Perceived (Sharing) Risk, Fidelity, Scenario, SEAM, Information Privacy, Situation

Since this study showed that trust is not the variable that mediates the relationship between interview style and risk perception, further research should investigate a

We first derive the transfer function of the equivalent linear time-invariant filter relating the input to the voltage sampled on the capacitor in the switched-RC kernel.. We show how

operational information) influence the level of trust (goodwill and competence) in buyer- supplier relationships?’ and ‘How do perceptions of information sharing (strategic and

variables the marginal effects are not statistically significant, meaning that the literacy of respondents has no effect on the perceived risk attitude of individual investors..